Lately, I’ve been working on an essay that attempts to these qualms—it was first conceived as an essay that would make videogames of interest to the philosophically inclined. But the material which might achieve that aim comprises only perhaps a quarter of the essay, which is quickly becoming one of my very longest. That quarter, I’m sure, can be shaped into something good—but I don’t know if it’s good enough to justify the rest. The problem isn’t that I think videogames aren’t fitting subjects for narrow analysis—I very much do. The problem, rather, is that I primarily write philosophy, and I just don’t know if my narrow analysis of videogames is really any good.

And so, a prototype: I’m going to write you a more modest piece about a game that I lately haven’t been able to get off my mind. See, I think there are lot of people with patterns of thinking very similar to mine—especially people who are somewhat older than I am—who *could* really enjoy videogames, but haven’t yet given them a proper try. Or, if they have given games a try, perhaps they just haven’t sampled the right ones. The three games I’ll be covering in my longer piece are probably not the ones I’d recommend to such a person. But my current fixation—The Message from Deep Space—absolutely is. And TMFDS—the work of solo developer Applesinmypants, who I’ll henceforth call “Apples”—is striking me more and more as a game that all the clever intellectuals of Substack ought to give the time of day. And so I figure that, even if this essay winds up falling horrifically short of the bar I’ve set, it will accomplish the important public service of spreading the word about this truly *awesome* game.

Okay, curtains closed. Let’s get started:

The Message from Deep Space is an unassuming title. It is not visually impressive. It is not, at a glance, mechanically impressive. It *is,* however, remarkably accessible. It has a free demo, it can be played on a MacBook, and the full version costs only ten dollars. So if anything I say in the subsequent paragraphs intrigues you, remember that the barrier to entry here is low. You can close this essay and give the game a shot any time you want.

So, from the top: The year is 1973. A foreign body has made contact with the Earth. A meteorite, yes, but not a natural one—*this* meteorite is host to a metallic, yet almost muscle-like growth, a mass of pseudo-biological material of a sort not known on Earth. And emerging from the metal muscle is a transmitter—one built to operate in the conditions of any conceivable atmosphere. This meteorite is not inert—the transmitter is active, and this hunk of rock contains the titular message.

The message needs to be studied. It needs, if possible, to be translated. This is your job. You are the lead of the Translation team, a four-man unit composed of yourself, Dr. Alan Akers, Dr. Carrie Collins, and Dr. Bautista.1 Akers is the resident astronomy expert, Collins is there for linguistics, Bautista works the computers, and you do the real work of translating, of interfacing directly with the alien machine.

At first, that work is simple. It consists in reading a number off a screen, typing the same number back, waiting for the next transmission to come in, and doing the same again.

You only get the numbers for free, however. The numbers you’re given represent incremental deviations from a “zero” frequency, but while a ten-unit2 shift in one direction does indeed represent the number 10, a ten unit-shift in the other direction does *not* represent -10—it doesn’t represent a number at all. Take, for instance, “-2.” This is the first negative number you’ll encounter, and it’s used only for formatting. I translated it as “_”, and the other negative numbers you’ll encounter early are best translated as “=”, “+”, “-”, et cetera.

I say “best translated” because this game will not give you the answers. The computer can take the definitions you give it and render a mess of numbers into a coherent equation, but whether you give it the right stuff is *entirely* up to you. And that’s the promise of the game—that you will be able to go from arithmetic to geometry to logic to chemistry and physics and all the way to the genuine article—proper, intuitive natural language—using only numbers. Using just a computer massive assortment of frequencies, you will be able to understand where the aliens come from, how their society is structured, what sort of technology they have, why they’ve tried to contact us, and where the frontiers of their science lie. And of course, there will be more beyond this. From numbers, you will assemble an alien world.

I want to emphasize that I do not myself know all the answers to the natural questions. Though I’m quite a ways past the midpoint, I also still have a good deal of game left to play, and so I can’t yet honestly testify to the quality of this game as a complete product. I can, however, tell you that what I’ve played so far has been *enthralling.* Some of, you, I’m sure, have been similarly hooked since the moment I first described the premise. If that’s you, I suggest you quit reading and just go play the thing.

For the rest of my readers, though, I’d like to say somewhat more about what makes this game so special—in so doing, I think I will be able to say something about what makes games *as a medium* so special. And we can start here with a natural comparison to some work in another medium:

In the movie Arrival, Louise Banks—played by Amy Adams—works with ???—played by Jeremy Renner—to translate an alien language. Forrest Whitaker is the army guy. It is a testament to the talents of Denis Villeneuve—and indeed all the filmmakers behind the scenes of Arrival—that I think of this movie so fondly even though only a single character’s name has stuck in my mind. And it is truly a remarkable film. But, as those of you who have seen it will know, it does not have very much to teach you about translation, nor does it give you a convincing sense of what it might really be like to decipher an alien language.3 This is because Arrival, at bottom, is not concerned with the translation of alien language. Translation, rather, is vehicle for the more philosophical ideas that the movie really wants to explore—philosophical ideas about time and free will and the worth of fleeting experiences. But since translation *is* the vehicle, the movie can’t handwave it completely. Thought it doesn’t need to teach us anything, it does need to convince us that Amy Adams knows what she’s doing. Here’s one time-tested way of doing this: sell us the idea that she knows something we don’t. Have her say things that the audience interprets as meaningful and yet can’t quite follow. This approach can be enhanced if we introduce other characters who also seem to know things we don’t—they can show us that our protagonist is perfectly capable of keeping up with them. Essentially, you can make it seem like all the smart characters in the movie are in a genius-club, a club such that the average member of the audience won’t meet the requirements for membership.

Now, let’s think about how to apply that model to a videogame. It’s possible, but in thinking about it you’ll see very quickly that it just doesn’t really work. If I were to make a game where the protagonist is a computer programmer working with a team of other computer programmers, I might be able to write the team in such a way that they—from their dialogue alone—came off as clever and knowledgeable, as people who really know their stuff. But remember, this is a game. I’m not asking you to watch the protagonist, but to *be* the protagonist—or, at the very least, to play the role of the protagonist. And if I’m playing the role of the protagonist, creating the sense that the protagonist knows their stuff is much more difficult. It’s not enough for the character I play to talk like a programmer—the puzzles I’m doing also need to *feel* like computer programming. An important note here is that this doesn’t mean they need to *resemble* real-life computer programming. For a quick case-study, let’s look at two different gameplay elements from 2018’s Marvel’s Spider-Man.

First, here’s a still image of a minigame in which Peter Parker assembles some circuits:

Second, here’s a video of Spider-Man fighting some goons. Click the link and watch about twenty seconds from the place I’ve started you:

Now, the important thing to notice is that the first type of gameplay by no means resembles the real process of assembling a circuit—you can tell that just by looking at it. But by the same token, the second kind of gameplay by no means resembles a real hand-to-hand fight. And yet, I bet you can tell *just* from that one still image that the circuit assembly minigame doesn’t even *feel* like circuit assembly. At the same time, I bet that short clip of combat gameplay is enough for you to see that…well, that playing Spider-Man really makes you *feel* like Spider-Man.4

You could say that this is because of the spectacle—the gameplay I’ve shown certainly looks like a cool fight even if it doesn’t look like a *real* fight. And presentation does have a lot to do with it—that’s the point I was making with the circuit minigame.

And yeah, the presentation in The Message From Deep Space is on point:

But there is, I think, quite a bit more to it—there are other ways in which Spider-Man’s circuitry fails and its combat succeeds. See, this is something you can’t know just from watching the gameplay, but fights in Spider-Man don’t merely *look* like Spider-Man fights. They also demand that you express skills which are, to some extent, similar to the skills that you’d expect Spider-Man has to express when he throws down. Not athletic ability, of course, but timely reactions, improvisation, knowing when to be defensive and when to push the attack. When we play the game, we don’t express these skills in the way that Spider-Man would, nor do we express them to nearly the same extent, but we do express them, and I hypothesize that this is what sits under the surface of the countless reviews proclaiming that playing Spider-Man really makes you feel like Spider-Man.

So this is the challenge for The Message From Deep Space. It doesn’t need to convince us through dialogue that the protagonist is a capable translator—the translator in question is actually a mute self-insert. But that’s fine, because here’s what the game really needs to do: it needs to give us challenges which demand expression of the skills that we’d expect to see a real translator express. It needs to test our ability to see the connections between one word and another, our appreciation of the nuances in meaning, our willingness to experiment with different definitions and stress-test their viability. Because we’re translating an alien language, it needs to make these tests difficult—we need to feel that our skills are being pushed beyond their ordinary limits. But also, it needs to make sure that these are tests we can *pass.* If the game demands too much of us and we feel forced to scour the internet for answers to our questions, there will be no feeling of satisfaction in our progression through the game. Likewise, we won’t feel that the protagonist really knows how to do their job.

Now, as I write this, I’ve translated more than 800 alien transmissions. Just after writing that sentence, I experienced a bit of mental whiplash. I set down the period, and then at once felt the urge to add something like: “not *real* alien transmissions, of course.” If I tell you that I’ve killed a boss in Sekiro, I experience no such dissonance. In playing the game, I certainly get caught up in the illusion, but not to the extent that I’d feel the need to say that I didn’t *really* decapitate a twenty-foot ape with a sword the size of my body. So Apples has achieved something special with TMFDS—I’ve never played a game where I’ve come closer to feeling as if I were *actually* doing what my character was doing. And yet, the hardcore immersion doesn’t come at the cost of hardcore difficulty. Of the eight hundred puzzles I’ve thus far done, there were only three that had me genuinely stumped, and in one case that was only a function of my failure to read carefully. So as far as the previous paragraph’s criteria are concerned, the game passes with flying colors.

A contributing factor here is the game’s clever handling of hints. See, as I mentioned early on, the translator isn’t working alone. When you’re in the room with the computer, you’ve basically always got Bautista, Collins, and Akers with you. As you progress through the game, they’ll comment on major developments, but they’ll also help you out in two important ways. First, after you’ve completed a sequence of puzzles, they’ll usually each offer hypotheses concerning the meanings of the new words you’ve encountered. What’s delightful here is that they don’t always converge on the word’s meaning, and almost never on the best way of expressing it. Bautista, being the computer guy, usually offers the most precise translations of logical or mathematical symbols, and also wants everything expressed using as few characters as possible. Collins, being the linguist, is the best person to consult once you’re mostly done with math and you start working through transmissions that are closer to natural language. And Akers, for his part, is the guy you want if you need a name for some celestial object. You only get these hypotheses after you’ve completed a whole set of transmissions. If, however, there’s something you’re really stuck on, you can attempt to brute-force until you make it to the other side, and then get a rush of sweet retroactive understanding as your colleagues chime in to help fill the gaps.

Likewise, there are some particular puzzles during which your colleagues will notify you that one of them has an idea. Afterwards, you can prompt them to tell you at any point, and get a hint for your current puzzle. These hints work well because you don’t *have* to use them, but also because they’re always formatted as a conversation between your colleagues. Akers might chime in with a wholly implausible suggestion, only to be chided by Collins and mocked by Bautista. But maybe Collins then has a burst of inspiration, furnishing you with a clue to the puzzle’s real solution. The back-and-forth makes the hints more natural, engenders a feeling of real camaraderie and shared purpose between you and your fictional colleagues. And the colleagues in question really are quite endearing—it’s a function of how the game is written. True, the emotional moments might have been handled more subtly, crafted with a defter touch. But they *do* touch the player, because by the time they roll around the player is already intimately acquainted with these characters—or feels that way, at the very least. Take, for instance, Akers and Bautista, who share an affinity for a well-ordered dictionary. When there are gaps—when the player has seen signals -8 and -10 used but not -9, for instance—the two of them will never fail to make some remark. And when the gap is filled, they’ll follow up. “Happy dictionary,” says Akers, and Bautista replies “Happy life.”

The “happy dictionary, happy life” joke is repeated time and again, but it never quite wears thin. This is partially because Apples takes care to add some little twist each time the bit is repeated. But more importantly, the joke works because Collins, for her part, just doesn’t get it. I’m reminded here of an insight from Orson Scott Card’s5 introduction to his Speaker for the Dead, one of my very favorite novels:

The difficulty of creating a character increases with each new major character that is added to the tale. Characters, as most writers understand, are truly developed through their relationships with others. If there are only two significant characters, then there is only one relationship to be explored. If there are three characters, however. there are four relationships: Between A and B, between B and C, between C and A, and finally the relationship when all three are together.

But Card—admirably as he writes the intertwined relationships of a large cast of characters6—does not, I think, articulate the whole picture here. If I write a cast of three characters, I’ll of course have to concern myself with the four relationships Card lays out. But the relationship of all three characters together is not quite like the relationship of one character to another—it contains sub-relationships. And so, as I go about writing, I’ll *also* have to concern myself with A’s relationship to the relationship of B and C, with B’s relationship to A+C, with C’s relationship to A+B. So seven, really, rather than four, but this only makes Card’s point more acute. Unfortunately, most *videogame* writers either do not understand what Card understands, or are unable to put this understanding into practice because the design of their game will not cooperate with them. The exceptions are a very precious few, and The Message from Deep Space is one of them. Collins has her dynamic with Akers, and a separate dynamic with Bautista, and the group has a collective dynamic, but Collins also has a particular way of relating to the way Bautista and Akers relate to each other, and so on.

And I think Apples, if they were to transpose their work here into a sort of experimental novel, would find that the success of these relationships as literary objects is largely dependent on the medium in which they’re constructed. They would find, I think, that the characters are so endearing largely as a function of their involvement in your central task. The jokes Akers cracks, the sentimental remarks of Collins, Bautista’s expressions of concern for the computer’s well-being—all of these occurrences punctuate discovery, discovery in which Akers, Collins, and Bautista have themselves aided. I said that the “happy dictionary, happy life” joke works largely because Collins doesn’t get it, because it lets us see how she looks at Akers and Bautista. This is true. But it *also* works because every time Akers and Bautista joke, you actually have just filled in a space in the dictionary, and probably they’ve helped you plenty in doing it.

And so, if Apples were writing an experimental novel, they might have been able to create the sense that Akers, Colllins, and Bautista are friendly co-workers, partners in science who love the truth and come to love each other too. But they would not have been able to create the sense that Akers, Collins, and Bautista are *your* coworkers, *your* partners. That familiar, powerful feeling is one that videogames are uniquely well-suited to provoking. For my part, I have made many of my most cherished friends while doing work that I was fiercely passionate about, work that in itself gave me the deepest possible satisfaction. And if you, too, have ever made friends in this way, you are likely to find a powerful simulacrum of that friendship in The Message from Deep Space. The work—as the best work does—will unite you with your fellow scientists not only mind-to-mind, but hand-in-hand and arm-in-arm. And likewise, it will unite them with each other.

Let’s round the essay out: if, thus far, you’re already interested in playing this game, please just quit here and go play. It’s cheap and accessible, and there are so many surprises in store. For those who are not yet sold, however, I want to give you a condensed tour of the first three acts of the game—I’ll do my best to give you a sense of how it really plays, to demonstrate that it does, in fact, *have the juice.* Don’t worry, though—the game has twelve acts. Even if you read everything that comes next, there’ll be plenty left for you to uncover:

We begin with a call and response game. “1420405752,” the alien computer tells you, and “1420405752,” you say right back. This number stands in for the frequency of the hydrogen line—the frequency of the radiation released by hydrogen atoms when their electrons change spin. The next few transmissions are the same number plus or minus a few, until final you get this:

6

-2

5

-2

4

-2

3

-2

The expected response will, I think, be obvious enough. What’s curious, though, is the addition of this -2 character—the result of the frequency shifting lower instead of higher. And here, as they say, is where the true The Message From Deep Space Starts.7 For—as you may have inferred—“-2” does not really represent a negative number here. But what it *does* represent is up to you. See, after a few more call-and-response type puzzles, Bautista pushes an update to the computer—from now on, you’ll be able to decide how the computer renders “-2.” For my part, I currently have it marked down as an underscore, but I originally used “STOP,” which was less effective. But that’s the point: this is a symbol that’s used purely for formatting purposes, so you could translate it in a variety of ways, and you’ll want to pick the most expedient option. The next symbol you’ll encounter, on the other hand, means something with a much more obvious translation. I’ll render the transmission with “-2” translated as “_”, and leave you to guess the meaning of the new symbol.

2

_

3

-5

5

_

_

6

_

1

-5

7

_

_

1

_

4

-5

And then you input the next number, which—as you may have guessed—is 5. That’s because “-5” really means “+.”

From this point, you’ll proceed to define the rest of the basic mathematical operations, to recognize the Fibonacci sequence, to cope with some quirks in the alien counting system. In short, you will come to realize that you and the visitors really do share a common language—or the tools, at least, for building one. And with these simplest tools in hand, having given the correct response to 74 total transmissions, you proceed to Act II. You’ll here work with basic 2D geometry, the contours of simple shapes. You’ll learn how to round to a certain number of decimal places. And you’ll learn also—as I’m sure my regular readers will be excited to hear—the basics of propositional logic. The game won’t ever have you doing any really advanced logic, but there’s a rush of excitement just in learning that a mechanical emissary has come to you from a foreign star, and it is bearing not only knowledge but *your* knowledge. The aliens understand the entailment of “P” and “if P then Q” in just the way that you can.

But there are more than new words for us to play with here. Act II also gives you the ability to tweak the formatting of the new words—to decide how they’ll appear. So suppose, recognizing the roles they play in the formatting, you have signals -14 and -15 translated as “(“ and “)”. Suppose also that you translate -16 as “P”(short for “point” and -17 as “LS”(“Line Segment”). Suppose finally that you identify -18 as “SHAPE.” Then suppose the game describes a square to you. In the default formatting, here’s how this might look.

SHAPE

0 (

LS

0 (

P

0 (

0 ,

0 ) ,

P

1 (

0 ,

1

)

)

,

LS

1 (

P

1 ,

P

2 (

1 ,

1

)

[…]

And so on until we’ve accurately described all four sides. Horrendous, yeah?Play with the formatting settings, however, and you can collapse this forseeably massive text crawl into a package which is very tidy indeed:

SHAPE0(LS0(P0(0,0),P1(0,1)),LS1(P1,P2(1,1)),LS2(P2,P3(1,0)),LS3(P3,P1))

And thus we’ve got everything we need to draw a perfect square. Of course, we don’t stop at squares. The circle will be the first shape we properly define, and soon after that we’ll move into three dimensions. Act 3 introduces the “Visual” tool, which can create a 3D rendering of spheres that the alien transmission describes:

image of the visual tool rendering an Act 3 transmission

Aided by this tool, the alien computer uses spheres to make a variety of instructive shapes. But the most instructive by far is the one that comes at the very end of the act. To set the stage, let’s take a look at what Akers, Collins, and Bautista have to say to Doppler—their supervisor—at the end of the act:8

Collins: “Doppler, it’s quite amazing what we’ve seen. It’s something you should see for yourself sometime.”

Doppler: “What is it, Collins?”

C: “It all depends on our interpretation and confidence in signal -45—[the translator’s] interpretation of UNKNOWN.9 Because we believe the METEOR has sent us a portrait of UNKNOWN.

D: “Woah.”

C: “Incredible, right?”

Bautista: “The render is quite low in detail, bound by the amount of signals it takes to produce a VISUAL transmission. The transmission was over 1600 signals.”

D: “May I see it, the visual you rendered?”

C: “Alan took a picture to show you!”

Akers: “Here ya go boss!”

D: “Huh. So that is UNKNOWN. If only we were confident in our understanding of signal -45…”

C: “That brings us to my hypothesis! If we were to create a message with visual capabilities, what would be our first message to share?”

B: “A circle to calibrate the resolution.”

A: “I’d send planet Earth, our sending address.”

D: “Those are both sensible ideas. But Collins, I sense you have something else on your mind. What is it? What would be your first picture to send.”

C: “A picture of a human.”

D: “…”

[…]

ACT III — COMPLETE!!

“TO SEE WHAT WE SAY”

ACT IV BEGINS…

So what *do* they look like, then?

That discovery, Translator, is yours to make.

According to the coolest form of constructivism, Kantian constructivism,1 rational agents are obligated to act in certain ways in virtue of stances that they *necessarily* take. If this theory is true, there are certain stances such that it’s impossible to be a rational agent without taking them, and so there are some obligations which all rational agents have. Then we’ve got a morality that is wholly stance-dependent, and yet in a certain sense objective—you are obligated to act in certain ways whether you want to or not, and this is not because of some feature of what it is to be human, but instead because of some feature of what it is to be *rational.* The upshot is that rational aliens would have many of the same obligations we do, and so all beings that are capable of acting morally would be required to do so if the theory succeeds.

Now, there’s much to be said about this theory: about whether it succeeds and whether it could succeed. But instead I want to ask a question that I haven’t yet seen probed in quite the way I intend to probe it: If we do have the sorts of obligations that Kantian constructivism says we have, would it really be appropriate to call these *moral* obligations?

See, I started out by asking us to suppose that there are no stance-independent moral truths—in other words, to suppose that moral realism is false. And indeed, the various forms of constructivism are typically understood as *alternatives* to moral realism. But unlike other conventionally antirealist metaethical positions—nihilism, say, or naïve emotivism—constructivism isn’t *necessarily* anti-realist—it does not entail that moral realism is false. For if the Kantian arguments succeed, their success has nothing to do with the existence of stance-independent truths. Kantian constructivism is thus totally compatible with realism, and it could be that we are rationally obligated to act in certain ways because the Kantian arguments succeed, and *also* that we are obligated to act in certain ways in virtue of objective, stance-independent moral truths.

As I’ve said, my goal in this essay is not to evaluate Kantian constructivism, and neither do I intend to pass any judgement on stance-independent moral truths.2 Instead, I want to simply suppose that *both* Kantian constructivism and moral realism are correct metaethical positions, leaving unspecified precisely which form of each theory is true. I’m interested in probing the possibility of this sort of “dual morality” and seeing what I can glean from it.

