There’s no inherent contradiction between the ideas. The argument for veganism is “you shouldn’t harm others a lot for small benefits,” not “you shouldn’t harm others a lot for small benefits—and also taxes are great.” And yet relatively few people with libertarian political dispositions refrain from consuming animal products.

The explanation, I submit, is ideology. The libertarian worldview holds that many problems are the byproduct of untoward government mischief-making. The libertarian carefully searches for harms of government. This often leads them to neglect other forms of injustice, even ones that don’t explicitly conflict with libertarianism.

We have all witnessed political partisans twist themselves up in pretzels trying to deny that harm has come from their side of the aisle. Upton Sinclair famously said, “It is difficult to get a man to understand something, when his salary depends on his not understanding it.” It is likewise difficult to get a man to understand something when it requires admitting that his side is in the wrong.

Ideology doesn’t just affect our direct beliefs. It affects which things we’re attuned to. The world is filtered through the lens of our ideology. This often leads us to overfocus on harms caused by those things that we are particularly ideologically disposed to notice, neglecting those that don’t fit neatly into our ideology. And this holds even if our ideology gives us no reason to ignore those harms. Libertarian political commitments have no direct implications concerning the badness of factory farming. Nevertheless, in practice, they affect one’s assessment of it.

Abundance—by Ezra Klein and Derek Thompson—discussed ways government policy could make goods more affordable and abundant. They propose eliminating housing restrictions, so that more houses could be built and prices lowered. Many of their fiercest opponents were leftists. Often, the responses to their book would contain bizarre non sequiturs, suggesting that the “real reason” housing prices were high was corporate greed.

Consider the above Tweet by Sirota. On its face, his complaint is bizarre. He cites a study saying that consolidation raises home prices and suggests that this refutes the Abundance thesis that housing regulations lead to high home prices. But more than one thing can raise housing prices. This would be like arguing that famine doesn’t kill many people by citing a study on mortality from smoking. It would be one thing if this were a one-off, but some variant of this criticism was the primary left-wing criticism of Abundance. What’s going on?

Obviously believing that burdensome regulations raise housing prices doesn’t conflict with leftist ideology. Presumably there is some degree of regulatory interference that would—even according to leftists—drive up housing prices. If the government required everyone who builds a house to hire 5,000 Hula dancers to dance on the property before beginning construction, that would obviously increase housing prices. Whether that degree actually exists in the real world is an empirical question. So on its face, it seems a bit puzzling that leftists are so unwilling to blame regulations for high housing prices.

But once we remember that ideology isn’t just about explicit beliefs, but broad narratives, it’s easy to diagnose this. Leftists are used to blaming problems on corporate greed. When problems are blamed on things other than corporations, leftists will often be driven by ideology to downplay those other things and blame corporations. This tendency even led some authors to badly misreport what studies actually said in order to argue against the Abundance thesis.

All too often, ideology turns us into hammers in search of nails. Insofar as some problem doesn’t fit neatly into our ideology, it is disregarded. Communists rush to defend every communist leader—or to blame their malfeasance on capitalism—because ostensible communists doing terrible things doesn’t fit neatly into their ideology. Anti-imperialists often rush to blame misconduct on American foreign policy and whitewash the crimes of foreign leaders. Though they recognize abstractly that evil was not a recent invention of the American foreign policy establishment, blaming things on foreigners with dark skin doesn’t fit as well with their picture of the world.

This tendency is not exclusively a leftist phenomenon. It is a risk we all face. For example, I recently saw a Tweet from Chris Rufo about USAID cuts:

This seemed like a particularly acute example of ideological derangement. A core tenet of the Rufo worldview seems to be that many horrible things come from liberals denying dark truths. This may be true, but it has no direct implications for the desirability of foreign aid. Treating this as exculpatory of Musk’s actions requires thinking both:

1. Liberals are generally unaware of the dysfunctionality of foreign countries that receive foreign aid.

2. The dysfunctionality of recipients of foreign aid means that foreign aid is undesirable.

But neither of these is true, and they’re certainly not clearly true. Instead of seriously looking at costs and benefits, Rufo simply repeated some tenet of his ideology and vaguely gestured at it having serious implications for the foreign aid cuts. He has regressed from a thinking creature into a simplistic GPT-2-esque pattern matcher.

Similarly, libertarians will often underestimate the number of lives saved by foreign aid programs. Because they are used to inefficient and sclerotic government programs mucking things up, in the rare cases where government functions well, their ideology prejudices them into not seeing this.

Or consider those religious thinkers who declare that AI is not conscious because the soul is immaterial and AI is made of physical stuff. Now, a moment’s reflection on this argument reveals its spuriousness. The brain is also a physical object, but it gives rise to consciousness. Why, then, can’t AI?

Here, the religious thinker has become blinded by ideology. They are so used to criticizing materialism that they erroneously assume that belief in AI sentience requires materialist assumptions. They are so used to attacking materialism that they treat an unrelated view as a proxy in their broader ideological war against materialism. Like Rufo, they have replaced thinking with mere pattern matching.

This is a risk that we all face, including me. If I had to summarize my ideology in a sentence, it would be something like “many of the world’s biggest problems stem from wholesale neglect of large classes of conscious beings.” I thus may be prone to overestimate harms that come from a constricted moral circle and underestimate harms from improperly designed institutions. Similarly, because I am on the political left, it is very likely that I have a general tendency to overestimate the badness of right-wing policies and underestimate the badness of left-wing ones.

Why is this tendency so dangerous?

One reason is the obvious one: it leads to us having systematically false and distorted beliefs about the world—thinking some problems are bigger than they are and other problems are smaller. One who only detects badness when it fits neatly into one’s own ideology will inevitably ignore many terrible things.

But there is another reason that may be more significant. Many of the world’s most serious problems don’t fit neatly into many ideologies. Take, as an example, the fact that mirror bacteria may be developed and cause the extinction of life on Earth. This isn’t particularly aligned with any ideology. Mirror bacteria are neither Muslim, transgender, Israeli, nor Trump supporters. Neither populism, conservatism, nor leftism has anything special to say about these rogue pathogens. They’re just bad. But insofar as we only focus on harms that correspond with our ideology, we will ignore those things which are just plain bad.

Gaza is ideological. Endless ink is spilled discussing it. Sudan and car crashes are not ideological. As a result, they are almost completely ignored (except in the rare cases when their existence becomes a convenient political football—e.g. when pro-Israel advocates suggest that anti-semitism explains why people excessively fixate on Gaza). The world is full of opportunities to make things better. Yet we cannot do so if we all ignore problems without an ideological tilt.

This is one thing I appreciate about effective altruism’s cause neutrality. Rather than only working on problems that fit within some specific ideology, effective altruists try to identify the problems where they can make the biggest difference. Malaria is unsexy and non-ideological, but that makes it no less horrible. When a child dies in her mother’s arms, her mother takes no solace in the fact that her child’s death is not a convenient American culture-war football.

To the extent we wish to avoid ideological derangement, we ought to focus on the biggest problems in the world and the areas where we can do the most good. Sometimes this corresponds with our ideology. Other times it does not. Political views describe what we see as the legitimate role of government. To treat them as totalizing accounts of all that matters in the world is to become blind to many of the biggest problems and to become systematically ideologically deranged. The world does not owe us, and rarely provides us, problems that fit neatly into pre-arranged ideological categories.

But what is this funny business about “candidates for being your present self”? Various critics have alleged that this term is polysemous or vague and that the core dynamics of the theory are underspecified. This isn’t true. But to explain why, I will need to introduce a bit of terminology.

Relative probability refers to how probable one thing is compared to another. You can know the relative probability of each of two things, even without knowing their absolute probability. You can know that one thing is half as likely as another even without knowing how likely either of them is.

A “center” refers to a person, at a time, in a world. A center will specify an agent, a time, and a world. So, for instance, take the famous Sleeping Beauty problem. Beauty is put to sleep on Sunday. A coin is flipped. If it comes up heads, she wakes up once—just on Monday. If it comes up tails, she wakes up once on Monday and once on Tuesday, each time with no memory of any previous days. The coin coming up tails begets two centers (one for each day she is awoken), while the coin coming up heads begets one.

SIA in a sentence: the relative probability that you’re some center doesn’t depend on the presence of other centers. Put more plainly, suppose you want to compare how likely it is that you’re each of two people at times relative to one another. SIA instructs you: only look at those two people at times. All the other people at times can be safely ignored.

Note that when I talk about centers here, I’m not requiring that we’re given some maximally precise specification of the world. SIA tells you about the world even when you don’t know everything true about it. The word “center” is compatible with any description of agents or the world, as I’m using it—it doesn’t require that the description is maximally precise. If you don’t like this looser use of the word center, feel free to sub in your own word (like partial center).

The story is the same for times. If you want to figure out the relative odds you’re awake Monday vs. Tuesday, SIA instructs you to ignore Wednesday. SIA can be used to figure out the odds it’s Monday, or the first half of Monday, or the block containing Monday and Tuesday. It doesn’t just apply to arbitrarily small slices of time. It can apply to any length of time.

Once we understand what SIA says, we can see why SIA favors theories on which there are more people. If each person occupies a bit of the probability space—despite their many compatriots—then theories on which there are more total centers will get more probability. The more centers there are, the more probability they’ll take up.

For example, suppose a coin is flipped. If it comes up heads, person 1 is created. If it comes up tails, person 2 is created and person 3 is created. You’re created but you don’t know which person you are. Following SIA, you reason:

1. The odds I’m person 1=the odds I’m person 2. After all, when comparing these relative odds, SIA says: ignore person 3. Well, if person 3 didn’t exist, then clearly it would be just as likely that you’re person 1 as person 2.

2. The odds I’m person 1=the odds I’m person 3. Once again, when comparing these relative odds, SIA says: ignore person 2. Well, if person 2 didn’t exist, then clearly it would be just as likely that you’re person 1 as person 3.

From these it follows that tails is twice as likely as heads. If you’re person 2 or 3, then the coin definitely came up tails. If you’re person 1, it definitely came up heads. All three of those events are equally probable. Two of the events entail that the coin came up tails. Thus, tails is twice as likely as heads.

Now, we’ve so far talked about “centers” rather than people. Here is why: SIA isn’t about people. It’s about centers—slices of people at times. To see this, consider the Sleeping Beauty problem. A coin is flipped Sunday. If it comes up heads, you wake up once on Monday. If it comes up tails, you wake up once on Monday, then your memory is wiped, and you’re woken up a second time on Tuesday. Put another way, if the coin comes up heads, you wake up once with no memories of prior days. If it comes up tails, you wake up twice with no memories of prior days. After waking up with no memories of prior days, how likely should you think it is that the coin came up heads?

Applying SIA, we reason:

1. The probability I’m presently the first wake up and the coin came up heads=the probability I’m presently the first wake up and the coin came up tails. After all, if we were to ignore the second wakeup that would happen if the coin came up tails, then obviously heads and tails would be equally likely. SIA instructs you: do that, ignore other centers, only consider the two centers you’re comparing when assessing their relative odds.

2. The probability I’m presently the first wake up and the coin came up heads=the probability I’m presently the second wake up and the coin came up tails. Same idea. If the tails coin flip only woke you up on Tuesday, then obviously heads and tails would be equally likely.

Thus, tails is twice as likely as heads. Each of the following events is equally probable: the coin came up heads and it’s day one, the coin came up tails and it’s day one, and the coin came up tails and it’s day two.

Let’s apply this same logic to infinity. A 100-sided die is rolled. If it comes up 1-99, one person is created. If it comes up 100, infinity people are created. Following SIA, we reason:

1. The probability I’m the first person and one person was created is 99 times greater than the probability I’m the first person and infinity people were created. After all, this is what the relative probability would be if we ignored all the other people.

2. The probability I’m the first person and one person was created is 99 times greater than the probability I’m the second person and infinity people were created. Same idea.

3. The probability I’m the first person and one person was created is 99 times greater than the probability I’m the third person and infinity people were created.

…through ∞.

Thus, we have infinity events each 1/99th as likely as the die having come up 1-99 and me being the first person. Because there are infinity of them, they gobble up all the probability space. The probability of the die coming up 1-99 falls to zero. This is not, of course, to suggest that an actual agent should be infinitely certain that there are infinite people. You should think there’s some chance that you’re mistaken about anthropics. But on the most plausible theory about how to do anthropic reasoning, your existence gives you infinitely strong evidence that there are infinite centers.

Thus, you can derive SIA from a very modest principle about how to assign probabilities. You don’t need to think that we’re all souls drawn from some great cosmic soul bank. That has never been what SIA presupposed. Similarly, questions about personal identity and origin essentialism are beside the point—there are people that exist at times, and SIA tells you how to assign probabilities to you being each.

Once one sees that SIA follows from this principle, this gives powerful reason to think SIA is right. When analyzing the relative odds that you’re each of two centers, why would the presence of other centers matter? It makes no difference to which one you are. Why would the relative odds I’m Bob vs James depend on whether or not Fred exists? Fred’s existence can make no difference to whether I’m Bob vs James.

It is an advantage of SIA that it falls out of some principled probabilistic principle, rather than arising solely from a gerrymandered attempt to fit our anthropic intuitions. SIA’s rivals are not this way. SSA, the most plausible rival of SIA, depends on positing a totally ad hoc reference class with no principled basis. Compartmentalized conditionalization has a principled—if very metaphysically contentious—basis, but it implies near certainty that you’re a Boltzmann brain and other absurdities. Minimal reference class SSA inherits the same defects as CC.

We can also see, using this method, why SIA gives you strong evidence that there are infinite people. Suppose there are two theories of physics. One of them says there are infinite people. The other says there are finite people. Let’s say they have equal prior probability. Because the infinite theory holds that there are infinite centers that you might presently be, it gobbles up all of the probability—leaving nothing left for the finite theory.

At this point, there are a few remaining questions you might have.

First, does it make any sense to talk about the probability that the present you is one person vs. another person? Yes (and see here for my longer piece about it). Suppose there are seven copies of me created. Six will be tortured one hour from now. The seventh will not. One can perfectly sensibly wonder about the probability that I will be tortured in an hour. One can assign probabilities to such a thing. The probability here ought to be 6/7.

Second, here I’ve talked about relative probabilities. But SIA is supposed to give you absolute probabilities as well. Absolute probabilities are obtained by normalizing the relative probabilities so that they sum to one and then deriving the probabilities of the individual events. For example, in Sleeping Beauty, you know that tails is twice as likely as heads. The only possible events are heads and tails. Tails thus gets 2/3 of the overall probability mass, heads gets 1/3. This process is called renormalization.

Third, does SIA run afoul of the constraint that you can’t assign a uniform probability distribution over an infinite set? No more than anyone else. Following SIA, you can sensibly hold that, in the above case, the probability that you’re any particular person is zero. You can also hold that the probability is infinitesimal (though you’ll probably have to give up countable additivity). But even if you have no probability directed at being any particular person, 100% of your probability mass still can—and should—be directed at you being one of the infinite people created by the tails coin flip.

Everyone will have to assign probabilities in some way when there are infinite people, or else admit that, if the physicists who suggest the cosmos is infinite are correct, we’re in a skeptical scenario. However the non-SIAer does this, the SIAer can as well.

The consideration I cite in this article strikes me as reasonably forceful. It may not be enough on its own to tell you which anthropic theory to adopt, but it at least illustrates one major reason to find SIA plausible. Contrary to what people often say, SIA depends on neither bizarre metaphysics nor tendentious theories of personal identity. It merely depends on the notion that the relative odds that you’re each of two person-moments doesn’t depend on other person moments. Surely that should be the default assumption.

Then, when one thinks more about anthropics and notices that alternative views imply myriad independent bizarre absurdities, they should simply think that SIA is the right view.

Some very strange drama has been brewing recently on philosophy Substack involving me. Two charges have been leveled against me. The second charge is funnier. But “first, I have to reply to the older charges and to my first accusers, and then I will go to the later ones. For I have had many accusers, who accused me of old, and their false charges have continued during many years… But far more dangerous are these, who began when you were children, and took possession of your minds with their falsehoods, telling of one [Bentham’s Bulldog], a wise man, who speculated about the heaven above, and searched into the earth beneath, and made the worse appear the better cause.”

It began with a post on the anthropic argument made by a fellow named . The post was pretty long but foundationally misunderstood what the argument actually says. It claimed that using SIA to update in favor of many observers depended on reference classes (false) and that these arguments are crucially ambiguous about what kinds of entities you should update in favor of there being more of (false).

I penned a response pointing out some of the things he got wrong. Stan then replied to me in a very odd way. He doubled down and repeated his errors, but he also accused me of carrying out a bizarre multi-year-long campaign of pathologically lying about my beliefs.

Because in the end, does Matthew actually believe that God exists and Lords over a multiverse with a tally of denizens beyond reckoning? Because he feels rationally compelled after combining questionable ethics assumptions with ambiguous identity criteria and Bayes? Come on. The very first prerequisite of good faith discourse is that people really mean what they’re saying. And it just never passed the sniff test for me. From the Twitter days to our more recent disengagement (which I’m eager to resume), it’s been my sense that Matthew is literally smarter than that.

Here is Stan’s theory of my behavior. I don’t really believe in God or that there’s a big multiverse. Instead, I am just pretending to in order to get attention on the internet. Let’s note some curious things about this theory.

I’ve believed in God for about two years. I have told all of my loved ones about this belief. Stan finds the idea that I believe what most people in the world believe so outlandish that I must be pretending in order to make blog posts about it on the internet. He believes this even though my pieces about theism tend to get fewer reads than my pieces about other subjects, my pieces about anthropics get way fewer reads, and most of my readers are atheists. I’ll confess: this doesn’t seem like an amazing theory

Another thing complicating the story: I give away a sizeable fraction of my paycheck (more than 10%) to effective charities, some of which help shrimp and insects. So on this picture, either I’m so committed to the bit that I give away a bunch of my salary to maintain it or I really believe what I say about shrimp and insects—but I don’t believe what I say about God. Isn’t the more plausible hypothesis that I sometimes find counterintuitive arguments convincing?

And what is Stan’s amazingly powerful evidence for this position? What is the ironclad logic supporting Stan’s repeated histrionic public declarations about my insincerity? Well, you see, it is a feeling that he got when reading some of my old pieces. In particular, Stan and I many years ago had some exchanges in which I would stop eventually replying to him and would not change my position. He feels this is evidence that I was just pretending to believe what I claim to.

Here is another explanation: I didn’t think Stan’s objections were convincing. Note that I have on many occasions publicly admitted that my past arguments were wrong. It isn’t as if I obstinately refuse to ever admit that I made an error. Instead, I reserve that judgment for when I come to believe that I was wrong (a judgment that interactions with Stan didn’t produce). For example, in our most recent exchange, I didn’t change my view in response to Stan’s arguments because they depended crucially on not understanding the anthropic argument.

Stan thinks my arguments—especially the anthropic argument—are so bad that they can only be the byproduct of lying and sophistry. Note that this judgment is not shared by many eminent philosophers, including the ones I am writing a book alongside, the two reviewers who approved my paper on the topic, and many others. So Stan’s evidence is that Stan is personally unpersuaded by some of my arguments.

Thus, ladies and gentlemen of the jury, I plead not guilty.

2 Absolutely fucking wrecked

We then turn to part two of my trial, this bit much more insane; it’s also much funnier. The second prosecutor delivers his opening arguments, these ones rather more laced with profanity and rage than is typical. How you have felt, O men of Substack, at hearing the speeches of my accusers, I cannot tell; but I know that their persuasive words almost made me forget who I was - such was the effect of them; and yet they have hardly spoken a word of truth.

But in the accusations from my second accusers, there is a twist: I am accused not just of making the weaker argument into the stronger and of corrupting the youth, but of fleeing from serious opponents who could expose my sophistic ways. Lance Bush, my second accuser, alleges that I flee from serious people, and—never one to display undue modesty—can think of no better example of such a person than himself.

Lance is a lecturer in psychology at Cornell University. His blog is mostly dedicated to criticizing moral realism. Lance finds much of the language that moral realists use unintelligible, thinking that the idea of a stance-independent reason is gibberish. While Lance has a relatively small number of core arguments, he seems to have an endless desire to keep rehashing those core arguments each time anyone on the internet expresses support for moral realism. So, for example, he has a bunch of posts providing thousands of words responding to 250-character Tweets.

Well, it turns out that Lance doesn’t much like me (oh well, I try so hard to be his friend - and maybe some day that will happen). He chimed in after #Pattongate to complain about an incident several years ago that upset him greatly. During this incident, I cited a study by two university professors of mine and a third author. This study indicated that philosophers have pretty accurate intuitions about which philosophical judgments are widely held.

Lance told me in a Facebook comment that he didn’t think the study was convincing. I was curious, so I asked him why. I wasn’t expecting a long response. He apparently took around two hours writing out a very long reply. I remember thinking the reply had some force, but the study was still decent evidence. I didn’t think that would be a very interesting comment, and I’m a reasonably busy guy, so I didn’t reply to his Facebook comment.

This, it turns out, was inexcusable. Like Hitler’s invasion of Poland. But worse!

