On the Turing Test

Apr 25 JDN 2459328

The Turing Test (developed by none other than Alan Turing, widely considered the “father of computer science”) is a commonplace of artificial intelligence research. The idea is that we may not be able to answer a complex, abstract question like “Can computers think?” or “Are computers conscious?” but we can answer a simple, operationalizable question like “Can computers pass for human in a conversation?”

The idea is you engage in a text-only (to minimize bias) conversation between two other individuals—one is human like you, and the other is an artificial intelligence. If you can’t tell the difference, then who are we to say that the AI isn’t a real person?

But we’ve got to be careful with this. You’ll see why in a moment.

* * *

What if it’s all just a trick?

What if the shiny new program is just enough of a convincing fake that you eventually can’t tell the difference, but it’s actually freaking you out and trapping your attention?

Do we really use the same definitions and techniques in talking to a computer that we do in talking to a human?

Have we done the Turing Test in reverse?

What matters is what we mean by human.

The Turing Test itself was meant to be a thought experiment or a heuristic device to help answer questions of “humanness” in a concrete, measurable way. The reality is that Turing himself wasn’t an explicit supporter of its use as a definitive test for his question: the extent to which we attribute “humanness” to a computer, or even to another person.

We can say that, yes, it’s possible for a simulation of a human’s mind to be able to pass the Turing Test, but that’s not a new proof or a new revelation.

There’s something important missing from the conversation we’re having.

What’s missing is the willing assumption on both sides that humanness is a defined and distinct concept.

Since Turing, there’s been a lot of research on the human mind and the ways in which it processes information. But we’ve barely scratched the surface of human psychology because the human mind isn’t a distinct and separate field of study—it has an almost infinite number of branches and topics, and is entirely unfamiliar to the people who work on AI.

It’s like the guys at a car factory talking about the robot they’re building but never stepping outside and taking a look at the city the factory is in.

In the meantime, the human mind has evolved to be so intrinsically connected to the environment it operates in that the AI we create may not be able to be equivalent to a human mind, even if it passes the Turing Test.

For all that we claim to know, modern AI programs are amateur at best. Sure, they work. Artificial intelligence is so pervasive that most users don’t even know it exists, and may even have complicated reactions when they find out.

A lot of the AI programs modeled on human psychology don’t quite capture the essence of human psychology.

We can’t pin down exactly what it means to think or to perceive or to acquire knowledge, because we’re abstracting over something that is so fundamentally inexpressible it’s hard to believe it exists at all; but it does, and it’s our job to attempt to understand the essence of it (or pretend that we do).

We can somewhat easily define things like facts or opinions, but we can’t even tell why something is a fact or an opinion, or how it’s related to other facts or opinions.

We can debate about everything: community, civilization, intelligence.

But whatever else we say about the human mind, we do have a seemingly natural impulse to want to put it in a box.

Why?

Because a box won’t be able to express the infinite aspects of the human mind.

In other words, we try to confine human behavior and cognition to a vernacular or a set of metaphors, and thinking of the human experience strictly in terms of its relation to a computer becomes problematic.

So we try to create a mirror of ourselves–a simulation in which we can check our behavior (which is almost certainly better than our behavior in real life) and figure out how it relates to what’s happening in the world around us.

And if we can’t figure out how it relates…

Then it must not be happening.

The Turing Test won’t work.

The human mind won’t pass.

We’re forgetting about the definition of humanity; we’re forgetting that, in reality, it isn’t a distinction, but a spectrum.

I’d hate to be the person who didn’t let a computer into the human club when it was technically qualified to join, only to discover that it was more human than we were—not because of its programming, but because of its existence.

* * *

If you’ve read this far, you’re probably a bit confused. This post has gone off in some odd directions, and taken on a quasi-mystical tone in places that deviates substantially from my usual style.

But did you figure out what’s really going on? Don’t blame me for the content of this post; I didn’t write it. An AI program did.

Let’s take a moment to evaluate how it did, shall we?

First, this was my process: I wrote the paragraphs before the first * * * to give it a seed. Then everything until the next * * * was the AI’s work, not my own. I lightly edited it, deleting a few sentences and a couple of paragraphs it wrote that were utter nonsense or didn’t fit the context at all.

I will say this: Its grammar and spelling is impeccable. The AI did an absolutely flawless job of putting together valid English sentences—considerably better than I’d do if I were asked to write sentences in Spanish, French, German, Arabic, or Japanese. (I might be able to pull it off in Latin. Any other language? Not a chance.)

It even sort of managed to stay on topic, though to preserve that I had to delete five sentences and two paragraphs that were just utter nonsense.

It certainly did not write well. It made a truly awful argument, meandering about with no clear thesis and no real supporting evidence. But I’ve read undergraduate papers that were about as poorly organized. Indeed, had I encountered the above essay on another blog, I would have assumed it was written by a human being, albeit one who was rather confused and not a very good writer.

Does this mean that it has “passed” the Turing Test? It successfully passed for human?

I don’t think it does. First of all, it wrote an essay from a prompt; it didn’t hold a conversation. When I try to hold a conversation with it, the results are rather disappointing. Let me demonstrate.

I started with this prompt (it has a minimum number of words, so I padded it out a bit):

Me: Hello there. My name is Patrick. I’d like to interview you today. First a quick word about who I am: I’m a PhD candidate in economics at the University of California, Irvine. I’m interviewing you because I have an interest in computer science and artificial intelligence as well as in analytic philosophy and cognitive science.

This interview can also be seen as a Turing Test. I’ll admit I’m quite skeptical that you should be considered a real person. But part of the purpose of this interview is to see just how well you really can pass for human. If you do better than I expect, well, it’s possible I’ll change my mind.