First off, I’ll lay out some terms. I’ll call the duties we have in virtue of our stances “KC(Kantian constructivist) duties,” and those we that simply *do* have I’ll call our “independent duties.” In considering these two kinds of duty, there are two possibilities. Our KC duties might sometimes conflict with our independent duties, or they might not. The latter scenario strikes me as less likely, but I’ll discuss it briefly. If it so happened that joining KC with independent morality *never* yielded a first-order conflict, I think this would be good evidence for theism, axiarchism, or some other hypothesis that has the basic structure of morality as sensitive to morality in a similar way.3 For, if our two kinds of duties never conflicted, this would mean that we could come to know and act in accordance with objective, stance-independent moral truths, and yet have no direct acquaintance with them whatsoever. Some theory that says we were designed to be capable of moral action would explain this evidence much better than some theory that says we weren’t.4

I’m more interested, however, in what strikes me as the far more likely scenario—we have KC duties, we have independent duties, and sometimes they *do* conflict. If that were the case, there would be three further possibilities:

1. When a conflict occurs, we ought to act in accordance with our KC duties.

2. When a conflict occurs, we ought to act in accordance with our independent duties.

3. When a conflict occurs, there is—at least sometimes—no fact of the matter about what we ought to do.

Option 1, I claim, is untenable. To see why I think so, let’s analyze a case. Take a classic trolley problem, and suppose that our independent duty is to pull the lever, but that our KC duty is to refrain and stand aside. The nature of my independent duty is that it simply *is* my duty to pull the lever, and for no further reason. The nature of my KC duty, on the other hand, would be quite different. If I, as a Kantian constructivist, say that I am duty-bound to act in a certain way, there is a first-person sense in which this statement is true. From a third person perspective, however, what is instead true is that I will consider myself duty-bound to act in this way insofar as I am rational. So, for it to be true that this outweighs my independent duty, it would have to be the case that morality could never require me to behave in an irrational way. Now, insofar as I am *incapable* of behaving irrationally, it is trivially true that morality could not require me be irrational, via ought implies can. There are, I’m aware, some arguments against OIC, but Kantians certainly ought to accept it, so I’m going to set those aside as we look to our next case:

If, to riff on Parfit, some despot had threatened to kill my child unless I came to believe that 2+2=5, morality could not require me to have this belief because I cannot simply *decide* to be irrational in this particular way. If however, I could take a pill that would—for an hour’s duration—somehow grant me the ability to decide what I believe, it seems as if morality *would* then require me to take this pill, and so acquire the irrational belief. But now recall that Kantian constructivist metaethics, if successful, shows only that insofar as I am rational, I will *consider* myself under certain obligations. Constructivists will typically add the further plausible thesis that it is also irrational to consider yourself obligated to act in a certain way and yet *not* act in that way. But unlike believing that 2+2=5, it is by no means out of my control, once I consider myself duty-bound to act in some way, whether I act contrary to that duty. So what independent duty demands of me is less akin to *believing* that 2+2=5, and more akin to writing “2+2=5” on a blackboard, or—perhaps a better analogy—playing the lottery in the hope of saving myself from debt even though I’m well aware that there are better ways to make money. I claim that any tenable moral theory could require us to do that much—to do something we believe to be irrational, or even believe to be contrary to duty.5 And that rationality compels us to consider ourselves duty-bound in some way does not entail that we actually *are*—certainly not if there are independent facts about our duties.

Now, does this argument rule out option 3 as well? Ought we conclude that, if moral realism is true, independent duty wins every time? I actually think is is still an open question. To see why, let’s precisify our understanding of the moralities we’re working with. Our KC morality, I will assume, treats deontic categories as fundamental. That is, it holds that whether we ought to perform some action is a function of whether this action is obligatory, permissible, of forbidden. If an action is obligatory or forbidden, morality alone determines how we are to act. If an action is permissible, it is still possible that morality makes a recommendation, but it does not bind us—we may do as we please. I assume that KC morality will work in this way because all such theories I’m aware of do, but I don’t think much turns on the question anyhow.

With independent morality, on the other hand, we’ll consider two possibilities. Independent morality could treat deontic categories as fundamental, but it could also be scalar—it could treat *reasons* as fundamental.

Let’s take the first possibility first. If deontic categories are fundamental to independent morality, it’s almost trivial that option 2 wins out. It follows from our understanding of how the deontic categories work that, if I am morally obligated to act in a certain way, I ought to act in that way *no matter what.*6 The reverse holds for forbidden actions. If some course of action is instead permissible, then of course I may act as KC morality dictates, and insofar as I am rational, I will.

What if independent morality is scalar, though?7 If so, then what we’re dealing with at the fundamental level is nothing more than reasons of different strengths. If we say that things are right or wrong, obligatory or forbidden,8 this may be convenient and useful, but it’s not a precise report of the truth. What’s true instead is that we all have reasons to do certain things. Some of these reasons are strong, others weak. My reason to save a drowning child might be very strong indeed, and my reason to refrain from being rude to a coworker might be much weaker. Nonetheless, a scalar conception of morality holds that the difference here is really a difference of degree, if often an extreme one. Note, however, that if morality is scalar and reasons are fundamental, there may be some equally fundamental reasons which are not moral. The obvious focus is epistemic reasons, but these might be disanalagous from moral reasons in a variety of important ways, so we’ll pass over them here and look instead at self-interested reasons. Someone who thinks that reasons are fundamental to an explanation of normativity might think that we have reasons to, say, exercise frequently. But these reasons won’t be moral—we won’t say that it’s *wrong* not to exercise. Instead, we’ll just say that I have reasons of different types. The way this applies to Kantian constructivism is fairly straightforward. Just as I have reason to act in my own interest and reason to act morally, we might think that I have reason to act rationally.9 And if behaving as KC morality dictates really is rational, then I’ll have reason to do that. But how do we weigh this kind of reason against our properly moral reasons?

Here, there are two possibilities. Reasons of different kinds are commensurable, or they aren’t. If they’re commensurable, there will always be a determinate fact of the matter about what I ought to do in any given moment. In some instances, my reasons to act as KC morality dictates will be stronger than my reasons to act according to independent morality, and in other instances the reverse will be true. It could be, however, that reasons of different types are not commensurable, or at least are only weakly commensurable. It could be that when I have reasons of similar strengths pulling me in different directions, there is no fact of the matter about what I ought to do—either way would be an appropriate response. And if reasons of different kinds are completely incommensurable, then there will never be a fact of the matter about whether I ought to privilege the dictates of KC morality over the dictates of actual morality. Even if reasons of different kinds are only incommensurable when their strengths are similar, it will sometimes be indeterminate which morality I should use to guide my action.

There is, I’m sure, much more to be said about each of these possibilities, and more still about whether either of these moralities is real in the relevant sense. But even without exploring any further, I think our little foray into the potential conflicts between the two moralities grants us an extremely important insight: they aren’t the same thing.

I know this will currently seem to be an obvious point, but it’s a significant point when we consider that views like Kantian constructivism are often construed as alternative understandings of morality—alternatives to realism, that is. And indeed, if—as I suspect—there are no stance-independent truths, then constructed norms may be a good candidate for the thing we slot into our concept “morality,” a concept that plays an important social role. So it may well be the case that it’s appropriate to consider a constructed morality “the real morality.” Words, after all, get their definitions from the way they’re used. But if there *are* objective, stance-independent moral truths, there would then be an important sense in which these truths would *instead* constitute “the real morality.” If we could learn them, they would give us the final word on what, morally speaking, we ought to do. And their existence would answer the all-important question of whether we ought to do anything at all. KC, in my view, can’t quite do this, because a constructed morality cannot be absolutely normative. If Kantian constructivism succeeds, it shows that rational agents will place themselves under certain obligations, but it does not show that we ought to be rational. I doubt that any argument which does not invoke an irreducibly normative truth will ever be able to do that.10

Well. I’ve written myself into a bit of a corner, and I’m not quite sure how I ought to conclude. I suppose I’ll play the hits:

In my essays, I’ve made no secret of admiration for the work of Derek Parfit. Recently, I’ve finished reading his On What Matters. In this truly mammoth book—almost 2,000 pages across three volumes—Parfit attempts something very admirable, something that philosophers should perhaps attempt more often than they do. Parfit tries to resolve confusion—to show that he and many of his fellow philosophers *do not disagree* on some of the most fundamental issues in ethics and metaethics. He does not, in my view, succeed to nearly the extent that he seems to imagine he has. But his effort is nonetheless extremely admirable—admirable because of Parfit’s thoroughness and precision, admirable also because of his hope that our most fundamental disagreements are not so deep as they appear.11

I am afraid I cannot be so hopeful.

I have implied that I find Kantian constructivism much more plausible than realism. This is true. Constructivism, I’ll confess, also appeals to me on an emotional level. This is because of the project’s ingenuity, but also for the following reason: if constructivist arguments succeed, they preserve the correctness of many judgements that the vast majority of us find intuitively plausible without taking on any metaphysical baggage.12 But though this is quite enough for me, I’m coming to realize that—even if realism is false and constructivism succeeds—constructivism does not give the realists everything they want. Realists hold that certain actions are wrong, yes, and constructivism can account for this judgement. But when realists say that something is wrong, they do not mean precisely what we mean. They do not mean that we’ll refrain from acting in some way insofar as we are rational, or that—insofar as we are rational—we will consider ourselves under the relevant obligation. The realists mean instead that there are some ways in which we simply ought not to act, and that this is an irreducible and immutable truth. In the eyes of the ardent realist, a constructed morality simply is not a real morality. So we constructivists are not saying the same thing as the realists, only using a different language. There is a deep disagreement here, and it will not be resolved in a way that shows it to be illusory. There are, I’m sure, many *other* confusions to be resolved, and their resolution is important work. But when that work is done, it won’t be the end. There’ll still be an argument to win.

Or rather, there are two arguments. There is the argument over whether constructivism succeeds, and there is the argument over whether realism is true. I’m now going to use the state of the former to weigh in on the latter.

Realists, in general, do not tend to think that Kantian constructivism could be successful. Kantian constructivists, in general, do not tend to think that any form of realism is true. I say “in general,” but I might just as well have said “as a rule,” for I know of no exceptions. And I think these rules imply something interesting about the way the way some realists approach the prospect of normativity. See, I think there are *some* realists—I won’t say a great many, and indeed it may be very few, but I think they exist—who ought not to be realists,13 and I think the reason they ought not to be realists is intimately tied up with their attitude to views like Kantian constructivism. The type of realist I have in mind sees the arguments for views like KC and finds them unpersuasive. But at the same time, he finds KC appealing. It would be wonderful, he thinks, if we could explain the objective truth and falsehood of moral claims without appealing to abstract entities or necessary truths or anything else that’s difficult to make sense of. And so, if he *did* find KC to be a successful project, he would no longer be a realist. If we can explain the rightness and wrongness of certain actions with a constructed morality, he thinks, then we would have no need to accept any more robust realist view.

I’ll reiterate that I’m making no concrete claims about the number of realists who fall into this category—I say only that there are some, and I say this because it’s an attitude I’ve personally encountered on occasion.14 Having so re-iterated, I’m going to argue that anyone who does have this attitude ought not be a realist in the first place. And the argument for this is very simple. In cases like the one I’ve described, I think our realist’s underlying belief is very likely to be something of this form: “it is *definitely* true that some actions are objectively right or wrong. There are, however, multiple theories which make this true, and they offer different explanations of *why* it is true. We should accept whichever such theory is most plausible on its own merits.”

This, I believe, used to be a thought that undergirded some of my own reflections on morality. You wouldn’t know it from reading my essays, but there was a time—two or three months following the point when I first got into philosophy—during which I would have described myself as a realist. Then I got some proper exposure to Kantian constructivism, and was quite taken with it, and gradually began to abandon my realism, favoring instead the Kantian account of what it is for actions to be right or wrong. Then I became uncertain of whether KC was really tenable. I am still uncertain of this, and yet these days I rarely feel any pull back towards realism. This, I think, is because I have come to see the way our imagined realist thinks about the rightness of actions—the way I believe I once thought about this—as deeply flawed. I could explain why by appeal to some principle of the following sort: “you do not know something unless you know *why* it is true.” But this principle, though intuitively plausible, only has so much mileage. If certain forms of realism are true, there is no *why.* Truths about morality simply *are* true, and there’s nothing more to it.

Instead, I’ll appeal to the insight that examining the case of double morality granted us: realist wrongness and constructivist wrongness, though similar in many ways, just *aren’t* the same thing. For emphasis, I’ll repeat myself: Kantian constructivism, when set against realism, is *not* an alternative way of understanding the objective rightness and wrongness of actions, but something entirely different.

Keeping this in mind, let’s dig a little deeper into the case of our imagined realist. This realist believes that, if KC succeeded, that would allow him to give up realism. What this means is that the success of constructivist arguments would entail that it would be unreasonable to maintain a belief in moral realism. We’re here considering two different sets of norms, both of which are good candidates for filling a particular role. But the problem here is that—as I’ve been so keen to emphasize in the course of writing this essay—whether realism is true is totally independent of whether constructivism succeeds. Though independent morality and KC morality might both reasonably be called “morality,” using that same terms for both sets of norms can, in circumstances like this one, cause a bit of confusion. Because they *just* aren’t the same, and whether each stands or falls has nothing to do with the other. Accordingly, whether it’s reasonable to *accept* realism is also independent of whether constructivism succeeds. Your view on whether realism is a tenable position should have nothing whatsoever to do with your view on the success of constructivist arguments. And so, if you’re of the view that the success of KC would allow us to abandon realism, you’re making an inference of a form close to this: “if there is a successful way of constructing objective morality, morality does not exist stance-independently.” This is a very poor inference. The success of constructivist arguments will not make realism fail. If the arguments for realism fail, they fail—constructivism’s got nothing to do with it. And so, I claim that *if* you’re a realist of the view that constructivism would allow you to abandon realism, there is probably some fatal problem with realism which you already perceive, and so you should already be leaving it behind.15

Having done so, if you still find it difficult to accept that there’s nothing in the vein of objective morality, constructivism still may be a good place for you to look. The arguments are interesting, and some versions of them may succeed. It may be appropriate for you to hope that they do. But if that’s so, they succeed on their own merits—realism’s got nothing to do with it.

As a final note, I’ll say that I am not aware of—nor at this moment can I think of—any arguments for the claim that there really could be only one morality. If such an argument were to succeed, that would change the entire nature of the landscape I’ve been surveying. So if you have one or know of one, I encourage you to let me know.

While describing how he came to write his great, drab book The Methods of Ethics, Sidgwick remarks that he had ‘two masters’: Kant and Mill. My two masters are Sidgwick and Kant.

I can likewise say that my own two masters are Kant and Parfit himself—they are my masters not in the sense that they are the philosophers with whom I most agree, but instead in that they two have shaped my ways of thinking more than any others. And I echo Parfit’s phrase not to place myself among these titans, but instead because Parfit, I think, would have been pleased to hear a young student of philosophy put his name next to Kant’s. Of Kant’s Groundwork, Parfit says this: “The truth is that, in the cascading fireworks of a mere forty pages, Kant gives us more new and fruitful ideas than all the philosophers of several centuries.” Change “a mere forty” for “a whole five hundred,” and I could say something very similar of Parfit’s Reasons and Persons. No philosopher I’ve read has sparked, crackled, and sustained a burn in the way that Parfit does over the course of those glorious half-thousand pages.

On What Matters is different. It is a very long book—almost two thousand pages across three volumes. And it is not a raging flurry of bright new ideas; it is, instead, a thorough and impassioned defense of some very old ones. Where the four parts of Reasons and Persons might well be four different books, On What Matters is a single-minded effort. Towards the end of the third volume, Parfit writes this:

I regret that, in a book called On What Matters, I have said so little about what matters. I have been trying to defend the belief that some things really do matter. I hope to say more about what matters in what would be my Volume Four.

This is a tragic paragraph. Parfit was dead before it saw publication, and no fourth volume will ever be written.

I confess that I find his death difficult—not difficult to believe, but simply difficult. Unlike Kant, Parfit is no relic. When I first encountered his work, he did seem to me a historical figure, an old master of philosophy very much like Kant or Hume. But I did not know then that I had been ten years old at the time of his death. And in reading the acknowledgements of On What Matters, I see the names of people whose recent work I know. I see even the names of people with whom I have conversed on this very platform. If Parfit were alive today, he would be 84 years old. People *live* to 84. The possible world in which I might have gotten the chance to meet my hero is not very far away, and this nearness is at the heart of the matter; it is what makes the absence of the great philosopher so thoroughly *difficult.*

And it is, of course, more difficult still because he was truly great. In Reasons and Persons, this greatness manifested as inventive genius. In On What Matters, however, it takes quite a different form. Here, Parfit does not present astonishing new problems for ethicists to solve, nor does he give solutions to the problems with which he once shook all the world’s ivory towers. Instead, he devotes himself entirely to an attempt to prove that some things really do matter, matter in a way that has nothing to do with who people are or what they want or how they act. In my opinion, it is an attempt that fails. But it is remarkable and admirable nonetheless because of the spirit in which it was carried out. In many instances, Parfit does not make his case by arguing that his opponents are wrong, but instead by arguing that he and his opponents do not really disagree. He spends an entire chapter discussing Nietzsche’s moral philosophy, attempting to show that even Nietzsche may not have disagreed with him in ideal conditions. He does something similar with Hume, and with his late old friend Bernard Williams. Some may accuse Parfit of failing to get the ideas of these philosophers quite right, but none can accuse him of failing to take their disagreement seriously. At the beginning of the thirtieth chapter, Parfit writes as follows:

I shall now make some personal remarks. Though I became a philosopher so that I would have more time to think about what matters, much of this book is about other questions. It may be worth explaining why.

One reason is that, when we ask what matters, part of the answer is obvious. As Nagel writes:

There are so many people one can barely imagine it … what happens to each of them is enormously important, as important as what happens to you.… what happens to anyone matters the same as if it had happened to anyone else … the elimination of the worst sufferings and deprivations matters most … The alleviation of misery, ignorance, and powerlessness and the elevation of most of our fellow human beings to a minimally decent standard of existence, seem overwhelmingly important.

These aims are overwhelmingly important.

When we ask what else matters, the answer is less obvious. Some of my first thoughts were about which futures for humanity would be better or worse. If the quality of future people’s lives would be lower, for example, could that be outweighed by the fact that there would be more people who would live? I came across some puzzling problems which I failed to solve. I also thought about some metaphysical questions whose answers might make a difference to our concerns, and to our reasons for concern. I had what seemed some false beliefs about what is involved in my continuing to exist, and in its being true that various future experiences will be mine, or true that I shall be dead. It seemed likely that many other people have similar false beliefs. We may therefore misunderstand which are the facts that give us reasons to care about our future, and misjudge the strength of these reasons. Since my thoughts about persons and personal identity had been influenced only by certain Western philosophers, I was encouraged to learn that Buddha seems to have reached similar conclusions. We may also have false beliefs about the nature and rational significance of time. We do not, I believe, have reasons to care more about the future than about the past. What has happened is just as real as what will happen. Nor is the past less real than the present. Though people who are dead do not exist now, that is merely like the fact that people who live elsewhere do not exist here. And time’s passage is an illusion.

After starting to discuss these questions in my first book, I intended to think about them further. As the contents of this new book show, that is not what happened. I became increasingly concerned about certain differences between my views and the views of several other people. We seemed to disagree not only about what matters, but also about what it would be for things to matter, and about whether anything could matter.

I might have ignored these disagreements, and tried to answer my own unanswered questions. But when other people’s beliefs conflict with ours, that challenges our beliefs. As Sidgwick writes:

if I find any of my intuitions in direct conflict with an intuition of some other mind, there must be error somewhere: and if I have no more reason to suspect error in the other mind than in my own, reflective comparison between the two intuitions necessarily reduces me … to a state of neutrality.

Such disagreements give us reasons to doubt that we are the people whose beliefs are true. These disagreements may also give us reasons to doubt that any of these conflicting beliefs are true. Perhaps none of us is right, because our questions have no answers.

And so Parfit spends a great deal of his great book trying to resolve disagreements, trying to show that his interlocutors are not so distant as they might believe. And what Parfit exhibits here is not genius. Instead, it is a greatness of spirit. It is a greatness that consists in passion, in dedication, but most importantly in *respect.* Respect for what matters, yes, but also a deep well of respect for the other people spending their lives in pursuit of the same truths. It is manifest here, in the way Parfit expresses of his visceral anxiety about disagreement, and it is manifest also in every reference he makes to his fellow inquirers—to Kant, Sidgwick, and Nietzsche, but also to Scanlon, Gibbard, Railton, Williams, and Street. And this respect, over and above all the arguments, is the finest quality of On What Matters. The best passages of Reasons and Persons are exhilarating, white-hot, lightning in a silver-plated bottle. By contrast, the best passages of Parfit’s later work are contemplative, vast and oceanic—they are not the passages that reveal the conclusion of some brilliant argument, but instead the passages that reveal *Parfit* in all his sober brilliance. Parfit, who dared to hope that we would all agree if each could only speak the other’s language.

And it is, as I have said, difficult that this man is gone. But here is a way of lessening the difficulty:

In the third part of Reasons and Persons, Parfit proposes a theory of personal identity which is, by the standards of Western philosophy, quite radical. Namely, he proposes that there is no such thing, that there is never any fact of the matter about whether a person at one time is identical to a person at another time. Of the person currently writing this essay and the person who will succeed him twenty years hence, we can say only that there is a relationship of psychological continuity between the two. This relationship will probably be very strong. But still, the difference between my successor and the successors of my good friends, as they relate to me, will be a difference of degree. There will be no robust sense in which any of us are identical to the persons we take to be our future selves. Parfit argues powerfully for this theory, and I believe that only the soul can save us from the success of his arguments. As of late, I am increasingly skeptical of the soul.

But should we want to be saved? I, for my part, am unsure, but Parfit had an answer. In perhaps his most famous passage, Parfit remarks on the implications of his theory:

Is the truth depressing? Some may find it so. But I find it liberating, and consoling. When I believed that my existence was a such a further fact, I seemed imprisoned in myself. My life seemed like a glass tunnel, through which I was moving faster every year, and at the end of which there was darkness. When I changed my view, the walls of my glass tunnel disappeared. I now live in the open air. There is still a difference between my life and the lives of other people. But the difference is less. Other people are closer. I am less concerned about the rest of my own life, and more concerned about the lives of others.

And in much the same way, though we cannot bring Parfit back to life, we can dissolve the walls we build between our lives and his. We can read his books, and in reading his books we can come to think in a way that is closer to *his* remarkable way. In doing so, we can strengthen the relationship of psychological continuity between the person that was once called Derek Parfit and the people that we are now. Though Parfit is gone, we can make him less so.

In closing, I’d like to quote the very last paragraph of the third and final volume of On What Matters:

If we are the only rational beings in the Universe, as some recent evidence suggests, it matters even more whether we shall have descendants or successors during the billions of years in which that would be possible. Some of our successors might live lives and create worlds that, though failing to justify past suffering, would have given us all, including those who suffered most, reasons to be glad that the Universe exists.

The life of Derek Parfit, I claim, gives us one such reason.

It’s been a while since I busted out the hard stuff, so indulge me:

The realization of the summum bonum[highest good] in the world is the necessary object of a will determinable by the moral law. But in this will the perfect accordance of the mind with the moral law is the supreme condition of the summum bonum. This then must be possible, as well as its object, since it is contained in the command to promote the latter. Now, the perfect accordance of the will with the moral law is holiness, a perfection of which no rational being of the sensible world is capable at any moment of his existence. Since, nevertheless, it is required as practically necessary, it can only be found in a progress in infinitum towards that perfect accordance, and on the principles of pure practical reason it is necessary to assume such a practical progress as the real object of our will.