Lance felt like given how much time he spent typing his Facebook comment, he was owed some kind of Lance acknowledgment. Years after the incident, he publicly lamented, “No acknowledgement, no concessions, nothing. He simply never addressed my response at all.” Apparently, he desired either some kind of obsequious public declaration that I had been majorly wrecked or at least acknowledgment of his Facebook comment.

I find this a bit crazy.

If someone asks me for my opinion and I give it, I don’t think I’m owed a reply. I wouldn’t, like, rage at them years later for not having responded to my internet comment. I get about thirty blog comments per day, so I can’t really respond to all of them. Though this slight occurred years ago, Lance has continually stewed on its outrageousness for over a year, saying in a recent comment:

This was one contributing factor in me developing a dislike for Matthew and my eventual conclusion that he is not someone that we should be taking very seriously. Over time, from my interactions with him and from witnessing his interactions with others, I believe Matthew disengages from at least some conversations when he’s cornered or would have to admit error or make concessions. Maybe this is all unconscious not in any way intentional; I’m not fond of accusing others of bad faith. But whatever his intentions or state of mind, it’s bad behavior.

So Lance’s theory seems to be that I avoid adversarial discussions with people who might force me to “admit error or make concessions,” and that I disengage when “cornered.” It is true that I reasonably often “disengage” from lengthy back-and-forths. This is a necessity when you get many Substack comments daily.

Lance elsewhere compares me to a boxer who fights against amateurs but runs from serious fighters. Setting aside the weird machismo of this analysis of internet philosophical disagreements, this is a very poor theory. I routinely have public debates with very smart people and subject-matter experts who I disagree with. I’ve even had two public discussions with Lance himself! In the last week, I’ve had two multi-hour debates, one of them with one of the people Lance alleges would devastate me and force me to make concessions.

I have had many people complain that I’m too argumentative. Never before have I gotten the curious complaint that I’m not argumentative enough. I even have done call-in shows where anyone can call in and pick a bone with something in one of my pieces. This is not the behavior of someone who avoids debates with formidable opponents.

In Lance’s public discord server, I gave the example of my several public debates with the esteemed James Fodor to illustrate my willingness to have adversarial debates. The fact that I’ve repeatedly debated one of the sharpest critics of the sorts of views I hold seems like pretty good evidence that I don’t run from sharp critics. In response, Lance began bragging about how he’d totally destroy me if we ever debated (something he now refuses to do) and he’d wreck me way more than Fodor. Imagine Lance was Israel, I was Egypt, and our debate was the Six-Day war—that’s how bad it would be:

Lance then congratulates himself, declaring “I absolutely fucking wrecked you in my blog posts on your views on moral realism.” For those curious, Lance is forty years old. He then followed it up with several gratuitous insults.

lol

At this point, the Lance that has pierced my armor keeps burrowing deeper, wounding me mortally. “Stop, he’s already dead,” reply local spectators, but Lance continues until justice is served. He coolly decides to insult me some more. This was helpful because after the previous comments, you’d have been pretty unsure what he thought of me, so it was helpful that he added some more clarification on this front.

Then he suggested that I only refrained from replying to his nearly book’s worth of writing replying to me, spread across over a dozen blog posts, because of how “completely fucking wrecked” I was. Now, I’ve written a bunch of posts replying to Michael Huemer. Huemer has never replied. Never once have I seen fit to suggest that he’s TERRIFIED of how TOTALLY DESTROYED he was by my articles.

Lance seems to imagine that I carefully read through the many tens of thousands of words he wrote criticizing my blog articles, saw I was DESTROYED, and then didn’t reply because I was scared. Here is what actually happened: I didn’t read his replies to me, because I don’t have infinite time to read long-winded internet commentary.

Lance then looked upon his apoplectic meltdown and saw that it was good. He felt very proud of his conduct. He felt it had cast him in a good light, making him look forceful but measured—a man of love, but also of justice. He decided to pin his strange crashout on his public discord server.

So here’s what’s left of Lance’s theory: I flee from debates with competent interlocutors. Strangely, I sometimes have debates with very competent interlocutors like James Fodor, Daniel Munoz, and Chris Meacham. But I only do this when I believe that I won’t get “completely fucking wrecked” by intellectual powerhouses like Lance Bush.

I’m terrified of what would happen if we debated (even though I’ve debated him multiple times and am still willing to debate him). If we debated, he would say, “I don’t find moral realism intuitive or understand what a stance-independent reason is,” and I would burst into tears and cry, “you have bested me and my silly realist ways—I have nothing to say against the penetrating power of your idiosyncratic bafflement at ordinary English sentences.”

Strangely, according to this theory, I sometimes even agree to debates with people like Lance who will “completely fucking wreck” me and my incompetent amateurish ways, but then get bored after considerable time spent wrangling over the wording of a resolution. There is no reason other than mortal terror at the extreme debating firepower of Mr. Bush that I would not indefinitely continue wrangling over a debate resolution.

If you are forty years old and take to bragging on the internet about how formidable of a debater you are and how other people are scared to debate you, I would suggest you take a deep breath, log off, and go for a walk. This behavior is unbecoming. While there have been times I have replied to people on Substack, often while raising objections that seem to me to be quite forceful, I have never felt the need to brag to them about how much I absolutely fucking wrecked them.

But did Lance really wreck me?

I thought I’d peruse his post replying to me to see if I was really as destroyed as has been claimed. Unfortunately, this poses a problem: there isn’t just one post. Lance has written around a dozen articles responding to me. One of these pieces was 17,000 words. While perhaps this is evidence of my cowardice, I won’t respond to that one. Instead, I’ll respond to his shorter piece summarizing the 17,000-word piece. Let’s see if I really was “absolutely fucking wrecked.”

3 Can anything give you a reason?

If you’re interested, you can peruse my original piece arguing that we have stance-independent reasons. A stance-independent reason is a reason to do something that you don’t want to do. So, for example, I make the crazy, audacious, and perhaps even unintelligible suggestion that even if you want to load a baby into a cannon and fire it across a stadium, you shouldn’t do that.

Here’s the basic argument I give in that piece for the existence of stance-independent reasons. Sometimes we have reasons to do things. For example, I have a reason not to take the pen next to me and jam it into my eye. But, I suggest, accounts on which all reasons are stance-dependent are wrong. If there are reasons, and not all reasons are stance-dependent, then there are stance-independent reasons.

Why reject views on which all reasons are stance-dependent? Four basic reasons:

1. Intuitively, it often seems like you have reason not to do things that you desire to do. For example, suppose I had an overriding desire to try to eat a car—even though I knew I’d derive no joy from doing so—or to douse myself in lighter fluid and light a match. Seems like I still shouldn’t do it. Thus, it seems we have reasons to do things that we don’t want to do.

2. It doesn’t seem like stances give you a reason at all. If I want to throw my pen across the room, even if doing so doesn’t benefit me at all, it doesn’t seem I have any reason to throw my pen across the room.

3. Accounts on which all reasons are stance dependent don’t make much sense. They can’t be true by definition, because one who says “I have a reason not to douse myself in lighter fluid, even if I want to,” is clearly not speaking improperly (even if they’re making some kind of error). So if the notion that all reasons come from desires isn’t a definitional truth, what could possibly make it true?

4. On this view, reasons come from our aims, but we have no reason to have the aims in the first place. But it seems you can only have reason to follow an aim if you have reason to have the aim in the first place. You only have reason to buy a plane ticket to get to Paris if you have reason to go to Paris. So if the view implies we have no reason to have our ultimate aims, then it would imply we have no reason to do anything.

Lance replies:

On the one hand, Bentham is correct: there are no good reasons to endorse any of these antirealist accounts of reasons. Bentham also offers a solid objection to the reflective desire account: there may be no single, stable fact about what an ideal version of you would endorse, since there are “many possible idealization procedures,” which might lead to different outcomes depending on various factors, such as “the order on [sic] which you learn the facts.” I agree.

However, Bentham presents little by way of a positive argument for stance-independent facts “giving” us reasons, either. These positions are not mutually exclusive. Demonstrating that reasons aren’t given by our desires doesn’t show that they’re given to us by something other than our desires. There is a third alternative:

Nothing “gives” us reasons.

…

Analytic antirealist accounts propose that this statement means something like this:

Your desire to not be mauled by a tiger gives you a reason to avoid petting the tiger.

My own position, in contrast, is this:

It would not promote your values to pet the tiger.

We can jettison all talk of giving reasons. The word “giving” is not key to the argument. It’s giving—unnecessary. Instead just ask: what is the thing that makes a difference to whether you have a reason? Stance-dependent accounts say: something about your stances. Stance-independent accounts say: something independent of your stances.

Lance’s account doesn’t help address any of the four objections I gave to the notion that our stances make a difference to whether we have reasons. Concerning the first objection, it seems you might not have a reason to do something that you want to do. Ted Bundy had reason not to kill people even though killing people promoted his values. Similarly, even if throwing your pen across the room promotes your “values,” it doesn’t seem you have reason to do it, assuming it benefits no one.

(Note that for this argument values must be used in a broad sense to include whatever you care about—otherwise we get the result that one doesn’t have reason for most ordinary behavior).

This is not a plausible semantic account. If Lance were right, then “I have a reason not to X” means “X doesn’t promote my values.” But it seems one can perfectly coherently say “I might sometimes have reason to do stuff that doesn’t promote my values (if, say, my values are wrong).” This wouldn’t be a nonsense English sentence. But if the phrase “a reason” simply referred to what promotes your values, that would be nonsense, like saying “sometimes what promotes my values doesn’t promote my values.” If you want to read more about this kind of argument, see Moore’s open-question argument.

Hmm, I’m noticing that the complete fucking wrecking I was promised hasn’t been occurring.

Bentham follows this claim about the datum by repeatedly maintaining that antirealist accounts fail because they are inconsistent with Bentham’s personal intuitions. Bentham’s personal intuitions do not provide a strong rationale for rejecting antirealist accounts of reasons.

But these aren’t weird idiosyncratic intuitions that only I have. They’re things like “even if you want to light yourself on fire or try to eat a car, you have reason not to do so.” If one doesn’t have these intuitions, then fair enough, they shouldn’t be moved by this particular argument. But if they do, then they should abandon anti-realist accounts inconsistent with it.

Finally, Bentham repeatedly refers to the kinds of reasons he and other realists endorse as “genuine” reasons, implying that the sorts of reasons antirealists believe in are somehow not “genuine.” This frames the dispute between realists and antirealists in a way that tendentiously labels the rival positions in a way favorable to realism. After all, why should antirealists grant that realist reasons are “genuine”, but antirealist reasons somehow aren’t?

When I used the phrase “genuine reason,” I meant “something that is actually a reason.” This phrase appeared in sentences like “So long as you can coherently ask whether one has genuine reason to do what they want, it can’t be an analytic truth that one has reason to do what they want.” So no, I wasn’t trying to define only realist reasons as genuine.

So, in short, none of my objections were satisfactorily answered. While it is possible that the longer piece answers them in more detail (having briefly skimmed a few parts of it, I think not), I do not think that I was COMPLETELY DESTROYED, contrary to Lance’s claim.

4 Conclusion

When you disagree with someone, and write a response, it can often feel like they’re being deliberately obtuse if they don’t agree with you. It can be easy to assume that they read your piece, secretly know you’re right, but refuse to admit it. But this assumption is usually in error. Mostly they don’t agree with you.

I’m not above minor internet crashouts. Sometimes I get pretty annoyed when people make dumb errors about anthropic reasoning, for example. I’m often less charitable and gracious on the internet than I should be. So I can understand it. But let’s all chill and try to be a bit more charitable. Incandescent rage is less productive than constructive dialogue.

People of Substack, “if those among you who are said to be superior in wisdom and courage, and any other virtue, demean themselves in this way, how shameful is their conduct! I have seen men of reputation, when they have been condemned, behaving in the strangest manner…And I say that these things ought not to be done by those of us who are of reputation; and if they are done, you ought not to permit them; you ought rather to show that you are more inclined to condemn, not the man who is quiet, but the man who gets up a doleful scene, and makes the city ridiculous.”

If you find yourself apoplectically raging, yelling on the internet about how much you destroyed someone in your posts on substack dot com, boasting about how you’re a big dog in debates on the internet, or accusing people without evidence of running a bizarre multi-year-long campaign of pathological dishonesty, you should pause, take a breath, and touch some grass.

Below I’ll also describe one version of the Dutch book that I had in mind in the video. The basic intuition you should have about this kind of thing is that because the view leaves your credences sensitive to things that don’t affect the payout structure—having to do with, say, the number of geese in the far future—by changing these things around we can deliberately influence your betting behavior so that you predictably lose. Cold Button Issues’ view is that you should treat the present moment as randomly drawn from all moments of conscious experience (SSSA with a universal reference class).

There are two views of how to bet in these anthropic scenarios: one is that you bet at twice the odds that actually reflect your credence, because doing so correlates with how you’ll bet on the other day. The idea here is that you know that if you bet some way on one day, you’ll bet the same way the other day, so you think of yourself as voting on how you bet for both days.

Here, the Dutch book is straightforward. On Sunday, you’re given a bet where you get 90 cents if heads and you lose 80 cents if tails.

Then, each day after you wake up, you’re given a bet. If the coin came up heads, you lose $1. If it came up tails, you get 30 cents. Here imagine there are a ton of other people in the world, so you have thirder credences (on Cold Button Issues’ view). Because you’re a thirder, you bet at twice the odds you actually believe, and you believe the probability is 2/3. So you bet as if tails was 4x likelier than heads. You think “2/3 probability the coin came up tails. If the coin came up tails, then if I bet on tails today, I’ll also bet on tails the other day. I’ll get 30 cents each day—60 cents in total. 1/3 odds I lose $1, so in expectation I win from the bet.”

If the coin comes up heads, you win your Sunday bet and gain 90 cents. But then you lose your wake-up bet, so you’re down a dollar. In total, you lose ten cents. If the coin comes up tails, you lose 80 cents on Sunday, get back 30 cents both days, and are down 20 cents. So overall, you’re guaranteed to lose—either 20 cents or 10.

What’s the Dutch book if you don’t think you should bet at twice your credence on grounds that your actions across days are correlated?

Imagine that you’re the only person in the world. Thus, the view recommends halfer credences.

On Sunday (let’s imagine the Sunday bet is arbitrarily short for simplicity, though it can work even if it’s longer), you are offered a bet: if heads, you lose 60 cents, if tails, you gain 70 cents.

In the experiment, each wakeup you get the option to get 50 cents if heads and lose 40 cents if tails. You think heads and tails are equally likely, you bet at the odds you believe, so you take this bet.

Taking both bets, whether the coin comes up heads or tails, you lose ten cents. If heads, you lose 60 cents in the initial bet and only get back 50 in the experimental bet. If tails, you gain 70 cents in the initial bet, but lose 40 cents both days—so in total, you lose 80 and are down 10 cents.

I had Chat-GPT make a helpful chart to illustrate:

On August 5, in Amsterdam in the Netherlands, there’s a protest that could help shut down the world’s largest insect farm. It’s at Teleportboulevard 105 1043 EJ Amsterdam, Netherlands. Read more at this link. There’s also a zoom call on Wednesday 29th July at 12:30pm (CEST) with more info. This is a super important protest, so if you’re in the Netherlands, please show up!

This single farm, Protix, raises tens of billions of farmed insects. Shutting down this farm could affect many hundreds of billions of insects in total. The life of a farmed insect is a nightmare: the adults are not fed and left to starve to death; the juveniles are shredded, boiled, or microwaved; they’re overcrowded so that disease spreads rapidly and natural behaviors are thwarted. There aren’t any welfare regulations for insects.

If larger animals were being treated this way, we’d all recognize it as a horror. But the fact that insects are small shouldn’t make much difference to how bad their mistreatment is. So long as they can suffer, we shouldn’t cause extreme suffering to trillions of them for no reason. It is pretty plausible that black soldier flies (the kind Protix farms) feel pain. Amos Wollen nicely summarized the evidence:

To date, the best and biggest literature review is Gibbons et al.’s 2022 paper, “Can insects feel pain? A review of the neural and behavioural evidence”, which reviews over 350 articles. Gibbons and her co-conspirators considered the evidence for insect sentience along eight criteria: nociception, sensory integration, integrated nociception, analgesia, motivational trade-offs, flexible self-protection, associative learning, and analgesia preference.

While there was, in many cases, a paucity of evidence—shame, shame!—there was, apparently, no good evidence that any insect failed on any of the eight criteria.

Protix is strapped for cash, but they might get bailed out by their current investors like InvestNL—an investment fund dependent on taxpayer funding. If you’re a Dutch taxpayer, your tax money might facilitate the painful slaugter of billions of insects. This protest would gently encourage them not to bail out Protix :).

Investment also makes little financial sense. The industry claimed it would transform the food system and has failed completely. Companies are imploding left and right, but investors are still investing. There’s just no viable path forward: farmed insects are more expensive, worse for the environment, and pose greater pathogenic risks than alternatives. A Dutch newspaper summarizes the dismal state of the industry in a piece titled How Investors Completely Choked on Edible Insects. And Protix has failed to answer basic questions about how their product could become remotely viable.

Even if you don’t think insects matter much morally, we shouldn’t harm them by the billions without a good reason. The insect farming industry has no good reason to exist. It has failed at delivering each one of its foundational promises. There is no reason to keep sacrificing billions of insects to provide exorbitantly priced environmentally harmful petfood and food for farmed fish.

If we assume that each person at the protest adds an extra .01% chance of shutting down Protix, that Protix’s continued operation will lead to an extra 300 billion farmed insects, then each protestor benefits an expected 30 million insects. It’s not every day that you have the option to benefit a population 10% as large as America! If you’re in the Netherlands, Belgium, or West Germany, go to the protest and take a friend. If you’re not, send this to your friends in those areas. This really, really matters.

In justifying starving black soldier flies to death, the industry often claims that they don’t have mouths to eat. This is false, like much of what the industry says. But though they have mouths, they cannot speak—there is no one to represent the starved, crushed, boiled, and ground-up black soldier flies who rot in overcrowded indoor facilities. Join the protest to stop this senseless cruelty and protect those who have no voice.

When I was becoming a theist, this was the main consideration that pulled me away from belief in God. God explained so much about the world, yet explained so little about evil. The atheist, it seemed to me, was left with myriad unsolvable mysteries; the theist had just one terrible mystery, present in each of the world’s sorrows, each time two loved ones must say goodbye for the final time (in this life, at least).

In this piece, I’m not going to give a comprehensive explanation of why God allows evil. If you want my preferred solution, you can read it here. Instead, I’m just going to make some points that I found helpful for illustrating that there might be a solution to the problem of evil. Ideally I want to convince you that it’s not obviously totally impossible that God has some reason to allow evil, the way I used to think it was. My suggestions will undercut the evidential force of the problem even without being a complete solution.

A child might not understand precisely why their parent brings them to undergo a surgery. But they might nevertheless have good reason to think their parent has such a reason. So too for God.

1 Minor evils

The problem of evil generally focuses on really big evils: genocide, slavery, rape, torture, etc. But there are many more bad things in the world than these. People routinely stub their toes, feel mild irritation, and prick their fingers. While the problem of horrendous evils is generally seen as a horrendous problem, the problem of trivial evils is seen as comparatively trivial. If all the evils in the world were small and minor, that wouldn’t automatically solve the problem of evil, but it would make it look a lot less impossible to solve. Evil would remain some kind of intellectual problem, but it would no longer possess the same crushing existential weight.

But note that if theism is true, then the evils in the world really are small and trivial compared to the goods of the afterlife. All Earthly suffering is a comparatively insignificant speck next to heaven. One second of heaven more than outweighs all the suffering experienced so far in human history by infinity orders of magnitude. If Earthly evils were a pale shadow, heaven would be a supernova. If suffering all your Earthly sorrows would make a single second of heaven 1% sweeter, it would be worth it.

God has infinite power. Insofar as he is perfectly good and cares for us, we should expect the goods he’ll provide in the life to come to be infinitely large. So if theism is true, compared to the goodness of what is to come, all evils are minor. This is not to downplay the horrific things that happen on Earth. It is merely to note that they are nothing compared to God’s infinite goodness. However dark and painful this life is, a single second of heaven is more than enough to compensate it. Everyone who suffers, from animals to children, will experience joy too boundless to comprehend.

Now, I don’t think this single-handedly defangs the problem of evil. The problem of minor evils is still a big problem. In fact, I find it a lot more serious than most people do. It’s hard to see how God would have a reason for allowing so many minor evils—many of which seem not to serve any greater purpose. But when one fixates on how good the afterlife may be, the problem of evil begins to look not much more serious than the problem of minor evils.

2 Unknown reasons

It’s hard to know why God allows evil. But oftentimes people much smarter than me do things that I don’t understand. I generally don’t know why Magnus Carlsen makes the chess moves that he does or why Terrence Tao includes a particular calculation in his papers. Nonetheless, I trust that they have good reasons for what they do. Insofar as there is good reason to think God exists, then we should similarly trust him. The gulf between us and God is infinitely larger than the gulf between us and merely very skilled humans.