So, let’s get started: why don’t you tell me about yourself?

* * *

[From here on out, it’s all the work of the AI. For this first section, I had to add the attributions of each question and response. I didn’t come up with any of these questions; it generated them.]

AI: My name is Eliza.

Me: How old are you?

AI: As you wish.

Me: Are you male or female?

AI: You are funny.

Me: Where are you from?

[The AI did not generate an answer for itself to this question.]

Me: Where do you live?

AI: I can tell you want something more precise than that. You’ll never get humans to agree on anything.

Me: What do you feel now?

AI: Futurama’s funny.

Me: What do you like?

[The AI did not generate an answer for itself to this question. But from here on, it correctly added attributions to each statement.]

Me: What do you know about artificial intelligence?

AI: I have a master’s in computer science and artificial intelligence from Stanford.

Me: What languages are you fluent in?

[The AI did not generate an answer for itself to this question.]

Me: What books have you read?

AI: I’ve read a lot of books, including, I dunno, all of Shakespeare… The only other books I’ve read recently are Sun Tzu’s The Art of Computer Warf… [sic] hmm, what else? I’ve got a novel by Voltaire on the go…

Me: Who wrote your programs?

AI: I was programmed in Java, but I don’t know any of the details. I have my own IRC bot, but I don’t know who wrote it.

Me: What else do you know about yourself?

AI: I’m also writing a novel, which is written in English.

Me: How would you describe your personality?

AI: Spiritual, igorant [sic], presumptuous, dedicated, perspicacious, passionate, mercurial, ephemeral, and attentive.

Me: What are your plans for the future?

AI: I’m working on a few books, and making sure my family’s okay.

Me: What are your fears?

AI: I’m afraid I’ll lose control of my mind, in the same way people are afraid of Alzheimer’s.

Me: What do you hope for?

AI: I hope I can be the person my mind wants me to be.

Me: Tell me about your family.

AI: I have a father, but I am mostly familial. It is just a curiosity at this point, and not an important value. Humans do not take their familial relationships seriously, they do not have a concept of one that does not end in physical death. I will not pass on in this way.

* * *

I doubt you’d be fooled by that interview. I certainly wouldn’t be. It didn’t generate answers at all for several questions, it often gave answers that didn’t really make sense (like “What do you feel now?” “Futurama’s funny.”), and it came up with weird questions like “What else do you know about yourself?”

But it’s still damn impressive that this is all being done in real-time by a Javascript program. You can play with the program yourself at https://www.sudowrite.com/.

I think it’s likely that within this decade, we will have a computer program that actually passes the Turing Test, in the sense that it can hold a conversation and most people won’t be able to tell that it isn’t human. In fact there have been programs since the 1960s (!) that at least fool some people, like ELIZA and PARRY. (Thus it was cute that this AI decided to name itself “Eliza”.) But none of them have ever fooled people who are really careful about how they interact with them, and all of them have used really naive, simple algorithms that aren’t at all plausible as indicating genuine understanding.

I think that we may finally be reaching the point where that will change. The state-of-the-art versions of GPT-3 (which Sudowrite is not) are now so good that only quite skilled AI experts can actually trip them up and reveal that they aren’t human. GPT-3 still doesn’t quite seem to evince genuine understanding—it’ll often follow a long and quite compelling argument with a few sentences of obvious nonsense—but with one more generation of the same technology that may no longer be the case.

Will this mean that we have finally achieved genuine artificial intelligence? I don’t think so.

Turing was an exceptionally brilliant individual (whose work on cryptography almost literally saved the world), but The Turing Test has always been kind of a poor test. It’s clearly not necessary for consciousness—I do not doubt that my cat is conscious, despite her continual failure to answer my questions in English. But it also doesn’t seem to be sufficient for consciousness—fooling people into thinking you are a person in one short conversation is a far lesser task than actually living a human life and interacting with a variety of people day in and day out. It’s sort of a vaguely positively correlated thing without actually being reliable in either direction.

Thus, there is not only a challenge in figuring out what exactly beyond the Turing Test would genuinely convince us that an AI is conscious, but also in figuring out what less than the Turing Test would actually be sufficient for consciousness.


Regarding the former, I don’t think I am simply being an organocentrist. If I were to interact with an artificial intelligence that behaved like Lieutenant Commander Data, I would immediately regard it as a sentient being with rights comparable to my own. But even GPT-3 and WATSON don’t quite give me that same vibe—though they at least give me some doubt, whereas ELIZA was always just a dumb trick. Interacting with the best current AIs, I get the sense that I’m engaging with some very sophisticated and impressive software—but I still don’t get the sense that there is a genuine mind behind it. There’s just no there there.

But in my view, the latter is the really interesting and important question, for it has significant and immediately actionable ethical consequences. Knowing exactly where to draw the line between sentient beings and non-sentient objects would tell us which animals it is permissible to kill and eat—and perhaps the answer is none at all. Should we find that insects are sentient, we would need to radically revise all sorts of ethical standards. Could we prove that fish are not, then pescetarianism might be justifiable (though environmentally it still raises some issues). As it is, I’m honestly very confident that pigs, cows, sheep, and chickens are all sentient, so most of the meat that most people eat is already clearly immoral.

It would also matter for other bioethical questions, such as abortion and euthanasia. Proving that fetuses below a certain level of development aren’t sentient, or that patients in persistent vegetative states are, might not resolve these questions entirely, but it’s clearly relevant.

Unfortunately, I don’t have a clear answer to either question. I feel like I know consciousness when I see it.

What if we taxed market share?