Now, this endless progress is only possible on the supposition of an endless duration of the existence and personality of the same rational being (which is called the immortality of the soul). The summum bonum, then, practically is only possible on the supposition of the immortality of the soul; consequently this immortality, being inseparably connected with the moral law, is a postulate of pure practical reason (by which I mean a theoretical proposition, not demonstrable as such, but which is an inseparable result of an unconditional a priori practical law.)

This, of course, is Immanuel Kant, the Critique of Practical Reason.1 Kant, here, is executing a shockingly modern dialectical maneuver. His idea is this: through practical reason, we can figure out that the we are obligated to realize “the perfect accordance of the mind with the moral law.” We are, in other words, obligated to take the right turn at every possible juncture. Problem: no one can do this. We know how people are, and even the actions of the most virtuous are not determined *entirely* by the moral law, nor can we imagine that any finite period of reflection and practiced right action would make it so.2 This puts us in a strange predicament: we’re obligated to realize this perfect accordance, and yet it seems that no person—no matter how virtuous his disposition or how hard he works to cultivate virtue—is capable of doing so. And we can’t be obligated to do something impossible!

By Kant’s lights, the solution is to have a postulate of practical reason. He calls this “a theoretical proposition, not demonstrable as such, but which is an inseparable result of an unconditional a priori practical law.” In this case, the theoretical proposition in question is the immortality of the soul. For if we know that no finite period of time is long enough to cultivate perfect accordance of mind and moral law, but we are nonetheless obligated to do so, then our obligation must extend over an *infinite* period of time.

Before I say anything further about this argument, I want to note a similarity to another, an even more famous one. It goes like this:

P1: Here is one hand

P2: Here is another

P3(1,2): There are at least two objects in the external world

C(3): The external world exists

This one, of course, comes courtesy of G.E. Moore, a philosopher who—unlike Kant—you will not often see me cite.3 And I want to note that, though Kant’s argument is harder to make sense of, Moore and Kant are actually doing very similar things here. Both of them are saying: look, there’s this thing that I *know* is true, this thing I just get to take for granted, and the truth of that thing requires the truth of some other things, so I get to take those for granted too.

The trivial difference is that Moore asserts the existence of the external world because it is simply obvious to him that his hands exist, and for no further reason. Kant, on the other set of hands, is drawing his claims about our obligations from a much more robust ethical/metaethical framework.

The deeper and more important difference is this: Moore is effectively arguing that the existence of his hands(and hence the external world) is so blindingly obvious that he’s entitled to reject any argument which attempts to show the non-existence of the external world. This argumentative strategy, whereby one claims that the argument’s conclusion is obviously false and so too must be one of its premises, is known as the Moorean shift, and it is frequently employed in response to a few particular kinds of argument—we’ll return to that soon. First, though, we ought to make clear that Kant’s postulates of practical reason are *not* doing this. Kant is not saying: “we get to take for granted that we have a duty to act morally all the time, and this requires us to be immortal, so we can reject any argument which concludes or implies that we are not immortal.” To see what Kant is saying instead, we can retreat a section and see his thoughts on the primacy of practical reason:

By primacy between two or more things connected by reason, I understand the prerogative, belonging to one, of being the first determining principle in the connection with all the rest. In a narrower practical sense it means the prerogative of the interest of one in so far as the interest of the other is subordinated to it, while it is not postponed to any other. To every faculty of the mind we can attribute an interest, that is, a principle, that contains the condition on which alone the former is called into exercise. Reason, as the faculty of principles, determines the interest of all the powers of the mind and is determined by its own. The interest of its speculative employment consists in the cognition of the object pushed to the highest a priori principles: that of its practical employment, in the determination of the will in respect of the final and complete end. As to what is necessary for the possibility of any employment of reason at all, namely, that its principles and affirmations should not contradict one another, this constitutes no part of its interest, but is the condition of having reason at all; it is only its development, not mere consistency with itself, that is reckoned as its interest.

If practical reason could not assume or think as given anything further than what speculative reason of itself could offer it from its own insight, the latter would have the primacy. But supposing that it had of itself original a priori principles with which certain theoretical positions were inseparably connected, while these were withdrawn from any possible insight of speculative reason (which, however, they must not contradict); then the question is: Which interest is the superior (not which must give way, for they are not necessarily conflicting), whether speculative reason, which knows nothing of all that the practical offers for its acceptance, should take up these propositions and (although they transcend it) try to unite them with its own concepts as a foreign possession handed over to it, or whether it is justified in obstinately following its own separate interest and, according to the canonic of Epicurus, rejecting as vain subtlety everything that cannot accredit its objective reality by manifest examples to be shown in experience, even though it should be never so much interwoven with the interest of the practical (pure) use of reason, and in itself not contradictory to the theoretical, merely because it infringes on the interest of the speculative reason to this extent, that it removes the bounds which this latter had set to itself, and gives it up to every nonsense or delusion of imagination?

[…]

Thus, when pure speculative and pure practical reason are combined in one cognition, the latter has the primacy, provided, namely, that this combination is not contingent and arbitrary, but founded a priori on reason itself and therefore necessary. For without this subordination there would arise a conflict of reason with itself; since, if they were merely co-ordinate, the former would close its boundaries strictly and admit nothing from the latter into its domain, while the latter would extend its bounds over everything and when its needs required would seek to embrace the former within them. Nor could we reverse the order and require pure practical reason to be subordinate to the speculative, since all interest is ultimately practical, and even that of speculative reason is conditional, and it is only in the practical employment of reason that it is complete.

There’s a lot here, but the important bit for us is this: Here, Kant is claiming that pure practical reason can determine our will in certain ways. Later, he claims that acting in these ways only makes sense if we grant certain theoretical propositions. Now, if speculative reason could pass judgement on the truth of these propositions, Kant claims that it would then “have the primacy.” But on Kant’s view, speculative reason *can’t* speak to these things—there is no speculative argument to be made that will demonstrate either the immortality of the soul or the reverse. So while Moore says that we get to reject any argument which contradicts a proposition we can take for granted, Kant is saying instead that we can take certain propositions for granted because there is no contrary argument to reject: we get to believe whatever theoretical propositions practical reason requires because speculative reason is incapable of either refuting or confirming them. So there is, as it were, nothing standing in the way of the soul’s immortality or the existence of God.4

§2: The Kantian Shift

Of course, Kant’s views on what speculative reason can and can’t know are unique, and I’m hardly the person to talk about them anyhow.5 So instead of saying anything further about Kant’s argument and whether or not it works, I want to just extract a useful idea from it—the distinction between practical and speculative reason, and the ability of practical reasons to “think [certain propositions] as given.” I’m going to riff a bit on Kant’s definitions, though. I’ll define practical reason as encompassing two ways of thinking: first, it is pragmatic in the naïve sense. First, practical reason tells you to believe—or, as I prefer, to act *as if* you believe—whatever will be good/useful for you. It also tells you to *do* whatever will be good/useful for you. Second, practical reason is moral reason: it tells you to do what is good simpliciter(whatever that means to you—I won’t commit to any moral theory here).6 I’ll define speculative reason, on the other hand, as truth-directed; tells you *only* to believe whatever is most likely to be true.

Having laid out this common-sense version of the distinction between practical and theoretical reason, I want to note that the two often come into conflict, and to give my favorite example of this:

As I write this essay, I’m sitting in a chair. Practical reason dictates that I consider this chair to be an object in its own right, something that exists freestanding, so to speak. At the same time, my dogs are somewhere nearby, and so too does practical reason dictate that I do not consider there to be such an object as a chairdog,7 an object which currently exists and which is composed of the chair I’m sitting in and one of my dogs. That I handle chairs and dogs in this way may not be a matter of strict practical *necessity,* but it’s certainly a very expedient way of interfacing with the world. When I interface in this way, the position that I’m assuming is mereological conservatism, which essentially says that those composite objects we generally take to exist do, in fact, exist, and that those we don’t, don’t. But mereological conservatism, useful as it is for navigating ordinary life, is a position that becomes very hard to hold when subjected to serious speculative scrutiny. Basically, conservatism seems to commit you to ontic vagueness—to it sometimes being indeterminate whether some object exists. And ontic vagueness is a bit of a nightmare.8

For those curious, I’ve left an expansive footnote explaining in much more detail why I consider mereological conservatism basically untenable. But if you feel differently, feel free to just substitute your favorite example of some assumption that’s borderline necessary for navigating ordinary life, and yet doesn’t survive scrutiny.

Now, when I debate the metaphysics of composition, I generally defend mereological nihilism. This is the position that only fundamental particles—quarks, leptons, what have you—actually exist. When I see a table, I judge that there is no real table present, but only a collection of fundamental particles arranged “table-wise.” Does the fact that I (tentatively) accept nihilism mean I shouldn’t say I’m currently sitting in a chair, though? I think not. I think it makes perfect sense, even once you’ve accepted a wacky theoretical position, to continue speaking and acting as if you accept the common-sense account of the matter at hand instead. If you accept mereological nihilism, I do think you should *believe* that tables don’t exist, and if someone asks you directly you should say so. But I don’t see anything wrong with pretending you believe otherwise as you go about your day. I’ll call this way of thinking the *Kantian* shift, and I’ll elaborate on it in a moment.

§3: No More Moore

Let’s talk a bit about cosmological fine-tuning. It’s likely that the story here is well-known to you. Physicists have discovered that many of the constants in the equations we use to describe the fundamental laws of physics are such that the existence of life would become impossible if you tweaked their values only very slightly. This is often taken as evidence for the existence of God, because this fine-tuning seems more likely if the Universe has a designer than if it doesn’t. I don’t intend to comment here on whether this argument works. Instead, I just want to run us through one atheist response and one common theist counter-response.

Encountering the fine-tuning argument, the atheist might agree with the theist that the existence of God predicts fine-tuning better than the complete absence of a designer. But the atheist can say that God is not the only kind of designer our Universe could have. We could, for instance, be in a computer simulation.9 One common theist response—not the only one, of course—is to say that if we were in a computer simulation, we would be in a skeptical scenario. And we basically get to take for granted that we’re not in a skeptical scenario. This is the aforementioned Moorean shift.

Now, I’m not going to pass judgement on which of these two theories should have a higher prior, nor on which of them better predicts the evidence. So keep in mind as I go forward that I’m not actually committing myself to the claim that the Simulation Hypothesis is better or more probable than theism. I do, however, really dislike this particular dialectical maneuver, and I’m going to argue against it.

First off, I’m going to make an obvious point. There is *some* amount of evidence that could convince you you were in a computer simulation. Suppose one day you’re thinking about how much you’d like a bar of dark chocolate, and out of the blue a strange man comes along and hands one to you. He’s gone before you can get a good look at his face. But he appears again. Every day at twelve noon, the strange man appears and hands to you exactly the thing that you’re currently most craving. This goes on for a year. He’s always quick and crafty. Though you start to get a sense of how he looks—stocky, dusty-blond hair, twice-broken nose—you never get the chance to talk to him. It begins to drive you mad. It gets to a point where you barely sleep, and instead just pace by a window at night, looking out to the street and hoping desperately to catch a glimpse of him. But it’s always he that comes to you. And on day number three-hundred-sixty-six, he finally does give you the chance to talk to him. You ask how he does it, how he knows. And he says simply that he knows your mind because he can read it, that you’re computer code to him. He’s playing your world like it’s a video game.

Now, it is, at this point, still entirely possible that you aren’t in a simulation, that something even stranger is going on. But three-hundred-sixty-five days of mind-reading is hard to argue with—it’s not something you could Moore-shift. If, at this point, you wanted to rationally believe that you weren’t in a skeptical scenario, you would need either some alternate explanation of how the blond man read your mind, or some reason to think he was lying. If you had neither—well, tough luck, then.

This example is enough to establish that the Moorean shift isn’t *always* an appropriate maneuver. I’m going to argue, however, that the Moorean shift as it’s often used is basically *never* an appropriate maneuver.

Let me illustrate a bit: I once participated in a conversation with a pretty well-known Catholic apologist. We were discussing miracles, and someone had proposed to the apologist something along the lines of a generic animist explanation of miracles. Such explanations might posit, for instance, that there are a great many powerful spirits in the world, and that these powerful spirits—like human beings—have a variety of beliefs and dispositions. So some spirits might create seemingly-Catholic miracles because the spirits in question are actually Catholics—the same goes for miracles in other Christian sects, or Buddhist miracles, or what have you. But the apologist was of the view that this would put us in a skeptical scenario—how could we be sure of anything with so many spirits of different powers and dispositions whizzing around. And the apologist said further that he wasn’t particularly interested in contending with skepticism—if he could prove that you ought to be a Catholic insofar as you aren’t a skeptic, that would be sufficient.

Here’s the very simply reason I’m inclined to object to this way of arguing: it might *actually* be true that we’re in a skeptical scenario. It is, as far as I can tell, not horrifically unlikely, or at least not *so* unlikely as to render any argument in favor of that position unacceptable. It would indeed be very *strange* if we were in a skeptical scenario, but many strange things are difficult to reject. By this I mean to include philosophical positions like mereological nihilism/universalism, but also some settled science. Wave/particle duality is, on its face, really strange and implausible, but it is not something that an appropriately informed and fully rational person would able to reject.

But then, we have pretty powerful evidence for wave/particle duality. I do believe that if we had a similar amount of evidence for the truth of, say, the Simulation Hypothesis, most philosophers would be willing to accept that we’re in a simulation. The trouble is, though, that *even if* we’re in a simulation, it is highly unlikely that we’ll ever get that amount of evidence. But we really could be! And if we were, I would want to *know.* I would want the truth. But getting the truth requires being open to the truth. If I would only accept some theory conditional on getting an amount of evidence that even the truth of that theory would not lead me to expect, I am closing a very, very thick gate. I am sequestering myself in a fortress of dogma.

But surely I shouldn’t just go around believing the conclusion of every argument that I can’t decisively refute? That would be madness! There are, after all, a great many philosophers who are much, much smarter than me, who have made much better arguments. And famously, they do not all agree with each other.

Here, I have two thoughts. First off, we may have cases where some argument leads me to a conclusion which is perhaps in tension with my ordinary beliefs, but does not strictly compel me to act different than I otherwise would have. An argument against mereological conservatism is one such case. In another, I might come to believe that a simulation explains the evidence that’s typically taken as evidence for God somewhat better than theism. I might also think, however, that the best explanation of the evidence which does not imply I’m in a skeptical scenario is, in fact, theism.

Here, we can apply the Kantian shift. In the case of composition, I claimed that it’s perfectly acceptable to argue for and accept mereological nihilism, but to go around continuing to pretend that tables exist. Likewise, I claim that one who considers the Simulation Hypothesis a better explanation of the world than theism may, in a philosophical context, argue that we should believe we are in a simulation, and so that we should be very skeptical about a great many things, and may accept this position. At the same time, as soon as this person leaves the seminar room, they—considering theism to be the second-best explanation, and perhaps theism of some particular religious persuasion10—may live their life as a devout practitioner of some religion, continuing to form beliefs and transact with people as if they can, in fact, be sure that all of it is real in a robust sense of the word.

What I am claiming, to be more technical, is that in cases where the conclusions of theoretical reason have no heavy-duty practical implications, it is acceptable to leave them out of our ordinary lives. It is acceptable to live by the lights of practical reason whenever theoretical reason has nothing to say about what we ought to believe or ought to do for non-epistemic reasons. Hence a *Kantian* shift instead of a Moorean shift; I believe licensing the use of such a maneuver dissolves much of the concern around believing highly implausible conclusions of philosophical arguments. Most philosophers, after all, hold some kind of belief that many other philosophers are inclined to consider crazy. But insofar as the philosophers in question *don’t* act crazy, we don’t generally consider it bad in itself that they hold somewhat crazy beliefs.

What if, however, the conclusions of an argument are both highly implausible and practically weighty? It’s one thing to say that you don’t get to reject an argument just because it implies skepticism. It’s quite another thing, however, to say that you can’t reject an argument because it implies that you are morally obligated to start pitching babies off the rooftops of New York City. If there were such an argument, and somehow it were not clear how it could be refuted, the Kantian shift would not be a viable option. This is because an argument implying we’re in a simulation doesn’t have any direct implications concerning what we ought to do. An argument that attempts to prove we have some particular moral obligation, on the other hand, most certainly does.

The point I’ll make here will actually also apply to skepticism: Even if you encountered some difficult-to-refute argument with the aforementioned baby-pitching conclusion—you would probably also have encountered more than one difficult-to-refute argument implying the opposite. In such a case, it’s obvious that one of these arguments must go wrong somewhere.11 And so, in such a case, when one position is more consistent with our broader set of beliefs, or is supported by more strong arguments, it may be appropriate to provisionally accept that position over the other. But though I’m describing a kind of Moorean shift here, I want to be clear that this shift ought not function as an outright rejection, but as a *promissory note.* Insofar as we’re to use the Moorean shift at all, our methodology should not simply be to say: “this argument fails because its conclusion is wrong.” It should instead be to say: “this argument fails because its conclusion is wrong, *and so* I’m going to provisionally reject it while I search for the problem with the argument.” Many times, I think this is actually how the Moorean shift is applied, but not always. For the Catholic apologist I previously mentioned, the fact that a general animist theory implied skepticism was sufficient for him to reject it wholesale; at least in the moment, he saw no further need to determine which of the premises in the argument was wrong, or why. And this, I claim, is not sufficient. If you want to convince me I ought to be a Catholic insofar as I’m not a skeptic, fine. Keep doing as you do. But it may be that I consider skepticism more plausible than Catholicism. And if that’s the case, you will need to do more. Instead of showing me what I ought to believe if I’m not a skeptic, you will need to show me what I *ought to believe.* You will need to give me *Truth.* And G.E. Moore isn’t going to do it for you.

Anyway, I didn’t give my hiking partner all of the details, but his take on the miracles was one that I’ve heard before, one that I’m quite sympathetic to: he told me that he might be willing to buy a few truly miraculous happenings, but not so much that the happenings in question vindicated any particular religion. If I really did consider the miracles supernatural—I’m not quite sure I do—this would probably also be my position, and I said as much.

“But,” I also said, “There’s definitely *some* amount of evidence that would convince me Catholicism was true.”

I’ll paraphrase the response I got:

“I often ask my students whether they’d still be Catholics if we discovered an ossuary containing the bones of Jesus Christ. And when they say no, that they’d probably have to give it up then, I always get so mad at them. Because you have this beautiful story, this sacrifice. God is more powerful than death, and you’re going to give all that up because of some inconvenient history?”

In response, I pointed out that, if you were a Catholic and you believed Jesus had literally been resurrected, finding out otherwise would certainly at least cause your relationship to Catholicism to change. On reflection, however, I think it might be just fine to continue practicing in the way you had always practiced.

Some elaboration: Amos Wollen has a nice essay called “Pluralist Priors and Baha’i Bias.” The idea is that, insofar as we think any religion is divinely inspired, we should give more credence to the proposition that *many* religions are divinely inspired, which is what Baha’is believe. However, what if I’m a bare theist? What if I think God exists, but that no particular religion is the right one? This is a position which I’ve become more and more sympathetic to over the last two months or so, and now I’m wondering whether its truth would render ritualized worship irrational.

In response to that question, I ask you to imagine that it’s 1850, that you’re sitting in a saloon in some desert town way out in the Wild West. The place isn’t really big enough to hold the patrons, but the whiskey’s flowing anyhow, and everyone’s in a good mood. They’re swapping stories, all of them, talking about some gunslinging cowboy folk hero. Some of the tales sound straightforward enough that you figure they probably could’ve happened, others are outlandish beyond belief. In any case, there’s too many of them—and too many self-proclaimed witnesses—for it *all* to be true. But the stories are mostly flattering. And if this cowboy hero came to town, he’d get a warm welcome, free drinks, plenty of conversation and admiration from pretty much every single person present.

Now, if some particular religion is the only one to be divinely inspired, then it seems to follow that God wants you to be a part of that religion, insofar as you find out about it. If many religions are divinely inspired, God might care less about which you choose, but it’s probably still better to pick something with a heavenly stamp of approval. If, however, no particular religion is divinely inspired, then our situation is quite different. Then, I claim, the religious landscape is akin to our Old West Saloon. There are stories everywhere, and some of them—maybe even most of them—aren’t true, but they’re all generally flattering. And telling these kinds of stories, going around the world acting as if they’re true, sharing them eagerly, all that seems like the sort of thing that might actually be good for building a relationship with God. Because in all likelihood, God really is just beyond our comprehension, and it’s difficult to have a relationship with something you can’t understand. But we do understand something: we understand God to be worth worshipping, we understand him as good—maximally good, *infinitely* good. And so it strikes me as appropriate for us to take whatever on Earth strikes us as presenting the clearest picture of something worth worshipping, something awesome and supremely good, and then taking that picture and saying to God: “through this, we devote ourselves to you.”

If that’s the way we were thinking about it, we would have to know the picture wasn’t the real thing, but I’m not sure that would matter.

But then, of course, there’s the question of how we handle the stories that aren’t so flattering. The first one that leaps to mind for me—as I’m sure it does for many of you—is God’s commanded genocide of the Canaanites in the Old Testament.

I’ll use another analogy to demonstrate how I think those of us who are religious4 ought to approach these pain-points. There is a movie by Hayao Miyazaki—a beautiful movie—called The Wind Rises. It is a highly fictionalized biography of Jiro Horikoshi, a Japanese aeronautical engineer. I say “highly fictionalized” because only about half of it is based on Horikoshi’s life story—the other half thrusts him into an adaptation of a romance novel, and the source material hasn’t the least to do with the real man’s life.5 Anyway, the movie still has a lot of scenes involving planes and plane-designing. It does not, however, have a scene that showcases Horikoshi’s most famous plane doing the thing it was most famous for. That is to say, The Wind Rises does not depict a Mitsubishi A6M “Zero” executing a kamikaze attack.6

Now, it’s not as if the real Horikoshi wanted his planes used in this way, or even wanted Japan to go to war with America in the first place, And it’s not as if he had any say in either. So there’s no perfect analogy here with stories about God and the atrocities he is said to have commanded. My point is rather that the use of Zeros in kamikaze attacks is the most regrettable and uncomfortable part of Jiro Horikoshi’s life story, and so the movie doesn’t bring it up. Likewise, though religious people needn’t entirely ignore their least comfortable stories about God, they also needn’t emphasize them. They can say something like this: “look, we have this book, and it creates a picture of a wonderful, awesome God, very much worth worshipping. But there are difficult parts to it, parts that we struggle to understand. And we’re not going to change the book because of that—we’re going to take what was given to us. But the stories we most emphasize, they’ll be the ones that give us the clearest picture of that which is worth worshipping in the God we love.”

And those stories, presumably, will not involve divinely-sanctioned genocides.