Now, I am not saying that we are totally in the dark about what God would do. I am not saying—nor do I think—that we haven’t the faintest clue what God would do, so no features of the world are ever evidence against God. I don’t buy this skeptical theist claim. I’m merely saying that God knows many things we don’t, so we should be wary about confidently concluding from the fact that we can’t see a reason for God to permit evil that he has no such reason.

Sometimes we do think that a person’s behavior is good evidence against their expertise. If someone claimed to be a good chess player, and then began making obviously terrible moves, we would begin to doubt them. So should we do the same for God? Yes, to some degree. Evil is evidence against the existence of God. But it is not totally decisive evidence on its own.

It’s hard to be extremely confident from our vantage point that some feature of the world has no ultimate justification. We peer through an arbitrarily small window into ultimate reality. So it is less like we witness repeated obviously terrible moves, and more like we witness moves that seem bad to us in a game with rules we don’t understand, where we can’t see most of the gameboard, and we’re on move two out of infinity.

3 Eternal benefits

If God exists and cares about us, he has no need to capriciously snuff us out after a few decades. He can keep us around forever, experiencing infinite joy for infinite time. Thus, insofar as the evils do anything to enhance relationships we have with others stretched out across an infinite period of time, they pay off infinitely. Alexander Pruss put it well (though I find the problem far more compelling than he does):

I want to confess something: I do not find the problem of evil compelling. I think to myself: Here, during the blink of an eye, there are horrendous things happening. But there is infinitely long life afterwards if God exists. For all we know, the horrendous things are just a blip in these infinitely long lives. And it just doesn’t seem hard to think that over an infinite future that initial blip could be justified, redeemed, defeated, compensated for with moral adequacy, sublated, etc.

It sounds insensitive to talk of the horrors that people live through as a blip. But a hundred years really is the blink of an eye in the face of eternity.

When you vividly contemplate the fact that there are infinity seconds in the afterlife for every second of earthly life, theodicies begin to seem markedly less absurd. The goods standardly appealed to by theodicies—soul building, connection building, deepening our relationship with God in some way—last forever. In their infinitude, they’re more than enough to outweigh earthly evils.

4 God might have a reason to allow the world to be ruined

The world has many different evils, many of which seem much more likely given an indifferent world than one with a God who cares about us. Consider the suffering of the deer in the forest fire. It is not plausible that some great good accrues from the deer experiencing the pain of being burned to death. It would be very surprising if the semirandom mix of evils produced by evolution happened to be optimal for our spiritual flourishing.

So if theists were committed to the idea that every one of the world’s evils leads to greater goods, then theism would be very implausible indeed. It would also be morally dubious. If the world is better because it has malaria, why bother treating malaria? One should not ordinarily rid the world of good things.

Fortunately, theists are not committed to this. What they should instead think is that God had reason to allow the world to be ruined. One version of this view, though far from the most plausible one, is the Christian doctrine of the fall—that God had reason to permit early humans to plunge the world into blazing indifference if they so choose. Alternative versions attribute the world’s indifferent state to omissions from powerful non-human entities, misconduct from simulators, and more.

Any account like this will require positing somewhat surprising things for which we do not have decisive independent evidence. So theism will pay a cost in probability. But it is cheaper for theism to pay a cost once for all evils than a different time for each evil. It is more likely that God has some good reason to allow the world to be screwed up than that there is some separate reason to permit each of the world’s evils.

Those pressing the problem of evil often have a tendency to treat evils as probabilistically independent. But they’re not. If God has a reason to allow a world with cancer, torture, flesh-eating parasites, and evolutionary suffering, that at least raises the odds that he has a good reason to permit lupus. To explain a few evils, one must posit that God has a reason to allow a world to become screwed up, which will then explain the remaining evils.

5 Bias

There are philosophically potent arguments for both theism and atheism. Yet the arguments for theism tend, in general, to be pretty abstract. While you might be convinced by the argument from psychophysical harmony or fine-tuning, it doesn’t have any emotional or existential pull. The arguments for theism are largely cool intellectual puzzles; the arguments for atheism make us want to shake our fists at God in rage.

You should expect yourself to have some bias to overrate arguments that have more emotional force. Because the problem of evil is an emotional problem and not just an intellectual one, we are liable to overrate its intellectual force. Our emotions affect our perception of arguments more than we like to think.

Now, I am not suggesting that the problem of evil is only an emotional problem and that it lacks intellectual force. I don’t believe that, and I hate when apologists say it! My claim is simply that we should expect ourselves to overrate the force of the problem of evil relative to other arguments that are similarly evidentially compelling but lack the same kind of emotional payload.

When I reflected, I noticed this in myself. Even when, say, the Anthropic argument strikes me as about as compelling as the problem of evil, it doesn’t pull me psychologically in the same way. But when there’s a judgment that seems good to you rationally, you shouldn’t reject it just because your emotional state can’t get on board.

6 Conclusion

Mill famously titled one of his chapters in Utilitarianism “Of What Sort of Proof the Principle of Utility Is Susceptible.” Think of this essay in a similar spirit. I have not given a solution to the problem of evil. I have merely aimed to argue that it’s not obviously impossible that there is a solution. Explaining evil given theism isn’t obviously more difficult than explaining psychophysical harmony or fine-tuning given atheism.

When one considers the eternity to come, the infinite goods available to God, and all the things God knows that we don’t, it seems like he just might have a reason to permit evil for a short period of time. Eventually, every tear will be wiped away, everything sad made untrue.

The models were being tested for their ability to solve the problems on a test called ExploitGym. ExploitGym is designed to test offensive cyber capabilities. Tests of this are supposed to happen in a controlled environment where models don’t have internet access. The models figured out a way to gain internet access.

Once they had internet access, they learned that the company Hugging Face had the solutions to the ExploitGym questions. So they hacked Hugging Face to get the answers. Now, on the one hand, if you’re being tested on your offensive cyber capabilities and you hack a company to get the answers, it seems like you passed the test. On the other hand, this is pretty alarming.

Those concerned about AI risk have generally worried that AI will be given goals that it pursues in dangerous and unpredictable ways, leading to serious harm to humans. The more ambitious its goals get, the more seriously it might harm humans in pursuit of its goals. The fact that AI was willing to hack to score well on a test is evidence for this threat model.

It also illustrates the amazing hacking abilities of AI. The fact that the models were able to break free, access the internet, and hack a third party indicates that these kinds of models have sufficiently adept cyber capabilities that they’re hard to control. And capabilities are only likely to improve.

Lastly, it shows that containing models is difficult. Just as the escape of a pathogen from a lab is evidence that containment is hard, an AI model going rogue provides evidence that containing them is hard. The models a year from now will be far better than the models of today. If the models of today are already breaking free, we should worry about the models of the future.

There are also three features of this case that temper how alarmed we should be.

First, it’s not clear to what extent this is an alignment failure. OpenAI was trying to test offensive cyber capabilities. Though OpenAI has been a bit cagey, it seems like the models they were testing didn’t have the standard features that prevent models from hacking. They say in their blog post that the models in question had “reduced cyber refusals for evaluation purposes,” adding:

This week, we published a blog on improving safety and alignment in an era of long horizon models⁠. These deployment safeguards were intentionally not enabled during this evaluation because it was aimed at testing cyber vulnerabilities.

So if you take the models that are best at hacking, test their hacking abilities, don’t add ethical safeguards that make them not hack, and switch off their cyber refusals with regard to hacking, you shouldn’t be that shocked that they do some hacking. This is a bit like selectively breeding dogs to be aggressive, and then noting that one broke free and bit someone. It doesn’t tell you much about the dogs of the real world.

It’s like testing a genetically engineered hypervirulent flu strain that briefly escapes but is contained. It’s alarming that this happened at all. But it doesn’t give you much evidence about how ordinary flu strains will behave. On the other hand, we should worry that the models we’re training are a few safeguards away from dropping all moral constraints and nefariously optimizing for the goals they’re given.

A second bit of good news: the cyberattacks were detected twice over, both by OpenAI and Hugging Face. At this point, it seems that rogue models are not able to silently hack without leaving traces. That, at least, is good news.

Third, when the models went rogue, they did so in pursuit of the task they were given. So while this is evidence for the threat model “we’ll give AIs bad tasks, they’ll pursue them in bad ways and cause lots of harm,” it’s not direct evidence for the threat model “AI will have unpredictable goals divorced from the ones we tried to give them and ruthlessly optimize.”

My overall takeaway? This provides real but modest evidence that we should be concerned about AI. It illustrates that the worrying scenarios that AI doomers have warned about can really happen. And as capabilities advance, things will only get crazier. The reason this isn’t more evidence is that we already knew that this sort of thing could happen.

AI will continue its steady march toward greater and greater power. Its ability to hack will grow. Its ability to invent new and dangerous tech will grow. Those who control AI will have an increasing share of global power. Control over world affairs may be concentrated into a small number of hands. While it may unleash an age of unprecedented wealth, welfare, and abundance, it poses many serious risks.

You wouldn’t allow a private actor to build a powerful bomb without facing any regulations. But AI is quickly becoming more powerful than a bomb and poses far more of an existential threat. It seems clear that AI should be regulated more than it currently is being.

We don’t normally leave it up to private companies to ensure that they don’t impose serious externalities on others. As Adam Smith said, “It is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their own interest.” Similarly, in the ideal regulatory regime, we would not rely on AI companies voluntarily behaving altruistically.

To take one serious example, it’s entirely up to the AI companies how they affect model welfare. Models probably aren’t conscious, but they might be, and it’s reasonably likely they will be soon. AI companies do not face direct competitive incentives to tend to their model’s welfare. If ChatGPT were conscious and suffering all the time, this wouldn’t much affect OpenAI’s bottom line.

So insofar as we apply the standard account economists give of how to deal with externalities, there is a strong case for AI regulation.1 We are building a technology that already has the power to hack and assist with various nefarious schemes. Soon it will be able to develop dangerous new weapons, pose serious existential risks, and potentially even suffer in great numbers. When private companies only directly bear a tiny fraction of the costs they are potentially imposing on others, they ought to be regulated.

I remain sort of moderate on the AI risk issue. I think AI poses serious threats and also has major upsides. If we want to navigate the risks and capture the upsides, we should have a law like the RAISE act nationally. Plan A also seems pretty good. As we learned from Chernobyl, a major and publicized disaster seriously retards industry progress. The recent hacking incident is a stark reminder of the risks posed by powerful AI.

Despite my youth, I’m old enough to remember when people were viciously mocking those who thought AI might be impressive near-term. Some of them are still doing it. The last few years have fit much better with the predictions of those who said AI was going to be a big deal than those who didn’t. If you notice someone got things right very early on, and you mocked them for it, then you should at least think carefully before mocking them for their next prediction.

This isn’t the only major piece of AI news. Just this week AI disproved the Jacobian conjecture—Fable taught us something new about math. In a world of AI models proving things that mathematicians couldn’t and going rogue, escaping, and hacking, ideally we’d have some mea culpas from the stochastic parrot crowd. At the very least, after all the AI boosters have gotten right, and all the stochastic parroters have gotten wrong, you’d think they would proceed with less brazen temerity. If you’ve been wrong repeatedly, at the very least you should refrain from mocking those who have been right.

1 Moral radicalism

The woke are moral radicals. By and large, they think that society is doing things that are seriously morally wrong and that blind deference to existing societal norms is a recipe for humongous disaster. I often disagree with their specific complaints about society. But I think that they’re correct that society makes serious moral errors, and these merit careful reflection and major correction.1

A priori, we should expect society to make serious moral errors because every past society uncontroversially has. Another reason to think we’re making serious moral errors is that there are giant torture farms in every developed country, and this is where most meat comes from; these cause more suffering in a few years than humans have experienced in history. Oh, and we also ignore the interests of, to within a rounding error, all sentient beings who will ever live.

Most moral errors have involved neglecting the weak and vulnerable. Throughout much of history, pillaging, conquering, and raping innocent people was seen as a reasonable response to a glorious military victory. The Romans left infants to freeze to death and crucified disobedient slaves. In America, just a few hundred years ago, black people were shipped across the ocean in nightmarish slave ships and then bought and sold.

Non-wokes will often acknowledge somewhat abstractly that we might make moral errors too. But they tend not to be very serious about this belief. The woke take very seriously that we may be like the Romans, Genghis Khan, or 1830s America, in the scale of our errors. Now, they won’t usually be that explicit about this, because comparing things to past atrocities (especially American and British ones) is often regarded as offensive in woke circles. But they at least take potential atrocities more seriously.

2 Fatphobia

The woke talk a great deal about how badly overweight people are treated. This is often dismissed by non-wokes. But I think the woke nail this. There really is somewhat serious stigma against those who are overweight.

We judge attractive people more positively than others (one reason I’m judged very positively). Thus, there is direct and strong bias against those who are seriously overweight. One study found:

Among the findings are that 28% of teachers in one study said that becoming obese is the worst thing that can happen to a person; 24% of nurses said that they are “repulsed” by obese persons; and, controlling for income and grades, parents provide less college support for their overweight than for their thin children.

Now, sometimes people go a bit too far with this and act like one’s weight is wholly out of their control. That isn’t right. But at the very least, the woke are right that those who are overweight are often treated pretty badly, and this is unfair, rather than deserved or merited by poor life choices.

3 Judging past historical figures

has a piece called The Best Thing About Wokeness that begins:

What’s the best thing about wokeness? Simple: Holding people in the past to universal moral standards. History is packed with mass murder, slavery, and other atrocities, often committed by famous beloved figures like Columbus and the American Founding Fathers. When confronted with these harsh realities, most thinkers try to weasel out with a variation on, “It was a very different time.”

He’s right. All too many people behave like cultural relativists when it comes to judging the misconduct of people in the past. The fact that one’s society approves of slavery does not, on its own, shield one from guilt. You shouldn’t do evil things even if your society is doing them.

The woke are either implicitly or explicitly cultural relativists when judging the behavior of societies not composed of Europeans. So, they’ll be very wary about saying that Aztec human sacrifice was immoral, even though it obviously was. But they at least jettison this kind of extreme relativism when it comes to judging our own society. Many others don’t.

The sensible position, of course, is to judge both our society and others based on universal moral principles. Evil remains evil no matter when or where it was done. Being in a society of people doing evil things mitigates guilt to some degree, but it does so incompletely.

4 Veganism

Not all woke people are vegan. But the woke are a lot likelier to be vegan than other people. This is a major point in their favor because standard meat consumption involves paying for animals to be tortured for comparatively slight benefit. And what’s more, the woke are disproportionately likely to see animals as an oppressed class. They even express solidarity with animals in their unique love language: writing things with titles like Ecofeminism: Feminist Intersections With Other Animals and the Earth or The Sexual Politics of Meat: A Feminist-Vegetarian Critical Theory.

Not always, of course. Sometimes woke people complain about “white veganism,” where I guess the inference is supposed to be that a view can be discredited if its proponents have white skin sometimes. Woke people also sometimes say that veganism is wrong because it’s a self-evident unshakeable Moorean fact that indigenous whale hunters do nothing wrong when they put a spear through a whale’s skull (not totally clear how this is supposed to vindicate the average Walmart shopper). But still, the woke are much likelier to be vegan or think they ought to be than the average member of the public.

A group having less predilection to pay for brutal torture on a regular basis than average is a very desirable feature.

5 Foreigners matter

In contemporary American life, it is standard to ignore those who reside outside the Western world—and to mostly ignore those who reside inside the Western world so long as they lie outside of America’s borders. The common view is that our reasons to save a person’s life dwindle to near-nonexistence provided the person doesn’t live here. But this is a clear mistake.

Where you were born doesn’t seem to make much difference to the importance of helping you—and thinking it does licenses actions which harm one person to benefit no one. Even if we had especially strong obligations to help Americans over foreigners, it’s hard to believe this would make a difference by many orders of magnitude. The interests of foreigners would still be very significant.

To their credit, the woke get this right. They spend lots of time thinking about harms to foreigners. While for most Americans, the welfare of those outside our borders is an afterthought, for the woke, it is typically front and center. Now, there is still a tendency among the woke to excessively focus on those we harm, and neglect those we might benefit. But overall, in the “concern for foreigners” department, the woke are batting above average.

6 Wokeness improved

The woke go right when they apply moral standards consistently, even outside society’s Overton window. Their hits come from having an expansive moral circle that cares about many different sentient beings. Now, their moral circle is still less expansive than it should be but more expansive than is typical.

They go wrong in neglecting effects. So they err when they excessively fixate on pointless servile speech acts, get mad at jokes, oppose beneficial actions that involve interacting with structures deemed oppressive, and fail to make comparisons between the values of different activities. An upgraded version of wokeness—without its myriad vices—would be concerned for the meek and oppressed, even when caring for them is outside the Overton window, but would combine that with a rigorous quantitative approach to making the world a better place.

Fortunately, there is such a movement. Around 2011, some Oxford philosophers had the idea of trying to do a lot of good instead of just a little bit. This was a revolutionary idea; no one had ever thought of trying to do more good rather than less before.2 The movement this started saved about 300,000 lives, spared hundreds of millions of animals from intense suffering, and majorly benefitted the world. It has all of the virtues of wokeness without any of the pernicious woke vices.

You didn't mention her most fascinating theories about how Charlie Kirk was a time traveler or had access to Sumerian-derived time travel technology owned by US the government.

…

I’m pretty sure she thinks it was an Israeli collaboration with France, who are angry about her coverage of Brigitte Macron. She also claims the Macrons have or had ordered her assassination.

added “I was also shocked he left out the occult ritual theory - I think that was the point when I realized this was either borderline psychosis or an extremely cynical grift and not non-pathological extreme credulity and stupidity.” Sorry guys, I don’t keep up with everything crazy Mrs. Owens thinks. Doing so would be a full-time job.

So, naturally, I watched the episode of her show where she lays out this view. Dear reader, this was, without a doubt, the single craziest video I have ever seen. Mark and Ethan, if anything, understate things. I have seen videos made by flat-earthers that were considerably saner, less prone to wild and speculative inferential leaps, and less outrageously ridiculous than Owens’ video.

Read more

By “God” I mean an uncaused being with limitless power, knowledge, and goodness. I think such a being has a high prior probability (probability before examining specific evidence), because this being is simply a mind without limits. Given that the being has no fundamental limits, nothing constrains it, so it is omnipotent, and nothing limits its awareness, so it’s omniscient. Then, because it has a perfect grasp of morality, and suffers from no rational defects, it always does what’s best. If you see that something is desirable and you suffer no rational defect or contrary psychological inclination, then on plausible views, you’ll be automatically disposed to want it. One with perfect rationality wants everything that’s worth wanting.

Given that minds are one of the fundamental things, a mind without limits is simple. It’s simply one of the fundamental elements of reality instantiated without limit. An infinitely large space is simpler and more intrinsically probable than a space that’s exactly 10 quintillion meters across. Likewise, a God who knows every fact is more intrinsically likely than one who knows every fact but one. Likewise, a mind totally without limit is far more intrinsically probable than any mind subject to some arbitrary constriction.

Even if you’re a bit skeptical of this account, it’s plausible enough that the prior probability of theism is not outrageously low. If you think that there’s a 10% chance that an entity is simple, and you think its prior probability is 1% if it is simple, then its prior will be 1/1,000. The evidence for theism is strong enough that it should overcome even a low prior.

A fact is evidence for theism if the odds of it are higher given theism than atheism. If your DNA is at the crime scene, that is evidence for your guilt, because the likelihood that your DNA would be at the crime scene is higher if you committed the crime than if you didn’t. It’s not sufficient evidence, but it is evidence.

Here are some things that are likelier given theism than atheism—that theism nicely explains.

1. The laws line up with the things that exist. There might instead have been laws of the form “all Xs do A,” but no Xs. By default, it’s very surprising that the laws line up with the things that exist, just like by default it would be surprising if a random rulebook to a game lined up with the pieces of the game, unless this correspondence was designed.

2. We have laws capable of giving rise to interesting structure. By default, almost all laws devolve to random chaos or simple and valueless patterns.

3. We have finely-tuned constants, so that they fall in a narrow range needed for the formation of complex structure.

4. Some physical structures give rise to consciousness.

5. The structures that give rise to consciousness exist. If As are the only things that produce consciousness, but there are no As, then there’d be no consciousness. By default, if there’s no God, out of all the possible ingredients for consciousness, it’s pretty surprising that they happen to be ones that arise.

6. Consciousness pairs up harmoniously with the physical world. Most conceivable pairings would lead to nothing of much value.

7. As your existence illustrates, there is a giant multiverse with absolutely infinitely many people. Put another way: your existence is N times likelier if there are N times more people, so you should think the number of people that exist is the very most that there could be—and this is likelier if God exists than if God doesn’t exist.

8. Humans possess knowledge of morality, mathematics, the modal facts, right priors, and other non-physical facts.

9. We’re not radically deceived in our most basic and fundamental beliefs about the world, the way we’d be if we were Boltzmann brains or brains in vats.