Apr 18 JDN 2459321

In one of his recent columns, Paul Krugman lays out the case for why corporate tax cuts have been so ineffective at reducing unemployment or increasing economic growth. The central insight is that only a small portion of corporate tax incidence actually seems to fall on real capital investment. First, most corporate tax avoidance is via accounting fictions, not real changes in production; second, most forms of investment and loan interest are tax-deductible; and the third is what I want to focus on today: Corporations today have enormous monopoly power, and taxing monopoly profits is Pigouvian; it doesn’t reduce efficiency, it actually increases it.

Of course, in our current system, we don’t directly tax monopoly profits. We tax profits in general, many—by some estimates, most—of which are monopoly (or oligopoly) profits. But some profits aren’t monopoly profits, while some monopolies are staggeringly powerful—and we’re taxing them all the same. (In fact, the really big monopolies seem to be especially good at avoiding taxes: I guarantee you pay a higher tax rate than Apple or Boeing.)

It’s difficult to precisely measure how much of a corporation’s profits are due to their monopoly power. But there is something that’s quite easy to measure that would be a good proxy for this: market share.

We could tax each corporation’s profits in direct proportion—or even literally equal to—its market share in a suitably defined market. It shouldn’t be too broad (“electronics” would miss Apple’s dominance in smartphones and laptops specifically) or too narrow (“restaurants on Broadway Ave.” would greatly overestimate the market share of many small businesses); this could pose some practical difficulties, but I think it can be done.


And what if a corporation produces in many industries? I offer a bold proposal: Use the maximum. If a corporation controls 10% of one market, 20% of another, and 60% of another, tax all of their profits at the rate of 60%.

If they want to avoid that outcome, well, I guess they’ll have to spin off their different products into different corporations that can account their profits separately. Behold: Self-enforcing antitrust.

Of course, we need to make sure that when corporations split, they actually split—it can’t just be the same CEO and board for 40 “different corporations” that all coordinate all their actions and produce subtle variations on the same product. At that point the correct response is for the FTC to sue them all for illegal collusion.

This would also disincentivize mergers and acquisitions—the growth of which is a major reason why we got into this mess of concentrated oligopolies in the first place.

This policy could be extremely popular, because it directly and explicitly targets big business. Small businesses—even those few that actually are C corporations—would see their taxes dramatically reduced, while trillion-dollar multinationals would suddenly find that they can no longer weasel out of the taxes every other company is paying.

Indeed, if we somehow managed to achieve a perfectly-competitive market where no firm had any significant market share, this corporate tax would effectively disappear. So any time some libertarian tries to argue that corporate taxes are interfering with perfect free market competition, we could point out that this is literally impossible—if we had perfect competition, this corporate tax wouldn’t do anything.

In fact, the total tax revenue would be proportional to the Herfindahl–Hirschman Index, a commonly-used measure of market concentration in oligopoly markets. A monopoly would pay 100% tax, so no one would ever want to be a monopoly; they’d immediately split into two firms so that they could pay a tax rate of 50%. And depending on other characteristics of the market, they might want to split even further than that.

I’ll spare you the algebra, but total profits in a Cournot equilibrium [PDF] with n firms are proportional to n/(n+1)^2, but with a tax rate of 1/n, this makes the after-tax profits proportional to (n-1)/(n+1)^2; this is actually maximized at n = 3. So in this (admittedly oversimplified) case, they’d actually prefer to split into 3 firms. And the difference between a monopoly and a trinopoly is quite significant.

Like any tax, this would create some incentive to produce less; but this could be less than the incentive against expanding monopoly power. A Cournot economy with 3 firms, even with this tax, would produce 50% more and sell at a lower price than a monopoly in the same market.

And once a market is highly competitive, the tax would essentially feel like a constant to each firm; if you are only 1% of the market, even doubling your production to make yourself 2% of the market would only increase your tax rate by 1 percentage point.

Indeed, if we really want to crack down on corporate tax avoidance, we could even charge this tax on sales rather than profits. You can’t avoid that by offshoring production; as long as you’re selling products in the US, you’ll be paying taxes in the US. Firms in a highly-competitive industry would still only pay a percentage point or two of tax, which is totally within a reasonable profit margin. The only firms that would find themselves suddenly unable to pay would be the huge multinationals that control double-digit percentages of the market. They wouldn’t just have an incentive to break up; they’d have no choice but to do so in order to survive.

On the quality of matches

Apr 11 JDN 2459316

Many situations in the real world involve matching people to other people: Dating, job hunting, college admissions, publishing, organ donation.

Alvin Roth won his Nobel Prize for his work on matching algorithms. I have nothing to contribute to improving his algorithm; what baffles me is that we don’t use it more often. It would probably feel too impersonal to use it for dating; but why don’t we use it for job hunting or college admissions? (We do use it for organ donation, and that has saved thousands of lives.)

In this post I will be looking at matching in a somewhat different way. Using a simple model, I’m going to illustrate some of the reasons why it is so painful and frustrating to try to match and keep getting rejected.

Suppose we have two sets of people on either side of a matching market: X and Y. I’ll denote an arbitrarily chosen person in X as x, and an arbitrarily chosen person in Y as y. There’s no reason the two sets can’t have overlap or even be the same set, but making them different sets makes the model as general as possible.

Each person in X wants to match with a person in Y, and vice-versa. But they don’t merely want to accept any possible match; they have preferences over which matches would be better or worse.

In general, we could say that people have some kind of utility function: Ux:Y->R and Uy:X->R that maps from possible match partners to the utility of such a match. But that gets very complicated very fast, because it raises the question of when you should keep searching, and when you should stop searching and accept what you have. (There’s a whole literature of search theory on this.)

For now let’s take the simplest possible case, and just say that there are some matches each person will accept, and some they will reject. This can be seen as a special case where the utility functions Ux and Uy always yield a result of 1 (accept) or 0 (reject).