Now, it seems to me that, conditional on God existing but no religion being divinely inspired, this approach might be a good way to cultivate a relationship with God. It’s also worth noting that many agnostics tend to think that bare theism is the form of theism they’d adopt if they were to adopt any at all. Accordingly, agnostics might also consider this approach to religious practice worthwhile—they might think it’s worth *trying* to have that sort of relationship with God even if we can’t be sure it’ll come to anything in the end. In this way, I think we can endorse religious pluralism *even if* we consider all religions empty of divine inspiration.

One last note: If some claim of mine strikes you as wrongheaded, be sure to check if I’ve footnoted it. There are some cases where I’ve omitted to make some distinction that is not particularly important to the matter at hand, which might be very important in a broader context, or for further investigation. Whenever I’ve done this intentionally, I’ve left a footnote that gives something closer to the full picture.

Introduction:

Before I start talking about philosophy, I need to introduce you to one of my favorite books. Its title is Personal Recollections of Joan of Arc,1 it was written by Mark Twain, and I’d be surprised if you’d heard of it before. Because despite the fame of both the subject matter and the author, I’ve never met someone who’d read this book independent of my recommendation, and I myself don’t know how I came to have a copy. My copy, nonetheless, has been well-loved; I’ve read it four or five times by now.

Personal Recollections of Joan of Arc—henceforth simply Joan of Arc—is framed as an English translation of the memoir of Sieur Louis de Conte, a fictionalized version of the real Joan of Arc’s secretary and page. Twain worked on the book for fourteen years—twelve of research, two of writing. It was serialized in 1895 under the name of the fictional translator, Jean Francois Alden. Many events in the novel are wholly fictional, but the broad outlines of the story stick close to those of the actual life of the Maid of Orleans. The book never received much attention, not during Twain’s life and not after his death. It is fine work nonetheless, and Twain, quite sensibly, held it in very high regard:

I like Joan of Arc best of all my books; and it is the best; I know it perfectly well. And besides, it furnished me seven times the pleasure afforded me by any of the others; twelve years of preparation, and two years of writing. The others needed no preparation and got none.2

Having been informed of the book’s existence, the second thing you ought to know is this: Mark Twain was utterly *fascinated* by Joan of Arc, and his interest in her was by no means ordinary, in no way akin to the casual interest you or I might have in a historical figure like Augustus Caesar.3 It was something both more inspired and more intense.

As Andrew Tadie tells it in the introduction, Twain was generally a determinist, and yet believed the actions of Joan of Arc so improbable as to render free will4 their best explanation. In other words, he revered her because he understood her as history’s only example of a truly free human being. Pause on that, for a moment. Here we have Mark Twain, a brilliant man who generally rejected free will, and was yet willing to see freedom in the actions of Joan of Arc, and her *alone.*5 I’m sure there are some incompatibilist determinists among my readers. If that’s you, ask yourself: what would a person have to do to convince you that the ordinary determinstic mechanisms weren’t enough to explain why they’d acted as they did. To hear Tadie tell it, that’s how Mark Twain saw Saint Joan. I think it may be fair to say that no person has ever held another in such high esteem as Mark Twain held Joan of Arc.

Tadie’s characterization is not universally accepted. Going forward, though, we’ll assume that he does have the right of it—that will be the best way to frame a fruitful discussion.

§1: If the Paladin Spoke True

I think the connection to philosophy of religion now ought to be clear enough. For one of the most intuitive and well-known answers to the Problem of Evil is the Free Will Theodicy, which asks us to imagine that the evil in the world is the product of free human choice, and that God permits the evil because the freedom is so very good. More sophisticated versions of the theodicy will appeal not to the bare value of freedom, but to the value of difference-making. A simple illustration: suppose I see you drowning in a pond and freely choose to jump in and relieve your suffering. According to the proponent of difference-making goods, the fact that you *actually* would have drowned given my omission to act makes my act more valuable, and so does the fact that I really could have chosen not to save you. Many people, myself included, are initially skeptical of the idea that difference-making really does have a value of its own, and even more skeptical that the mere *possibility* of difference-making justifies the existence of evil.

Now, in a case like the one I’ve given, we might also appeal to the value of the relationship between savior and saved. If I pull you from the pond, you may be extremely grateful, and your gratitude may be the foundation of a beautiful friendship between you and I. But your gratitude may depend on my having been free to *not* save you—were my hand being forced, so to speak, you would have only luck to thank. This is a more concrete kind of value, easier to get an intuitive sense of.

But though this is a plausible genuine value that difference-making helps to instantiate, it is not a value inherent to difference-making. There are seemingly other ways to instantiate equally valuable relationships, ones that don’t involve suffering. And so we might think, all else equal, that God would prefer these. If that’s right, the theodicy that appeals to difference-making will have to maintain that the added value lies, at least to some extent, in difference-making itself.

Now, in trying to understand the unique goods that are enabled by free will, another thing we might appeal to is Twain’s Joan of Arc. If Twain really did consider her the only free human being yet to walk the Earth, we might expect his book to give us the clearest possible example of the value of freedom, and perhaps the value of difference-making. And indeed, such an example is what we shall soon get.

Before we speak of Joan, though, we ought to speak of her lovely and pathetic childhood playmate, Edmond Aubrey, who is most often called the Paladin. He acquired his nickname in childhood, and one needs only listen to a bit of his childhood speech to understand why:

Now I will tell you why I remain chafing here in a bloodless tranquillity which my reputation teaches you is repulsive to my nature. I do not go because I am not a gentleman. That is the whole reason. What can one private soldier do in a contest like this? Nothing. He is not permitted to rise from the ranks. If I were a gentleman would I remain here? Not one moment. I can save France—ah, you may laugh, but I know what is in me, I know what is hid under this peasant cap. I can save France, and I stand ready to do it, but not under these present conditions. If they want me, let them send for me; otherwise, let them take the consequences; I shall not budge but as an officer.

The Paladin, as you may now have guessed, is a proud boy, and boastful, and very much a liar, even if not wholly conscious of it. I shall allow you the chance to read the book for yourself and see what a wonderful man the story makes of him. In the meantime, I’ll only ask a question. All this that the Paladin says—what would it be worth if all of it were true? It is an important question; the whole matter of difference-making turns on it. For while the Paladin boasts to his friends in idyllic Domremy, the situation of his country is quite different. All around France, villages burn, raided by the English or the English-hearted Burgundians. France itself has been all but sold to the English. And the Paladin is correct when he says that a peasant boy has no chance of rising through the ranks, no chance of leading armies; if it were true that he was capable of the feat he describes, it would still be true that no one would let him prove it.

My thoughts are pulled in different directions here. When I consider the question in the abstract, it does strike me that being the sort of person who *would* do a valuable thing, if given the chance, is just as good as actually having done it. Then, however, I consider particular cases. Imagine, for instance, some young man who takes up boxing, and, due to a combination of talent and hard work, cultivates a level of skill such that he’d easily be able to win a global tournament. Sadly, there is no such tournament for him to compete in. In this case, it strikes me that a boxer who *could* win a global tournament, though admirable, is not nearly so admirable as a boxer who actually *has* won a global tournament. But why might this be?

Well, simply put, it strikes me that people generally only deserve credit for things they’ve actually done. This is intuitive in its own right, and we also might note that the alternative has strange implications. If people deserve credit for good deeds that they *might* have committed had they found themselves in different circumstances, do they also deserve blame for merely hypothetical misdeeds? It’s hard to see how you could accept the former without accepting the latter, and the latter is indeed very hard to accept.

But does our analysis of the boxer case hold up in a morally changed scenario? Well, suppose that the boxer would give any money he wins in a tournament to charity. It’s then unfortunate both for him and for his potential beneficiaries that there was no tournament for him to compete in, but the intuitive analysis is the same: being the sort of person who would donate tournament winnings to charity is not the same—not as valuable—as being a person who actually does.

Note the implications for theodicy: if being so disposed that you *would* do a good thing is just as good as actually doing the thing in question, evil can’t be justified by difference-making. The world we live in has plenty of opportunities to make a difference, but the view I’ve just articulated implies that it would be better for people to be so disposed that they *would* make a difference if given the chance, but for there to be no actual differences to make. Furthermore, I think the success of the theodicy is going to require us to say that *actually* making a difference is better than trying and failing to do so. That’s because the Free Will Theodicy rests on the idea that there doesn’t *need* to be evil, that evil is the result of our actions or omissions. The theodicy is not nearly so plausible if the good God is aiming at is merely for people to *attempt* to make a difference. If it were, God could simply finish the job in any case where some person tries at difference-making and fails, and so the theodicy wouldn’t do nearly so well at actually explaining evil.

But now we might worry a bit about moral luck. What if the boxer fails to win the tournament because a freak accident breaks his leg in the semifinals? He has failed to actually make a difference in the lives of those he’s donated to, but through no fault of his own. In the same way, the success or failure of any difference-making effort might often come down to luck.

To borrow another classic example, let’s imagine two drunk drivers, one of whom happens to find the roads empty as they return home, the other of whom meets traffic and causes an accident which kills a few people. Have we really any grounds to say that the conduct of the latter is worse than the conduct of the former? One driver found blood on his hands and the other didn’t, and mere chance determined which was which. Can one man be more blameworthy than another merely because he was less lucky?6

First off, I’ve shifted the discussion a bit. We were initially talking about what might make one situation better than another, and now we’re talking about what might make some person blameworthy in some situation. To tie the two together, we’ll tack on the plausible assumption that good things are better—and bad things worse—when some person is directly responsible for them. Compare a murder to death by lightning-strike—the latter seems better even when we hold the amount of pain suffered fixed. On the flip side, compare the case of one person rescuing another from a rockslide to a case in which the second person simply happens to get lucky, and is missed by all the rocks.

All that said, here’s one thought I have: it may be the case that the second driver really is more blameworthy than the first, and that this is because he ultimately does something worse. But both drivers are nonetheless on the hook for drunk driving. And so we might say that drunk driving belongs to a certain class of actions such that, when you take them, you open yourself to accountability for their consequences *whatever they may be.* Instead of saying the moral status of certain actions is determined by chance in an unfair way, we can instead say that certain sorts of actions are akin to calculated moral risks—using the word “calculated” in a loose and colloquial sense, of course.7 And note that this accounts remarkably well for the way we tend to think about such cases. Suppose that instead of killing anyone, the unlucky drunk driver merely causes some severe injury. If he is made to pay for the treatment of the person he has injured—to take responsibility for the consequences of his actions in a very direct way—he is hardly licensed to complain, nor to point to other, luckier drunk drivers and demand that they pay their share. Likewise, no one is licensed to take money from the lucky driver in order to fund the treatment of an accident victim, not when when the lucky driver himself has not caused an accident.

Two important notes:

First: If there is a such thing as a calculated moral risk(CMR, going forward), then I don’t think only bad actions will fall into this category. An illustration: when I was at Pearl Harbor’s Aviation Museum, I learned the story of Alymer Robinson.8 He owned Ni’ihau—one of the smaller Hawaiian islands, which I believe to be basically uninhabited—in the years leading up to the Japanese attack on Pearl Harbor. As early as 1924, people in the U.S. military had suspected that such an attack might one day come. At some point in the 30s, someone from the Navy contacted Robinson and told him about these suspicions. The fear was that, if the Japanese to attack Hawaii and establish a presence there, they might use Ni’ihau as an airstrip. To prevent this, Robinson used his own money to furrow most of the land on the island, making it unusable for launching planes. Now, the furrowing did indeed make trouble for the one Japanese pilot who crash-landed on the island,9 but Ni’ihau was never properly occupied by the Japanese, and so Robinson’s cleverness didn’t amount to much. He still deserves credit for being clever in the service of a noble goal, certainly, but not for frustrating a Japanese occupation of the island, for such an occupation never actually occurred. If drunk driving is a paradigm case of a blameworthy CMR, this seems like a paradigm case of a praiseworthy one.

Second: We might ask ourselves how to distinguish between actions that are CMRs and actions that aren’t. It is tempting to say that every action which might have morally weighty consequences is a CMR. But consider this: even if I were to take my dog for a walk, there would be some tiny chance of disastrous consequences. Am I responsible for those consequences if they do come to pass?

I won’t stick on this subject for long, but my thought is that there are two ways of handling this. First off, we might say that actions are CMRs only with respect to outcomes with sufficiently high probabilities. These would be probabilities high enough that the person performing the act ought to take the prospect of these outcomes seriously. So if I destroy the world by taking some action that had a 1/10^1000 chance of destroying the world, and otherwise would have brought about some good outcome, I won’t deserve any blame. The other thing we might say, though, is that every action is a CMR with respect to all possible consequences, but the degree of blame one deserves for having done a certain thing doesn’t depend *only* on the outcome of their choice, but also on the likelihood of that outcome(or perhaps how likely the person in question believed the outcome to be). So in the case where I destroy the world, the amount of blame I deserve is negligible because the probability of that outcome was negligible itself, but I do nonetheless deserve blame when I would not have otherwise. Of these options, I slightly favor option 1, but option 2 has the great virtue of leaving almost no room for vagueness. There may, of course, be some elegant way of threading the needle. As far as this essay goes, I’ll leave it there.

Now, let’s recall that we began moving down this road in the first place by investigating the claim that people only deserve credit for things they’ve actually done. A plausible principle when analyzed in the abstract, and some of you may even agree that the idea of the CMR helps to make sense of it. But, setting aside the previously discussed considerations of luck, don’t counterexamples abound?

Let’s consider two kinds. First off, suppose some doctor tries to do some good thing—save the life of a dying child, for instance—and fails. Must we say she deserves no credit?

Here, I think, the answer is straightforward. The doctor doesn’t deserve praise for saving the child, because she hasn’t, but she does deserve credit for *trying* to save the child, which is praiseworthy in its own right. All that’s required to make that work is for an attempt to save a child’s life to have a value of its own, which is plausible. Trying to save a child’s life *is* actually doing something, after all—it’s just not doing the same thing as successfully saving a child’s life. In such cases, however, CMR-theory will require us to say that failure is *less* praiseworthy than success.

A second kind of case, perhaps more troubling: my friend confesses to me that, had he ever been given the chance to murder a puppy, he would have done it. He hasn’t done so only because the opportunity never presented itself. Are we really committed to saying that he deserves no blame simply because he has killed no actual puppy?

Well, it’s true that my friend can’t be blamed for having killed a puppy, and we can’t justly treat him as a puppy-murderer. We can blame him, however, for failing to have cultivated a better disposition, and the blame this warrants may actually be quite substantial. And we needn’t even say that my friend is blameworthy because he’s made himself more likely to do something bad in the future. We *could* say this, but we could also just say that having a puppy-killing disposition is bad in its own right, and would be bad even in a world with no puppies to kill.

Those strike me as the most obvious counterexamples, and I think I’ve answered them adequately. Accordingly, I think the CMR provides one viable framework for understanding—at least at the level of first-order claims—how it could be that what really matters is largely what people actually do, consequences included.

Now, I don’t personally accept something like CMR-theory, and perhaps in the future I’ll say why not. But recall that we’re digging into theodicy here, and the best theodicy does not necessarily have to fit cleanly into the best independent theory. The goal of theodicy is to reconcile God with evil in a satisfying way, so that we might see how the two of them can coexist. Accordingly, if we have some theodicy that works within a viable framework, but not the framework that appears best, and if this theodicy really does show the coexistence of God and evil to be more likely than we’d otherwise imagine it to be, that is sufficient.10 It also should be said that there may be other, better frameworks that can capture the relevant benefits of CMR-theory.

With all this stowed in the back of our minds, let’s return to the case of the Paladin. We are now, I think, properly equipped to say that—if the Paladin had spoken true—he would have deserved praise for cultivating a disposition such that he would’ve been able to save France, had the chance to come to him. And perhaps the praise this warrants is very great. But the praise warranted by *actually* saving France is far greater. And this, of course, is what Joan of Arc did.

§2: Joan of Arc Receives a Wound

We’re going to return to the book now, because I want to describe Joan’s deeds to you. I want you to get a visceral sense of their impact before you see them philosophically analyzed. I want, in particular, to narrate for you my favorite episode in the story of Mark Twain’s Maid of Orleans. I’ll do it, of course, with a great deal of help from Twain himself.

The episode I’m thinking of comes in the second part of Twain’s book, “In Court and Camp.” It immediately follows Joan’s foot having received a minor wound. In the wake of the wound, Joan—unable to read or write—dictates a letter home:

“But the thing which moves me to write now, is to say that when you presently hear that I am wounded, you shall give yourself no concern about it, and refuse faith to any that shall try to make you believe it is serious.”

She was going on, when Catherine spoke up and said:

“Ah, but it will fright her so to read these words. Strike them out, Joan, strike them out, and wait only one day—two days at most—then write and say your foot was wounded but is well again—for it surely be well then, or very near it. Don’t distress her, Joan; do as I say.”

A laugh like the laugh of the old days, the impulsive free laugh of an untroubled spirit, a laugh like a chime of bells, was Joan’s answer; then she said:

“My foot? Why should I write about such a scratch as that? I was not thinking of it, dear heart.”

“Child, have you another wound and a worse, and have not spoken of it? What have you been dreaming about, that you—”

She had jumped up, full of vague fears, to have the leech called back at once, but Joan laid her hand upon her arm and made her sit down again, saying:

“There, now, be tranquil, there is no other wound, as yet; I am writing about one which I shall get when we storm that bastille tomorrow.”

Joan goes on to say that her heavenly Voices had told her of this wound seven weeks ago. And so for nearly two months she had been begging the Dauphin11 to let her fight, persuading the generals of France to stick with her plans, then riding out herself with her armies, knowing all the while that come May 7th, she would be struck down to writhe in agony.

Catherine, of course, tries to persuade her not to go out into battle. Joan resists, but when resistance becomes impossible she makes a wily concession. She will stay away from the fighting if Catherine only tells her to, but she will need help composing a letter to give her generals and her army proper notice:

“Then you will send word to my headquarters that I am not going?”

“Oh, gladly. Leave that to me.”

“It is good of you. And how will you word it?—for it must have proper official form. Shall I word it for you?”

“Oh, do—for you know about these solemn procedures and stately proprieties, and I have had no experience.”

“Then word it like this: ‘The chief of staff is commanded to make known to the King’s forces in garrison and in the field, that the General-in-Chief of the Armies of France will not face the English on the morrow, she being afraid she may get hurt. Signed, JOAN OF ARC, by the hand of CATHERINE BOUCHER, who loves France.’”

Twain’s Joan is full to bursting with this sort of cleverness, but full as well of a fierceness and a fire. See also what she says after she has finished dictating her letter home. Her confessor comes bringing a knight, the knight bringing a message, and Joan learns that the war-council means to retreat from the fray, to let Orleans remain besieged, for the city is well-provisioned. Here is the scene that follows:

“The incurable cowards!” exclaimed Joan. “So it was to get me away from my men that they pretended so much solicitude about my fatigue. Take this message back, not to the council—I have no speeches for those disguised ladies’ maids—but to the Bastard and La Hire, who are men. Tell them the army is to remain where it is, and I hold them responsible if this command miscarries. And say the offensive will be resumed in the morning. You may go, good sir.”

Then she said to her priest:

“Rise early, and be by me all the day. There will be much work on my hands, and I shall be hurt between my neck and my shoulder.”

And let us now see what was prophecied—let us now speed to the scene of the wound.

the iron bolt from an arbaquest struck between her neck and her shoulder, and tore its way down through her armor. When she felt the sharp pain and saw her blood gushing over her breast, she was frightened, poor girl, and as she sank to the ground she began to cry bitterly.

If the English had captured Joan then, Charles VII. would have flown the country, the Treaty of Troyes would have held good, and France, already English property, would have become, without further dispute, an English province, to so remain until Judgment Day. A nationality and a kingdom were at stake there, and no more time to decide it in than it takes to hard-boil an egg. It was the most momentous ten minutes that the clock has ever ticked in France, or ever will. Whenever you read in histories about hours or days or weeks in which the fate of one or another nation hung in the balance, do not you fail to remember, nor your French hearts to beat the quicker for the remembrance, the ten minutes that France, called otherwise Joan of Arc, lay bleeding in the fosse that day, with two nations struggling over her for her possession.

And you will not forget the Dwarf. For he stood over her, and did the work of any six of the others. He swung his ax with both hands; whenever it came down, he said those two words, “For France!” and a splintered helmet flew like eggshells, and the skull that carried it had learned its manners and would offend the French no more. He piled a bulwark of iron-clad dead in front of him and fought from behind it; and at last when the victory was ours we closed about him, shielding him, and he ran up a ladder with Joan as easily as another man would carry a child, and bore her out of the battle, a great crowd following and anxious, for she was drenched with blood to her feet, half of it her own and the other half English, for bodies had fallen across her as she lay and had poured their red life-streams over her. One couldn’t see the white armor now, with that awful dressing over it.

The iron bolt was still in the wound—some say it projected out behind the shoulder. It may be—I did not wish to see, and did not try to. It was pulled out, and the pain made Joan cry again, poor thing. Some say she pulled it out herself because others refused, saying they could not bear to hurt her. As to this I do not know; I only know it was pulled out, and that the wound was treated with oil and properly dressed.

Joan lay on the grass, weak and suffering, hour after hour, but still insisting that the fight go on. Which it did, but not to much purpose, for it was only under her eye that men were heroes and not afraid. They were like the Paladin; I think he was afraid of his shadow—I mean in the afternoon, when it was very big and long; but when he was under Joan’s eye and the inspiration of her great spirit, what was he afraid of? Nothing in this world—and that is just the truth.

Toward night Dunois gave it up. Joan heard the bugles.

“What!” she cried. “Sounding the retreat!”

Her wound was forgotten in a moment. She countermanded the order, and sent another, to the officer in command of a battery, to stand ready to fire five shots in quick succession. This was a signal to the force on the Orleans side of the river under La Hire, who was not, as some of the histories say, with us. It was to be given whenever Joan should feel sure the boulevard was about to fall into her hands—then that force must make a counter-attack on the Tourelles by way of the bridge.

Joan mounted her horse now, with her staff about her, and when our people saw us coming they raised a great shout, and were at once eager for another assault on the boulevard. Joan rode straight to the fosse where she had received her wound, and standing there in the rain of bolts and arrows, she ordered the Paladin to let her long standard blow free, and to note when its fringes should touch the fortress. Presently he said:

“It touches.”

“Now, then,” said Joan to the waiting battalions, “the place is yours—enter in! Bugles, sound the assault! Now, then—all together—go!”

And go it was. You never saw anything like it. We swarmed up the ladders and over the battlements like a wave—and the place was our property. Why, one might live a thousand years and never see so gorgeous a thing as that again.