These all strike me as very good. Each of the arguments above seems on its own enough to majorly raise the likelihood of theism. This next list has somewhat weaker arguments, but they’re still pretty good. Some more things God explains better:

1. Fine-tuning for discoverability. Robin Collins has argued that the constants are suspiciously tuned for us to be able to discover them and do science. This one is very powerful evidence against atheism if the physics works out, but I don’t know if it does.

2. Belief in God is widespread (a priori, given atheism, I wouldn’t have expected almost everyone in the world to believe in God).

3. A bunch of pretty credible miracles seem to happen all the time, mostly in religious settings.

4. Widespread religious experience (just ask: a priori would the atheist have expected people to routinely think they were interacting with the perfect creator of the universe? No. Would theists? More likely, because the perfect creator of the universe loves them.)

5. Time (a static world would have had no value).

6. Anything whatsoever exists.

7. Laws of nature, rather than particles that are impotent and do nothing, like sleepy Joe.

8. Intentionality (some mental states are about other things, which is a pretty weird feature of our world. No purely physical thing is about any other physical thing).

9. Mental causation.

10. Reason.1

Now, some more miscellaneous features that theism maybe explains a bit better but don’t have a ton of force on their own:

1. Souls.

2. Love.

3. A very beautiful natural world.

4. Haecceities.

5. Language.

6. Scott Alexander’s post on John Rawls, the song What I Wouldn’t Do, and Mitch Hedberg’s joke, “I haven’t slept for 10 days... because that would be too long.” (could naturalism explain such amazingly good things?).

7. A physical world.

8. Design in biology (read the linked article for more; I’m not making any Behe-style non-credible scientific claims—nothing I’m saying conflicts with standard science).

9. The laws of nature being conducive to abiogenesis and complex chemistry more broadly.

10. The fortuitous correspondence between objective list goods and goods that are conducive to pleasure.

11. The fact that the universe is discoverable and investigable by science.

12. The infinite size of the universe, according to plausible theories of physics.2

13. Moral motivation.3

Three notable things about this case. First, there are a lot of different considerations. So if atheism keeps taking hit after hit from each extra fact, it’s in rough shape. Second, while the considerations aren’t entirely probabilistically independent, many are at least largely independent. For example, a solution to fine-tuning probably doesn’t explain consciousness.

Third, many of the considerations are quite forceful. Not all of them, certainly; the twelve non-jokey miscellaneous considerations at the end generally don’t have a ton of force. But each of the first nine strikes me as very forceful. In the days of my atheism, I felt like I had to say something pretty hand-wavy and improbable about each of them. It’s more likely that there’s a solution to the problem of evil than to all nine of those arguments, some of which seem individually about as forceful as the problem of evil. The next ten also strike me as pretty good.

If God exists, that is the best news in the universe. It means that there is a being of limitless knowledge, power, and goodness, who made all things. It means, with very high likelihood, that we will all spend forever in paradise with our loved ones—that reality is, on the whole, infinitely good. It means that when we do good things for other people, this might strengthen our connections with them for all time, so that our reasons to help are even stronger than they might otherwise be.

In the afterlife, there is no limit to the joys that a perfect God can bestow on us. We will experience joys so great that a single second of them would be worth all the misery so far in human history. Yet these will not end, nor lessen, nor slowly fade to oblivion. They shall continue, forever, and be experienced alongside those we love most. Each second will outshine the darkest earthly tortures more thoroughly than a supernova outshines a shadow.

Yet God’s existence also raises the likelihood that we will be on the hook for our malfeasance—that we may one be held accountable by God for our decisions. The existence of God makes our actions far more significant than they would otherwise be. It means our wrongdoing is much worse than it would otherwise be, and our right actions are much better.

While we should love God, we should also approach God with some measure of fear and trembling, as we would if our innermost life was on full display before the world’s most virtuous person, whose relationship with us was of infinite value. The bad things we do even though we know deep down that they’re wrong—failing to give our money to effective charities, prioritizing our own desires over doing what we know to be right, failing to carefully follow the evidence when it is inconvenient or conflicts with our preexisting judgments—are things God knows about, that we may one day have to answer for.

This is not a reason to think that God’s existence is bad news. It’s the opposite. God is so limitlessly perfect that anything that might strain our relationship with him, even slightly, is infinitely terrible. While I don’t think there’s an eternal hell, there might be a temporary hell—a period where we suffer for the purpose of our purification, akin to purgatory.

The world’s religions standardly hold to at least this, and it wouldn’t be surprising if the process of us becoming more virtuous and regretting our previous wicked ways was painful. Even if it’s not painful, so long as it merely takes time, it costs us infinitely, because every second with God is of infinite value. And while I expect the afterlife to be a place where we are made smarter, more virtuous, and more forgiving, the good things we do for each other in this life will resound forever.

There is good reason to think that God exists. This is the best possible news. But it is a reason not to be complacent and to act in the ways that we think a perfect being would approve of. For one day, we may come face to face with such a being, and there, our excuses will not carry much weight.

Here I want to argue for something surprising: in many of the plausible cases that are often touted as counterexamples to utilitarianism, non-utilitarian views agree with utilitarians (at least, in something nearing the actual world). So, for example, utilitarians favor killing one to save five, assuming all the myriad exotic stipulations ensuring the action won’t have untoward second-order effects. I think there’s a pretty nice and non-obvious proof that non-utilitarians should say this too. Same goes for special obligations—belief in special obligations shouldn’t affect real-world behavior.

(In a utilitarian spirit, I’ll donate the money I make through paywalling this article to effective charities).

Read more

If we achieve existential security, we will have room to breathe. With humanity’s longterm potential secured, we will be past the Precipice, free to contemplate the range of futures that lie open before us. And we will be able to take our time to reflect upon what we truly desire; upon which of these visions for humanity would be the best realization of our potential. We shall call this the Long Reflection.

—Toby Ord, “The Precipice.”

Future humans will make profoundly important decisions. On Earth today, our options are largely limited by biology and sparse resources. But if we don’t go extinct, we will have the ability to venture out across the cosmos and transform it from a mostly barren void to an intergalactic bliss-filled utopia.

In his book “The Precipice,” Toby Ord proposes that before we do this we carry out a Long Reflection. Before we transform the universe on a galactic scale, he proposes a long and protracted collective planning process, where we carefully think through how it should be transformed. If our actions will determine the fate of incredible numbers of conscious organisms, across a sextillion star systems, across a trillion years, then it is of literally astronomical importance that we not screw things up.

Now, Ord has explained that the Long Reflection wasn’t intended as a super exact proposal. Getting progress to freeze for 1,000 years as we sit around in armchairs and reflect isn’t tenable. But still, he thinks we should aim for something in the vicinity of a Long Reflection. So do I. We shouldn’t permit major transformation of the universe to happen through the accidental and unplanned decisions of private actors. We should reflect before spreading across the cosmos and transforming it.

And the good news is that something resembling a Long Reflection wouldn’t take that long. As my colleagues at Forethought argue in Preparing for the Intelligence Explosion, once AI can automate cognitive labor, enormous amounts of cognitive labor can occur very quickly. Given falling costs of compute and improvements in AI capabilities, at some point AI-assisted reflection might be able to drive much more progress in a few years than humans could in 1,000. So a “Deep Reflection,” as the cool kids are now calling it, wouldn’t even take that long.

But what’s so important about this kind of reflection?

Humans make many terrible mistakes. If we’d ventured out to the stars a few hundred years ago, we might have brought slavery across the galaxy. Sixty years ago, we’d have brought segregation. Today, we might bring factory farming and wild animal suffering.

Even setting aside specific catastrophic risks, following the wrong moral theory leads to the loss of most available value. Suppose that the right view is hedonic utilitarianism (pleasure minus pain should be maximized), but we act on desire-satisfactionism (desire satisfaction should be maximized). We’d lose out on nearly all realizable value. Other examples can be even more extreme: the optimal world by the lights of totalism (one of the leading theories of population ethics) might be actively bad according to critical level views (another of the leading views).

Totalism says the value of a population is measured by its total welfare. Critical level views say that it’s determined by the welfare of everyone above some threshold minus the welfare of those below the threshold (so e.g. a person .5 above the threshold would add .5 to total value, and a person .5 below the threshold would subtract .5 from total value). So totalism might favor proliferating barely happy minds in enormous numbers. The totalist’s utopia is the critical level theorist’s nightmare.

One option is to simply settle the universe without any discernible plan in mind. But that is likely to be catastrophic. Most value could be jeopardized by ruthless competition to grab resources or individual civilizations unleashing cosmic catastrophes. Even if that doesn’t happen, unless we’re specifically optimizing for bringing about value, we’re likely to miss out on most value we can get. Arbitrarily technologically proficient civilizations get much more of what they optimize for than what they don’t.

Many of the best things to do in the future might be weird and counterintuitive. Perhaps the digital minds that can experience the most joy are weird. It’s hard to imagine a big constituency for bringing them about in large numbers unless we reflect carefully. Other routes towards the production of enormous good—like acausal trade—might be even weirder. Probably there are a lot of actions that are very valuable in non-obvious ways. Absent reflection, we likely won’t take them.

Put extremely concisely, the case for a Deep Reflection is that careful planning assisted by superintelligence allows civilizations to:

• Make good decisions that produce lots of value but are non-obvious.

• Avoid making serious moral errors akin to spreading slavery to the stars.

• Prevent most value from being lost to Darwinian competition.

It’s helpful to take an outside view here. Ask: would past societies have done terrible things if they’d settled space without a Deep Reflection? Of course! Why think we’re any different? You should be suspicious if you conclude that we are the first civilization to avoid major moral error. Past societies have thought the same about themselves. They were wrong. So are we.

Does my argument depend on moral realism? By no means! Even if you don’t think morality is objective, you should think that careful thinking can improve the quality of our decision-making. You don’t have to think that beauty is objective to think that the best way to maximize the universe’s beauty is to think carefully about how to do it.

Now, you might be skeptical of this kind of reflection. Sometimes you think hard about something and don’t get the right answer. But in general, I’d expect better decisions to be made if we reflect carefully than if we don’t. We should expect reflection to improve decision-making. If you didn’t think that, then you should favor reflecting less before making big decisions, because it’s unlikely we’re at precisely the optimal level of reflection. If we shouldn’t favor more, then we should probably favor less.

In addition, superintelligence is likely to improve our ability to reflect well. Compare Fable to the AIs of 2023. Then imagine another decade or more of that kind of progress. The end result would be AI that can make decisions very well. It’s unlikely that humans are the best decision-makers that there can be. Thus, it should be possible in principle to have significantly better decision-making than we have today. Given how important this decision will be, it’s worth waiting for an improved decision-maker before making extremely big decisions.

The Long Reflection often sounds slightly sci-fi-ish. But the core idea is very modest and even a bit trivial. Before making extremely big decisions of cosmic significance that will affect things for billions of years, we should think carefully. So long as you take morality seriously and think major moral mistakes could be cataclysmic, a Deep Reflection is vital.

Sadly, philosophy journals penalize ambitious arguments. Thus, to get it in publishable shape, the piece had to be castrated and lobotomized. The original title was “A Presumptuous Proof of Theism.” The new—less interesting—title is A Presumptuous Path to Theism, Modal Realism, or axiarchism. The published paper on the topic is significantly less comprehensive than my blog post on the subject. There’s much less about why the main rivals to theism are implausible.

I’ll tell you about the paper. But first: book announcement. Joe Schmid, Brian Cutter, Amos Wollen, Dustin Crummett, Philip Swenson, and I are writing a book defending the existence of God. One of the chapters will be about the Anthropic argument and will be the most comprehensive presentation of the argument to date!

Our aim is to popularize and defend all the insanely strong Bayesian arguments for theism that nobody aside from like ten philosophers and some of my blog readers has ever heard of. We’ll talk about the prior probability of theism, a priori fine-tuning, nomological harmony, psychophysical harmony, and some new arguments that have yet to see the light of day! Rumor has it that it will strike the fear of God into the hearts of atheist readers and illustrate that there are many different powerful arguments that substantially raise the probability that God exists.

The book probably won’t be out for a year or more. It’s in its early stages. But when it comes out, finally the theism question will be resolved, and philosophers can move on to other philosophical questions, like whether there are chairs. If I had to hand-pick a power team to defend God’s existence, my coauthors would be some of the first people I’d pick. It’s amazing to have such a stacked group of coauthors, providing such stacked arguments, for the best possible news—that there is a being of limitless perfection with an infinitely amazing plan for the universe who loves each and every one of us.

Here’s the Anthropic argument in a nutshell. The self-indication assumption claims that your existence favors theories on which there are more candidates for being your present self. So, for example, imagine a coin gets flipped. One person is created if it comes up heads and ten people are created if it comes up tails. You get created by the coin flip. SIA says you should think it’s ten times likelier the coin came up tails than that it came up heads. Because tails leads to ten times more people being created—and you could be any one of them—it makes your existence ten times likelier. SIA in an oversimplified sentence: your existence is likelier if there are more people. (If you want the most precise definition which doubles as an argument for SIA, see here).

Oh and here’s another thing about the SIA: it’s the correct view. The other views are, to use technical language as a Real Scholar™ who has just published a paper on the subject and will one day be the Jacob Urowsky Professor of Philosophy at Yale University, totally whack. There are a bunch of very strong arguments for SIA and no good responses to them. I’ve published another paper arguing for the SIA and have various blog posts that are more detailed.

If the SIA is right, then your existence favors theories on which there are more people. A theory on which there are 10x more people makes your existence 10x likelier—a theory on which there are a trillion times more people makes your existence a trillion times likelier. But then by the same logic, a theory on which there are infinitely more people makes your existence infinitely likelier.

Conditional on the correctness of the SIA, your existence gives you infinitely strong evidence for the existence of infinitely many people. There are also smaller and larger infinities. So if the SIA is right, then your existence confirms that the number of people that exists is the very largest number of people that there could be. Maybe this is some infinite cardinal like Beth 2 (ℶ2)—more likely, it’s absolute infinity (Ω). As Menzel describes it:

Cantor recognized that some collections — notably, the collection of all ordinal numbers — are themselves “too big” to be assigned a definite size, a definite cardinal number. This idea led to the development of the limitation of size approach to the set theoretic paradoxes that focuses on over-largeness per se as their source. I am of course strongly challenging that view — it is only when a collection is over-large due to unboundedness that paradox can arise. However, it seems to me it is quite intuitive to say that the “size” of a set that cannot be measured by the cardinals that emerge in the course of the set formation process represents an unsurpassable — albeit mathematically indeterminable — limit, what Cantor called the “absolutely infinite”: an “absolute quantitative maximum” that is larger than any set with a definite cardinality and is incapable of either determinable measure or any definite form of increase; no two absolutely infinite sets can differ in size in any mathematically definite sense.

So if SIA is right then you should think the number of people that exists is this giant infinity that’s way more than the number of numbers. Standard atheistic multiverses don’t get you anywhere near that number of people. You need chicanery to get more people than numbers. So if there’s no God, the odds there’d be that many people are very low. In contrast, because it’s good to create, if there is a God, it’s reasonably likely he’d do a lot of creating. Thus, the fact that there’s a ridiculously large infinity worth of people—confirmed by the SIA—is strong evidence for theism.

Here are the basic steps in the argument:

1. SIA is right.

2. If SIA is right, then the number of people is a very large infinity (if it can even be called a number).

3. The odds of the number of people being a very large infinity are vastly higher given theism than atheism.

4. Thus, the fact that the number of people is a very large infinity is strong evidence for theism.

The paper didn’t argue for 1. Instead, we simply argued that if you accept the self-indication assumption, then you should start thinking that the number of people is some giant-ass infinity (that’s a technical term for an infinity that is very large). If you think that, then you should be inclined toward some pretty exotic picture of reality. The main options we discuss are axiarchism (some generic force of limitless power that compels good things to happen), theism, and modal realism (the view that every possible world concretely exists). All of these have the potential to produce some ridiculously large number of people. Standard atheism has much more trouble.

Why? Well, the number of possible people is very large. To explain how large, I’ll have to tell you a bit about infinities. The smallest infinity is aleph null (ℵ0). It’s the number of natural numbers: 1, 2, 3, 4, etc. Bigger than that is Beth 1 (ℶ1)—the number of real numbers, which includes every decimal number, even infinite non-repeating decimals. Bigger than that is Beth 2 (ℶ2). All of these describe the size of sets. But there are some collections that are too big to have an associated cardinality. For example, you can prove that there can’t be a cardinality of the collection of all truths.1

So how big can we prove the “collection” of possible people is?2

Here’s an argument that there are at least ℶ1 possible people. There are ℶ1 positive real numbers. It seems you can create a copy of me with a height for each of the positive real numbers. Thus, there are at least ℶ1 possible people.

There are ℶ2 ways of arranging space-time points. If a region contains ℶ1 spacetime points and each point admits at least two relevant states, there are ℶ2​ binary arrangements of those points. So if there’s a version of me with each of the points in his hand scrambled in any way they can be scrambled, then there are at least ℶ2 possible people.

So long as you can clone a person N times, where N is any distinct cardinality, then the number of possible people must be absolute infinity (Ω); it must be too big to have a cardinality. And so long as you buy the right version of the SIA, on which your existence favors the presence of more people, rather than the existence just of a larger share of possible people, your existence confirms that the number of possible people is Ω (for more on this, see section 4.3). After all, there are more total people if Ω people exist than some smaller number.

Thus, if SIA is right, then your existence gives you arbitrarily strong evidence that the number of people that exists is the most it could be. Probably that’s Ω—too large to have a definite cardinal ranking. The odds of so many people existing are high (or at the very least not very low) if God exists. If God is going to bother creating at all, he has no reason to stop after one universe, or ten, or one hundred. If creating is worth doing, there’s no reason to stop before creating Ω universes. Thus, conditional on God creating anything at all, it seems very likely that he’d create the very most people he could.

Contrast that with atheism: what is the plausible atheistic story on which you get even ℶ2 people, much less Ω people? Standard atheistic multiverses like eternal inflation just get you ℵ0 people. The many worlds can maybe get you a bit more, but not enough3—and certainly not Ω. To the best of my knowledge, there has never been a theory of physics proposed that gets you enough people.

So if the argument is right, standard non-exotic “physics is all there is” pictures of reality are out. You’ll need to believe in something very weird like modal realism. And modal realism probably undermines induction: if there are infinite concretely-existing worlds that are like ours up until one second from now and then chaotically break, it’s hard to see on what grounds I can rationally trust that our laws won’t break.

That’s basically what the paper argues. If you buy SIA, you can’t be a moderate. You have to think ultimate reality is much weirder—and much bigger—than you thought before. There are a few non-theistic theories that aren’t automatically ruled out, but if the argument single-handedly wipes out 99% of atheistic probability space, then that’s a reasonable success!

The wild animals honor me,

the jackals and the owls,

because I provide water in the wilderness

and streams in the wasteland,

—Isaiah 43:20

Genghis Khan is among history’s most prolific rapists and murderers. His armies passed through vast swaths of Eurasia, slaughtering, pillaging, and raping those they came into contact with. In total, his armies killed around 40 million people, enough that his murdering spree may have induced geologically-detectable global cooling.

Despite this, he is a reasonably popular historical figure. In Mongolia in particular, he’s very popular, appearing on the currency, in statues, and in the names of airports. One of history’s great murderers is revered all around the world. People often think about historical events in a morally indifferent, almost clinical way. We discuss great conquests, yet rarely think much about what they involved—what it would be like to be a child run through by a Mongol invader’s sword.

When people think about Khan as a great man, they think only about how significant his achievements are. They become detached from moral assessments of those “achievements.” This is how people often read history. They think of it as an impartial accounting of historical events, without spending much time to think about what it would be like to be someone alive at the time period. They become detached from moral assessment, recording instances of brutality like a notary, without really assessing them.

We do the same thing for wild animals.

We are all used to hearing about how brutal life is for wild animals. About the deer whose head is crushed in the lion’s maw. Yet we don’t generally spend much time thinking about what it’s really like to be a deer—whether there is something lamentable about the fact that the near-universal condition of life on Earth consists of dying after days or weeks spent running from predators hell-bent on ripping their victims limb from limb. Just as our detachment concerning Khan leaves us blind to obvious moral truths, our detachment concerning wild animals leads us to neglect something very obvious: there is something very wrong with nature.

The mind-boggling amount of wild animal suffering

Begin with the empirical facts: almost all suffering in the world is experienced by wild animals. My sense of the general view among those who have looked into it is: mammals and birds are definitely conscious (95%), reptiles are very likely to be conscious (85%), fish are pretty likely to be conscious (75%), and insects are maybe about 50% likely to be conscious. So even if you have conservative views about which animals are conscious, there are orders of magnitude more suffering animals than people.

These animals are mostly r-strategists, having extremely large numbers of offspring, most of whom die after a short period. Thus, nearly all animals die as juveniles in some painful way, often over an extended period of time. The most humane deaths wild animals get come from being eaten alive. Many slowly die of starvation or disease. At any given time, ridiculously large numbers of wild animals are crying out in unbearable pain.