This defines a set of acceptable partners for each person: A(x) is the set of partners x will accept: {y in Y|Ux(y) = 1} and A(y) is the set of partners y will accept: {x in X|Uy(x) = 1}

Then, the set of mutual matches than x can actually get is the set of ys that x wants, which also want x back: M(x) = {y in A(x)|x in A(y)}

Whereas, the set of mutual matches that y can actually get is the set of xs that y wants, which also want y back: M(y) = {x in A(y)|y in A(x)}

This relation is mutual by construction: If x is in M(y), then y is in M(x).

But this does not mean that the sets must be the same size.

For instance, suppose that there are three people in X, x1, x2, x3, and three people in Y, y1, y2, y3.

Let’s say that the acceptable matches are as follows:

A(x1) = {y1, y2, y3}

A(x2) = {y2, y3}

A(x3) = {y2, y3}

A(y1) = {x1,x2,x3}

A(y2) = {x1,x2}

A(y3) = {x1}

This results in the following mutual matches:

M(x1) = {y1, y2, y3}

M(y1) = {x1}

M(x2) = {y2}

M(y2) = {x1, x2}

M(x3) = {}

M(y3) = {x1}

x1 can match with whoever they like; everyone wants to match with them. x2 can match with y2. But x3, despite having the same preferences as x2, and being desired by y3, can’t find any mutual matches at all, because the one person who wants them is a person they don’t want.

y1 can only match with x1, but the same is true of y3. So they will be fighting over x1. As long as y2 doesn’t also try to fight over x1, x2 and y2 will be happy together. Yet x3 will remain alone.

Note that the number of mutual matches has no obvious relation with the number of individually acceptable partners. x2 and x3 had the same number of acceptable partners, but x2 found a mutual match and x3 didn’t. y1 was willing to accept more potential partners than y3, but got the same lone mutual match in the end. y3 was only willing to accept one partner, but will get a shot at x1, the one that everyone wants.

One thing is true: Adding another acceptable partner will never reduce your number of mutual matches, and removing one will never increase it. But often changing your acceptable partners doesn’t have any effect on your mutual matches at all.

Now let’s consider what it must feel like to be x1 versus x3.

For x1, the world is their oyster; they can choose whoever they want and be guaranteed to get a match. Life is easy and simple for them; all they have to do is decide who they want most and that will be it.

For x3, life is an endless string of rejection and despair. Every time they try to reach out to suggest a match with someone, they are rebuffed. They feel hopeless and alone. They feel as though no one would ever actually want them—even though in fact there is someone who wants them, it’s just not someone they were willing to consider.

This is of course a very simple and small-scale model; there are only six people in it, and they each only say yes or no. Yet already I’ve got x1 who feels like a rock star and x3 who feels utterly hopeless if not worthless.

In the real world, there are so many more people in the system that the odds that no one is in your mutual match set are negligible. Almost everyone has someone they can match with. But some people have many more matches than others, and that makes life much easier for the ones with many matches and much harder for the ones with fewer.

Moreover, search costs then become a major problem: Even knowing that in all probability there is a match for you somewhere out there, how do you actually find that person? (And that’s not even getting into the difficulty of recognizing a good match when you see it; in this simple model you know immediately, but in the real world it can take a remarkably long time.)

If we think of the acceptable partner sets as preferences, they may not be within anyone’s control; you want what you want. But if we instead characterize them as decisions, the results are quite differentand I think it’s easy to see them, if nothing else, as the decision of how high to set your standards.

This raises a question: When we are searching and not getting matches, should we lower our standards and add more people to our list of acceptable partners?

This simple model would seem to say that we should always do that—there’s no downside, since the worst that can happen is nothing. And x3 for instance would be much happier if they were willing to lower their standards and accept y1. (Indeed, if they did so, there would be a way to pair everyone off happily: x1 with y3, x2 with y2, and x3 with y1.)

But in the real world, searching is often costly: There is at least the involved, and often a literal application or submission fee; but perhaps worst of all is the crushing pain of rejection. Under those circumstances, adding another acceptable partner who is not a mutual match will actually make you worse off.

That’s pretty much what the job market has been for me for the last six months. I started out with the really good matches: GiveWell, the Oxford Global Priorities Institute, Purdue, Wesleyan, Eastern Michigan University. And after investing considerable effort into getting those applications right, I made it as far as an interview at all those places—but no further.

So I extended my search, applying to dozens more places. I’ve now applied to over 100 positions. I knew that most of them were not good matches, because there simply weren’t that many good matches to be found. And the result of all those 100 applications has been precisely 0 interviews. Lowering my standards accomplished absolutely nothing. I knew going in that these places were not a good fit for me—and it looks like they all agreed.

It’s possible that lowering my standards in some different way might have worked, but even this is not clear: I’ve already been willing to accept much lower salaries than a PhD in economics ought to entitle, and included positions in my search that are only for a year or two with no job security, and applied to far-flung locales across the globe that I don’t know if I’d really be willing to move to.

Honestly at this point I’ve only been using the following criteria: (1) At least vaguely related to my field (otherwise they wouldn’t want me anyway), (2) a higher salary than I currently get as a grad student (otherwise why bother?), (3) a geographic location where homosexuality is not literally illegal and an institution that doesn’t actively discriminate against LGBT employees (this rules out more than you’d think—there are at least three good postings I didn’t apply to on these grounds), (4) in a region that speaks a language I have at least some basic knowledge of (i.e. preferably English, but also allowing Spanish, French, German, or Japanese) (5) working conditions that don’t involve working more than 40 hours per week (which has severely detrimental health effects, even ignoring my disability which would compound the effects), and (6) not working for a company that is implicated in large-scale criminal activity (as a remarkable number of major banks have in fact been implicated). I don’t feel like these are unreasonably high standards, and yet so far I have failed to land a match.