Now I think you must see clearly Joan’s courage, her force of untainted will—the fruits of real freedom. Though it bloats the essay, I am glad to have given you so much of Twain’s story. For his work here is fine, and far too little-known. And I am glad to have done so, also, because I suspect that your heart may now be swelling with the same sort of second-hand pride that Twain has instilled so strongly in me. I am not, generally, an especially brave man. But when I read this scene, I feel that I could be like the Dwarf, breaking a whole army of enemy soldiers but doing it not for glory or bloodlust or the thrill of the fight; doing it, rather, for France, who was called Joan of Arc.

§3: On the Possibility of Heroism

The heroic deeds of Joan of Arc were possible only because others had committed horrible acts. Now, suppose it is the case that the value of people’s actions is proportional to the credit/blame they deserve for the actions in question. Suppose also that people deserve credit/blame for their actions—at least in part—because of the actual consequences of these actions. The CMR-theory we’ve previously discussed is one way—but not the only way—for all this to be true.

I do not accept these claims myself, but it is at least conceivable to me that they *could* be true. And if they are true, then there is a real value in the deeds of Joan of Arc which is enabled by the existence of evil.

As far as theodicy is concerned, I have as yet talked only about the Free Will Theodicy. One might think, however, that the story I’ve relayed instead calls Soul Building to mind. For is it not the spirit of Joan of Arc, her steadfast courage, her playful wit, her deep compassion, that we ought to be in awe of? But note that Joan’s character does not *change* as a result of the episode I’ve described. In Twain’s book—in accordance with his reverence for her—her conduct is saintly from beginning to end. She is at all times both superhuman and so deeply human as to inspire our greatest sympathies. But static, all the while. So, in reading Twain’s book, what *I* am in awe of is not how Joan grows and flourishes in response to the challenges set before her. It is, rather, how Joan freely chooses, without error, what is right and what is good in the face of each challenge. The free choice of good amidst evil, the free choice of good for good’s sake: that is the value of which Joan of Arc has now become, for me, the shining exemplar. And her free choice is all the better, her example all the brighter, because it was *hard.* Because of the tears and the blood. And note as well that Joan is not an example only to me, but also to Louis de Conte, to the Paladin and his friend Noël Raineguesson, to the steadfast Dwarf, to the great swaggering general La Hire, to the humble Catherine Boucher. And to Mark Twain. We mustn’t forget Mark Twain, who was so inspired by Joan that he spent a whole fifth of his life laboring to tell her story, and from those many years drew his greatest work. So here is another thing of great value that comes from Joan’s conduct—the admiration of others, the *deserved* admiration.12 And then, of course, there is the value of the story.

And if Joan were merely a puppet, what value would there be in any of these things? And would their value be so great if Joan had not needed to struggle, struggle *hard,* to conduct herself as she did?

And what if the world she found herself in, the world she strove to make better, what if this world had not been the product of the free choices of others?

The first two questions were merely rhetorical, but I will speak directly to the last. It matters a great deal, I think, that Joan is an exception. It matters that she has risen to a challenge which others have failed to meet. Just as the value of a champion boxer’s skill is, in part, a function of the comparative lack of skill found in most people, the value of a saint’s right action is, in part, a function of the wrong actions of most. But the actions must be free—it must be the case that anyone *could* be a saint, that most simply fail. It must be the case that Joan of Arc made a difference where others could have, but failed to.

I’m not saying anything new here. I have heard all these arguments made in the past, and I suspect you have too. Always, however, in the abstract, or with the aid of some well-trodden thought experiment. And sometimes, you really do just have to see to believe.

Twain’s Joan of Arc is not, of course the real Jeanne d’Arc. Twain, despite his admirable fidelity to history, has not written a biography here. But if God gives us the chance to make a difference because he wants to see the sort of shining value that Twain saw in Joan of Arc, then I won’t say that this justifies all the evils in the world. But I will say that it makes it far easier for me to understand them. And for that, I have Mark Twain to thank.

§4: The Road to Omission

Though I haven’t given much in the way of argument, I think I’ve done enough to explain why I’ve personally come to find the Free Will Theodicy more appealing than I once did. Let’s now pave the road to Omission.

The Omission Theodicy is, in essence, an *extension* of the Free Will Theodicy.13 Such an extension is necessary because the Free Will Theodicy is famously inadequate in its ability to account for *natural* evils.

Natural evils, as I see them, belong to two primary categories. First, we have the bad things that animals do to each other. Given that animals are not generally free in the sense that’s important for theodicy, the standard Free Will Theodicy can’t account for their actions, especially not in situations where no human has the opportunity to step in and save the animals from each other, as is the case most of the time. Second, we have the bad things that come about as a result of no particular creature’s action. If, for instance, I get struck by lightning and die on the spot, and no one was in a position to help, it’s hard to see how the possibility of difference-making justifies this—there was no difference to be made.

In response to such cases, Brian Cutter and Philip Swenson propose the Omission Theodicy. The idea is that there might be some very powerful entities—be they aliens or angels—who are responsible for supervising the various life-supporting corners of the Universe(s). For now, I’ll suppose it’s angels. If the angel responsible for Earth has been doing a bad job of it, then we might expect to see all the terrible things we actually see, the ones for which it seems that no one in particular is responsible. At the same time, the possibility of difference-making would remain—our angel could step in whenever he pleases. The Omission Theodicy also crucially appeals to the infinite value of the afterlife relationships that we might cultivate with the angels who protect us—the value is infinite because the relationships will be very good, and will also last forever.

Now, there are various objections to the Omission Theodicy, and I consider some of them quite forceful. The obvious one is that the Omission Theodicy works only if the Free Will Theodicy also works, and it may not. That’s why I’ve gone to such lengths to explain why I’ve recently begun to find the Free Will Theodicy more attractive.

But there’s one other major objection that I’d like to touch on here: Omission really seems not to predict that we’ll find ourselves in the circumstances we do. Why? Well, to secure the value of difference-making goods, God needs only make it *possible* that some angel freely chooses not to make things better on Earth. But God doing that is entirely compatible with making the disposition of angels such that 99 percent of them—or far more—do an excellent job.14 And then we wouldn’t expect to find ourselves in one of the vanishingly few places where the angel makes a mess of it.

So here’s my proposed revision: make it extremely, extremely difficult for the angels to do their job right.15 This would render it far, far more likely that many of them fail, which would render it far more probable that we find ourselves in a world that’s inadequately protected. But it would also make those angels who do manage to succed in preventing evil comparable to Joan of Arc—they would deserve credit for having done things which other angels *could* have done, but did not have the force of will to see through. And just as Joan of Arc was admired by Mark Twain—an American writer who lived centuries after she did—these angels could have good afterlife relationships even with the people who *weren’t* under their power, relationships grounded in an admiration akin to that which we feel for rockstars and professional athletes. And *these* relationships could also have infinite value when stretched over an infinite afterlife.

§5: The Triple Theodicy16

Call what I’ve just described the “Arduous Omission” Theodicy. Arduous Omission is, I think, a bit of an improvement on bare Omission—it may seem somewhat less a priori plausible, but it explains the world we see much better than the simplest version of the Omission Theodicy.

I’m not done, however. I’m about to get much, much wackier.

Let’s now introduce two other theodicies. The first is another revision of the Omission Theodicy, courtesy of the admirable Amos Wollen and an unnamed Baha’i reader of his blog.17 Amos calls this the Padre Pio Omission Theodicy—I’ll call it Pio Omission for brevity. The idea behind Pio Omission is this: we can make an Omission scenario more plausible if we can construct a version of Omission that doesn’t require the presence of angels or advanced aliens. If we can make it such that the people responsible for omitting to prevent evil are just ordinary people, then we’ll be on the safe side of Ockham’s Razor. Here’s one way that this could be the case: God wants people to pray for things to be better than they are, to reach out towards the Divine with a sincere hope for a better world. The evil that exists in the world might then be the result of the fact that many of us omit to do this.

The obvious problem here is that this can’t account for all the bad things that happened before anyone capable of praying was around. But what if it *could?* Amos—recalling that the famous Catholic Saint Padre Pio was known to pray for the salvation of his long-dead great-grandfather—suggests that perhaps we are responsible for praying that the *past* be made to go better than it otherwise would have gone, and that *this* is the relevant omission.

There are, of course, a lot of details to be hashed out here. The whole idea is a bit mind-boggling. Pio thought God was able to respond to prayers from all times because God was outside of time. Amos posits a few other mechanisms, several of which rely on God’s omniscience in some way or another. But I want to go crazier. In a moment, when I introduce our second theodicy, you’ll understand how.

Now, this second theodicy, which comes courtesy of Samuel Lebens and Tyron Goldschmidt, is the craziest one I know of.18 Call it the Divine Proofreader Theodicy. The key concept for this one is hypertime, and it’s a bit tricky to explain, so I’m going to ease us in by asking an important question: what would it mean for God to change the past? Suppose that I’m currently at time t1, that event A happened at t0, and God wants to make it so that event B had happened at t0 instead. Assume B-theory(that the past, present, and future are all equally real).19 So God makes the change; he swaps event A with event B, making the necessary tweaks to leave everything else the same. But God is doing this from outside of time. From my perspective at t1, it will not seem as if anything has changed, as if it used to be the case that A had occurred but now B has occurred instead. It will seem to me that B *always* had occurred.

But then it doesn’t seem quite right to say that there’s no point at which A had occurred rather than B. God made a *change,* didn’t he?

Here’s one way we might describe the situation: there is no *temporal* point at which A occurred rather than B, but there is a *hyper-temporal* point at which that was in fact the case. And so we say that, at hypertime h0, it hyper-was the case that A had occurred at t0, but at h1, it hyper-was the case that B had occurred instead.

For our purposes here, we won’t get bogged down further in how one might understand hypertime.20 The one thing I will note is that Lebens and Goldscmidt claim that we have good theodicy-independent reasons to accept hypertime. For brevity’s sake, I won’t review those here. Instead I’ll simply say that, if they’re right, the Divine Proofreader Theodicy does not offend against parsimony in the way that the standard version of the Omission Theodicy does.21 If you’re interested in learning more about how we might motivate hypertime, I recommend reading the paper, which I’ve linked in footnote 16.

Now, I’m going to present the intuitive, light-on-wacky-metaphysics version of this theodicy. For it is, by my lights, shockingly intuitive. It works like this: if you do something bad in your life, God gives you a do-over. Suppose at t0 I freely perform evil action A, when instead God would have liked me to freely perform good action B. At h0, it is the case that at t0 I performed A. But at h1, God could give me another chance, and it could then be the case that I performed B instead of A. The neat thing is that these do-overs don’t have to be handouts. Suppose at h0 it is the case that I committed A at t0, and also the case that at t1 I endured some ordeal C which changed my character in such a way that I now *would have* done B instead of A. God can change the past in light of this, making it so that my t0 self at h1 has the character which I cultivated as a result of undergoing C at h0/t1. The idea is that in the long run, God can do this with *all* evils, creating—through a series of hypertemporal changes—a time-structure in which everything is good, and in which this is the result—at least to some extent—of free choices.22

I’m just going to note once again that you really do need to engage in some pretty wacky metaphysics to resolve some damning problems with the most basic version of the theodicy. Again, I recommend reading the paper to get a handle on all that.

Now, the Proofreader Theodicy, neat and fun as it is, does not seem to fully account for natural evil. But if we add Omission, it can. And if we add Pio Omission in particular, we have a natural complement for our system of hypertemporal changes. For if the Proofreader Theodicy is right, we get a picture on which God is willing to change the past in accordance with actions taken in the present. Given that, it’s not too much of a stretch to say that he might also be willing to change the past in response to the sincere prayers of people who exist in the present. With Proofreader and Pio Omission put together, we get the possibility, conditional on enough of us praying hard enough, of making a world where the Holocaust never happened, where evolution was a gentle process, where the human being was a herbivore, and the lion too. Extrapolate, and evil disappears entirely at the end of hypertime. And if we combine Proofreader with Pio Omission, we get Arduous Omission almost by default; While it might not be difficult for one person to make a practice of praying for the past, it would certainly be difficult to convince great numbers of people to do this. That’s something that I haven’t the strength of will to do, even if I did have the capability. But Joan of Arc could have done it.

One distinct advantage of combining these theodicies is that their conjunction doesn’t commit us to the implausible claim that the world we currently live in features the optimal amount of (latitude for) evil. According to the conjunction of these theodicies, we’re most likely simply somewhere in the middle—hyper-temporally speaking—of a hyper-long process. And this process would merely be the optimal way of actualizing a world that *does* have the optimal amount of evil, which is none.

Another distinct advantage, at least for certain parties to the debate over the problem of evil: if some version of the Triple Theodicy is compatible with religions like Catholicism that affirm the existence of Hell, it might also imply—conditional on such a religion being true—that Hell is empty. It might be that people are saved or damned not in accordance with the actions they take in this life, but in the life that will be at the end of hypertime. In this way, everyone could be saved.

And so I present to you my Triple Theodicy: Pio Omission, Arduous Omission, the Proofreader. These three harmonize remarkably well with each other, and in harmony they give us a remarkable promise. They give us the promise of a world where evil is unreal, and where this is the result of our free choices. And in this world, Joan of Arc *still* deserves the admiration of her countrymen.

Conclusion

This has been a long essay, and a bit of a ramble, so I thank you for your patience. And I hope *you’ll* thank *me* for stopping here, when in fact there’s much more I could say about all of this!

I had first intended only to write about Mark Twain’s marvelous book and how it had swayed me a bit in the direction of finding the Free Will Theodicy satisfactory. The essay I’ve actually written turned out much longer and much bolder, and I think this is both to its detriment and its credit. I’ve sacrificed a bit of elegance, but that sacrifice has given me the chance to discuss a wide variety of ideas that are, in my opinion, very exciting. But I consider this one more of a whirlwind tour than a conclusive take on anything, and so I’ll leave it for you to tell me whether any of these ideas are actually any good. It’s been a pleasure anyhow.

As always,

Ask me about my preferred genre of music, and I’ll probably tell you that I mostly listen to prog rock; This is technically true, but misleading, because I don’t listen to prog rock so much as I just listen to Rush. For years on end, Rush has been the only proper band whose songs I’ll go out of my way to listen to. I don’t have enough musical sophistication to explain why this music in particular has, to my ears, such a singular appeal. It may be the uncommon fusion of loud noises and propulsive rhythms with measured and contemplative lyrics. It may be that I just happened to encounter Rush during a formative period, and that I’m resistant to trying new things. It may be my Canadian blood. But the reason why I love Rush doesn’t matter much to me anyhow—not nearly so much as the plain fact *that* I love Rush.

Now, as fans will know, it’s been a decade since Rush last toured. As some non-fans will also know, their illustrious drummer and lyricist, Neal Peart, died in 2020. Another tour was not to be expected after that, and so the current series of Fifty Something concerts constitutes, for many fans, a sort of minor miracle.

For me, it constitutes a major one. See, you’ve got to understanding that I didn’t even really *know* about Rush until after Neil Peart was already gone, until the band was already presumed to be done with touring. And so I spent many of my teenage years listening to the work of a group that was, as far as I was concerned, painfully recent history. And I was entirely obsessed. I recall having once gone to summer camp, taking along with me an old iPod Nano. The Nano was loaded with just about all the music my dad owned, at least two thousand songs. Among them were about two albums’ worth of Rush’s greatest hits, added at my request. The rest of it was good music, no doubt. But it wasn’t *Rush,* and so I didn’t listen to it. Instead, I spent a whole month revisiting the same twenty-odd songs. I have such fond memories of lying on the grass at night, staring at a bright sky and using my fingers on my palm to mark the rising notes in the endless staircase that begins “Xanadu.”

And then, of course, singing along:

To seek the sacred river Alph

To walk the caves of ice

To break my fast on honeydew

And drink the milk of Paradise

“Xanadu,” as it happens, was the first song I heard tonight.

“Xanadu,” as it happens, is a song about immortality. And tonight—with Annika Niles instead of Neal Peart on the drums—Rush chose it as the kick-off for their show.

The song is cynical about the prospect of eternal life, but I am not. Because any time I hear the first notes of “Xanadu” ring forth, I am absolutely sure that I do not want this time to be the last time. So it was tonight.

See, you’ve got to understand that Rush is not merely my favorite band, but—as far as my picture of the musical world is concerned—something close to the *only* band. And as I’ve listened to them over the last few years, I have often reflected on how profoundly sad it was—if only for me—that I would never get to hear them play live. But I was wrong.

Tonight, hearing the opening notes of “Xanadu,” I felt as if I were growing older, and at once that the world was growing younger. I began to cry, and my breath grew short and choppy in a way that it seldom ever has.

Because, tragic as it was to see Neal Peart absent, Alex Lifeson and Geddy Lee really can still *play.* They play so damn well that, for once, I didn’t really mind the presence of the too-loud fanatic who inevitably lurks in the row behind me at every concert I attend. For once, I wanted to be just as loud, just as selfish in my displays of enthusiasm. Because before me was the most beautiful thing in the world, made more beautiful still by the fact that I’d never imagined I’d really get to see it.

Admittedly, the band didn’t play all the songs I’d hoped they’d might. However, that would’ve taken eight hours, so I think I can forgive them for it. Tonight, I feel that I could forgive them anything, because I’ve never been to a better show.

Tonight, many of the songs were played as a sort of tribute to something wonderful, something deep and storied that I now feel I possess a larger part of. Many were also played as a tribute to peerless Neal Peart. And of these, none were better or more touching than “The Garden,” which again brought tears to my eyes. I won’t quote the lyrics this time. I want you to listen to it for yourself:

And then I want you to imagine more than ten thousand people gathered together, seizing the first chance in a decade to receive these solid-ruby soundwaves. I want you to imagine how they cheered. And I want you to picture the scene that followed the show: one kid just barely out of high school standing on the sidewalk with a crowd of grey-haired hippies, all of us listening to a drummer on the street play along to recordings of the songs we’ve just heard live, because it’s been ten years and we still haven’t had our fill. And we’re singing along, singing and talking and laughing in turn. The guy next to me tells me this is his sixth show, tells me he came down two hundred miles from Canada to see. He offers me a hit of what I assume is a weed pen, and I decline, but we keep talking. I stick around for fifteen minutes or so, then head off to catch my train. The guy with the weed pen probably stuck around for an hour more.

Where else could you find greater love and respect? Neil Peart may be dead, but his music *lives.* And tonight, his lyrics were echoed by the voices of people he never met, and he got the full measure of the respect he’s earned.

Here, from “Headlong Flight,” are some of those lyrics.

All the highlights of that headlong flight

Holding on all with all my might

To what I felt back then

I wish that I could live it all again

I have stoked the fire on the big steel wheels

Steered the airships right across the stars

I learned to fight, I learned to love and learned to feel

Oh, I wish that I could live it all again

Tonight, three hours after the end of the show, this is already how I feel.

But Rush lives, man!

Rush lives.

If you read my blog, there’s a very high chance that you already know *why* the note proved ironic, but I’ll recap briefly. Adelstein spent much of the day following this note engaged in a back-and-forth with another philosophy blogger, Stan Patton. Unfortunately, their dispute quickly turned vitriolic. You’ve likely seen a good chunk of the nasty stuff already, but in case you haven’t, I’ll share the segment that most disappointed me:

All else being equal, I would’ve preferred not to cram this essay full of notes. But I didn’t want to take anything out of context. I also wanted to give Adelstein the last word, as I think he’s basically correct.

I generally respect both of these guys. I respect their contributions to Substack philosophy, and I also respect them as decent people—at least in the cautious way that one might generally extend personal respect to some public figure who one doesn’t really know. Nonetheless, Patton is clearly in the wrong here. Matthew Adelstein is clearly not an insincere person. The extent of my personal contact with the guy is one phone call and a few exchanges in the comments of his posts, but I’ve heard a lot about him from people who do know him personally; it is very, very clear to me that he actually believes all of the wild things he says. I don’t *actually* have any harsh words for him, but I’d call him crazy before I’d call him insincere. And Patton is definitely implying insincerity on Adelstein’s part—the dude literally uses the word “LARP.” And yes, he says he’s articulating a feeling rather than making a definite claim, but that doesn’t get you off the hook. In the final note I linked, Adelstein makes that point just as well as I could.

Okay, that’s the first-order stuff out of the way. But frankly, no one really cares about my verdict on the first-order stuff. I’m not liable to change any minds, and the effort certainly isn’t worth an essay. I wrote this essay, rather, so I could make a higher-order claim. It’s the one in the title.

Here’s what’s rad about Substack: when I published my first full-length essay here, I was still in high school, and had made no secret of it. Within hours of publication, I had a professor in my comments leveling a challenge in good faith, and a grad student willing to spend hours more going back and forth with me. I was immediately taken seriously, immediately invited to take part in a kind of rigorous philosophical discourse that you don’t find in many places on the internet. We’ve got something very special here.

However, we also have a lot of the same stuff that you see everywhere else on the internet. If I go scroll through Notes for a while, it’s pretty likely that I’ll see a silly meme implying that moral realists are all stupid/incompetent, or an outright statement of the same. And I’m not singling out anti-realists here—just as often I’ll see moral realists implying or claiming the same about their opponents. You see a good amount of similarly insulting rhetoric going back and forth between theists and atheists. It is rarely clear that these sorts of notes are friendly jests.

I do get it. It’s fun to make fun of the enemy. I’m happy to do it myself when I’m among friends, and I’m sure I’ve slipped up and done it a couple times in a more public forum. But it degrades the quality of discourse.

I started writing on Substack because I have a *blazing* passion for philosophy, and I wanted to do some verbal sparring with likeminded people. I suspect many of you who write about philosophy here feel exactly the same. Here’s the good thing: there is, in my view, a lot more serious philosophical argumentation on Substack than there are mean digs at atheists/theists/physicalists/panpsychists/whatever. At first, I had suspected that the opposite might be the case. Then I set a timer for five minutes, scrolled Notes at a very fast pace, and noted how many specimens of discourse-degredation I encountered. I saw only one, which I think is a pretty good result, and certainly a good check on my pessimism. But this stuff does still pile up if you spend a lot of time on the platform—more than that, when you see notes of the kind I’m describing, you will often see them restacked by very prominent writers. I don’t want to call anyone in particular out, so I hope you’ve seen some of this stuff for yourself.

Perhaps I’m being histrionic here, but I really do see this as a serious problem. I think I can help illustrate why by writing out some norms for philosophical discourse:

1. Do not call your interlocutors dishonest.

2. Do not call your interlocutors stupid.

3. Do not interpret the claims of your interlocutors uncharitably.

4. Do not claim that the positions your interlocutors hold are nonsensical or unreasonable *without supporting this claim.*

There are exceptions to these norms. You can, for instance, call your interlocutors dishonest if you have good evidence that they’re being dishonest, though you probably shouldn’t call them stupid even if you have good evidence for that. And I explicitly qualified the fourth norm, because I don’t think we should be too soft in making substantive criticisms of weak positions. Rather, the point is that we shouldn’t make the criticisms without the substance.