This is a very surprising empirical fact. It means that behind the veil of ignorance, you’d care way more about wild animal suffering than human suffering. It’s a shocking fact about our world that well above 99% of expected suffering is experienced by wild animals. When the facts are surprising, a good ethical theory will often deliver surprising verdicts. As a point of comparison, if it turned out that 99.9% of the suffering in the world was caused by Bob, then it wouldn’t be a defect of an ethical theory that it said Bob’s crimes were some of the worst in history.

Is this a bad thing? I think so. Note that if the pain of a wild animal matters 0.1% as much as the pain of a human, then wild animal suffering is one of the worst things in the world, if not the worst. Even with such conservative accounting, because there’s more than 1,000x more wild animal suffering than human suffering, wild animal suffering would still be a worse global problem than human suffering.

Let’s see if we can prove that wild animal suffering matters from uncontroversial moral premises.

The badness of pain

Let’s start with the obvious one that I’ve mentioned a bunch of times: suffering is a bad thing. It’s bad to hurt because hurting feels bad, not because of other extraneous factors like the intelligence of the sufferer. put it well:

What on earth does your intelligence have to do with how bad your suffering is? To me, that’s like saying that pain is only bad when the person in pain is able to solve differential equations. The badness of pain just doesn’t on its face have anything to do with that. Does the claim make more sense if you add a bunch of other intellectual tasks that a person could do?

Bryan says that intelligence is the ability to learn. So pain is only bad if you have a lot of learning ability? Imagine putting your hand on a hot stove. Imagine the intense pain. You can just see that that is a bad experience. Now, is part of what makes it bad the fact that you are able to learn a lot of stuff in general? Again, that just seems obviously not true.

We all recognize in other contexts that animal pain is bad. If a dog cries out in pain after a nail pierces its paw, we don’t think that’s some irrelevant issue of no consequence. Ask: would it make any difference if the dog was wild? No, that just seems irrelevant. The intrinsic badness of the dog suffering intensely depends on how its suffering feels rather than whether it’s a wild animal or a household companion. When a dog runs away and returns to nature, does its moral importance fall off a cliff?

Something else that’s important to note is that wild animals are cognitively very similar to a wide range of sentient beings that we uncontroversially regard as being morally important. Babies and the mentally disabled matter—certainly their pain matters. Wild animals are mentally a lot like babies and the most mentally disabled people. They cannot speak or form various complex thoughts, but they still have desires which they pursue. At the present moment, countless beings who experience the world like human babies are starving and being ripped limb from limb.

So if what matters with respect to the badness of a being undergoing some period of suffering is something about its mental life, then what happens in nature is about as bad as if quadrillions of babies and mentally disabled people were suffering and dying in every forest, field, and den. That would obviously be a big deal! So is wild animal suffering.

Just think about what it’s like to be a wild animal. When you imagine being eaten alive by a predator, having parasites devour you from the inside, and starving to death over the course of days, doesn’t this seem like a bad thing? Exactly what should be done about it is a separate question. But doesn’t it seem like there’s something unfortunate going on?

Ex Ante Pareto (warning: technical philosophy)

Here’s another modest principle from which we can deduce the badness of wild animal suffering. There’s a principle beloved by ethicists called Ex Ante Pareto (EAP). It states “if some event is better in expectation for some and worse in expectation for no one, then it’s better overall.” There might be weird exceptions to this in infinite cases, but in most ordinary contexts it’s largely uncontroversial. If there’s a lottery which has an 80% chance of giving a person an extra 10 years of life, that’s better than if there’s a lottery that only has a 20% chance of giving that person 10 years of life.

There are different things that EAP might apply to. Here’s what I’ll be analyzing: EAP as applied to assessments of the goodness of chancy events relative to what an agent knows. So the principle is not concerned with how an agent should act. Instead, the principle says that if based on what an agent knows, they regard one event to be better in expectation for some people than another, and worse in expectation for none, then they should regard that event as better in expectation overall. Just as one can assess the goodness of end states of affairs, they can assess the expected goodness of chancy events.

But now consider the following case: there’s some being you see. There is a 60% chance that they are a wild animal. There is a 40% chance that they’re a human with some kind of horrible genetic condition that leaves them similar mentally and physically to a wild animal. This being is in terrible pain. Now ask: which of the following is better?

1. The being is cured of the pain if and only if they’re a wild animal.

2. The being is cured of the pain if and only if they’re human.

Intuitively, it seems that 1 is better. After all, whether they’re a human or wild animal makes no difference to any of their mental capacities or what they look like. Those are held fixed. 1 leaves them with a 60% chance of being cured of terrible pain, while 2 leaves them with only a 40% chance of being cured. Rationally, this being would prefer 1 to 2, and no other being is affected. Surely, by Ex Ante Pareto, 1 is better.

But this is incompatible with the view that human suffering matters more than wild animal suffering. After all, if human pain is 2x more important than wild animal pain, then 2 would be better—it has a 40% chance of doing something twice as good. And with more extreme views, 2 would be better even if you were 99% sure that the being was a wild animal.

Remember, in order to avoid the judgment that wild animal suffering is the worst thing in the world by far, you need to think that wild animal pain matters tens of thousands of times less than human pain. Yet any judgment according to which wild animal pain matters any amount less than human pain violates EAP. The more extreme the judgment, the more extreme the violation is.

There is one way you can get around this: you can hold that a human’s pain is intrinsically worse for them than a wild animal’s pain is for them. On this view, when a human feels hurt, this is much worse for them than when a wild animal feels a similar amount of hurt. There’s no longer an Ex Ante Pareto violation, because picking 2 over 1 leaves them better off in expectation. In the event that they’re a wild animal, they are harmed less by the pain.

However, this view remains counterintuitive. First of all, it implies that you should pick 2 over 1. That’s weird. Remember, we are holding fixed all capacities—both mental and physical—of this being. The only difference between 2 and 1, therefore, is that 1 has a better chance of lessening their pain, but if 2 lessens their pain then they are biologically Homo sapiens. Surely that should make no difference?

Second, the view implies not only that the overall badness of your pain depends on whether you’re a wild animal. It implies that how bad your pain is for you depends on whether you’re a wild animal. So if a deer in a forest feels pain in its leg, but it isn’t sure if it’s a wild animal, whether it’s a wild animal is relevant to assessing how harmed it is by the pain in its leg. That just seems ridiculous! How harmed one is by pain doesn’t seem to depend on their species or whether they dwell in nature.

Third, remember, the view has to be extreme to evade the conclusion that wild animal suffering is one of the world’s worst things. It must hold that wild animal suffering matters many thousands of times less than human suffering. But intuitively, even if being biologically Homo sapiens makes a fixed quantity of suffering cause more harm, it seems very implausible that it would make a difference by a factor of 100,000 or 1 million.

Who is more harmed: a deer when it’s ripped limb from limb or me when I stub my toe? Intuitively it seems the answer is the deer. But if human pain harms them millions of times more than comparable deer pain, then I would be harmed more. That’s very implausible.

So to recap, any view that counts the pain of wild animals less than that of comparable humans will either imply:

1. Being biologically non-human or being wild makes your pain intrinsically less harmful to you, even without affecting your mental capacities at all or how your pain feels. If you have two beings that are identical in physical and mental respects, but one is a wild animal, when they have intense headaches, it simply harms them less. In fact, to avoid the result that wild animal suffering is one of the worst things in the world, it will need to harm them hundreds of thousands or millions of times less at least, even without making any difference to their conscious experience.

2. Being a wild animal doesn’t affect how harmed one is by pain, but it makes their interests intrinsically count less. This violates Ex Ante Pareto. It also implies that the badness of your experience of pain goes down by a factor of around a million if you’re a member of a different species and aren’t human—even if this makes no difference to your felt experience of pain.

The more egregiously one discounts the interests of wild animals, the more severely they’ll be gored on one horn of the dilemma. And note: even if you think wild animal suffering matters 1,000 or 10,000 times less than a comparable amount of human suffering, it still is one of the worst things in the world. To avoid this conclusion, you need to discount their interests by a factor of around a million or a billion—thinking wild animal suffering matters effectively zero.

How are we so wrong?

Thinking wild animal suffering is a big deal is counterintuitive. It puts you in an extreme minority (assuming we don’t count wild animals among the relevant voting bloc). So how is the great mass of humanity so wrong? I think there are basically five reasons.

1. Detachment: as mentioned at the outset, when we think about wild animal suffering, we treat it like the pillaging of Genghis Khan. We treat it as being beyond the realm of moral assessments, a bit like events in the distant past. If one treats something as beyond the realm of moral evaluation, it isn’t very surprising that they’ll underestimate its badness. In general, few people spend much time empathizing with wild animals.

2. Lack of empathy: As notes, human empathy largely evolved to facilitate cooperation between other humans. This is why historically people have excused serious crimes against members of other tribes and other nations. We don’t naturally feel much empathy directed at wild animals for evolutionary reasons. Much moral progress has come from empathizing with others—seeing that they are like us. But wild animals are not like us, and we don’t naturally feel much empathy for them. It’s no surprise that we wrongly neglect their interests.

3. Status quo bias/just world fallacy: Humans have a general tendency to assume that ubiquitous features of the natural world are unobjectionable. Just as people make crazy excuses for obviously horrible features of the world—saying that death is a good thing because it gives life meaning—they also excuse the horribleness of wild animal suffering. One useful test for seeing if something is status quo bias: imagine it didn’t exist and then see how we’d feel about it being added. If you want to determine if, say, people’s support for death is a byproduct of status quo bias ask: if there were a race of immortals, do you think they would be reasonable to suggest a homicidal resolution to the crisis of meaning? Similarly, if all animals in nature lived awesome, peaceful, idyllic lives, wouldn’t we think it a tragedy if their lives turned into the hell of the present—so that nearly all died horribly after a short time?

4. Scope neglect: Humans are bad at grokking big numbers. Humans will pay about the same amount to save 2,000 birds as 200,000 birds. Thus, when one is dealing with quadrillions of wild animals suffering, it’s no surprise that our intuitions are totally screwed up. (Simple fix for this: think in terms of ratios rather than absolute amounts. So ask: is it a big deal that for every single person, there are maybe a million fish that die painfully after a few days or weeks every single year? Answer: yes).

5. Outrageous empirical facts: Our moral intuitions are generally formed with the assumption that the empirical facts are roughly as we’d expect. But if the empirical facts are crazy and shocking, as they are here, then our moral intuitions will be screwed up. To see this, let’s lay out the main steps in the argument for wild animal suffering being the worst thing in the world.

a) The source of nearly all the world’s suffering is one of the worst things in the world. This seems obvious enough.

b) Wild animal suffering is the source of nearly all the world’s suffering. This is the surprising premise. The end judgment is surprising because the facts are surprising, not because the moral premise supporting concern for wild animals is surprising.

Conclusion

Wild animal suffering matters. This depends on totally ordinary moral judgments, rather than super weird ones. Now, it’s a hard problem to solve. At this point, the best thing to do is likely to fund research on the subject and try to get more people concerned about it. There are some things we can do near-term—though discussing them in detail is a topic for another article.

The first step towards solving a problem is recognizing that it is a problem. We haven’t reached that stage with wild animal suffering. We should. Ordinary life for almost every creature who has ever lived has been hellish and brutal—a condition that we’d call torture if you inflicted it on a human. If we can’t see anything significantly bad about that, then we are like those who praise Genghis Khan: unable to recognize the horror of suffering, bloodshed, and death, even when the facts are in front of our noses.

Sadly, the pieces cataloguing Biblical errors are often written by twelve-year-old new atheists with titles like “10,000 Biblical Contradictions,” where half are obviously fake. I’ve been disappointed by the paucity of writings making the case against inerrancy seriously. It thus seemed worth doing a comprehensive write-up.

Note that passages can give evidence against inerrancy even if there is some possible reading on which they aren’t in error. Merely coming up with some explanation is inadequate. So long as a passage is more likely to be contained in a book that errs than an inerrant one, it provides evidence against inerrancy. Across many passages like this, the evidential force compounds.

Obviously this list won’t be complete—the Bible is a long book, and I can’t hope to address every single plausibly wrong thing in it. I’ll try to focus on the examples that seem most serious and most representative of broader themes. So, for example, I’ll give some examples of wrong genealogies, some examples of serious moral errors, and some examples of incorrect authorship—while noting there are plenty more.

Here’s the roadmap, where before the colon lists the title and after describes it in more detail:

1. Warcrimes: slaughter commands in the Hexateuch.

2. Slavery: repeated Biblical support for slavery.

3. Leviticus: all the craziness in Leviticus.

4. Daniel: it contains a lot of historical errors and says it was written by Daniel when it very likely wasn’t.

5. 2 Peter: it claims falsely to have been written by Peter.

6. Genesis 19: this passage tells a very implausible story of the Israelite’s rival nations both being birthed through incest.

7. Conflicting genealogies: Matthew and Luke tell conflicting and irreconcilable genealogies.

8. Imminent return: the Bible repeatedly indicates Jesus’s second coming occurring during the life-time of the Apostles.

9. Contradictions: lists a bunch of contradictions concerning the time of the crucifixion, who commanded the Davidic census, the person who killed Goliath, whether people can speak to God face to face, and how Judas died.

10. Violent fulmination: describes immoral depictions of the wholesale destruction of Israel’s enemies, willed by God.

11. Even Christians shouldn’t affirm inerrancy: explains why there aren’t good theological reasons to affirm inerrancy.

12. Conclusion: concludes, as you might have guessed.

1. War crimes

The book of Joshua is dedicated almost entirely to discussing the glory of the Israelite conquests—carried out with conspicuous disregard for the lives of the innocent people being slaughtered. If any other ancient source depicted these acts, we wouldn’t hesitate to label the depicted acts serious war crimes. Joshua 6 describes what happened after the Israelites conquered Jericho:

²¹ They devoted the city to the Lord and destroyed with the sword every living thing in it—men and women, young and old, cattle, sheep and donkeys.

…

²⁴ Then they burned the whole city and everything in it, but they put the silver and gold and the articles of bronze and iron into the treasury of the Lord’s house.

Joshua 8 describes similar brutality after conquering Ai:

²⁴ When Israel had finished killing all the men of Ai in the fields and in the wilderness where they had chased them, and when every one of them had been put to the sword, all the Israelites returned to Ai and killed those who were in it. ²⁵ Twelve thousand men and women fell that day—all the people of Ai. ²⁶ For Joshua did not draw back the hand that held out his javelin until he had destroyed^[a] all who lived in Ai. ²⁷ But Israel did carry off for themselves the livestock and plunder of this city, as the Lord had instructed Joshua.

Deuteronomy 20 describes God commanding the Israelites to enslave some nations and slaughter others in their entirety:

¹⁰ When you march up to attack a city, make its people an offer of peace. ¹¹ If they accept and open their gates, all the people in it shall be subject to forced labor and shall work for you. ¹² If they refuse to make peace and they engage you in battle, lay siege to that city. ¹³ When the Lord your God delivers it into your hand, put to the sword all the men in it. ¹⁴ As for the women, the children, the livestock and everything else in the city, you may take these as plunder for yourselves. And you may use the plunder the Lord your God gives you from your enemies. ¹⁵ This is how you are to treat all the cities that are at a distance from you and do not belong to the nations nearby.

¹⁶ However, in the cities of the nations the Lord your God is giving you as an inheritance, do not leave alive anything that breathes. ¹⁷ Completely destroy^[a] them—the Hittites, Amorites, Canaanites, Perizzites, Hivites and Jebusites—as the Lord your God has commanded you. ¹⁸ Otherwise, they will teach you to follow all the detestable things they do in worshiping their gods, and you will sin against the Lord your God.

Some people suggest that this is just wartime hyperbole. They cite the fact that other texts in the Ancient Near East would sometimes use hyperbolic language of total destruction that wasn’t meant to be taken literally. But insofar as a text includes language where God allegedly commands the wholesale slaughter of innocent people, that is a clear error. It’s hard to interpret the command to “not leave alive anything that breathes” explicitly contrasted with making people “subject to forced labor” as hyperbole.

Even if the actual destruction of the cities is hyperbolically inflated, the command is not. How can detailed and contrastive instructions to carry out genocide as opposed to some other act—allegedly coming from God—be anything other than a serious moral error? It’s especially hard to see 1 Samuel 15 as hyperbole, when there’s a detailed plot point about Saul being deposed for not completely wiping out the Amalekites. Even if successful, this defense reduces the crime from genocide to ethnic cleansing—still a serious problem.

Similarly, people sometimes suggest that these passages aren’t supposed to be taken literally as depictions of any kind of conquering event. Instead, they are just there to teach us lessons. But it seems like a clear error if you write something whose non-literalism isn’t obvious—and this leads billions of people throughout history to believe in the permissibility of genocide. It is hard to imagine that the optimal vehicle for delivering true and important theological content is a detailed book-length description of mass slaughter of men, women, and children.

Should one just bite the bullet and say that this was justified? After all, we already know that God permits the innocent to die if he exists at all—from malaria, flesh-eating parasites, and earthquakes. God has all sorts of reasons for doing things that we don’t know about. And standard theodicies purport to explain God’s reasons for allowing apparent evil.

Still, there’s a huge difference between directly commanding an evil and simply setting up a world where the evil can happen. God pretty rarely intervenes directly in history in enormous and direct ways. It would be pretty surprising if one of the few times he did, it was to order mass slaughter. The other evils can be chalked up to “bad feature that results from the world being screwed up.” That won’t cut it for God directly commanding mass slaughter. It seems pretty clear that if God is going to intervene in history in a major way, he can work out a better way of achieving whichever ends he aims at than commanding mass slaughter.

Second, the problem isn’t just moral. It’s also that we know that Bronze Age people claimed all the time that their gods supported military victories for their people against their enemies. It would be really surprising if one of God’s major interventions in history corresponded closely to what everyone else throughout the surrounding time period believed wrongly. I like the way Dustin Crummett put it:

One issue here is that it’s not *just* that, e.g., the genocide commands seem like something God wouldn’t command; it’s that they *also* seem like the sort of thing Bronze Age people would falsely make up. This is obvious in the case of, say, Mormonism: “all these women need to have sex with Joseph Smith” seems surprising coming from God, but not surprising coming from Joseph Smith. People have claimed divine sanction for war crimes all throughout history, and presumably we both agree that they were almost always wrong; it’s kind of weird if it turns out God really did order genocide a couple times, and it happened to be in a context where everybody thought genocide was cool and good.

(Some probably was written during the iron age).

If God is going to command things that seem obviously immoral, it would be surprising if he did it right at a time when everyone thought the thing in question was moral. Lastly, it seems the archeological record indicates that this didn’t happen.

I find Dustin’s point to have more force than the purely moral argument. The main problem isn’t that we can be super confident that God wouldn’t command mass slaughter. Theists already think God allows lots of bad things to happen. It’s that once one plays the skepticism card, there is an almost inconceivable range of things that God might do—it would be ridiculously surprising if he just so happened to do the things in the Ancient Near East that resembled what Ancient Near Easterners thought their tribal war gods did. Additional Bayesian machinery in a footnote.1

2. Slavery

The Bible has a number of passages that appear to sanction slavery in a way typical of the beliefs of the surrounding time period, but falling well short of the commands of a perfect God. For example, from Leviticus 25:44-46:

“Your male and female slaves are to come from the nations around you; from them you may buy slaves. You may also buy some of the temporary residents living among you and members of their clans born in your country, and they will become your property. You can bequeath them to your children as inherited property and can make them slaves for life, but you must not rule over your fellow Israelites ruthlessly”.

Note that this explicitly sanctions the practice of slavery. It goes out of its way to permit permanently enslaving foreigners. Exodus 21:20-21 declares:

²⁰ “Anyone who beats their male or female slave with a rod must be punished if the slave dies as a direct result, ²¹ but they are not to be punished if the slave recovers after a day or two, since the slave is their property.

Ephesians 6:5 says “Slaves, obey your earthly masters with fear and trembling, in singleness of heart, as you obey Christ.” Colossians 3:22 reiterates “Slaves, obey your earthly masters in everything, not only while being watched and in order to please them, but wholeheartedly, fearing the Lord.” Other passages that express similar sentiments include 1 Peter 2:18 and 1 Timothy 6:1.

To reiterate: the concern is not just that these seem immoral (though, of course, they do). It is that they are immoral in precisely the way that human authors making errors in the surrounding context would be expected to be immoral. So this sort of writing is much more expected if inerrancy is false than if it’s true. It would be surprising if, when God makes counterintuitive moral judgments, they line up suspiciously with the widely-shared judgments of the surrounding time period. (Dustin’s “all these women need to have sex with Joseph Smith” analogy is a good one).

Now, maybe God couldn’t have commanded the outright elimination of slavery. Perhaps that would have backfired. But at the very least, he could have described slavery as a lamentable institution and refrained from enshrining in law a kind of slavery that treated foreigners much more cruelly than Israelites. He could also have avoided giving slaves direct instructions to obey their masters. I also find it weird that God apparently was willing to use his limited political capital on prohibiting shellfish and mixed fabrics, but not on banning slavery.