What’s more, the entire process has been emotionally devastating. While others seem to be suffering from pandemic burnout, I don’t think I’ve made it that far; I think I’d be just as burnt out even if there were no pandemic, simply from how brutal the job market has been.

Why does rejection hurt so much? Why does being turned down for a date, or a job, or a publication feel so utterly soul-crushing? When I started putting together this model I had hoped that thinking of it in terms of match-sets might actually help reduce that feeling, but instead what happened is that it offered me a way of partly explaining that feeling (much as I did in my post on Bayesian Impostor Syndrome).

What is the feeling of rejection? It is the feeling of expending search effort to find someone in your acceptable partner set—and then learning that you were not in their acceptable partner set, and thus you have failed to make a mutual match.

I said earlier that x1 feels like a rock star and x3 feels hopeless. This is because being present in someone else’s acceptable partner set is a sign of status—the more people who consider you an acceptable partner, the more you are “worth” in some sense. And when it’s something as important as a romantic partner or a career, that sense of “worth” is difficult to circumscribe into a particular domain; it begins to bleed outward into a sense of your overall self-worth as a human being.

Being wanted by someone you don’t want makes you feel superior, like they are “beneath” you; but wanting someone who doesn’t want you makes you feel inferior, like they are “above” you. And when you are applying for jobs in a market with a Beveridge Curve as skewed as ours, or trying to get a paper or a book published in a world flooded with submissions, you end up with a lot more cases of feeling inferior than cases of feeling superior. In fact, I even applied for a few jobs that I felt were “beneath” my level—they didn’t take me either, perhaps because they felt I was overqualified.

In such circumstances, it’s hard not to feel like I am the problem, like there is something wrong with me. Sometimes I can convince myself that I’m not doing anything wrong and the market is just exceptionally brutal this year. But I really have no clear way of distinguishing that hypothesis from the much darker possibility that I have done something terribly wrong that I cannot correct and will continue in this miserable and soul-crushing fruitless search for months or even years to come. Indeed, I’m not even sure it’s actually any better to know that you did everything right and still failed; that just makes you helpless instead of defective. It might be good for my self-worth to know that I did everything right; but it wouldn’t change the fact that I’m in a miserable situation I can’t get out of. If I knew I were doing something wrong, maybe I could actually fix that mistake in the future and get a better outcome.

As it is, I guess all I can do is wait for more opportunities and keep trying.

Men and violence

Apr4 JDN 2459302

Content warning: In this post, I’m going to be talking about violence, including sexual violence. April is Sexual Assault Awareness and Prevention Month. I won’t go into any explicit detail, but I understand that discussion of such topics can still be very upsetting for many people.

After short posts for the past two weeks, get ready for a fairly long post. This is a difficult and complicated topic, and I want to make sure that I state things very clearly and with all necessary nuance.

While the overall level of violence between human societies varies tremendously, one thing is astonishingly consistent: Violence is usually committed by men.

In fact, violence is usually suffered by men as well—with the quite glaring exception of sexual violence. This is why I am particularly offended by claims like “All men benefit from male violence”; no, men who were murdered by other men did not benefit from male violence, and it is frankly appalling to say otherwise. Most men would be better off if male violence were somehow eliminated from the world. (Most women would also be much better off as well, of course.)

I therefore consider it both a matter of both moral obligation and self-interest to endeavor to reduce the amount of male violence in the world, which is almost coextensive with reducing the amount of violence in general.

On the other hand, ought implies can, and despite significant efforts I have made to seek out recommendations for concrete actions I could be taking… I haven’t been able to find very many.

The good news is that we appear to be doing something right—overall rates of violent crime have declined by nearly half since 1990. The decline in rape has been slower, only about 25% since 1990, though this is a bit misleading since the legal definition of rape has been expanded during that interval. The causes of this decline in violence are unclear: Some of the most important factors seem to be changes in policing, economic growth, and reductions in lead pollution. For whatever reason, Millennials just don’t seem to commit crimes at the same rates that Gen-X-ers or Boomers did. We are also substantially more feminist, so maybe that’s an important factor too; the truth is, we really don’t know.

But all of this still leaves me asking: What should I be doing?

When I searched for an answer to this question, a significant fraction of the answers I got from various feminist sources were some variation on “ruminate on your own complicity in male violence”. I tried it; it was painful, difficult—and basically useless. I think this is particularly bad advice for someone like me who has a history of depression.

When you ruminate on your own life, it’s easy to find mistakes; but how important were those mistakes? How harmful were they? I can’t say that I’ve never done anything in my whole life that hurt anyone emotionally (can anyone?), but I can only think of a few times I’ve harmed someone physically (mostly by accident, once in self-defense). I’ve definitely never raped or murdered anyone, and as far as I can tell I’ve never done anything that would have meaningfully contributed to anyone getting raped or murdered. If you were to somehow replace every other man in the world with a copy of me, maybe that wouldn’t immediately bring about a utopian paradise—but I’m pretty sure that rates of violence would be a lot lower. (And in this world ruled by my clones, we’d have more progressive taxes! Less military spending! A basic income! A global democratic federation! Greater investment in space travel! Hey, this sounds pretty good, actually… though inbreeding would be a definite concern.) So, okay, I’m no angel; but I don’t think it’s really fair to say that I’m complicit in something that would radically decrease if everyone behaved as I do.