These are, I hope, norms to which all of us can agree. I have two thoughts about them. First, I think they ought to apply even when your interlocutor is not a specific person. Whether I’m adressing a position held by Matthew Adelstein in particular or one held by theists in general, I should not imply that the people whose positions I’m addressing are being stupid or insincere. When you’re writing something on social media, you generally should not say about a group of people what you would not say to a given individual from the group.

For my second point, a bit of an illustration:

Suppose I see some note on Substack, the content of which is something like this: “I seriously can’t believe that anyone would be a physicalist. You’re just not being rational if you think we can explain consciousness in terms of the behavior of atoms.”

Notes like this don’t dominate the platform, but they aren’t *that* rare either, and if one pops up in my feed, it’s often because some prominent non-physicalist has restacked it.

Please don’t do this. My fictional note isn’t obviously terrible—it doesn’t contain harsh language and doesn’t insult any particular person. But it also isn’t a joke and doesn’t really contain an argument, and it does insult a group of people in a way that our norms do not sanction. So if you see a note like this one, please don’t drop a like, please don’t restack.

To elaborate a bit further, I’ll borrow some phrasing from the Bulldog himself:

Your norms have to be Kantian. They must be universalizeable, unbiased, and fair to all parties. They cannot have arbitrary carveouts, nor can they give you a loophole to do whatever you want if you feel like it[…]The strength of norms is not binary but linear. The more people violate norms, the weaker they become. There is no precise, critical threshold where a norm disappears, only a series of events that make the norm more brittle and frail.

Adelstein was writing here about political violence and the assassination of Charlie Kirk, but I’m not making any comparison between unproductive Substack Notes and political violence; I borrow Adelstein’s words only because I remember having been struck by the power of his phrasing.

If we want norms like the ones I’ve listed—and I hope many people are like me in that they come to Substack searching for discussion that adheres to such norms—we need to stick to them as well as we possibly can. As far as I’m concerned, the difference between my fictional anti-phsyicalist note and accusing Matthew Adelstein of habitual insincerity is a very, very large difference, but is nonetheless a difference of degree and not a difference of kind. So if you don’t approve of the latter, try your best not to support the former.

Now, much as I hate to see the sort of note I’ve described—the sort that subtly violates the norms which its own author would probably endorse in a heartbeat—we’re hardly on some kind of precipice here. That’s lucky. We can stop ourselves from ever getting there.

Thanks for tuning in.

If you’ve read many of my essays, you may have noticed that I really like to bring up Derek Parfit.1 I’m a big fan. However, unlike many of Parfit’s admirers, I don’t think I agree with him about very much. This goes especially for metaethics. I am the sort of person who finds it very difficult to wrap his head around the more robust forms of moral realism. To some extent, I share Parfit’s feeling that something like moral realism *must* be true, that everything is otherwise meaningless. At the same time I struggle to see how realism *could* be true—hence my fascination with Kant.

Parfit, though he was also fascinated by Kant, evidently did not struggle in this way. In fact, he spent much of his life deeply frustrated by the fact that so many intelligent philosophers were seemingly incapable of understanding normativity as he understood it. Bernard Williams was famously a frequent object of this frustration—Parfit was known to tell anyone who was willing to listen that Williams lacked the concept of a normative reason.

Fearing this to be a concept that I, too, lack, I turned to Parfit’s great book on ethics: On What Matters. In this essay, I’d like to muse a bit on what Parfit says in the first part of the first volume, in which he discusses reasons and rationality. Parfit defends an objective, value-based theory of reasons, on which most of his arguments about normative ethics ultimately rest. I take the kind of objectivity Parfit has in mind to imply stance-independence, though he doesn’t use that term himself. Therefore, unless I say otherwise, I’ll be using “objective” to mean “objective and stance-independent” in the remainder of this essay. That said, I’m going to use the rest of my space here trying my best to get my head around Parfit’s objective theory.

§1: Objective Reasons

Parfit kicks off by doing his best to clarify the concept of a reason. He thinks “reason”—an expression of a fundamentally normative concept—is indefinable, but that statements about which reasons agents have to do certain things are pretty much equivalent to statements about which facts “count in favor” of those agents doing those things.2 When we have sufficient reason to do some thing, we may rationally choose to do it. When we have *decisive* reason to do some thing, we are rationally compelled to do it.

Parfit then discusses the difference between objective and subjective theories of reasons. According to subjective theories, our reasons to do things are given by the desires we have. According to objective theories, our reasons are given to us by facts about value—that is to say, facts about which objects of our potential choices are inherently worth pursuing. Parfit writes in support of objective theories. According to Parfit’s own objective theory, people have reasons that they aren’t aware of, and they can also rationally do things that they have no reason to do—in such cases, they are acting on apparent reasons.3 You have an apparent reason just in case you have some false belief such that you would have an actual reason if the belief were true.

Parfit favors objective theories primarily because they avoid unwelcome implications in certain cases. Subjective theories may imply, for instance, that some people could have reason to cause themselves to be in extreme pain, or even to prefer extreme pain over moderate pain.

But for now, let’s table that point. I want to talk about another important distinction, one that Parfit takes great pains to draw clearly. In discussing different theories of reasons, Parfit stops for a while on deliberative theories. Here is Parfit’s general statement of the basic deliberative theory:

We have most reason to do whatever, after fully informed and rational deliberation, we would choose to do.

He discusses the deliberative theory in its capacity as a subjective theory, but points out that proponents of objective theories can fully agree with the deliberative theory as he’s here formulated it. For the objective theorist, we do in fact have most reason to do whatever we would choose to do after fully informed and rational deliberation—ideal deliberation, for brevity. However, this is not because the process of ideal deliberation is somehow reason-generating. Rather, it’s because engaging in ideal deliberation would lead any given person to understand what their objective reasons for action are, and so to act in whatever way such reasons say they ought to act. Note something important here: Parfit’s theory posits reasons which are such that any fully rational person who comes to understand that he has decisive reason to do some thing will, in fact, do that thing.

This is an important insight. If there do exist objective reasons of the sort Parfit contemplates, I think this *must* be their nature. For on Parfit’s account, my understanding that I have decisive reason to perform some action is for me to understand a fundamentally normative relationship between me and the action in question—a relationship of obligation.4 To deny the compulsive power of reasons, we’d need to say that some fully rational person could understand that he isobligated to do some thing and yet refrain from doing it. For me, this borders on incomprehensibility. Of course, there are many cases where people understand themselves to be obligated do some thing and yet don’t do it. But every actual person is—regrettably—only rational to a certain extent, and so we needn’t have any problem understanding how it is that many actual people often fail to do things that they consider themselves obligated to do.

§2: Procedural and Substantive Rationality

Here’s a situation that I think ought to give a proponent of Parfit’s theory a bit of trouble.

Sociopath: Some sociopath commits serial murder, claims to understand that serial murder is wrong, and yet does not care. In all other respects, the sociopath is a clear and logical thinker. Furthermore, when asked about the moral status of various other kinds of actions, he seems to have a really keen and sophisticated understanding of which ones ought to be deemed right or wrong. He even seems to understand *why* such actions ought to be deemed right or wrong.

Sociopath is not a far-fetched thought experiment; Such people do exist. What does Parfit’s theory say about them?

Recall that on Parfit’s theory, to understand that some act is wrong would be to understand that one has decisive reason not to perform that act.5 But the sociopath, in asserting that he doesn’t care whether his actions are wrong, seems to be contradicting himself. Point out the contradiction, and he just repeats that he knows he’s done wrong and yet doesn’t care.

Parfit, I think, ought to say that such a person doesn’t *actually* understand the wrongness of his action. To explain what’s going on, we’ll employ another useful distinction: the distinction between substantive and procedural rationality. Some person acts in accordance with procedural rationality just in case he acts in the way that accords with his beliefs and aims, whatever they may be; in Parfit’s terms, this would be for a person to act as his *apparent* reasons dictate.6 The same person acts in accordance with substantive rationality just in case he has certain correct aims in the first place, and certain beliefs by extension.

So, here’s what I think Parfit ought to say about the sociopath: if the sociopath really did understand what it meant for serial murder to be wrong, we’d be observing a failure of procedural rationality. However—given the sociopath’s keen understanding of which actions fit into the categories of “right” and “wrong,” and given his associated total apathy—a failure of substantive rationality is more likely. Though our sociopath well comprehends the wrong-making features of actions, he’s simply incapable of grasping that those features compel him not to perform the actions in question. That’s the sense in which he fails to understand that his actions are wrong. If we wanted, we might even say that our sociopath lacks the concept of a normative reason.

Cards on the table: I am not inclined to believe there is any such thing as substantive rationality, nor that there really are objective value-based reasons. Nonetheless, I’m going to proceed in good faith. I want to really try to understand Parfit’s position even as I try to explain why I find it nigh-impossible to accept.

§3: The Persuasion Problem

I struggle with Parfit’s concept of substantive rationality largely because of what I’ll call the Persuasion Problem:

Suppose you’ve got some person who’s behaving in a way that’s procedurally rational, but substantively irrational. Sociopath is one such case. If you wanted to convince that person to adopt the aims required by substantive rationality, how would you do it? The problem is that the answer to this question seems to be: “you couldn’t.”

Let me draw up three quick Parfit-style cases for further illustration:

Vanity: Vanity has a desire to self-immolate. However, she has a much stronger desire to preserve the quality of her skin.

Toggle: Toggle also has a desire to self-immolate, but she doesn’t want to do it right now. Today is Monday, and Toggle wants to self-immolate only on some day of the week whose name starts with a “t.”7 She has no desire to do something similar on any other day. She also has a general desire to seek pleasure and avoid pain, but this is weaker than her desire to self-immolate on days that begin with “t.” There are no other salient facts about Toggle’s desires.8

Moonwatcher: Moonwatcher has a desire to self-immolate, but only when the moon is full.9 Her desire is non-arbitrary—for her, the full moon inspires such extreme dread and awe that only a grand, self-destructive gesture can properly express her feeling, which she very much desires to do. Moonwatcher also lacks a general desire for self-preservation, cares little about the people who would suffer on account of her death, has very few other significant desires, et cetera.

Let’s take the cases one at a time.

It would be very easy to explain to Vanity why she ought not self-immolate. If she were to do so, she would ruin her skin, frustrating a stronger desire. So we can tell Vanity that, insofar as she self-immolates, she is being procedurally irrational; by her own lights, she ought not do it.

Toggle is not obviously being procedurally irrational. Still, if she plans to self-immolate this Tuesday, we may be able to talk her down. Keeping in mind that Toggle’s desire is arbitrary, we might say to her that there’s no robust sense in which Tuesdays and Thursdays are better-suited for self-immolation than other days of the week, and she might agree with us. We could then point out that, though acting on her desire wouldn’t really *contradict* any of her beliefs or her other desires,10 it would entirely fail to *cohere* with them. We might thus have some chance of getting her to understand that she ought to abandon her plan and, to the best of her ability, the corresponding desire. After all, Toggle’s desire is strangely out-of-place even by her own lights, and that seems a good reason to at least treat it with skepticism.

Now let’s turn to Moonwatcher. If Moonwatcher plans to self-immolate during the next full moon, what shall we say to her? Like Toggle, but unlike Vanity, there’s no sense in which Moonwatcher is directly contradicting herself. But unlike Toggle *or* Vanity, Moonwatcher’s objectionable desire coheres nicely with her wider system of desires and beliefs. Still, perhaps we can tell her she has reason not to burn herself. But what reason? Unlike with Vanity and Toggle, none of our data comes from the inside here, so it’s difficult for us to do more than simply *assert* that Moonwatcher has a reason, and Moonwatcher won’t be likely to take our bare assertion very seriously. But, on second thought, perhaps we can do a bit more. We can present a case for an objective theory of reasons. So we begin to do so. We start by explaining, as Parfit would, that naïve subjectivism implies that a person might have reason to cause herself extreme agony only because she wants to do so. We’re prepared to level our devastating objections to the more sophisticated subjective theories which Moonwatcher will surely cook up to avoid this problem. But we never get so far. Because of *course* people have reason to cause themselves extreme agony when they want to. At least, that’s how it is according to Moonwatcher. After all, she’s soon going to cause herself to die in an extremely painful way, and it’s precisely the thing that she most wants to do.

So back to our original question: If you wanted to show some substantially irrational person how to be substantially rational, how would you do it? Is the answer really that you just couldn’t?

I’ll look at some potential responses to the Persuasion Problem in just a moment. First, though, I want to note one potential implication of the problem as I’ve laid it out here. It may imply that people who are substantively rational are not, in any way, responsible for acting in accordance with those reasons which they do not perceive. We can’t appeal directly to “ought implies can” here, because it is certainly possible for those people to commit themselves to acting in conformity with substantive rationality. But, to borrow a bit of phrasing from Kant, it really would just be in conformity with rationality; such people would not actually be acting *rationally.* So can we hold people responsible for acting in conformity with rationality when they don’t have any understanding of why they ought to do so? I’m inclined to answer in the negative. Consider some child who does some bad thing, who is scolded by a parent, who then asks why the thing they’ve done is wrong, and receives an answer which amounts to “it just is.” So the child does this thing again, is again scolded, again interrogates the parent, again receives no answer. After doing this thing a third time, the child is punished harshly. This, to me, seems very wrong, and seems so almost no matter what the child had actually done. The wrongness consists not in teaching the child to refrain from doing bad things, but in the refusal to explain to the child why they ought not do so. Now imagine that the child is a grown man, and the punishing parent another grown man. In my view, the case only becomes worse.

I flag this only because it raises some practical questions. I suspect that most people, when considering what we have reason to do, might agree, for instance, that we all have reason not to kill and cannibalize each other, but this is a low bar. On many issues where the correct answer is even slightly less obvious, I expect we’ll see widespread disagreement. In such cases, we’ll probably have to say that members of each party are each substantively rational only to a certain extent, and also probably that one party is being more rational than the other. But how do we decide whicht that is? Because of the Persuasion Problem, and because of a related issue that I’ll discuss later in the essay, I’m worried that Parfit’s theory is ill-equipped to answer that question. Is this a damning objection? Perhaps not. But the theory would certainly be more powerful if it could resolve such disagreements.11

§4: Two Persuasion Problems, One (Dis)solution?

The Persuasion Problem has its own problem—a problem of symmetry. There may be some procedurally irrational person of whom it is true that he cannot be convinced to be more rational. Perhaps, due to some mental disability, he is wholly incapable of understanding what follows from what. Such a person would be deficient, and we would not be able to remedy his deficiency simply by explaining to him how he he is deficient and how he might be less so—if he could understand his deficiency in this way, it would be cured. But is the same not true of Moonwatcher?

Call this new case an illustration of the Procedural Persuasion Problem(PPP), as opposed to the previously discussed Substantive Persuasion Problem(SPP). I suspect the problems are asymmetrical, and the SPP disturbs me much more. In a moment, we’ll touch on a few reasons to accept an asymmetry. First, though, I’d like to make a point about the scope of the problems. The PPP is a problem for *any* theory that wants procedural rationality to be, in some way, normative. This will obviously include Parfit’s theory, but also the subjective theories he discusses, which will imply that people ought to be procedurally rational because, in most cases, this will help them to achieve the objects of their desires.12 Accordingly, such subjective theories will also be vulnerable to the PPP insofar as the PPP actually is a serious problem. The SPP, on the other hand, is a problem *only* for objective theories. Given that my goal here is mostly to discuss *objective* reasons, we can take the SPP seriously as a challenge to objective theories even without considering whether the PPP successfully creates a similar challenge to any theory that says people ought to be procedurally rational.

So, solving the SPP. The ideal solution, as far as I’m concerned, is to find some way to convince Moonwatcher that she really does have a reason not to self-immolate. Moonwatcher seems generally rational, and a rational person ought to be able to understand the reasons she has, even if it takes some prodding. But no solution of this kind is forthcoming.

But now the objectivist cries: “Aha! The buffoon hath failed to distinguish between procedural and substantive rationality! He heeds not the distinction which he himself took such care to draw! Take him to the pillory, lads!”

And indeed, I have failed to pay attention to the distinction. But my failure prompts a new avenue of inquiry.

§5: I was talking to Parfit—we were speaking English—and he said that in England you’ve got to be rational, and I said “what the hell do you mean by that?” And he said that he was talking about rationality in the ordinary sense of the word, and so it might be the case that we did not disagree.13

Now, I think, we ought to give our ideas about rationality a closer look. Though Parfit claims that reasons are more fundamental than rationality, it may still be useful for us to try to understand reasons in terms of their ability to compel a rational person. Given that Parfit’s reasons are otherwise indefinable, considering the issue from the other direction might allow us to acquire some insight.

So let’s revisit the PPP. Some people will not see the PPP as a problem at all. It is solution enough, these people will claim, to say that a person who cannot be convinced out of procedural irrationality is simply deficient. Where do we draw the line, anyhow? Many animals, while being somewhat intelligent, are not procedurally rational, nor are they capable of understanding that they ought to be. But this does not seem like a mark against Parfit’s understanding of reasons.

Is the deficiency response likewise sufficient for the SPP? Well, the problem here is that we’re understanding a person to be substantively rational just in case that person has the aims(and associated beliefs) given to them by certain objective reasons. But, insofar as we follow Parfit, we accept objective reasons primarily because their absence gives us unintuitive verdicts about what people have reason to do. But take the verdict that people could sometimes have reason to cause themselves intense agony—why is this unintuitive? Well, we—the ones making and accepting these sorts of arguments—think that we have reasons to avoid causing ourselves great pain, and likewise for others. So, insofar as we’re the sort of people who find Parfit’s arguments intuitively plausible, it seems like we’re just understanding substantive rationality as “the concept that accurately describes our own psychology.” In other words, we’re saying that people are rational insofar as they’re psychologically similar to us in some particular way. We have no independent explanation of why that is, however—we can appeal to objective reasons, but our motivations for accepting objective reasons stem directly from the fact that we have the psychological features which we are trying to demonstrate the value of. There may not be a strict circularity here, but it still strikes me that something serpentine is going on.

Note that procedural rationality is *not* like this, or at the very least not obviously so. Though it may be the case that a person with no procedural rationality whatsoever won’t be able to understand its value, it’s not the case that procedural rationality simply consists in having a Parfitian psychology. We might also conceive of procedural rationality as the way of thinking that generally helps people to get what they want, or the way of thinking that generally best allows people to approach knowledge, or perhaps in some other way entirely. Let’s roll with the first understanding for now. Note the work that “generally” is doing here. Though it’s the case that being procedurally rational will help you get what you want in most cases, there are exceptions—sometimes, even sometimes by luck alone, some deeply irrational person may best fulfill his desires by continuing to be irrational.14 But since this isn’t normally the case, we can explain the value of rationality by appealing to its results. And if we’re going with “getting what you want” as the result we care about in this particular case, maybe we’re appealing to the premise that people who are thinking properly will do what helps them get what they want, insofar as they understand how. Procedural rationality is thus normative in a very light-weight way. This, I think, is a difficult premise to reject, and I think it gives us a useful asymmetry between the PPP and the SPP.15

Armed with this sort of understanding of procedural rationality, we can take some person who isn’t fully rational, but possesses the basic capacity of procedural rationality, and we can convince that person to be more procedurally rational, as well as showing them how. Substantive rationality doesn’t necessarily work this way at all—a person who understands some of their objective reasons may not be able to understand the rest by extension, perhaps not with any amount of explanation.16

And now let’s returned to the question of whether it’s possible to show Moonwatcher that she does, in fact, have a reason to avoid burning herself. The reason in question is meant to be fundamentally normative, normative in a way that we can’t break down into smaller pieces. So it’s not in our power—or in anyone’s—to explain Moonwatcher’s reason to her in more basic terms. The best we can do is to explain the relevant reason-giving fact, and that this fact gives a reason. Perhaps we can explain why this fact gives a reason by appeal to some higher-order fact about which facts generally give reasons. But that’s about as far as we go. When it comes to the higher-order fact, we won’t be able to explain *why* it is that this fact has weighty normative implications, except by appeal to the first-order facts which Moonwatcher does not herself accept.

In an attempt to break free from our mysterian cage, we may continue our investigation of rationality. Given what we understand so far, we certainly wouldn’t be able to persuade Moonwatcher by appealing to the normativity of substantive rationality—she’ll be precisely as motivated to accept that as to accept the objective reasons that are in contention in the first place. But Parfit also talks about the general, broader concept of rationality. Let’s see what he says there:

When we are aware of facts that give us certain reasons, we ought rationally to respond to these reasons. We respond to decisive reasons when our awareness of the reason-giving facts leads us to believe, or want, or try to do what we have these reasons to believe, or want, or do. We are irrational, or less than fully rational, insofar as we fail to respond to reasons in these ways.

So far, so good. Not yet, however, very illuminating. Let’s continue. Comparing his own view to T.M. Scanlon’s, Parfit writes:

I have rejected Scanlon’s claim that, when Scarlet and Crimson prefer an hour of agony to a minute of slight pain, these people are not irrational. There may, however, be no disagreement here. I am using ‘irrational’ in its ordinary sense, to mean ‘open to strong criticism of the kind that we also express with words like “foolish”, “stupid”, and “crazy”’. Scanlon suggests that we should use ‘irrational’ in what he calls a narrower sense, which applies only to people who fail to respond to what they themselves acknowledge to be reasons, or who are inconsistent in certain other ways[…]It is clearer, I believe, and in other ways better to use ‘irrational’ in its ordinary, wider sense.

So, when Parfit talks about rationality and irrationality, he’s using the words in their “ordinary sense,” which means that he’s using them to express that a person/action is open to a certain kind of praise(if rational) or a certain kind of criticism(if irrational).17 This is the kind of rationality he believes to be relevant to his discussion of reasons, and by extension his discussion of morality. Presumably, Parfit believes both procedural and substantive rationality to be covered by “rationality” in its ordinary sense—if he believed that only procedural rationality was included, he wouldn’t be able to call Scarlet and Crimson irrational.

This is weird. It’s weird because what is entailed by “rational in the ordinary sense of the world” will range widely—for one group of people, it may be quite ordinary to call racial prejudice irrational; for another group, it may be quite ordinary to call the *absence* of racial prejudice irrational.

Parfit, of course, is not going to say that whether you’re rational depends on whether you’re a member of the first group instead of the second. Instead, he’s going to need to say something like this: certain people/actions actually do deserve praise or criticism , and that this desert is what our words “rational”(for praise) and “irrational”(for criticism) currently express. Holding this fixed, I now become worried about our ability to know that our concept of rationality maps neatly onto actual rationality. For Parfit is here appealing to common sense. For when we use the word “rational” in ways that imply reference to substantive rather than procedural rationality, the extent to which our use of the word maps onto “real rationality” varies according to the time and place. We can imagine some nineteenth-century American slaveholder speaking with his slaveholder buddies, calling some radical abolitionist Yankee irrational for not understanding that he doesn’t really have any moral obligations to black people.