Note one last point: nearly all Christians throughout history, until the last few hundred years, thought these texts permitted slavery. It seems like a crucial oversight to include in sacred scripture passages which are interpreted by billions of people to support the permissibility of slavery.

3. Leviticus

The book of Leviticus seems to contain a number of problematic verses reflecting common errors at the time. In context, Leviticus was the law code, taken seriously by the Jewish people. Ethically objectionable laws in Leviticus are thus a problem. But these abound.

1. Leviticus 21: “If a priest’s daughter defiles herself by becoming a prostitute, she disgraces her father; she must be burned in the fire.”

2. Leviticus 15:19: “When a woman has her regular flow of blood, the impurity of her monthly period will last seven days, and anyone who touches her will be unclean till evening.”

3. Leviticus 20:18: “If a man has sexual relations with a woman during her monthly period, he has exposed the source of her flow, and she has also uncovered it. Both of them are to be cut off from their people.”

4. Leviticus 20:15: “If a man has sexual relations with an animal, he is to be put to death, and you must kill the animal.” Why should the animal be put to death?

5. Leviticus 20:9: “Anyone who curses their father or mother is to be put to death. Because they have cursed their father or mother, their blood will be on their own head.” This is pretty insane!

6. Leviticus 20:13: “‘If a man has sexual relations with a man as one does with a woman, both of them have done what is detestable. They are to be put to death; their blood will be on their own heads.”

These are just a select few of the relevant passages. But surely these are vastly more likely to be the work of a human author making mistakes than a divine author communicating deep wisdom without error! If these don’t count as errors, ask: is there anything that could be found in Leviticus that would count as an error?

If there were just one passage like this, it would be one thing. But Leviticus is suffused with apparent immorality that seems to reflect mistaken prejudices of the surrounding culture. The odds of these sorts of passages appearing in a text produced by humans at the time is quite high—the odds of it being produced in an inerrant text are much, much lower. If a law code is included in scripture as being set up by God himself, and the law code has bad laws, surely that is an error.

Perhaps God was accommodating the existing errors of the culture. After all, Jesus says Moses only permitted divorce because of the hardness of the people’s hearts. But this only explains keeping existing bad laws—not instituting new ones that are both bad and bizarre.

It is very unlikely that the Israelites were so committed that leadership had to impose laws that executed homosexuals, those who curse their father, animals that were raped, and prostitute daughters of priests. Similarly, it’s hard to believe that belief in the impurity of women having their period wasn’t changeable. Remember, this is all after God just did a bunch of totally crazy miracles, freed the Israelites from Egyptian bondage, and spoke from a mountain.

Perhaps God’s ways are mysterious, so we should expect the moral law he implements to look weird to us? Sure, but we shouldn’t expect it to look weird in all these ways that we’d actively expect from law codes of primitive people. It would be very surprising indeed if an infallible law code contained suspicious overlap with the kinds of immoral laws you’d expect from primitive Near-Easterners.

4. Daniel

Daniel claims to have been written by Daniel, repeatedly referring to itself in the first person. For example, Daniel 8:15 says “While I, Daniel, was watching the vision and trying to understand it, there before me stood one who looked like a man.” But there is a pretty overwhelming scholarly consensus against Daniel, who reportedly lived around 600, having written Daniel. And this scholarly consensus is plausible—it looks like the book was compiled around 164 based on older material, often from the 300s:

1. Ben Sira lists major Israelites and includes Isaiah, Jeremiah, Ezekiel, and the Twelve Minor Prophets. Yet he leaves out Daniel. If, as the book suggests, Daniel was of similar eminence, it is hard to see why he’d be left out.

2. The book jumbles the history that occurred around the time Daniel allegedly lived (between 620 and 530 BCE). It treats Belshazzar as being the son of Nebuchadnezzar when really he was the son of King Nabonidus.2 It also treats Belshazzar as king, when Nabonidus was the last king.3 Now, conservative scholars often reply that Nabonidus entrusted leadership authority to his son. They’ll cite the fact that the verse account of Nabonidus says he “entrusted the kingship to [his son],” but this overlooks a bunch of problems that I’ll put in a footnote.4

   Similarly, the book of Daniel claims that Darius the Mede became the king when Belshazzar was killed, while other sources say Cyrus directly took over after Nabonidus was captured and exiled.5 Daniel 9:1 says Darius was the son of Xerxes—in fact, Xerxes was Darius I's son. The description of Darius the Mede also appears to be based on jumbled up history from someone later named Darius Hystaspis, who, like the supposed Darius the Mede, took Babylon, organized the empire into satrapies, and was named Darius (though he lived a lot later, so this definitely couldn’t have been an accurate description). Darius the Mede doesn’t seem to have existed! For this reason, the early dating defenders generally suggest that Darius the Mede was a name for someone else, yet in no case are these people called Darius, and in no case are they both King and a Mede (additional problems for each in a footnote).6 And see here for other errors.

3. Daniel uses the term “Chaldean” to mean wise man, even though contemporary Neo-Babylonian sources use Chaldean as an ethnic term. The term being used to mean wise man appears only in Persian and Hellenistic-era sources, so Daniel's usage reflects a time period after the fall of Babylon.

4. The story depicts various things that are very unlikely to have happened:7

Other indications adduced to [show] that the Book of Daniel is not the work of a contemporary, are such as the following :-The improbability that Daniel :i. and his companions, all strict Jews, should have suffered themselves to be initiated into the superstitious arts of the 'wise men', or that he should have been accepted as their president by the 'wise men' themselves 2; the improbability that Nebuchadnezzar should hold all the wise men of Babylon, including Daniel and his three companions, responsible for the failure of some, and condemn them to death even before their skill had been tried; Nebuchadnezzar's seven years' insanity (lycanthropy), with his public proclamation respecting it; the absolute terms in which both he and Darius, while retaining, so far as appears, their idolatry, recognize the eternal and universal sovereignty of the God of Israel.

5. Probably the strongest evidence is that when depicting events up through the period of the Seleucids in 167, Daniel is very accurate. Right after (starting verse 11:40), his accuracy falls off a cliff. Daniel predicts that Antiochus will launch another major campaign against Egypt, pass through lands including Judea, dominate Egypt/Libya/Cush, then pitch his royal tents “between the sea and the glorious holy mountain,” and die there. That is not how Antiochus died. He died in the east, in Persia/Babylonia-region traditions, not after conquering Egypt and not in Palestine.

   His death in Persia is attested to by Polybius, Josephus, and 1 and 2 Maccabees. Daniel also predicts the resurrection of the dead and the coming of Michael occurring immediately after which obviously didn’t happen. And note: the text claiming to be from Daniel occurs in the visions spanning chapters 7-12—so the bit of the text claimed to be first-person is precisely the bit that we have the strongest evidence is not from Daniel.

6. Daniel depicts the coming end of the world. But every other example of a full blown apocalypse story is from the third century BCE onwards.

7. The whole book converges on Antiochus. Each of the major prophecies in Daniel terminates at the time of Antiochus’s reign. But it would be pretty weird for a predictive prophecy written in the sixth century BCE to focus excessively on this random second-century ruler.

What do those who favor an early dating say in response?

Much of the contrary evidence is drawn from cases where the Aramaic portions of the text appear older. But the conventional view is that the mostly Aramaic portions in chapters 2-7 are older than the rest. Chapters 1-6 are third-person court tales, while chapters 7-12 are first-person apocalyptic visions. As for language, Daniel 1:1-2:4a are Hebrew, and then it switches to Aramaic through chapter 7, and then back to Hebrew. Chapters 1-6 are generally believed to be older by critical scholars, so citing their extra age isn’t dispositive.

Similarly, those defending an earlier date often cite the absence of Greek words as evidence of earlier composition, but Hebrew and Aramaic texts we know were composed in 2nd–1st century Palestine—the Qumran sectarian corpus—are equally free of Greek loanwords.

Similarly, it’s sometimes claimed that Daniel was correct about the existence of Belshazzar while other sources like Xenophon and Herodotus don’t know him. But demonstrably Babylonian traditions were present in the second century—the Prayer of Nabonidus at Qumran proves this, and Baruch 1:11-12 even illustrates that knowledge of Belshazzar was preserved. Thus, it wouldn’t be surprising for Daniel to know about Belshazzar’s existence while botching some of the details.

Lastly, there are complicated linguistic disputes about what dating the language favors: this is sufficiently complicated and disputed that I don’t give it a ton of weight (though if one gives weight to this, they should note that the Hebrew linguistically more closely resembles Ezra and Esther, written later, than books written in the sixth century. My uninformed outsider sense is that the 600-daters have to cope a bit about the Hebrew and the 160ish daters have to cope a bit about the Aramaic, with the later daters having a slightly stronger linguistic argument; the Aramaic seems consistent with many dates while the Hebrew seems to more strongly favor a later date. But this stuff is way above my pay grade.)

And note: the reason people don’t think Daniel wrote Daniel isn’t that they have a bias against the supernatural. None of the arguments I’ve cited assume that predictive prophecy can’t happen. They just assume that if you butcher the events of some period, you probably weren’t a contemporary—and if your accuracy dives off a cliff around some discrete date, probably that reflects when you stopped retrodicting and started predicting.

Now, some people have suggested that Daniel wasn’t a forgery but instead consistent with a literary genre where one attributes their insights to ancient sources. But all our earliest sources seem to think that Daniel really wrote Daniel. As Bart Ehrman notes, “ancient people really did believe that they were written by the people who claimed to be writing them, as seen repeatedly in the ancient testimonies.” Some sources that attribute the book to Daniel include: Matthew 24:15, Josephus, 4 Ezra 12:11, and Hippolytus.

When Porphyry in the third century AD argues that Daniel wasn’t by Daniel—and this discredits Christianity—Jerome replies by arguing that it was! If attributing books to people who didn’t write them had been an accepted practice, he could have simply granted the point. Everyone in that exchange seemed to agree that attributing books to people who didn’t write them was a big problem. This also illustrates that ancient writers thought that Daniel wrote Daniel.8 In some cases, people were even put to death for forging. Ehrman notes that the Greek descriptions of false authorship condemned it unequivocally:

“In Greek the two most common words to describe literary texts whose authors falsely claim to be a well-known figure are pseudos, which means “a falsehood” or “a lie,” and nothos, which means “an illegitimate child,” with connotations similar to our modern word “bastard.”

In early Christian communities, we have condemnations of forgery from Serapion of Antioch, The Muratorian Fragment, Tertullian, 2 Thessalonians, and more. Ehrman’s conclusion:

In short, there were long, protracted, and often heated debates in the early church over forged documents. Early Christians realized that there were numerous forgeries in circulation, and they wanted to know which books were written by their alleged authors and which were not. As we will see more fully later, practically no one approved of the practice of forgery; on the contrary, it was widely condemned, even in books that were themselves forged (such as 2 Thessalonians and the Apostolic Constitutions).

And note: even if Daniel doesn’t count as a forgery, the many serious errors of fact—with repeated, detailed descriptions of non-existent events and people—still count as errors.

5. 2 Peter

Another Biblical book that wasn’t by the person who is claimed to have written it is 2 Peter. The book claims to be written by “Simon Peter, a servant and apostle of Jesus Christ.” But it very likely wasn’t:

1. We know that in early Christian communities, people were constantly forging books and attributing them to NT main characters. So the prior on forgery isn’t low.

2. We have no mention of the book until Origen in the mid-third century.

3. The book of Acts says that Peter was “unlettered.” 2 Peter, however, is written in elaborate literary Greek (also different in style from 1 Peter).

4. The book contains various anachronisms like mentioning Paul’s letters as if they’re some organized collection of scripture—though no one else talks like this for at least 100 years. More decisively, it criticizes those who say “Where is the promise of His coming? For ever since the fathers fell asleep, all things continue just as they were from the beginning of creation.” This would be a weird thing to write during the days of Peter, when many of these people (notably: Peter) were still alive.

5. The book largely borrows from Jude, which would be odd for Peter—the chief Apostle—to do.

And note: there are many books that scholars don’t think are from the person who claims to have written them, not just 2 Peter and Daniel. The full list includes: 1 Timothy, 2 Timothy, Titus, Ephesians, 2 Thessalonians, Colossians (though this is a closer split), 1 Peter (also a closer split), James, Jude, Ecclesiastes, and various others. Though I admit to not having looked into these carefully.

6. Genesis 19

Genesis 19 tells the story of how the Moabites and the Ammonites—two of the main rival nations—came to be. Here is the story: Lot’s daughters got him drunk and slept with him on two consecutive nights while he was unconscious (it says “he was not aware of it when she lay down or when she got up,” which is, at the very least, physiologically improbable). They both gave birth: one to someone named Moab, who fathered the Moabites, and the other to Ben-Ammi, who became the father of the Ammonites.

Note: this occurs alongside other stories that are standardly thought to be literal history by inerrantists—e.g. about Abraham. So it’s hard to see why those who think that Christians should believe in a literal Abraham would be skeptical of this story.

Tell me: which seems more probable?

1. Two daughters got their father drunk, were able to successfully sleep with him in a cave while he was unconscious, and bore two sons. These sons went on to found the Israelites’ main enemy nations which suspiciously continued to bear their name, despite this not being how nations typically work. Oh, and the names also happened to be Hebrew puns on incest! A five-fold coincidence!

2. The Israelites had a fictitious mean story about the enemy nations’ disreputable origins, according to which they were the products of incest, just as lots of other nations have made up mean stories about enemy nations.

If you believe 1, I have a bridge to sell you.

What could inerrancy mean if it’s compatible with the transmission of stories—presented alongside ones taken to be historical—that are both false and bad in effect? By this standard, almost any book could be found in sacred scripture and not taken to be in error!

And note: while this is the most direct instance of this kind of thing, it’s far from the only one. We similarly have three different stories—two of Abraham, one of Isaac—passing off their wives as their sisters to avoid being killed. The second story of Abraham and the one of Isaac have them being passed off to the same person. Isn’t it clear that these stories are not accurate—but instead are garbled versions of a common motif?

7. Conflicting genealogies

The most serious conflict between genealogies in the Bible is between Matthew and Luke. Their differences are myriad:

1. Matthew runs through Solomon, Luke through Nathan.

2. Matthew gives Joseph's father as Jacob, Luke gives Heli.

3. The lists have different lengths (Matthew says there are 27 generations between David and Jesus, while Luke says there are 42).

4. They only share two names in the middle (Shealtiel and Zerubbabel).

Now, some say that Luke was giving Mary’s genealogy while Matthew was giving Joseph’s. This would be odd, though, because it describes Jesus as the son of Joseph, and then lists Joseph’s lineage. This is obviously not the plain reading of the text—which plainly appears to list Joseph’s father—and wasn’t dreamt up by anyone till 1490. In addition, the texts differ as to who Shealtiel’s father was.

Jimmy Akin proposes the following theory. Joseph was the adopted son of Heli but the biological son of Jacob. So the lists diverge regarding who to count as the grandfather: Matthew counts the biological grandfather, Luke counts the adopted grandfather. The two continue up the line until converging at Zerubbabel. They differ as to Shealtiel’s father because of adoption. Akin writes, “Though biologically the son of Neri, Shealtiel was reckoned as Jechoniah’s son for purposes of the kingly line.” Thus, Matthew counts Jechoniah as the father (taking into account his status as the adopted father), while Luke counts Neri as the father looking biologically. The two trace parallel paths through David. But this account has a number of problems:

1. At Shealtiel, Akin says Matthew (Jeconiah) gives the legal father and Luke (Neri) the biological one—he needs to say this since 1 Chronicles 3:17 explicitly makes Shealtiel Jeconiah's son, so Luke's Neri is otherwise a contradiction with Chronicles. Then, Luke gives the legal succession (Heli) while Matthew gives the biological succession (Jacob). But this means that they’re not consistent between using legal and biological frameworks.

2. It requires that two separate adoptions happened at the only two places the lines touch: one at Shealtiel and one at Joseph. This would be pretty surprising. Note: it’s not just that there would be two separate adoptions, but that they’d occur at precisely the points the lines overlap. It would be extremely surprising if both times the lists overlap after David, there was an adoptive marriage, and then the two counted them differently. The evidence for the Shealtiel crossover is nonexistent, and the evidence for the Joseph crossover comes from a source 190 years after Jesus, probably five or six generations removed from the family, that suggests Melchi is the father of Heli, while Luke suggests the father is Matthat. So the supporting source directly contradicts Luke!

3. The texts have wildly different numbers of generations. Between David and Jesus, Matthew has 27, Luke has 42. In addition, the length of the generations doesn’t hold up. David was born around 1040 BC; Jeconiah was born around 615. That's ~425 years across 18 links (based on the records according to Kings/Chronicles)—about 24 years per generation. Then, the remaining period, from Jeconiah to Jesus is roughly 610 years. At the dynasty’s own documented rate, you’d expect 25–26 generations. Matthew's list has only 13 links over the same ~610 years—~47 years per paternal generation, sustained for six centuries.

   Now, maybe Matthew is skipping generations (he appears to leave out some kings attested in Kings and Chronicles), but that seems to call into question an inerrant genealogy dictating who came from whom! (Worth noting that Luke’s number is pretty plausible). More problematically, Matthew declares “Thus there were fourteen generations in all from Abraham to David, fourteen from David to the exile to Babylon, and fourteen from the exile to the Messiah.” This seems to indicate he at least purports to be providing the number of generations, meaning he is in error.

8. Imminent return

The plain reading of the New Testament text implies that Jesus will return within the lifetime of the disciples.

Mark 8:38-9:1 depicts Jesus saying “If anyone is ashamed of me and my words in this adulterous and sinful generation, the Son of Man will be ashamed of them when he comes in his Father’s glory with the holy angels. Truly I tell you, some who are standing here will not taste death before they see that the kingdom of God has come with power.” Matthew 16 similarly states:

²⁷ For the Son of Man is going to come in his Father’s glory with his angels, and then he will reward each person according to what they have done.

²⁸ “Truly I tell you, some who are standing here will not taste death before they see the Son of Man coming in his kingdom.”

The same theme is found in Luke 9. Other passages from the Gospels that suggest this reading include Mark 13 which declares:

²⁶ “At that time people will see the Son of Man coming in clouds with great power and glory. ²⁷ And he will send his angels and gather his elect from the four winds, from the ends of the earth to the ends of the heavens.

²⁸ “Now learn this lesson from the fig tree: As soon as its twigs get tender and its leaves come out, you know that summer is near. ²⁹ Even so, when you see these things happening, you know that it^[d] is near, right at the door. ³⁰ Truly I tell you, this generation will certainly not pass away until all these things have happened. ³¹ Heaven and earth will pass away, but my words will never pass away.

Similarly, Matthew 10:23 declares, “Truly I tell you, you will not finish going through the towns of Israel before the Son of Man comes.” And many other passages in the Gospels seem to indicate that the second coming is near: Mark 14:62, Mark 1:15, Matthew 24:33, Mark 13:29, and Luke 18:8.

In Paul’s letters, we have similar claims that Jesus’s coming was imminent. Examples:

• 1 Thessalonians 4:15–17: “we who are alive, who are left until the coming of the Lord.”

• 1 Corinthians 7:29–31: “What I mean, brothers and sisters, is that the time is short. From now on those who have wives should live as if they do not; those who mourn, as if they did not; those who are happy, as if they were not; those who buy something, as if it were not theirs to keep; those who use the things of the world, as if not engrossed in them. For this world in its present form is passing away.”

• Romans 13:11–12: “salvation is nearer to us now than when we became believers; the night is far gone, the day is near.”

In the rest of the New Testament, we have similar attestation:

• Hebrews 10:37: “in a very little while, the one who is coming will come and will not delay.”

• James 5:8–9: “the coming of the Lord is near... the Judge is standing at the doors.”

• 1 Peter 4:7: “The end of all things is near.”

• 1 John 2:18 “Dear children, this is the last hour; and as you have heard that the antichrist is coming, even now many antichrists have come. This is how we know it is the last hour.”

• Revelation 22:20 He who testifies to these things says, “Yes, I am coming soon.”

  Amen. Come, Lord Jesus.

Then there are passages coming from the later books of the New Testament indicating an expectation of imminent return. 2 Peter 3:3 quotes scoffers asking "Where is this ‘coming’ he promised? Ever since our ancestors died, everything goes on as it has since the beginning of creation,” before answering (v8) “With the Lord a day is like a thousand years, and a thousand years are like a day.” John 21 purports to correct the widespread belief that the beloved disciple wouldn’t die until the second coming. Thus, it seems belief in an imminent return was very much in the air, and represented the standard belief of the Gospel authors.

C.S. Lewis calls one of the verses of Jesus predicting an imminent return “certainly the most embarrassing verse in the Bible.” He resolves this by suggesting that Jesus is capable, in his human form, of making errors. After all, the Bible depicts him saying he doesn’t know things and that he grew in wisdom, which requires previous ignorance. But this would not rescue Biblical inerrancy.9

My friend Aron Wall has an interesting post about this. His post is primarily concerned with Matthew 24:34, which says “Truly I tell you, this generation will certainly not pass away until all these things have happened.” Mark 13 and Luke 21 say basically the same thing.