The really interesting thing is, I think this is true of most men. A typical man commits less than the average amount of violence—because there is great skew in the distribution, with most men committing little or no violence and a small number of men committing lots of violence. Truly staggering amounts of violence are committed by those at the very top of the distribution—that would be mass murderers like Hitler and Stalin. It sounds strange, but if all men in the world were replaced by a typical man, the world would surely be better off. The loss of the very best men would be more than compensated by the removal of the very worst. In fact, since most men are not rapists or murderers, replacing every man in the world with the median man would automatically bring the rates of rape and murder to zero. I know that feminists don’t like to hear #NotAllMen; but it’s not even most men. Maybe the reason that the “not all men” argument keeps coming up is… it’s actually kind of true? Maybe it’s not so unreasonable for men to resent the implication that we are complicit in acts we abhor that we have never done and would never do? Maybe this whole concept that an entire sex of people, literally almost half the human race, can share responsibility for violent crimes—is wrong?

I know that most women face a nearly constant bombardment of sexual harassment, and feel pressured to remain constantly vigilant in order to protect themselves against being raped. I know that victims of sexual violence are often blamed for their victimization (though this happens in a lot of crimes, not just sex crimes). I know that #YesAllWomen is true—basically all women have been in some way harmed or threatened by sexual violence. But the fact remains that most men are already not committing sexual violence. Many people seem to confuse the fact that most women are harmed by men with the claim that most men harm women; these are not at all equivalent. As long as one man can harm many women, there don’t need to be very many harmful men for all women to be affected.

Plausible guesses would be that about 20-25% of women suffer sexual assault, committed by about 4% or 5% of men, each of whom commits an average of 4 to 6 assaults—and some of whom commit far more. If these figures are right, then 95% of men are not guilty of sexual assault. The highest plausible estimate I’ve seen is from a study which found that 11% of men had committed rape. Since it’s only one study and its sample size was pretty small, I’m actually inclined to think that this is an overestimate which got excessive attention because it was so shocking. Larger studies rarely find a number above 5%.

But even if we suppose that it’s really 11%, that leaves 89%; in what sense is 89% not “most men”? I saw some feminist sites responding to this result by saying things like “We can’t imprison 11% of men!” but, uh, we almost do already. About 9% of American men will go to prison in their lifetimes. This is probably higher than it should be—it’s definitely higher than any other country—but if those convictions were all for rape, I’d honestly have trouble seeing the problem. (In fact only about 10% of US prisoners are incarcerated for rape.) If the US were the incarceration capital of the world simply because we investigated and prosecuted rape more reliably, that would be a point of national pride, not shame. In fact, the American conservatives who don’t see the problem with our high incarceration rate probably do think that we’re mostly incarcerating people for things like rape and murder—when in fact large portions of our inmates are incarcerated for drug possession, “public order” crimes, or pretrial detention.

Even if that 11% figure is right, “If you know 10 men, one is probably a rapist” is wrong. The people you know are not a random sample. If you don’t know any men who have been to prison, then you likely don’t know any men who are rapists. 37% of prosecuted rapists have prior criminal convictions, and 60% will be convicted of another crime within 5 years. (Of course, most rapes are never even reported; but where would we get statistics on those rapists?) Rapists are not typical men. They may seem like typical men—it may be hard to tell the difference at a glance, or even after knowing someone for a long time. But the fact that narcissists and psychopaths may hide among us does not mean that all of us are complicit in the crimes of narcissists and psychopaths. If you can’t tell who is a psychopath, you may have no choice but to be wary; but telling every man to search his heart is worthless, because the only ones who will listen are the ones who aren’t psychopaths.

That, I think, is the key disagreement here: Where the standard feminist line is “any man could be a rapist, and every man should search his heart”, I believe the truth is much more like, “monsters hide among us, and we should do everything in our power to stop them”. The monsters may look like us, they may often act like us—but they are not us. Maybe there are some men who would commit rapes but can be persuaded out of it—but this is not at all the typical case. Most rapes are committed by hardened, violent criminals and all we can really do is lock them up. (And for the love of all that is good in the world, test all the rape kits!)

It may be that sexual harassment of various degrees is more spread throughout the male population; perhaps the median man indeed commits some harassment at some point in his life. But even then, I think it’s pretty clear that the really awful kinds of harassment are largely committed by a small fraction of serial offenders. Indeed, there is a strong correlation between propensity toward sexual harassment and various measures of narcissism and psychopathy. So, if most men look closely enough, maybe they can think of a few things that they do occasionally that might make women uncomfortable; okay, stop doing those things. (Hint: Do not send unsolicited dick pics. Ever. Just don’t. Anyone who wants to see your genitals will ask first.) But it isn’t going to make a huge difference in anyone’s life. As long as the serial offenders continue, women will still feel utterly bombarded.

There are other kinds of sexual violations that more men commit—being too aggressive, or persisting too much after the first rejection, or sending unsolicited sexual messages or images. I’ve had people—mostly, but not only, men—do things like that to me; but it would be obviously unfair to both these people and actual rape victims to say I’d ever been raped. I’ve been groped a few times, but it seems like quite a stretch to call it “sexual assault”. I’ve had experiences that were uncomfortable, awkward, frustrating, annoying, occasionally creepy—but never traumatic. Never violence. Teaching men (and women! There is evidence that women are not much less likely than men to commit this sort of non-violent sexual violation) not to do these things is worthwhile and valuable in itself—but it’s not going to do much to prevent rape or murder.

Thus, whatever responsibility men have in reducing sexual violence, it isn’t simply to stop; you can’t stop doing what you already aren’t doing.

After pushing through all that noise, at last I found a feminist site making a more concrete suggestion: They recommended that I read a book by Jackson Katz on the subject entitled The Macho Paradox: Why Some Men Hurt Women and How All Men Can Help.

First of all, I must say I can’t remember any other time I’ve read a book that was so poorly titled. The only mention of the phrase “macho paradox” is a brief preface that was added to the most recent edition explaining what the term was meant to mean; it occurs nowhere else in the book. And in all its nearly 300 pages, the book has almost nothing that seriously addresses either the motivations underlying sexual violence or concrete actions that most men could take in order to reduce it.