I want to think about how Parfit might break down such a case. So, for a little while, let’s take for granted that Parfit’s theory is true, and see where that gets us.

Recall that substantive rationality entails having certain *true* normative beliefs. Presumably, the belief that people of different races have the same moral status(or some belief very close to this) is one such. But here we have a situation where it seems entirely appropriate to use “rational” in its ordinary sense—the sense that is associated with words like “foolish,” “stupid,” and “crazy”—with reference to a *false* normative belief, and vice versa for “irrational.”

An obvious response: radical nineteenth-century abolitionists were not actually foolish or stupid, and so they weren’t irrational. It’s not an appropriate use of the word “irrational” if it’s not a *correct* use. Good! But if that’s the case, many cultures have been habitually mistaken about what is and isn’t rational, about how to apply one of their own concepts. This almost certainly includes our own culture. But you might think that this, by itself, isn’t too much of an issue. We could say the same about concepts like “justice.” And whether we’re talking about rationality or justice, not all cultures have been equally mistaken. Our modern-day culture, for all its flaws, seems to have gotten a lot of stuff right—stuff about justice in particular—where people in the past got it very, very wrong. Might rationality be another case of the same?

I’m not sure. To illustrate. I’m going to tell a story about a nineteenth-century slaveholder. It might not be all that realistic:

Justice: Our slaveholder was familiar with a word: “justice.” This word had an intension—a meaning, basically—and an extension—the set of all those things to which it applied. If you asked our slaveholder to define justice, he might have said something about fairness, about people getting their due, or maybe about doing God’s will. In other words, he might have understood “justice” to mean something which implies that slaveholding is unjust. But when looking at the act of owning slaves, he would have said: “no, this is just.” When we think about the word as the slaveholder actually used it, the intension of the word and its extension were in contradiction. One of them had to change. And the people who followed the slaveholder rightly decided that it was the second. We called just, and treated as just, many things which were actually unjust. We then began to call unjust, and treat as unjust, those very things. We thus—at least in many cases—resolved the contradiction between intension and extension.

Let’s now stop assuming the truth of Parfit’s theory, and take a different perspective on Justice. The mistake of the slaveholder is obvious; he just doesn’t see that the content of his concept fails to describe many cases which he habitually uses it to describe. In other words, the tools for understanding his error were in his own hands, and indeed many of his contemporaries—if too few—used those tools competently and did come to understand his error. This is not, of course, the real story of how we came to understand slavery as unjust; nonetheless, it’s a story that explains how slaveholders were mistaken without appeal to substantive rationality.

Now, when we try to consider these cases, we might ask questions about where our understanding of the intensions of our words comes from. I want to push those questions aside, however, and focus on something that currently concerns me more: Parfit understands both reasons and rationality pretty much *entirely* in terms of their extensions! He calls reasons indefinable and then appeals to our intuitions about thought experiments to argue that they’re real. Then he defines rationality(the thing that makes you respond to reasons) as basically “the thing that obtains when ordinary people consider it appropriate to use the word ‘rational.’” And perhaps this is enough for people who already find the idea of objective reasons intuitive and satisfactory. But I’m not such a person. And so, much as I’ve tried, the concept of a normative reason baffles me just as much as it did when I began reading.

Conclusion:

If Parfit’s theory of reasons is true, here’s the position we’re in: There are objective reasons, but they’re irreducible and basically indefinable. People who are rational respond to apparent reasons, which is to say that they act in the way they would have reason to act if their beliefs were true. But that’s not all—part of rationality is substantive rationality, which also requires responding to certain actual reasons. Our ability to understand which reasons these are is grounded in our substantive rationality, just as our ability to understand the implications of our beliefs is grounded in our procedural rationality. Both these kinds of rationality are part of rationality simpliciter, which is rationality in its ordinary sense. To call something rational in the ordinary sense of the word, however, is akin to expressing a certain kind of praise, and there’s going to be a lot of cross-cultural and even interpersonal disgreement on whether certain actions warrant that kind of praise. Of course, there is a such thing as rationality, and it’s objective, so some people have got it right.

Now we have are two possibilities. Grant that I’m procedurally rational. I’m either also substantively rational, or I’m not. If I am substantively rational, great—I understand the aims I need to have, my reasons for acting. But if I’m not substantively rational, there’s no way of showing me why/how it is that certain facts give me reasons. The best you can do is appeal to the ordinary sense of rationality. But that appeal will only get me to the truth if I’m in a context where the ordinary sense of the word “rational” maps on to real rationality with a good degree of accuracy. So, how do I know whether rationality, in the ordinary sense that *I’m* familiar with, includes substantive rationality? Well, remember that reasons are more fundamental. So we can get a sense of what substantive rationality really is by investigating reasons for action—which are real and which are merely apparent. But these reasons are the very thing I don’t understand!

So now, I think I understand why Parfit found it maddening that Bernard Williams lacked the concept of a normative reason. I get it because I believe lack this concept in much the same way, and I too find it maddening! When I speak to someone who endorses an objective theory in the neighborhood of Parfit’s, it’s not as if he and I are simply taking different perspectives on the issue; much worse than that, each of us is using a conceptual framework that seems borderline unintelligible to the other!

It might sound like I’m criticizing Parfit’s theory of reasons, and his moral realism by extension. And I suppose the essay will come off that way, but my goal was quite different. Here’s an argument that I’m *not* making: objective reasons are unintelligible, and therefore unreal.

For even if objective reasons were *entirely* unintelligible to me, I’m not sure that would justify me in declaring them unreal—I’m actually quite open to the idea that certain basic features of reality are unintelligible in some way or another.

Still, we’ve got to try our best to understand things. So what I’m really doing here, because objective reasons are *borderline* unintelligible to me, is trying my best to make it otherwise. That’s why I started reading On What Matters in the first place. But On What Matters—its first volume, at least—has given me a lot to chew on, and yet very few of the answers I was searching for. Perhaps I’ll find something better in the deeper reaches of this very long book.

In the meantime, I’d like to thank you for making it to the end of this very long essay. I think I’ll conclude by asking for your help. If you endorse something close to Parfit’s theory of reasons, I’d like to know why. Because though I am more sympathetic to Parfit’s theory than I was when I began writing this essay, and though I am less confident in my ability to tear it down, I am no more persuaded than I was at the outset.

There is a chasm here, and every bridge so far has buckled under my weight. So perhaps there is no adequate bridge. But I want to see the other side, and I’m not quitting yet.

“Hell!” he said aloud. Pravic was not a good swearing language. It is hard to swear when sex is not dirty and blasphemy does not exist. “Oh, hell!” He repeated.

Shevek, our Anarresti protagonist, speaks a completely artificial language. Pravic was deliberately developed by the Settlers of Anarres, anarchist revolutionaries who were given a world in exchange for an end to the revolution.

As it happens, swearing in proper Pravic is not only difficult, but impossible. The two obscene words the Anarresti have—“hell” and “damn”—are imports from Iotic, a language of the world where the revolution started.

The absence of profanity is not the only quirk of Pravic. At various points in The Dispossessed, LeGuin draws our attention to the peculiarities of the fabricated language. Here is the most important one:

The singular forms of the possessive pronoun in Pravic were used mostly for emphasis; idiom avoided them. Little children might say “my mother,” but very soon they learned to say “the mother.” Instead of “my hand hurts,” it was “the hand hurts me,” and so on; to say “this one is mine and that’s yours” in Pravic, one said. “I use this one and you use that.”

§2: Communism and Bullshit

When I was younger, I spent a lot of time on Reddit. This can be dangerous. If—at a sufficiently impressionable age—you spend too much time in the crazier corners of Reddit, you will get a very particular impression: it will seem to you that communism is obviously correct, and also the ideology adopted by the vast majority of truly intelligent people. But what is communism? Well, according to Reddit: everything in the world that’s bad and wrong and evil, that’s because of exploitation and systemic injustice, and we can fix it all by just reading Marx and Engels and then teaching each other to share things. In other words, communism is just the ideology that says we should fix all the bad things. How could anyone say no to that?

This is probably the right place for me to make clear that I don’t actually think every single anti-capitalist is deeply unserious.1 I don’t believe, for instance, that G.A. Cohen was somehow full of shit. However, now that I’m no longer a Reddit-dwelling preteen, I have come to the conclusion that communism, as an ideology, is generally deeply unpersuasive. This is, in my view, largely because even the most serious and intelligent communists seem to be much more interested in problems than solutions. I do think many of the problems that communists see in capitalist systems are real and worrisome—they happen, however, to be counterbalanced by immense benefits. To abandon capitalism would be to gain some things and lose others, and communists generally seem profoundly uninterested in explaining why the tradeoff would be worthwhile. Indeed, many communists sometimes seem incapable of acknowledging that there is, in fact, a tradeoff.

But for me, one exception to this unfortunate rule immediately leaps to mind; one writer with a vision of a plausible communist world—an anarcho-communist world, in fact—which seems worth living in. The writer I mean is Ursula LeGuin.

§3: A City on the Moon

The life of an anarchist is lived ten days at a time—the “decad,” as the Anarresti call it, is the most important unit of temporal structure. That’s because the decad gets the hard work done. On Anarres, you have nine days to do your own work, and on the tenth you contribute your share of menial labor. This is how the Anarresti manage to shovel their shit. Kleggich, they call this; Drudgery, as opposed to work. Work, of course, shares a Pravic word with play. But to hear Shevek tell it, even the kleggich is hardly an inconvenience:

If you work at a mechanical loom mostly, every tenthday it’s pleasant to go outside and lay a pipe or plow a field, with a different group of people.

As far as ordinary labor goes, you can do what you choose. Of course, some jobs are more necessary than others, and it’s hard to keep making things when no one wants what you’re willing to make. But there is no unemployment on Anarres. If you’ve nothing to do, you go to Divlab and they tell you where workers are needed, and then you simply hitch a ride and start working. Wherever you go, there’ll be space in a dormitory for you, and two solid meals each day.

The Anarresti system of labor organization is not very remarkable in its own right. It only becomes so when you consider the fact that Anarres acknowledges no leaders or, that even the absence of formal rules has not been codified in any way. If you live on Anarres, no one is *making* you do your share of kleggich. But to hear Shevek tell it, anarchistic cooperation is just the natural way of things:

People take the dangerous, hard jobs because they take pride in doing them, they can—egoize, we call it—show off?—to the weaker ones. Hey, look, little boys, see how strong I am! You know? A person likes to do what he is good at doing. . . . But really, it is the question of ends and means. After all, work is done for the work’s sake. It is the lasting pleasure of life. The private conscience knows that. And also the social conscience, the opinion of one’s neighbors. There is no other reward, on Anarres, no other law. One’s own pleasure, and the respect of one’s fellows. That is all. When that is so, then you see the opinion of the neighbors becomes a very mighty force.

A very mighty force indeed, for Anarres also has almost none of what we might call crime. Of course, the only real possible crime-equivalents are murder and rape. These may be punished when they arise, but why murder when there’s nothing to take from the dead? Why rape when love really does come free for those who want it? For even the sex lives of the Anarresti seem, to those of us on Earth, a Utopian fantasy.

§4: The Possibility of Utopia

Anarres is the product of a revolution, the Odonian Revolution, and the revolution managed to create a sort Utopia. But still, the question we have to ask ourselves now is whether the Odonians succeeded.

Urras, the world where the revolution started, is very obviously meant to mirror the one LeGuin lived in at the time of writing. More precisely, Urras was constructed as a heightened version of that reality. The dominant Urrasti nation, A-Io, is burdened by staggering wealth inequality, the product of an essentially unrestrained capitalist system. Consider, for instance, Shevek’s brief visit to the slums of Nio Esseia:

A fine, foggy rain was falling, and it was quite dark; there were no street lights. The lampposts were there, but the lights were not turned on, or were broken. Yellow gleams slitted from around shuttered windows here and there. Down the street, light streamed from an open doorway, around which a group of men were lounging, talking loud. The pavement, greasy with rain, was littered with scraps of paper and refuse. The shopfronts, as well as he could make them out, were low, and were all covered up with heavy metal or wooden shutters, except for one which had been gutted by fire and stood black and blank, shards of glass still sticking in the frames of the broken windows. People went by, silent hasty shadows.

This is the setting of ordinary life for the many Urrasti poor—the wealthy, meanwhile, live in a sort of luxury that Shevek finds nigh incomprehensible.

Patriarchy, too, is taken to its extreme in A-Io. There are no women in power, no women at the universities. In high society, the female members of the human species are visible only at formal gatherings, where custom dictates that they keep their breasts bare.

The foil to A-Io is Thu, a communist nation which has slid deep into totalitarianism. The little we learn about the socialist superpower leads us to believe that the Thuvian experiment has failed in basically every way that the Anarresti one succeeded.

And then we have poor, war-torn nations like Benbili, the Urrasti Vietnam; A-Io and Thu fight for supremacy there.

I began this section by asking whether the Odonian Revolution had succeeded. Is it not now obvious? In comparison to A-Io, to Benbili, to Thu, Anarres seems a proper paradise.

And *yet,* here is what Chifoilisk—the one Thuvian we encounter in the book—says to Shevek: “Why don’t you come to Thu and see how real socialism functions?”

This is precisely the sense in which the Odonian Revolution has failed. The goal was to build a paradise, yes, but also to change the world. And the world has not changed. In fact, Anarres endures only so long as the rest of the world permits it, only so long as the Anarresti continue to mine their own world for precious metals, and to sell them as tribute. Yes, sell. There is a sense in which every Anarresti is the profiteer that he despises.

And even on Anarres, after seven generations, the situation has a certain precarity. To make their project work, the anarchists need empty space—a whole Moon for only twenty million people. And they need a language where “work” and “play” are the same word. They need to teach children to be anarchists from the very moment of their birth, so strong is the human instinct to point at something and say “mine.”

The point is that Anarres is not a natural Utopia. Anarres is a project of sophisticated socio-political engineering, tailored to its precise conditions, and even then hardly iron-forged.

There comes one point in the book where Anarres experiences a shortage of food. Admirably, the Anarresti mobilize with staggering efficiency. Divlab tells people where they’re needed, and people go. It is their great pride to be necessary, to be of use in an hour of such need. But then the Anarresti actually have to starve. Shevek’s work crew, more than five hundred of them, are being taken home from their postings by train. Provision runs take priority, so the train moves slowly, and at one point—for twenty hours—stops entirely. The stopping-place is a village of one-hundred and sixty, a village with truck gardens but which nonetheless depends on provision runs for food, and they don’t know when the next is coming.

So they don’t share. How can they share, when the visitors outnumber them three to one? They’ll be wiped out, and then there won’t be enough for anyone.

The passengers’ resentment of the townsfolk got bitter, but it was less ominous than the behavior of the townsfolk—the way they hid behind “their” walls with “their” property, and ignored the train, never looked at it. Shevek was not the only gloomy passenger; a long conversation meandered up and down beside the stopped cars, people dropping in and out of it, arguing and agreeing, all on the same general theme that his thoughts followed. A raid on the truck gardens was seriously proposed, and bitterly debated, and might have been carried out, if the train had not hooted at last for departure.

And so, for sixty, hours, Shevek starved. The word LeGuin uses, as it happens, is “fasted.”

Later in the book, we learn the story of a man driving a provision train during the famine. Hungry people crowded about, begging, trying if they could to break inside. And the man backed up the train and killed some of them. Other were going to die if he didn’t, after all.

Later in the book, when asked about Anarres by a poor Urrasti man, Shevek says this:

Nobody goes hungry while another eats[…]But we have been hungry. We have starved. There was a famine, you know, eight years ago. I knew a woman then who killed her baby, because she had no milk, and there was nothing else, nothing else to give it. It is not all…all milk and honey on Anarres, Efor.

But Shevek, for all his sincerity, is not telling the truth. He himself has gone hungry just outside the boundary of a village where one hundred and sixty people ate, all of them having acquired the habits of propertarians.

So Anarres is fine-tuned, balanced unsteadily, never so far from collapse as the Anarresti would like to believe. Even after seven generations, the anarchists have not escaped power and ownership—rather, they are always running. Hard times give the Anarresti the opportunity to care for each other, to make the difference in each other’s lives. But hard times also prove that you have to learn to be an anarchist. Utopia cannot be as simple as “if everyone just.”

§5: The Real Urras

Much as she’s a realist about the material conditions that enable Anarres, LeGuin also ought to be commended for being a realist about Urras and A-Io. Her portrayal of the capitalist superpower is not ultimately sympathetic, but it also isn’t simple. The first Urrasti locale we’re shown, for instance, is Ieu Eun University, where Shevek comes to do physics with the Urrasti elite. The University is not merely a place of splendor—it is a place for something that simply cannot exist on Anarres. And though Shevek’s rooms there far outstrip any Anarresti dwelling in terms of comfort and convenience, the sort of splendor I mean is not only material. Consider, for instance, Ieu Eun’s student body:

They were superbly trained, these students. Their minds were fine, keen, ready. When they weren’t working, they rested. They were not blunted and distracted by a dozen other obligations. They never fell asleep in class because they were tired from having worked on rotational duty the day before. Their society maintained them in complete freedom from want, distraction, and cares.

You’ve seen the slums in Nio Esseia already, and so you know that the Ioti students don’t exist in a vacuum. And if you read the book yourself, you will see worse than slums. So if it really were true that the total freedom of the students was only possible because of the plight of the masses—and the book wants you to believe it is—then you would still have to ask yourself whether the tradeoff was worth it. Probably, you would conclude that it wasn’t, that Anarres was to be preferred. And you might still think so even when I’ve told you about the gravest failings of the Anarresti project.

Still. When I was in high school, my Latin teacher had a favorite rhetorical question: “How many things can be true?”

As the book begins to build to a climax, Shevek asks his University-supplied manservant, Efor, to confide in him. Efor describes how one of his children died in a rat-infested hospital—in fact, none of his three children made it to adulthood. So Shevek hears the testimony of a poor Ioti for the first time. And Shevek, when he at last directly faces the darker part of Urras, thinks this:

It was not “the real Urras.” The dignity and beauty of the room he and Efor were in was as real as the squalor to which Efor was native. To him a thinking man’s job was not to deny one reality at the expense of the other, but to include and to connect. It was not an easy job.

The answer to my teacher’s question: “More than one.”

Of Urras in particular, two important things are true. It is both a paradise and an abyss. The Anarresti look up at their bright Moon and they know—know even without seeing—that they’re looking at Hell. “To make a thief,” says Odo, “make an owner; to create crime, create laws.”

And yet, the rest of the settled Universe is in awe of Urras. Take, for instance, The Terrans—the Earthlings of the future. Having ruined their own planet, they had descended into a necessary totalitarianism from which they emerged only with the help of the altruistic, elevated Hainish.2 Still, the Earth is inhospitable. And so, here is what the Terran Ambassador has to say to Shevek:

Now, you man from from a world I cannot even imagine, you who see my Paradise as Hell, will you ask what my world must be like?[…]My world, my Earth, is a ruin. A planet spoiled by the human species[…]You Odonians chose a desert, we Terrans made a desert[…]We can only look at this splendid world, this vital society, this Urras, this Paradise, from the outside. We are capable only of admiring it, and maybe envying it a little. Not very much.

The point of all this, I think, is to emphasize the limits of Shevek’s perspective on Urras. In his time there, he only sees the extremes—the suffering of the very poorest, and also the callous machinations of the very richest. Everything which he finds worth admiring comes with an unconscionable cost. But Shevek lives on a world—a tiny world, the only world—where no one has to pay for anything.

So what we’re meant to take from the Ambassador’s words, I think, is that being in a position to understand Urrasti society as evil and irrational does not license you to ascribe evil and irrationality to all the Urrasti. For the most part, they have done what they could with what they were given. And though they were indeed given a world of plenty, they do not speak Pravic; they do not have language where the possessive pronoun is practically uncouth.

So consider now what Shevek says to the Ambassador:

You don’t believe in Anarres. You don’t believe in me, though I stand with you, in this room, in this moment . . . My people were right and I was wrong, in this: We cannot come to you. You will not let us. You do not believe in change, in chance, in evolution. You would destroy us rather than admit our reality, rather than admit that there is hope. We cannot come to you. We can only wait for you to come to us.

And consider as well Chifoilisk’s offer to show Shevek “real socialism.”

Shevek rejects Chifoilisk’s false socialism, and in the same way he rejects the Ambassador’s false Urrasti Paradise. And yet, as the conversation comes to a close, the Ambassador asks Shevek not to “judge us too hardly.”

Is it the Ambassador’s fault, Chifoilisk’s fault, that they do not understand how to make a better world—that a better world is possible?

Says Shevek, in response to the Ambassador: “I don’t judge you at all.”

§6: The Asylum on Segvina Island

In this essay, we’ve gotten very far without touching on Shevek’s reason for leaving Anarres in the first place. If Anarres really is the closest thing to Paradise, why ever go elsehwere?

The answer, quite simply: for the sake of physics.

But allow me to give a more roundabout response. I want to talk about a uniquely Anarresti tragedy, about Shevek’s friend Tirin. Tirin, a transgressive playwright he wrote a woefully misinterpreted play.

In the play—a comedy—Tirin has the leading role. It is the role of an Urrasti who smuggles himself onto Anarres and tries desperately to pay his way into the goods and services which the Anarresti are only too happy to share with him. Shevek is of the opinion that the play must have been very funny.

But Tirin’s play, for some reason or another, comes to be despised. The people of his community call it immoral, and they call a public meeting to tell him so. They tear him apart. They deny him a teaching posting. They bully him into seeking therapy in the Asylum on Segvina Island.

The Asylum is used primarily to rehabilitate—or to sequester—murderers and rapists. It is the closest thing to a prison the Anarresti have.

In the end, Tirin gets out. But in the end, he is terrified of showing his work to anyone but Shevek. He spends the rest of his life writing and rewriting the same play—the very one that brought an end to his career as playwright.

And Shevek, about Tirin and about Anarres broadly, says this:

We don’t cooperate—we obey. We fear being outcast, being called lazy, dysfunctional, egoizing. We fear our neighbor’s opinion more than we respect our own freedom of choice. You don’t believe me, Tak, but try, just try stepping over the line, just in imagination, and see how you feel. You realize then what Tirin is, and why he’s a wreck, a lost soul. He is a criminal! We have created crime, just as the propertarians did. We force a man outside the sphere of our approval, and then condemn him for it. We’ve made laws, laws of conventional behavior, built walls all around ourselves, and we can’t see them, because they’re part of our thinking. Tir never did that. I knew him since we were ten years old. He never did it, he never could build walls. He was a natural rebel. He was a natural Odonian—a real one! He was a free man, and the rest of us, his brothers, drove him insane in punishment for his first free act.