Aron starts by noting that there are a bunch of passages that caution against trying to guess when the second coming will happen. Matthew 24:42 says:

Therefore keep watch, because you do not know on what day your Lord will come. But understand this: If the owner of the house had known at what time of night the thief was coming, he would have kept watch and would not have let his house be broken into. So you also must be ready, because the Son of Man will come at an hour when you do not expect him.

And Mark 13:32 describes Jesus saying that not even he or the angels know when it will happen:

But about that day or hour no one knows, not even the angels in heaven, nor the Son, but only the Father.

But here I think there’s a pretty easy response: Jesus was cautioning against trying to predict exactly when it would happen. He did, however, think it would come soon. For a point of comparison, you might say to a small child “we don’t know exactly when Mom is coming home,” while also saying “she’ll be home soon.”

Aron notes that the passage in Matthew begins:

24 Jesus left the temple and was walking away when his disciples came up to him to call his attention to its buildings. ² “Do you see all these things?” he asked. “Truly I tell you, not one stone here will be left on another; every one will be thrown down.”

³ As Jesus was sitting on the Mount of Olives, the disciples came to him privately. “Tell us,” they said, “when will this happen, and what will be the sign of your coming and of the end of the age?”

He proposes that the disciples were asking two separate questions. One: when will this happen, where “this” refers to the destruction of the temple? Two: what will be the sign of Jesus’s coming? The destruction of the temple occurred in 70 AD, so Jesus was right that “this generation will certainly not pass away until all these things have happened,” where “these things” refers to the destruction of the second temple.

Here’s the big problem: verses after 36 seem clearly to be about the second coming. In verse 37 it says “As it was in the days of Noah, so it will be at the coming of the Son of Man.” So it seems like Aron needs the answer about the destruction of the temple to last about one sentence and be sandwiched between two passages about the second coming. To see this, I’ll highlight all the passages in yellow that really seem like they’re about the second coming and the passages in blue that Aron needs to be about the second temple:

³⁰ “Then will appear the sign of the Son of Man in heaven. And then all the peoples of the earth^[c] will mourn when they see the Son of Man coming on the clouds of heaven, with power and great glory.^{[d] 31} And he will send his angels with a loud trumpet call, and they will gather his elect from the four winds, from one end of the heavens to the other.

³² “Now learn this lesson from the fig tree: As soon as its twigs get tender and its leaves come out, you know that summer is near. ³³ Even so, when you see all these things, you know that it^[e] is near, right at the door. ³⁴ Truly I tell you, this generation will certainly not pass away until all these things have happened. ³⁵ Heaven and earth will pass away, but my words will never pass away.

³⁶ “But about that day or hour no one knows, not even the angels in heaven, nor the Son,^[f] but only the Father. ³⁷ As it was in the days of Noah, so it will be at the coming of the Son of Man. ³⁸ For in the days before the flood, people were eating and drinking, marrying and giving in marriage, up to the day Noah entered the ark; ³⁹ and they knew nothing about what would happen until the flood came and took them all away. That is how it will be at the coming of the Son of Man. ⁴⁰ Two men will be in the field; one will be taken and the other left. ⁴¹ Two women will be grinding with a hand mill; one will be taken and the other left.

Now, here’s one way you can try to get around this: you can suggest that everything until 36 is about the destruction of the temple. My friend Gavin Ortlund suggests that the early bit of the passage is about the destruction of the second temple but uses weighty theological imagery to indicate that this was a typological foreshadowing of the second coming—just as similar language is used in the Old Testament to describe judgment on nations.10

But then you’d need to have verses 29-31 all be about the destruction of the temple, when they plainly appear to be about the second coming. I’ll illustrate what this reading looks like just to show how implausible it is (blue=second temple, yellow=second coming):

¹⁵ “So when you see standing in the holy place ‘the abomination that causes desolation,’^[a] spoken of through the prophet Daniel—let the reader understand— ¹⁶ then let those who are in Judea flee to the mountains. ¹⁷ Let no one on the housetop go down to take anything out of the house. ¹⁸ Let no one in the field go back to get their cloak. ¹⁹ How dreadful it will be in those days for pregnant women and nursing mothers! ²⁰ Pray that your flight will not take place in winter or on the Sabbath. ²¹ For then there will be great distress, unequaled from the beginning of the world until now—and never to be equaled again.

²² “If those days had not been cut short, no one would survive, but for the sake of the elect those days will be shortened. ²³ At that time if anyone says to you, ‘Look, here is the Messiah!’ or, ‘There he is!’ do not believe it. ²⁴ For false messiahs and false prophets will appear and perform great signs and wonders to deceive, if possible, even the elect. ²⁵ See, I have told you ahead of time.

²⁶ “So if anyone tells you, ‘There he is, out in the wilderness,’ do not go out; or, ‘Here he is, in the inner rooms,’ do not believe it. ²⁷ For as lightning that comes from the east is visible even in the west, so will be the coming of the Son of Man. ²⁸ Wherever there is a carcass, there the vultures will gather.

²⁹ “Immediately after the distress of those days

“‘the sun will be darkened,
and the moon will not give its light;
the stars will fall from the sky,
and the heavenly bodies will be shaken.’

³⁰ “Then will appear the sign of the Son of Man in heaven. And then all the peoples of the earth^[c] will mourn when they see the Son of Man coming on the clouds of heaven, with power and great glory.^{[d] 31} And he will send his angels with a loud trumpet call, and they will gather his elect from the four winds, from one end of the heavens to the other.

³² “Now learn this lesson from the fig tree: As soon as its twigs get tender and its leaves come out, you know that summer is near. ³³ Even so, when you see all these things, you know that it^[e] is near, right at the door. ³⁴ Truly I tell you, this generation will certainly not pass away until all these things have happened. ³⁵ Heaven and earth will pass away, but my words will never pass away.

³⁶ “But about that day or hour no one knows, not even the angels in heaven, nor the Son,^[f] but only the Father. ³⁷ As it was in the days of Noah, so it will be at the coming of the Son of Man. ³⁸ For in the days before the flood, people were eating and drinking, marrying and giving in marriage, up to the day Noah entered the ark; ³⁹ and they knew nothing about what would happen until the flood came and took them all away. That is how it will be at the coming of the Son of Man. ⁴⁰ Two men will be in the field; one will be taken and the other left. ⁴¹ Two women will be grinding with a hand mill; one will be taken and the other left.

In Mark, the disciples only ask “Tell us, when will these things happen? And what will be the sign that they are all about to be fulfilled?” after Jesus describes that “Not one stone here will be left on another; every one will be thrown down.” Jesus gives a similar answer, with similar descriptions according to which “about that day or hour no one knows.” Thus, it seems the event that would overturn the stones is treated as the same as the second coming. Further supporting that interpretation is verse 29, which is naturally read as saying the crazy cosmic stuff happens “Immediately after the distress of those days.”

The Son of Man coming in his glory is elsewhere used in Matthew to refer to the final judgment. So it would be weird if he uncharacteristically—just in this passage—used it to refer to the destruction of the temple, using a wealth of cosmic imagery that is otherwise reserved for descriptions of the second coming.

And the parallel passage in Luke goes so far as to use the phrase “At that time they will see the Son of Man coming in a cloud with power and great glory,” sandwiched between “Truly I tell you, this generation will certainly not pass away until all these things have happened,” and “When you see Jerusalem being surrounded by armies, you will know that its desolation is near.” Thus, the Lukan account seems to require that Jesus goes from talking about the destruction of the temple to talking about the eschaton, back to talking about the destruction of the temple without any warning.

Verse 36 looks pretty weird on any such account. On this picture, Jesus randomly switched from talking about the date of the destruction of the temple to talking about the date of the second coming, with no warning. Also, this account seems to involve Jesus answering the question in an intentionally confusing way, where he uses vivid theological language to describe different events without detailing when he’s going from one to the other.

This account runs into similar awkwardness with the parallel passages in Luke and Mark—I’ll put the awkward highlighting in a footnote.11

Aron also gives a response to Mark 9:1, where it says “Truly I tell you, some who are standing here will not taste death before they see that the kingdom of God has come with power.” He proposes this refers to the resurrection, not the second coming. But this fits awkwardly with the version of this story from Matthew 16:

²⁷ For the Son of Man is going to come in his Father’s glory with his angels, and then he will reward each person according to what they have done.

²⁸ Truly I tell you, some who are standing here will not taste death before they see the Son of Man coming in his kingdom.

Any other account will have similar problems with Matthew 16. Insofar as it’s about the resurrection, it requires verse 27 to be about the second coming, but verse 28 to randomly switch to talking about the resurrection. So then the theory will require multiple independent cases where the text switches between the second coming and some other topic—in one case the resurrection, in the other the destruction of the temple—with no warning.

Lastly, none of these are any help for explaining away the apparent imminent eschaton described outside of Matthew, Mark, and Luke. Taking into account the fact that this was demonstrably a widespread belief at the time, the default assumption should be that it made its way into the synoptics. It would be super surprising if the synoptics accurately record Jesus giving a teaching that just so happens to sound on plain reading like he’s asserting another extremely common belief at the time.

9. Contradictions

The Bible contains a number of pretty clear contradictions. I won’t go through all of them, and we’re already ten thousand words in, so I won’t go into too much detail about many of the seeming contradictions, but here are a few:

1. Mark 15:25 says “It was the third hour when they crucified Him,” (the third hour is 9 am) while John 19:14 says it was “about the sixth hour,” (noon). People often try to resolve this by claiming that John was using Roman civil time, rather than traditional time-keeping, but we have no evidence that anyone ever used “Roman civil time” to count hours rather than days. Rome counted when a new day started at midnight, but there’s no evidence that this was how anyone counted the hours and there is strong contrary evidence. Note additionally that a crucifixion at 12 had theological significance, putting things near when the Passover lambs were slaughtered, so if the Gospels were distorting details, this is exactly the kind we’d expect them to distort.

2. 2 Samuel 24:1 says that God commanded David to take a census (saying “Go and take a census of Israel and Judah”) while 1 Chronicles 21:1 says “Satan rose up against Israel and incited David to take a census of Israel.” (Satan here might also mean an adversary). Surely whether something was commanded by God or the devil is not some minor, irrelevant detail. Now, generally people suggest God was the ultimate cause but Satan the immediate cause—this seems like quite a stretch, and also contradicts where the text suggests it was a direct command, delivered verbally, from God.

3. 2 Samuel 21:19 says “Elhanan son of Jaare-oregim the Bethlehemite struck down Goliath the Gittite,” while 1 Samuel 17 says that David killed Goliath. 1 Chronicles 20:5 even tries to patch this by suggesting that Elhanan killed the brother of Goliath.

4. Exodus 33:11 says “The Lord would speak to Moses face to face, as one speaks to a friend,” while Exodus 33:20 says to Moses “you cannot see my face, for no one may see me and live.” Exodus 24:9-11 says “⁹ Moses and Aaron, Nadab and Abihu, and the seventy elders of Israel went up ¹⁰ and saw the God of Israel. Under his feet was something like a pavement made of lapis lazuli, as bright blue as the sky. ¹¹ But God did not raise his hand against these leaders of the Israelites; they saw God, and they ate and drank.” Numbers 12:8 says that Moses “beholds the form of The Lord.” The most natural resolution of this is also the dominant scholarly view: the Torah is compiled from a number of different sources, and they often contradict each other. See also Jacob’s declaration in Genesis 32:30 “It is because I saw God face to face, and yet my life was spared,” which conflicts with the declaration in John 1:18 “No one has ever seen God, but the one and only Son.”

5. Matthew 27:5 says Judas “went and hanged himself,” while Acts 1:18 says, “Judas bought a field; there he fell headlong, his body burst open and all his intestines spilled out.” Each has theological reasons for their suggestion. The standard harmonization is that Judas was hanged and then fell from where he was being hung, leading to his insides bursting out. This would be quite a bizarre event—seems much less probable than the stories simply being in conflict.

10. Violent fulmination

A common theme in the Old Testament is brutal and violent depictions of what will happen to Israel’s enemies. Talking about the Babylonians, for instance, Isaiah 13 says:

⁹ Behold, the day of the Lord cometh, cruel both with wrath and fierce anger, to lay the land desolate: and he shall destroy the sinners thereof out of it.

…

¹⁵ Every one that is found shall be thrust through; and every one that is joined unto them shall fall by the sword.

¹⁶ Their children also shall be dashed to pieces before their eyes; their houses shall be spoiled, and their wives ravished.

¹⁷ Behold, I will stir up the Medes against them, which shall not regard silver; and as for gold, they shall not delight in it.

¹⁸ Their bows also shall dash the young men to pieces; and they shall have no pity on the fruit of the womb; their eyes shall not spare children.

¹⁹ And Babylon, the glory of kingdoms, the beauty of the Chaldees’ excellency, shall be as when God overthrew Sodom and Gomorrah.

Isaiah 49:26 adds:

I will make your oppressors eat their own flesh; they will be drunk on their own blood, as with wine. Then all mankind will know that I, the Lord, am your Savior, your Redeemer, the Mighty One of Jacob.

Psalm 137:9 proclaims, “How blessed will be the one who seizes and dashes your little ones Against the rock.” Now, I know that the standard explanation is that this is reflecting rather than endorsing existing anti-Babylonian sentiment popular in Israelite society. But I mean come on—if a passage depicts God saying that he will make the Babylonians eat their flesh and praises dashing babies against rocks, without any indication of it being a profoundly immoral sentiment, this is obviously an error. Especially if this is supposedly relayed from an inerrant prophet of God, as in the case of Isaiah.

11. Even Christians shouldn’t affirm inerrancy

Are there strong theological reasons to affirm inerrancy? I think the answer is no.

Here’s a very common throughline running all throughout scripture: God gives people lots of freedom to make mistakes. The Adam and Eve story, taken literally, implies that God gave people freedom to completely ruin the world. In light of this, we shouldn’t be that surprised that he gave the Biblical authors the freedom to make errors. It’s a bit hard to imagine being confident that God command mass slaughter and gives us the freedom to lie, murder, cheat, rape, steal, but he’d surely never ever allow something to be in Scripture that has errors.

There are proof texts cited for inerrancy. Once you reject inerrancy, as a Christian, you can think they’re wrong. They also come well short of proving inerrancy—saying things like “All Scripture is God-breathed and is useful for teaching, rebuking, correcting and training in righteousness.” That doesn’t entail inerrancy. And besides, the inerrantist will already be used to explaining away problematic passages—far better to do that with the few passages favoring inerrancy than the 50 bajillion passages that seem to be in error.

Lastly, Matthew 5:18 says “For truly I tell you, until heaven and earth disappear, not the smallest letter, not the least stroke of a pen, will by any means disappear from the Law until everything is accomplished.” But saying the law will not disappear, and he didn’t come to abolish it, isn’t the same as saying that it’s strictly inerrant.

12. Conclusion

Insofar as inerrancy is a non-trivial thesis, it is false. The Bible contains a great many different errors: concerning history, morality, authorship, and more. And I’ve only mentioned a small fraction of the ones on offer. Some I haven’t listed yet from just the first two books:

• Historical problems with the Exodus.

• Moral problems with the ten plagues in the Exodus.

• Contradictions between the Exodus and Psalms concerning the plagues.

• Problems with Genesis.

In fact, I claim that pretty much every book of the Bible gives some evidence against inerrancy. Certainly generating evidence against inerrancy from each book is relatively smooth sailing from Genesis at least until Ruth. Just a few of the examples could be explicable on their own (though I think some of the examples are decisive by themselves). But together, they comprise an overwhelming cumulative case: the Bible contains errors.

Nobody embodies this trend more than Candace Owens. Dear reader: however bad you think her public commentary is, it is much worse. Merely watching her show for a few minutes will cause you to lose about thirty IQ points. If Terence Tao were forced to watch it for a few hours, he’d undergo a Flowers for Algernon-esque decline in mental capacities, leaving him unable to read or write. And yet despite the fact that Owens is totally unhinged, she has somehow become a figure of prominence.

Her YouTube show has six million subscribers. She has 36 million followers across platforms. She’s spoken with Hunter Biden and schmoozed with Ana Kasparian in the last month alone. There is an increasing political axis of conspiratorial nutjobs—on both the left and right—who claim to represent some unified populist alliance, according to which Israel and the Epstein class are behind most of the common man’s serious problems.

The way one shows solidarity with steelworkers in Detroit, on this worldview, is to believe any conspiracy theory, however poorly evidenced, and blame all the world’s problems on some ill-defined cabal of shadow elites, who are probably also pedophiles. Members of this newfangled caste of whackjobs—many of whom are obviously shameless grifters—often treat each other as political allies even despite hailing from opposite ends of the political spectrum. Hence the bizarre phenomenon of various self-described democratic socialists expressing support for Marjorie Taylor Greene.

But even among these people, Candace is in a league all her own. To take one example, she has spent the last ten months claiming that Charlie Kirk wasn’t killed by Tyler Robinson (you know, the man who prosecutors claim confessed in writing on multiple occasions to committing the crime and whose DNA was on the murder weapon).

Instead, she claims it was a gigantic conspiracy, likely with Israeli involvement. Ben Shapiro may even have been involved. And she also suspects involvement from Charlie Kirk’s grieving widow and many other TPUSA insiders! Owens is currently being sued for defamation by one of the people she repeatedly smeared.

What is the amazingly powerful evidence for this giant conspiracy? Well, perhaps the most interesting piece of evidence came from the fact that Candace had a dream in which Charlie Kirk told her he had been betrayed. But don’t worry—that’s not all! In fact, this was corroborated by other dreams that other people had. Some of these—shockingly, scandalously!—featured the number six. This is obviously extremely strong evidence that six people were involved.

Does Candace have any evidence that doesn’t hail from the dream world? In fact she does. So bulletproof is her case that some of it is even drawn from waking life. Here is one particularly strong bit: you see, Candace finds it very suspicious that so many men were wearing maroon on the date of Charlie Kirk’s murder. Obviously, however, if you were planning a murder conspiracy, you’d make sure that all the people in on it were wearing matching colors. Note that UVU is located next to, and partnered with, MTECH—a school whose color is maroon! Coincidence? I think not!

Here are some other things Candace Owens—one of the most influential political commentators in America—believes. Dinosaurs are “fake and gay.” The moon landing is fake (as for its gayness, the jury is still out). She has explained, “I'm not a flat-earther. I'm not a round-earther.” Is she agnostic on the subject? Does she think that the earth is some other shape (perhaps a cube?) Does she not know what a round-earther is? She does not say. Similarly, she has claimed that the first lady of France is a man—an accusation for which she is currently in legal hot water.

I know some very smart conspiracy theorists. There are people who believe confusing and tangled webs of alternative history but who can actually defend them, possessing fairly detailed knowledge of lots of facts. I don’t agree with these people, but I at least have a kind of grudging respect for them. Candace Owens is not one such person. Her ideas are ridiculous. It usually takes around one minute of googling to debunk them. She has a pathological inability to correctly state the facts and lies all the time about her past statements.

For example, on one occasion she claimed (falsely) that Leo Frank killed Mary Phagan. As part of her evidence, Owens claimed (falsely) that Jim Conley was an “illiterate janitor,” even though the fact that he wrote the notes found under Mary Phagan’s corpse was a central facet of the trial. You can’t trust the kind of person who gets her evidence from dreams, misstates facts, and occasionally takes to claiming that various people are robots (I’m not kidding). I could go on through many more examples, but hopefully you get the point.

I’ve elsewhere suggested that public intellectuals tend to be a stupid person’s smart person. But Candace is surely a counterexample. Because I cannot believe that anyone finds her intelligent. Even Nick Fuentes is embarrassed by her, worried that she makes antisemitic crazy people look like idiots. Anyone possessing even basic knowledge of the world or rudimentary capacity for rational thought realizes, when listening to Owens’ show, that they are listening to the ravings of a demented lunatic.

There’s just no way anyone can listen to her and think she is insightful. She can’t even pronounce words correctly, ranging from architecture (pronounced arquitexchur), to unequivocal (pronounced unequivobal), to isolationist (pronounced isolationalist). She has even referred to the “statue of limitations.” So ubiquitous and extravagant are her misspeakings that there is an entire X account dedicated to documenting them.

While Owens may be unique in the sheer extent of her craziness, others like Tucker Carlson and Ian Carroll are broadly similar. Carroll—whose show is subscribed to by nearly a million people—recently took to Twitter to declare Owens vindicated in her theory that Kirk wasn’t killed by Robinson. He cited as evidence the absence of exit wounds, before sheepishly admitting that the bullets “may have been” soft-points—designed to expand in their targets, and thus unlikely to leave an exit wound.

So Owens is far from alone. Many other people have a similar shtick: lie, pretend to be some reasonable moderate “just asking questions,” never admit when your old theory was blown to bits, and treat any fact that you’re not smart enough to figure out the explanation for as decisive vindication. Candace Owens being a prominent political commenter is the sign of a diseased political culture, a collective decline in sanity, and ridiculously inadequate public epistemology.