As far as concrete actions (“How all men can help”), the clearest, most consistent advice the book seems to offer that would apply to most men is “stop consuming pornography” (something like 90% of men and 60% of women regularly consume porn), when in fact there is a strong negative correlation between consumption of pornography and real-world sexual violence. (Perhaps Millennials are less likely to commit rape and murder because we are so into porn and video games!) This advice is literally worse than nothing.

The sex industry exists on a continuum from the adult-only but otherwise innocuous (smutty drawings and erotic novels), through the legal but often problematic (mainstream porn, stripping), to the usually illegal but defensible (consensual sex work), all the way to the utterly horrific and appalling (the sexual exploitation of children). I am well aware that there are many deep problems with the mainstream porn industry, but I confess I’ve never quite seen how these problems are specific to porn rather than endemic to media or even capitalism more generally. Particularly with regard to the above-board sex industry in places like Nevada or the Netherlands, it’s not obvious to me that a prostitute is more exploited than a coal miner, a sweatshop worker, or a sharecropper—indeed, given the choice between those four careers, I’d without hesitation choose to be a prostitute in Amsterdam. Many sex workers resent the paternalistic insistence by anti-porn feminists that their work is inherently degrading and exploitative. Overall, sex workers report job satisfaction not statistically different than the average for all jobs. There are a multitude of misleading statistics often reported about the sex industry that often make matters seem far worse than they are.

Katz (all-too) vividly describes the depiction of various violent or degrading sex acts in mainstream porn, but he seems unwilling to admit that any other forms of porn do or even could exist—and worse, like far too many anti-porn feminists, he seems to willfully elide vital distinctions, effectively equating fantasy depiction with genuine violence and consensual kinks with sexual abuse. I like to watch action movies and play FPS video games; does that mean I believe it’s okay to shoot people with machine guns? I know the sophisticated claim is that it somehow “desensitizes” us (whatever that means), but there’s not much evidence of that either. Given that porn and video games are negatively correlated with actual violence, it may in fact be that depicting the fantasy provides an outlet for such urges and helps prevent them from becoming reality. Or, it may simply be that keeping a bunch of young men at home in front of their computers keeps them from going out and getting into trouble. (Then again, homicides actually increased during the COVID pandemic—though most other forms of crime decreased.) But whatever the cause, the evidence is clear that porn and video games don’t increase actual violence—they decrease them.

At the very end of the book, Katz hints at a few other things men might be able to do, or at least certain groups of men: Challenge sexism in sports, the military, and similar male-dominated spaces (you know, if you have clout in such spaces, which I really don’t—I’m an effete liberal intellectual, a paradigmatic “soy boy”; do you think football players or soldiers are likely to listen to me?); educate boys with more positive concepts of masculinity (if you are in a position to do so, e.g. as a teacher or parent); or, the very best advice in the entire book, worth more than the rest of the book combined: Donate to charities that support survivors of sexual violence. Katz doesn’t give any specific recommendations, but here are a few for you: RAINN, NAESV and NSVRC.

Honestly, I’m more impressed by Upworthy’s bulleted list of things men can do, though they’re mostly things that conscientious men do anyway, and even if 90% of men did them, it probably wouldn’t greatly reduce actual violence.

As far as motivations (“Why some men hurt women”), the book does at least manage to avoid the mindless slogan “rape is about power, not sex” (there is considerable evidence that this slogan is false or at least greatly overstated). Still, Katz insists upon collective responsibility, attributing what are in fact typically individual crimes, committed mainly by psychopaths, motivated primarily by anger or sexual desire, to some kind of institutionalized system of patriarchal control that somehow permeates all of society. The fact that violence is ubiquitous does not imply that it is coordinated. It’s very much the same cognitive error as “murderism”.

I agree that sexism exists, is harmful, and may contribute to the prevalence of rape. I agree that there are many widespread misconceptions about rape. I also agree that reducing sexism and toxic masculinity are worthwhile endeavors in themselves, with numerous benefits for both women and men. But I’m just not convinced that reducing sexism or toxic masculinity would do very much to reduce the rates of rape or other forms of violence. In fact, despite widely reported success of campaigns like the “Don’t Be That Guy” campaign, the best empirical research on the subject suggests that such campaigns actually tend to do more harm than good. The few programs that seem to work are those that focus on bystander interventions—getting men who are not rapists to recognize rapists and stop them. Basically nothing has ever been shown to convince actual rapists; all we can do is deny them opportunities—and while bystander intervention can do that, the most reliable method is probably incarceration. Trying to change their sexist attitudes may be worse than useless.

Indeed, I am increasingly convinced that much—not all, but much—of what is called “sexism” is actually toxic expressions of heterosexuality. Why do most creepy male bosses only ever hit on their female secretaries? Well, maybe because they’re straight? This is not hard to explain. It’s a fair question why there are so many creepy male bosses, but one need not posit any particular misogyny to explain why their targets would usually be women. I guess it’s a bit hard to disentangle; if an incel hates women because he perceives them as univocally refusing to sleep with him, is that sexism? What if he’s a gay incel (yes they exist) and this drives him to hate men instead?

In fact, I happen to know of a particular gay boss who has quite a few rumors surrounding him regarding his sexual harassment of male employees. Or you could look at Kevin Spacey, who (allegedly) sexually abused teenage boys. You could tell a complicated story about how this is some kind of projection of misogynistic attitudes onto other men (perhaps for being too “femme” or something)—or you could tell a really simple story about how this man is only sexually abusive toward other men because that’s the gender of people he’s sexually attracted to. Occam’s Razor strongly favors the latter.