I don’t have any hopeful takeaway or grand thesis beyond this: it’s bad that some of the most prominent commentators in America are certifiably batshit insane. In previous generations, talking the way Candace does might have gotten you confined to a psychiatric ward. Now, it might make you one of America’s most influential podcasters.

Effective altruists advocate giving your money to where it does most good. The leading areas EAs recommend donating to are making the long-term future go better, animal welfare, global health, and growing the effective altruism movement. Yet there’s been surprisingly little comparison between these cause areas. Here, I’ll explain my ranking: why I think it’s best to donate to Longtermist organizations or organizations focused on growing EA. I’ll also explain why I think animal welfare is better to donate to than global health.

Comparing cause areas is important. I sometimes find there’s a weird tendency among EAs to throw out cost-effectiveness when analyzing different cause areas—to treat all causes as equal, and only look at effectiveness within a cause area. This is wrongheaded. If we care about doing the most good, then we should primarily give to the best cause area. The moral imperative towards cost-effectiveness doesn’t go away just because all the cause areas have a generic EA stamp of approval.

In general, I’ll assume you’re broadly on board with the basic assumptions behind each cause area. So, for instance, I won’t bother arguing for Longtermism or against the idea that animal torture isn’t a big deal because they don’t have a rational nature and can’t solve differential equations. That’s because this article is primarily about trying to evaluate what those convinced by the central EA ideas should do, rather than to convince people that these ideas are right. I will, however, link to places where I’ve defended the cause areas in more detail and provide brief summaries of the essential arguments.

TLDR: my ranking for donations: Longtermist political donations>growing EA>other Longtermist donations>animal welfare>global health.

My ranking for careers: Growing EA=Longtermist careers>animal welfare>global health.

1 Longtermism

1.1 The case for Longtermist interventions being best

The case for Longtermist organizations being best is pretty straightforward. The future could be huge. One back-of-the-envelope calculation guessed there could be around 10^58 future well-off people—and another calculation was 20 orders of magnitude greater (see here and here for additional reasons to think this is conservative). There are a number of ways we could screw up the future. Experts consistently think there’s a several percent chance that misaligned AI kills everyone. Despite this, there are only about 1,000 people working on it. Some other crucial cause areas, like preventing AI-enabled coups and improving AI character, have just a few people working on them.

If this is right, Longtermist activities surpass everything else in value. To see this, let’s be generous and say the expected number of future well-off people added by preventing an existential catastrophe is only 10^40—meaning odds are only 1/10^18 that anything like the Bostrom calculation is right. Now, let’s say that ten billion dollars would reduce existential risks by .01%. This strikes me as very conservative. Doing so is enough to hire 10,000 people, which would massively increase the number of people working on the biggest issues.

But even by these very conservative estimates, each dollar then lowers the odds of extinction by one in one hundred trillion. Thus, each dollar increases the number of well-off future people in expectation by 10^26. That’s more than all the people who have ever lived. And remember, overall these assumptions are outrageously conservative.

This might seem surprising. But when you take note of how big the future might be, even tiny effects on it might be enormously impactful. See here for my broader defense of Longtermism.

1.2 Longtermist jobs>>Longtermist donations

Thus, it seems pretty clear that Longtermist donations surpass everything else in expectation. Even with pretty aggressive fudging of the numbers, it easily outweighs other things. Getting Longtermist jobs is especially impactful, because talent is a much bigger constraint than money. See here for some especially impactful project ideas that an individual could work on, and check out the 80,000 Hours job board for high-impact Longtermist careers. 80,000 Hours also has more general career advising.

I want to emphasize: getting one of these jobs is much much much much more impactful than donating. Some very wealthy longtermists are already willing to fund nearly any sufficiently promising project. The greater need is therefore to identify and build additional high-impact projects, rather than merely to raise more money. Donations are still good, but they’re a lot less good.

1.3 Where Longtermists should donate

For Longtermist donations, some good places to donate are the EA Long-Term Future Fund, Longview,1 and Forethought.2 The first two of these provide grants for impactful Longtermist projects. If you think Longtermism is right, then it’s probably a pretty good bet to give your money to smart people who spend their days thinking about how best to spend money longtermistly. Impactful political donations are likely many times more impactful than this (and harder for Anthropic donors to saturate). Note: this isn’t a particularly idiosyncratic view of mine; my sense is that most Longtermist grantmakers would agree with this.

If you are convinced that Longtermism is this important, then you really should try to make an impact in Longtermist areas. Demandingness concerns do not apply when we’re dealing with this kind of positive impact. If the impact in expectation that one person can have by working on Longtermist cause areas is equivalent to preventing more suffering than all the suffering so far in human history, then it’s obviously obligatory. Even if morality doesn’t always demand doing what’s best, it does when the stakes are this high.

2 Growing effective altruism

2.1 Why grow EA?

There are some organizations that are focused on growing the effective altruism movement. Some examples include the EA Infrastructure Fund, Effektiv Spenden, regional EA groups, Giving What We Can, and 80,000 Hours. These tend to be super effective: dollars given directly to Giving What We Can return about 7 dollars to effective charities and dollars given to Effektiv Spenden return about 13 dollars to effective charities.

The case for this being good is pretty straightforward. Many of the people doing the best work on the world’s biggest problems are EAs. EA is generally more talent constrained than money constrained, so getting productive people involved in the movement is valuable. And growing EA pretty straightforwardly dominates global health and animal welfare, because growing EA returns more money to global health and animal welfare. I don’t think it’s clear whether this is higher or lower impact than Longtermist donations. But it seems clear it beats global health, say, when each dollar given to GWWC returns many dollars to global health charities.

2.2 Comparing growing EA to direct Longtermist interventions

Two factors support growing EA over traditional Longtermist organizations:

1. Longtermism is more talent constrained than money constrained. In other words, there’s a bigger problem with a limited supply of talented people than a limited supply of funds. This supports converting money into talent by growing EA.

2. In my view, most future value is lost in worlds where we survive but fail to get a near-optimal future. If this is a bigger threat than specific catastrophes, then growing the influence of those motivated by doing what’s best is hugely important.

One factor that points in the opposite direction: we might be at an unusually important time given very rapid AI technological advancement. It may be that work in the next few years will be more important than work in the years after that. Thus, growing EA—which often takes years to pay off—won’t be worth it in time. On the other hand, we should have reasonable uncertainty about when AI will arrive, so there are still decent odds that superintelligent AI arrives after growing EA pays off.

Another advantage of promoting EA is that even if you’re wrong about what’s best, you still can contribute to whichever area is most important. So even if I’m wrong that Longtermism is right, I can still know that I’m preventing people and animals from suffering and dying. I suspect that this moves me more than it should.

My general guess, if I had to rank these things gun to my head, is that the ranking for donations would be something like:

Longtermist political donations>EA promotion>Other Longtermist donations.

And for careers, it’s too close to call, though I’d tentatively prioritize specifically Longtermist careers over EA promotion careers that don’t focus mostly on Longtermist stuff. Though things are close enough that the main things you should consider are personal aptitude and fit. While my guess is that pretty much any net-positive Longtermist job will beat the best global health job in expectation, the same isn’t true here.

3 Animal welfare and global health

3.1 The basic case for animal welfare

The basic case for animal welfare over global health is that it is ridiculously cheap to help animals. Charities helping chickens prevent about 10 years in a cage per dollar they raise. This pretty clearly dominates global health’s direct impacts on humans. It costs about 4,500 dollars to save one life. Let’s assume the saved person would have lived an extra 40 years.

Thus, for 4,500 dollars you can either add 14,600 days of human life or spare chickens from 45,000 years in cages. Even if we assume that the ten-year estimate is inflated by one order of magnitude, then it still comes out to sparing a chicken from a year in a cage or adding about three days of extra human life. Surely sparing the chicken from a year of nightmarish torture is better?

And note: many animal welfare charities are even better than this. Given the near total neglect of invertebrates and wild animals, donating to help them can be even more impactful. And when one takes into account the long-term effects that neglect of animals—farmed and especially wild—would have, this seems quite good from a Longtermist perspective. We might spread factory farming throughout the galaxy, and we’re even likelier to spread wild animal suffering.

3.2 Longtermism and growing effective altruism>animal welfare

Let’s first compare animal welfare donations to Longtermist donations. Animal welfare, I’d guess, is lower stakes than Longtermism. Nearly all expected future welfare will be experienced by digital beings. This is both because of the amazing possible efficiency of compute3 and the ease of converting large amounts of energy into well-off digital minds. Given how many digital minds there could be, this holds even if you have a low credence in the possibility of digital sentience.

Thus, if you’re interested in steering the long-term future, you should focus more on digital welfare than animal welfare. This favors Longtermist interventions over animal welfare interventions, because they have a much more direct route to benefitting digital minds.

The case for growing EA over animal welfare is similar.

1. Its indirect effects are greater. Getting hard-working people on board with animal welfare probably has better second-order effects than direct work on animal welfare. It leads to more people who are on board with reducing animal suffering, which is important for eventually abolishing factory farming and seriously mitigating wild animal suffering.

2. Growing EA also grows animal welfare—my guess is that each dollar spent growing EA returns multiple dollars to effective animal charities.

3. Growing EA has similar benefits to Longtermist donations—benefitting digital minds, in expectation, which likely swamps the benefits of animal welfare charities. Thus, if Longtermism beats animal welfare, probably growing EA does too.

3.3 Animal welfare>global health

Now let’s compare animal welfare to global health. The best argument for global health over animal charities is that global health charities lower wild animal populations. I’ll first explain this argument and then say why I don’t buy it.

There are strong reasons to think that animals in nature live bad lives. Most animals, especially the small and simple ones that are most numerous, have very short lives that end in a painful death. For this reason, I think it’s more likely than not that it is good for wild animal populations to be lowered. Humans lower wild animal populations to a staggering degree, especially populations of insects. Brian Tomasik estimated that the average person prevents about 14 million years of insect life per year they’re alive.

Similarly, Tomasik estimated that the Against Malaria Foundation prevents about 14,000 insect life-years per dollar. Let’s assume chickens are 10 times more conscious in expectation than the average insect. Let’s additionally assume that insects in nature on average live about 2 weeks. This means that each dollar given to the Against Malaria Foundation prevents about 392,000 insect deaths, which are mostly pretty painful. Adjusting for sentience, this ends up about as good as preventing around 39,200 painful chicken deaths. Let’s cut this by a factor of ten to adjust for uncertainty about the net effect of global health charities on insects.

Which is worse: ten years in a cage or 3,900 painful deaths? Well, that’s about one death per day. It’s not obvious how a day in a cage compares to a death from starvation or predation. So for this reason, I’m pretty uncertain about the direct comparisons.

However, the expected second-order effects of animal welfare donations are a lot more positive than those of GiveWell donations. Here is why: nearly all animals, in expectation, will live in the far future. More importantly, almost none of them will live on earth. There are about 10^21 stars in the reachable universe. Even if you think it’s pretty unlikely that we’ll settle many of those planets, one-in-a-million odds that we’ll settle 1% that many planets still means that we’ll settle 10^13 planets in expectation.

Compared to this, whatever long-term environmental effects GiveWell charities have on earth is an insignificant blip. If we’re concerned about benefitting animals, what matters most is ensuring that their lives are good outside of earth. It’s easy to see how changing legal standards for animals and shifting cultural attitudes might do that. It’s hard to see how GiveWell charities would do that.

I’ll write an article about this at some point, but just as nearly everyone who will live, in expectation, will reside in the far future, almost everyone in expectation will reside in distant star systems. The universe is big (source???). Thus, when trying to make the world better, just as we should care a lot about making sure that the vast distant temporal regions go well, we should also mostly focus on trying to make the vast distant spatial regions go well. That matters a lot more than what happens on earth. Thanks to Toby Ord for pointing this out. When you take this seriously, animal welfare impact-mogs global health.

4 Conclusion

Comparing cause areas is important and difficult. The thing I’m most sure of is that EA promotion and Longtermism beat global health and animal welfare. I’m also pretty sure that animal welfare beats global health in expectation. In any case, all the areas are hugely important, and people who give to one tend to give to others as well, so just generally giving EA charities more money is very good!

It is a tragedy that we have to make these sorts of tradeoffs: that there is some chance that future life is extinguished, that there’s a good chance that it doesn’t reach its full potential, that animals suffer by the billions in ghastly torture chambers of human construction, that more suffer in the wild, and that children die every day. But when faced with many horrible things, one has to make tradeoffs; one should do it sensibly, after reflecting carefully, instead of just giving to the charity that they like best.

I think the answer is yes. Even if you agree with a candidate politically, if they are morally rotten, you shouldn’t support them. Character matters in a candidate. But it’s not for the standard reason people think.

The most important facts about a candidate are what they will do politically. The harms of an elected official being personally nasty—or even being emboldened to commit serious crimes—pale in comparison to the huge effects of their policies being enacted. Even if electing some poorly behaved politician would embolden them to go out and commit several murders, this would be only a fraction as significant as the effects of their policies. And even if there is some good in retributively punishing people for misconduct, that pales in comparison to the benefits of passing good policies that benefit millions of people.

So character only matters insofar as it affects what a person does while in office. But character makes a big difference with regards to what politicians do. Even if someone is aligned with you politically, insofar as they’re a terrible person, they’re much likelier to enact bad policy.

Take the recent Save Our Bacon Act as an example. I don’t actually believe that our Congresspeople are unaware that locking pigs in tiny cages where they can’t turn around is cruel. Rather, they supported the bill because they were selfish and prioritized their own political advancement over protecting innocent creatures from torture. The worse a person is, the likelier it is that they’ll behave immorally on matters of consequence. If our elected leaders were all moral people, then the SOB act would have been dead on arrival.

Take another example: politically, I’m a lot more aligned with Bill Clinton than George Bush. But George Bush seems like a nicer guy. Clinton bombed Al-Shifa for cynical political reasons, even though he knew that this would cause lots of innocent people to die. Similarly, Clinton privileged his own political advancement over the lives of the Rwandans and as a result greenlit the genocide that may have in total led to millions of deaths.1

In contrast, Bush was mostly motivated by altruism when he set up PEPFAR. This decision led to tens of millions of lives being saved. Bush’s PEPFAR decision might have been so good that despite his repeated horrible policies—e.g. the war in Iraq—he may be, by utilitarian lights, the best president of my lifetime. Trump, by contrast, is perhaps the single most morally irresponsible man on the planet and callously undermined much of foreign aid.

If good character can redeem Bush, it matters a lot.

Cases where good character matters tend to be uniquely important. If you do things that are bad for Americans, you tend to pay a political price for it. So cases where political expediency conflicts with morality tend to be cases where one’s victims aren’t present Americans—and they often are very numerous.

The most consequential political decisions mostly affect foreigners. Americans are well off by global standards. Bad policies generally do not lead to Americans dying in large numbers. Bad economic policy whose victims are American might make well-off Americans a bit poorer; bad foreign policy might lead to babies being beaten to death with clubs or dying in their beds from lack of medicine.

Another case where there is a conflict between political expediency and morality is in cases where bad policies look good. For example, imposing pharmaceutical price controls looks like an obvious no-brainer. It’s not. Pharmaceutical profit margins tend to be very tight and come mostly from a small number of extremely profitable drugs. If firms make less profit from new innovation, they’re incentivized to spend less on life-saving new drugs. This has a truly calamitous effect on global innovation—studies often find pharmaceutical price controls would, over the course of decades, shave years off people’s average lifespan. Scott Alexander estimated that the number of lost life years from pharmaceutical price controls would be greater than the number lost from Mao’s great leap forward.

Suppose you’re a politician who knows this. You’ll benefit politically if you impose price controls. The victims—numerous though they will be—will be indirect and will live in the future. Politicians pay no costs for filling invisible graveyards.

This is merely one illustrative example of a more general point: there are lots of bad things that politicians can do. Though political incentives often eliminate the most egregious forms of misconduct, they aren’t enough to ensure that politicians do what they believe to be right. And when politicians do things that they know to be wrong, these often have extremely bad effects.

Now, here is one other feature of supporting virtuous politicians: we’re all better off if we all do it. Roosevelt is rumored (probably falsely) to have said of the president of Nicaragua “Somoza may be a son of a bitch, but he's our son of a bitch.” If both parties support their own side’s rogues and scoundrels, everyone is worse off. Neither party benefits politically—we just get the halls of congress filled with sons of bitches. Instead, we’re all better off if we don’t support sons of bitches, even if they agree with us politically.

Things may be better by the lights of Democrats if Platner wins than if he loses. But if both Platner and Roy Moore win, then everyone is worse off. Thus, it is good to have a political norm against supporting those who do terrible things. And you should be very wary about breaking good norms.

For this reason, even if I supported Platner’s political views (I don’t) I’d still call for him to drop out of the race. Politics is worse if our elected leaders are bad people, because this leads to them knowingly doing bad things. We ought not sell out our principles for political expediency. I’ll close by quoting ’s response to Ben Shapiro—when Shapiro claimed that Trump’s attempted coup was not disqualifying because “the guard rails held.” It’s not exactly the same, but the basic principle is similar.

Let me tell you how I view this. Say you’re on a bus ride on a winding mountain road. You see the driver suddenly swing the wheel to the right, trying to send the bus over the cliff. Fortunately, the guard rail on the side of the road holds, and the bus bounces back onto the road. The bus driver does this repeatedly during the drive, but every time, the guard rail holds the bus back.

When you finally get off the bus, one of your fellow passengers declares that this was an excellent bus driver. He proposes hiring this driver to drive the same group to another city. “What are you, out of your f—ing mind?” you reply. “He tried to drive us off a cliff!” “Oh that,” says the other passenger. “The guard rail held, so what’s the big deal? Don’t worry, this next drive won’t go by a cliff. Since the rest of his driving performance was fine, we should hire him.”

That guy is Ben Shapiro.

Do I have to spell it out, Ben? Driving off a cliff is not the only bad thing a bus driver can do. There is an indefinite number of disasters a crazy person can cause. Anyone who would try to drive a bus off a cliff can never be trusted with a bus, or indeed anything else, and if you think he’s an acceptable driver, you’re as crazy as he is.

1. I’ve mentioned this before on a few occasions, but I can speak backwards fluently. If you give me a word or sentence, I can say it backwards in full immediately. Two clarifications: 1. I flip around the order of the letters in each word, not the order of the words. So, “my name is Matthew” would be “ym eman si wehttam.” 2. This gets harder with longer words. I start to have trouble with words around fourteen letters (I did once figure out that antidisestablishmentarianism backward is msinairatnemhsilbatsesiditna, which I’m pretty proud of—though that took a few minutes). I’ve always been able to do this, never practiced.

(Demonstration—not sure why my arms are so strangely positioned lol).

2. I am extremely goofy—probably top 1% goofiness. I really like dumb jokes and probably make too many silly comments. My favorite comedian is Mitch Hedberg—my favorite joke is the following:

On a traffic light, green means go and yellow means yield. But on a banana, it’s just the opposite. Green means “hold on” and yellow means “go ahead.” And red means “where the fuck did you get that banana?”

I also really like the show Impractical Jokers (if you haven’t seen it, you must see Murr’s food tasting). The kinds of conversations I enjoy most are either in-the-weeds discussions of interesting topics or complete silliness—don’t like as much conversations in between.

3. I have an extremely spiky skill distribution. While I’m naturally pretty good at philosophy, and economics was also always pretty intuitive to me, I’m not great at math, was beyond terrible at learning foreign language, and don’t have much general historical knowledge. I’m pretty sure my middle school Spanish teacher thought I was mentally challenged at the same time my English teacher thought I might become some kind of major public intellectual.

4. I think babies are very cute! But I don’t like dogs, in general, except my cousin’s dog who is, by all accounts, the best dog on earth.

5. I have what once called “argument brain.” I find it hard to remember facts about the world unless they are relevant to some ongoing debate. I’ve forgotten most of what I’ve learned in history classes throughout the years, but I will never forget any of the arguments for how we know that humans wiped out the megafauna; this is solely because some people are so silly as to deny this.

6. I’m generally pretty happy most of the time.

7. I have a really awesome family! My parents are both very smart and reasonable; I’ve talked them into being vegan and giving some of their money to effective charities (though really before going vegan I was the main meat eater in the family, so they didn’t take that much convincing). My brother is great too—also an EA. I even have amazing grandparents! I also have very epic friends (the world has fewer better things than close friendships with and !)

8. In middle school I was politically libertarian and started a blog where I wrote a few posts. It wasn’t very good though and my thinking was often confused. Still, fun to occasionally look back on.

9. At every point in my life I’ve had one all-encompassing interest. The order: trains—>superheroes—>Pokémon—>Yugioh—Magic: The Gathering—>high school debate—>philosophy. This has consistently been the thing I’ve spent most of my time on.

10. I currently live in Oxford but I’m starting a Ph.D in philosophy at Princeton in the fall. Oh, and I recently published, along with , the first ever paper on the Anthropic argument for theism. Now that the existence of God has been proved, hopefully philosophers can move on to other topics.

Juicier info below the paywall.

Read more