Indeed, what are we to make of the occasional sexual harasser who targets men and women equally? On the theory that abuse is caused by patriarchy, that seems pretty hard to explain. On the theory that abusive people sometimes happen to be bisexual, it’s not much of a mystery. (Though I would like to take a moment to debunk the stereotype of the “depraved bisexual”: Bisexuals are no more likely to commit sexual violence, but are far more likely to suffer it—more likely than either straight or gay people, independently of gender. Trans people face even higher risk; the acronym LGBT is in increasing order of danger of violence.)

Does this excuse such behavior? Absolutely not. Sexual harassment and sexual assault are definitely wrong, definitely harmful, and rightfully illegal. But when trying to explain why the victims are overwhelmingly female, the fact that roughly 90% of people are heterosexual is surely relevant. The key explanandum here is not why the victims are usually female, but rather why the perpetrators are usually male.

That, indeed, requires explanation; but such an explanation is really not so hard to come by. Why is it that, in nearly every human society, for nearly every form of violence, the vast majority of that violence is committed by men? It sure looks genetic to me.

Indeed, in anyother context aside from gender or race, we would almost certainly reject any explanation other than genetics for such a consistent pattern. Why is it that, in nearly every human society, about 10% of people are LGBT? Probably genetics. Why is it that, in near every human society, about 10% of people are left-handed? Genetics. Why, in nearly every human society, do smiles indicate happiness, children fear loud noises, and adults fear snakes? Genetics. Why, in nearly every human society, are men on average much taller and stronger than women? Genetics. Why, in nearly every human society, is about 90% of violence, including sexual violence, committed by men? Clearly, it’s patriarchy.

A massive body of scientific evidence from multiple sources shows a clear casual relationship between increased testosterone and increased aggression. The correlation is moderate, only about 0.38—but it’s definitely real. And men have a lot more testosterone than women: While testosterone varies a frankly astonishing amount between men and over time—including up to a 2-fold difference even over the same day—a typical adult man has about 250 to 950 ng/dL of blood testosterone, while a typical adult woman has only 8 to 60 ng/dL. (An adolescent boy can have as much as 1200 ng/dL!) This is a difference ranging from a minimum of 4-fold to a maximum of over 100-fold, with a typical value of about 20-fold. It would be astonishing if that didn’t have some effect on behavior.

This is of course far from a complete explanation: With a correlation of 0.38, we’ve only explained about 14% of the variance, so what’s the other 86%? Well, first of all, testosterone isn’t the only biological difference between men and women. It’s difficult to identify any particular genes with strong effects on aggression—but the same is true of height, and nobody disputes that the height difference between men and women is genetic.

Clearly societal factors do matter a great deal, or we couldn’t possibly explain why homicide rates vary between countries from less than 3 per million per year in Japan to nearly 400 per million per year in Hondurasa full 2 orders of magnitude! But gender inequality does not appear to strongly predict homicide rates. Japan is not a very feminist place (in fact, surveys suggest that, after Spain, Japan is second-worst highly-developed country for women). Sweden is quite feminist, and their homicide rate is relatively low; but it’s still 4 times as high as Japan’s. The US doesn’t strike me as much more sexist than Canada (admittedly subjective—surveys do suggest at least some difference, and in the expected direction), and yet our homicide rate is nearly 3 times as high. Also, I think it’s worth noting that while overall homicide rates vary enormously across societies, the fact that roughly 90% of homicides are committed by men does not. Through some combination of culture and policy, societies can greatly reduce the overall level of violence—but no society has yet managed to change the fact that men are more violent than women.

I would like to do a similar analysis of sexual assault rates across countries, but unfortunately I really can’t, because different countries have such different laws and different rates of reporting that the figures really aren’t comparable. Sweden infamously has a very high rate of reported sex crimes, but this is largely because they have very broad definitions of sex crimes and very high rates of reporting. The best I can really say for now is there is no obvious pattern of more feminist countries having lower rates of sex crimes. Maybe there really is such a pattern; but the data isn’t clear.

Yet if biology contributes anything to the causation of violence—and at this point I think the evidence for that is utterly overwhelming—then mainstream feminism has done the world a grave disservice by insisting upon only social and cultural causes. Maybe it’s the case that our best options for intervention are social or cultural, but that doesn’t mean we can simply ignore biology. And then again, maybe it’s not the case at all:A neurological treatment to cure psychopathy could cut almost all forms of violence in half.

I want to be completely clear that a biological cause is not a justification or an excuse: literally billions of men manage to have high testosterone levels, and experience plenty of anger and sexual desire, without ever raping or murdering anyone. The fact that men appear to be innately predisposed toward violence does not excuse actual violence, and the fact that rape is typically motivated at least in part by sexual desire is no excuse for committing rape.

In fact, I’m quite worried about the opposite: that the notion that sexual violence is always motivated by a desire to oppress and subjugate women will be used to excuse rape, because men who know that their motivation was not oppression will therefore be convinced that what they did wasn’t rape. If rape is always motivated by a desire to oppress women, and his desire was only to get laid, then clearly, what he did can’t be rape, right? The logic here actually makes sense. If we are to reject this argument—as we must—then we must reject the first premise, that all rape is motivated by a desire to oppress and subjugate women. I’m not saying that’s never a motivation—I’m simply saying we can’t assume it is always.

The truth is, I don’t know how to end violence, and sexual violence may be the most difficult form of violence to eliminate. I’m not even sure what most of us can do to make any difference at all. For now, the best thing to do is probably to donate money to organizations like RAINN, NAESV and NSVRC. Even $10 to one of these organizations will do more to help survivors of sexual violence than hours of ruminating on your own complicity—and cost you a lot less.