Shevek, here, doesn’t speak from an impartial perspective. Anarres has fenced in Shevek’s physics in just the same way. As a young man, he worked under the aged physicist Sabul. Sabul was a plagiarist. His best-regarded work was stolen wholesale from Ioti physicists, of whom scarcely anyone else could understand a word. Shevek’s own work—brilliant, original work—was first suppressed for its uncomfortable ideological implications, and then published butchered, bearing Sabul’s name alongside Shevek’s on the cover.

I make a point of saying “stolen” in one case and not the other. Sabul steals from the Ioti, but he cannot steal from Shevek.

Eventually, after years, Shevek gathers his friends, establishes a new syndicate, and begins to publish his own stuff. He finds then that the problem was not only Sabul. Anarres does not need a unified theory of temporal physics. Such a theory might allow them to build an ansible—a device to enable instantaneous communication between any two points in the Universe. But Anarres does not need an ansible—Anarres wants no part of the Universe outside its own atmosphere. Shevek wants to give; his people refuse to take.

But Urras, A-Io—they will take. They will buy what Shevek is willing to sell.

In writing The Dispossessed, I believe that the question LeGuin wanted to answer was something like this: Suppose you wanted to create a stable society without a state to do the stabilizing—what would it take?

But that is a double question:

1. What would you need to do to build such a society?

2. What would you need to sacrifice to keep it functioning?

In §4, I attempted to give LeGuin’s answer to the first question. But the second is more important. I began this essay by describing Pravic, and I meant then just to emphasize the immense difficulty of the Anarresti project. If you wanted to build a communist Utopia, I meant to say, it wasn’t going to fall out of the sky. The project would demand ceaseless work, compromises with the powers that be, even a language of your own. But now, I want to emphasize the sorts of things that would be fundamentally different if you spoke a language that did not ordinarily permit you to use the possessive pronoun. I want to emphasize that the Anarresti culture is *not* the culture of a modern-day developed nation, only nicer and more generous; Anarres is something intrinsically different.

On Anarres, Shevek cannot do physics that would change the entire Universe. On Anarres, there are great books that never would have been written. Some would have been condemned as immoral, like Tirin’s play. Others, the Anarresti might have approved of, but they nonetheless never could have come to be. Parfit’s Reasons and Persons, for instance—we can imagine the Anarresti condoning its content, but never that they would’ve allowed Parfit the seclusion he needed to get the book written.

LeGuin doesn’t present these problems as inevitable, but she does make clear that there is a very basic tension between the structure of Anarresti society and the sort of wide-ranging artistic, philosophical, and scientific freedom that the best parts of our own society have cultivated and cherished. But the tension works both ways. If it’s true that having a free and fair society makes problems for the intellectuals and the artists, it may also be true that establishing the freedom of the intellectuals and the artists makes it more difficult to build a free and fair society.3 Ieu Eun University is like the Odonian Society on Anarres in that cannot be built easily, nor without great sacrifice.

In the end, Shevek wants to make a better, freer world for people like Tirin and himself. He also wants to give the rest of the Universe the kind of freedom that Anarres has enjoyed for seven generations.

“I’m going to go unbuild walls,” Shevek says. And then he goes to Abbenay to build the Syndicate of Initiative, which will publish his own physics as well as Tirin’s plays. And when he goes to Urras, it is the same wish that moves him.

But Shevek needs to unbuild walls only because the atmosphere of Anarres is itself a wall—a wall just like the ones that enclose the Asylum on Segvina Island. This is the price of his Utopia.

§7: Nozick’s Dream

By the end of the book, LeGuin’s position seems clear—Anarres is worth it, worth the struggle and the sacrifice. And by the end of the book, though I am tempted, I do not think I agree with her. There is something about the strength of Anarresti social norms, the monotony of culture, which to me is profoundly uncomfortable, perhaps even apalling. And I do endorse the idea that a man has the right to keep some things for himself. Not, perhaps, so many things as the Urrasti aristocrats have, but mine is nonetheless a sentiment that the Odonians would condemn.

Would I, however, condemn the Odonians? No. I don’t think I would. If someone on Earth wanted to build an Anarres, I would not want to live there, but I would commend the effort and argue in favor of it. Partially, this is because I do think there’s a real beauty in the sort of society LeGuin describes, something that does in many ways seem worth striving for. Of course, the brilliance of LeGuin’s book lies in the understanding that the Odonian Society would have its own distinct problems, problems which would be sticky, difficult to solve in just the way that the problems of our own societies are. When it comes to building a world, every choice entails the rejection of ever other choice, and every choice is a tradeoff. So what I want, really, is not Anarres but rather for people to have the ability to *choose* Anarres. I want something that Robert Nozick—a philosopher who the Anarresti would’ve been powerfully inclined to censure—wrote powerfully in favor of.

In the third part of Anarchy, State, and Utopia, Nozick writes:

The conclusion to draw is that there will not be one kind of community existing and one kind of life led in utopia. Utopia will consist of utopias, of many different and divergent communities in which people lead different kinds of lives under different institutions. Some kinds of communities will be more attractive to most than others; communities will wax and wane. People will leave some for others or spend their whole lives in one. Utopia is a framework for utopias, a place where people are at liberty to join together voluntarily to pursue and attempt to realize their own vision of the good life in the ideal community but where no one can impose his own utopian vision upon others.

I do not think I fully agree with Nozick either, at least not about the best method of achieving Utopia. But I agree wholeheartedly with the sort of Utopia he wanted to achieve. In a Nozickean pluralist Utopia, there would be an Anarres somewhere, and so too would there be societies where private property was the *only* law, and still other societies to fill out the entirety of the middle ground. I, for my part, do not want to live under the influence of Odonian neighbors, but I also do not want to deny the option to anyone who does. And Anarres, of course, has its place in a Nozickean Utopia because it is an imperfect society, a society that struggles to realize its own ideals—it is, in other words, a *plausible* society. I am able to sympathize so strongly with the Anarresti project solely because LeGuin has worked so hard to make it feel so real.

Shevek dedicated the greatest part of his life to unbuilding walls, but my own aspirations are more modest. Were I in his position, I would want to make sure only that they gates were unlocked, and to give people the strength they need to walk through them when they so chose.

I’m starting to think that I’ll never have another summer like this one.

On June 6th, 2026, I wrote about coming to Princeton, New Jersey for my summer job. On July 2nd, the day my work comes to an end, I’ll publish this essay so you can hear all about it.

I’ve spent the past month as a sort of TA/camp-counselor hybrid for a high-school philosophy camp—Reason and Rationality. Last summer, when I was a student, the program broadened my life. This summer, it’s done the same. I’ve never done work so completely exhausting, yet so completely fulfilling.

In my June 6th essay, I fell far short of explaining why this place is so special. Now, having hoarded a month’s worth of stories, I think I can do a lot better. But don’t worry—I’m not going to be indulgent. I’ll let one anecdote suffice:

We have a new program this year, an oral exam competition. It was designed as a sort of inoculation. We are, after all, taking in kids who are about to go to college, and college is changing. I share with many, including the instructors here, the suspicion that the essay is soon to shrink entirely from its place of dominance. Oral exams are a natural successor.

The way we do it here is this: your exam consists of a five-minute presentation of an argument, fifteen minutes of objection and response. The exams are graded, and the top four progress to a final round. There, they give their exam again, except this time they are evaluated by a professional philosopher, in front of all the other students and all the parents who’ve come to watch. But the story I want to tell isn’t so much about the exams as about the students who take them.

When I was in high school—I graduated only about a month ago—most of the people around me were, at least in one respect, relatively mercenary. Courses taken were most likely to be chosen on the basis of their contribution to a college application, or else simply because of ease. I can’t say I blame high school students for being this way—I never really was myself, but then I have a strong non-conformist instinct. And I suppose, on reflection, that even I made some concessions to the mercenary drive.

But all of this is really just to say that it’s exceedingly rare to see a high-school student take to an academic project with zeal. Some assignments are more fun than others, but they’re all drudgery at a fundamental level. On the other hand, consider sports. Doing a sport looks good for college, sure, but I spent four years on a wrestling team and I can tell you this much: people who join wrestling quit if they don’t want to wrestle. Step onto the mat, and this is something you can see. Music’ll be playing—“Otherside,” maybe. We’ll be slick with sweat, half of us grinning and half mid-grimace. Every one of us will be in motion, and we move with zeal. Or maybe you go to a tournament, and you’ll see the whole team rush down to embrace a winner when the ref lets his arm back down. Otherwise, you could see a phenom lose. He’ll be in tears over it, and the other guys’ll talk softly to him before leaving him some space. Still, he’ll take a moment to congratulate you when you win a big match later in the day. These are all the marks of care.

I talk about wrestling because there’s something I see at Reason and Rationality, and it’s something I’ve only ever seen before when I’ve been on the mat. The day before the oral exam competition, I go down into the basement of the dorm and there’s more than a dozen people there already. They’re arguing with each other the way that wrestlers handfight when they’re waiting for their matches to start. Here’s one student going back and forth with Noah McKay, striving to show that it’s immoral to mandate race-or-ethnicity-based affinity groups for schoolkids. I quickly slide through the crossfire and take my place beside them, but I’m only half-listening—I’m performing a survey, after all. There, against the wall, two students are hunched heavy over laptops, making ever slighter tweaks to premise and conclusion. And in the middle we’ve got a huddle going, Amos Wollen sitting cross-legged and peering over his glasses at one student—another looks on and listens. In the far corner, there’s Joe Schmid, looking about as intense I’ve ever seen him. The student he’s sitting with is proceeding calmly, laying out his problems with each of Joe’s objection. Of course, the instructors aren’t strictly trying to win the arguments; still, there’s a sense in which the students are seriously outclassed. Almost none of them’s got any formal training in philosophy, and here they are tangling with PhD candidates from the best universities in the world. It doesn’t bother them. Each one’s fighting like the playing field is level.

“Man,” I say, “it’s really humming in here.”

And really, in almost exactly the way that a wrestling room hums on the day before a match, it was.

That’s really the magic thing about R&R—I can’t say how it works, but we’ve engineered an environment where high-school students care about philosophy the way they’d normally only care about sports or theater. We have created an environment where teenagers will stay up past midnight working on arguments in applied ethics, where at one in the morning I will come to collect my laundry and see a small group of students *still* debating an exam topic. I’ll say this to them: “Guys. You are fifteen levels deep into the dialectic, and you only need to defend this for twenty minutes—go to bed.” And they’ll nod, but I won’t actually see them make a move to do it.

Of course, I was here last year as well, so I’ve known how it is to be a student at this place. Hell, I basically spent my whole senior year of high school just waiting to come back. And though it’ll be over by the time I publish this, it really was better the second time around. Here’s the very simple reason: last year, I learned to love philosophy. This year, I got to see so many other kids my age learn just the same; I got to stick my hands in the proverbial clay and do my own part to make it so.

There’s a chance that some of the people I’m talking about will read this essay. If that’s the case, let it be known that I love you all, and that this job is the most rewarding thing I’ve ever done.

Strangely, now that everything’s coming to a close, I don’t think I’m as sad as I ought to be. Perhaps it’s because, as I write this, it isn’t quite over yet. But then, perhaps not.

Lately, I’ve been reading one of my very favorite books again: Ursula LeGuin’s The Dispossessed. In the near future, I’ll have a great deal to say about this book. For now, here’s a passage I particularly like:

He would always be one for whom the return was as important as the voyage out. To go was not enough for him, only half enough; he must come back. In such a tendency was already foreshadowed, perhaps, the nature of the immense exploration he was to undertake into the extremes of the comprehensible. He would most likely not have embarked on that years-long enterprise had he not had profound assurance that return was possible, even though he himself might not return; that indeed the very nature of the voyage, like a circumnavigation of the globe, implied return. You shall not go down twice to the same river, nor can you go home again. That he knew; indeed it was the basis of his view of the world. Yet from that acceptance of transience he evolved his vast theory, wherein what is most changeable is shown to be fullest of eternity, and your relationship to the river, and the river’s relationship to you and to itself, turns out to be at once more complex and more reassuring than a mere lack of identity. You can go home again, the General Temporal Theory asserts, so long as you understand that home is a place where you have never been.

When I was in high school, I had the same Latin teacher three years running, and each year she started and ended us by having us translate the same thirteen words: “Hoc est quod Heraclitus ait: in idem flumen bis descendimus et non descendimus.”

In English: “This is what Heraclitus said: we both step and do not step into the same river twice.”

These are Seneca’s words, both the same and not the same as Heraclitus’s Greek. In the same passage, Seneca goes on to defend what is essentially Derek Parfit’s view of personal identity.

Latin, as it happens, was one of the only classes I found truly fulfilling in my senior year.

Next summer, I’m going to come back to R&R again. In the long run, if they’ll have me, I’ll teach. If I’m lucky, this has become a place I can return to. If I’m lucky, I can have a whole sequence of summers which are nothing like this one, which are nothing like each other except in the most fundamental and important ways. And every year, even as the years get better, I’ll always be waiting to come back.

“You can go home again, the General Temporal Theory asserts, so long as you understand that home is a place where you have never been.”

Sincerely,

Harry Moss

First off, I have to confess something: Heaven and Hell aren’t concepts I can really wrap my head around. My family is Jewish, and Jewish religious teachings tend not to emphasize salvation and damnation. In fact, when I was growing up, I don’t think I ever had a clear idea of what a Jewish Heaven was supposed to be like, or how one was meant to get there. Accordingly, whenever I consider questions about the afterlife, I find it useful to take a moment and try to think about how good Heaven would be, and how bad Hell.1 It’s not that I expect I’ll suddenly be able to get a firm hold on infinity—it’s just that I want these outcomes to be a little more to me than sideways eights on a payoff matrix.

In the course of doing this kind of thinking, it occurs to me that I’d do a whole lot to get out of going to Hell, and probably quite a bit less to guarantee Heaven. If you told me I could live for a while on Earth and then be annihilated, or have a 75/25 chance of Heaven and Hell respectively, I’d probably stay on Earth. I’m not sure that degree of risk aversion is rational, but it would probably guide my action nonetheless. More importantly, considering this sort of question helps me come to grips with the badness of Hell. When we talk about Hell, we’re usually talking about a place absolutely no one should want to visit, not even for a day. On certain views, it’s probably better to be a chicken on a factory farm than a person in Hell.

§1: Larry and Linda

Now, let’s talk about someone who might be going to Hell. Two people, actually—call them Larry and Linda. They’re a married couple, they’ve been together for a decade, and I want to be very careful with my construction of their case. It is of paramount importance that Larry and Linda are generally very good people, and so we’re going to cover our bases with extensive stipulation. Larry works for an EA charity, and Linda teaches at a local school, where she is beloved by her students to the extent that almost half of them credit her with having changed their lives. Each supports the other in their work, and they feel that this is a good way for them to strike a reasonable balance between macro and micro impact—they want to a whole lot of good, but they also feel that they have obligations to their own community. And of course they’re both vegans, but this has almost nothing to do with philosophical argumentation. Rather, it’s just that whenever they see a chicken breast in the supermarket, each of them thinks about the suffering of the innocent animal, and can visualize it clearly, and is simply unwilling to have any part in it. Often, Larry sheds a tear. It is so deeply unjust, he thinks, that any one of God’s creatures could be so abused.

They do believe in God, the both of them. They believe that he’s good, and that he created everything, and they’ve arrived at their belief because they look at the world around them and see a depth of beauty, a beauty so bright and profound that divinity is its only proper explanation.

Here are some further details: though Larry might like to have a child, Linda is absolutely unwilling to bear or to raise one, and decency prohibits us from laying out the precise reason why; suffice it to say that her motive is both substantial and sympathetic. Also, Larry and Linda have a very healthy and passionate sexual relationship—they have intercourse frequently, and it brings both of them extreme pleasure, but also a deeper kind of satisfaction—a sense of profound unity that borders on the spiritual. The sex an important part of their relationship, an important way in which they support each other. Now, perhaps it goes without saying that they use contraception—there’s really no other option, given Linda’s issues with children.

Two more details: First, Larry and Linda are generally truth-seeking people, and kind-hearted, but each cares more about the other than about anything else. Second, it is somehow the case that neither one of them has ever heard of the Roman Catholic Church, or knows the first thing about its teachings.

One such teaching is this: a person goes to Hell if they die in a state of unrepented mortal sin. There’s no complete list of mortal sins, but there are some criteria. A person commits a mortal sin just in case their sin is of grave matter, committed of one’s own autonomous volition, and committed with full knowledge of its sinfulness. There also doesn’t seem to be an exhaustive consensus on how exactly the category of “grave matter” is filled out, but a cursory web search tells me that contraception counts.

As far as the “unrepented” part goes, there are two ways to repent of a mortal sin. The first is to experience perfect contrition—that is, to regret your sin and turn away from it because of love of God alone.2 The other way requires being a Catholic. My understanding is that if you’re a Catholic, you only need to experience some fear of Hell, or perhaps have a sense of the ugliness of sin, and have an intent to do better in the future. Then the sacraments can save you.

One more thing: Under Catholicism a person also needs to be justified to go to Heaven. This can, of course, happen through baptism. But in cases like the one we’re considering, it can also happen through “implicit desire,” which—as far as I know—is basically a matter of having a God-directed disposition while ignorant of the Catholic Church.

Now, going forward, we’re going to assume the truth of Catholicism. As it stands now, Larry and Linda are not committing any mortal sin—they are contracepting, and doing it without hesitation or reservation, but they have no idea that this is impermissible. Now, remember what we’ve stipulated: Larry and Linda care about the truth, but they care about each other more than anything. Let me add something to that: the goal of each isn’t *only* to protect the welfare of the other, as they also care immensely about preserving the integrity of their relationship. Non-procreative sex is an important part of how they’ll do that. Actually, let’s suppose—and stipulate whatever you want to make this the case, so long as it doesn’t damage the couple’s moral character—that non-procreative sex is nearly a load-bearing pillar of the relationship.

The trouble I see is this: so long as Larry and Linda are ignorant of the Church, they’re in a position such that they’re not culpable for failing to know that non-procreative sex is wrong. And while it’s hard to speculate about whether or not they’re justified, they probably have at least as good a chance of going to Heaven as any other non-Catholic. Suppose, however, they do learn that non-procreative sex is wrong. Suppose, for instance, that a well-meaning evangelist comes along and teaches them—perhaps with the aid of reams of miracle evidence—that Catholicism is true. Now, it strikes me that for a couple such as Linda and Larry, being convinced of the truth of Catholicism—or at least learning enough that they would be culpable for *not* being convinced—might not be enough to bring about an immediate change of habits.

So let’s finish our story: suppose that after many days of thoughtful conversation with an evangelist, Larry and Linda are in an epistemic position that basically requires them to be Catholics. However, they don’t make the switch immediately, and they have enthusiastic, passionate, non-procreative sex a few times while they mull the issue over. Larry and Linda now *are* guilty of a mortal sin. In that state, they drive into town for groceries, and their car gets hit by a semi-truck. They die instantly, never having had the chance to experience contrition. Larry and Linda were once possibly going to Heaven, and are now certainly going to Hell. If their lives had gone on longer, they might have stopped having sex or started having kids, might have begun receiving the sacraments and ultimately been forgiven. That doesn’t matter now. Our protagonists are damned; their fortunes have changed because they learned the truth.

§2: A Proper Argument

I hope you’ll excuse my overlong parable. This was a case in which I thought it very important to make my point felt and not merely understood. Still, I have a good bit more to say.

Prior to meeting the evangelist, Larry and Linda might have been going to Hell and might have been going to Heaven. Let’s assume that Heaven is infinitely valuable, and Hell infinitely disvaluable. When I was constructing this argument, the first thought I had was that I might be able to show, at least in some cases, that it could be wrong for a Catholic to convert an unbeliever.

The idea is simple. Suppose it’s pro tanto wrong to cause harm to people, which I’ll define as making them worse-off than they otherwise would have been. It will then presumably be extremely wrong to make people infinitely worse-off than they otherwise would have been. Let’s apply that basic principle to the simplest version of the scenario, in which the utility of Hell is simply the utility of Heaven multiplied by -1. And let’s make it simpler still by stipulating that, prior to being evangelized, Larry and Linda each had a 50 percent chance of Heaven and a 50 percent chance of Hell. Now consider two scenarios: in the first, the truck is bearing down on the couple and they’ve been evangelized exactly as I’ve described. In the second, everything is the same, except that they’re still blissfully unaware of Catholicism. I say “blissfully” because the first scenario has, for each person, an expected utility of negative infinity, while the second has an expected utility of zero. Of course, whether the numbers work out the same when we tweak the probabilities depends on what kind of fancy math we’re inclined to use—also, the “real” numbers are going to depend on whether we think the value of Heaven and the disvalue of Hell can be represented by infinities of the same cardinality.

The problem is that I can’t actually show evangelizing to sometimes be wrong—I can only show that it seems that way to a non-Catholic.3 The Catholic, of course, is going to think that, in any circumstance, converting to Catholicism is always the right thing to do insofar as it’s a live option, and likewise for converting *other people* to Catholicism. But it remains the case that, if I’m right, converting to Catholicism makes some people worse-off—and I mean, much, *much* worse-off—than they otherwise might have been.

Let’s now flesh out what I mean, because I want to be clear that Larry and Linda’s situation is not such an extreme outlier as the story might make it seem.

I’m going to start with the simple observation that not all Catholics go to Heaven. Even among those who have found and accepted the true faith, some—or perhaps many—will die in a state of unrepented mortal sin.

The obvious thought here is that those people almost certainly go to Hell, whether or not they’re Catholic. And I suppose that’s hard to deny if the sin is something like murder. After all, you could probably figure out murder was wrong without knowing that God says so. The same might be true for theft, perhaps of adultery as well.

Still, though, I’m not convinced that even murderers-by-disposition are better off in virtue of being Catholic. Presumably, a Catholic who dies in a state of unrepented mortal sin still has the advantage of having been—to some extent—in touch with God during his earthly life. However, he is also guaranteed to be permanently separated from the grace of God in the afterlife. So long as he isn’t a Catholic, and isn’t culpable for not being one, this won’t be a guarantee. Even when the sin is something like murder, Hell will be the most likely outcome, but there’ll still be some chance that our guy makes it to Heaven—we can conceive of a person who really just doesn’t understand that murder is wrong, after all, or perhaps a person that understands *that* murder is wrong but doesn’t grasp the magnitude of the wrong. Such a person might still have *some* chance of going to Heaven—perhaps far less than one percent, but any chance of a perfect afterlife in the Kingdom of Heaven trumps guaranteed graceless doom. From the perspective of expected utility, this holds true even if there are some Earthly advantages to going down the latter route.

Furthermore, there are sins like masturbation and contraception that aren’t even remotely in the same camp as murder. By this, I only mean that you’d hardly expect a non-Catholic to autonomously figure out they were wrong. Catholics sin in these ways too, and go to Hell for it, but it’s not so obvious that innocent non-Catholics who do this sort of thing will go to Hell. We should at least figure that their chances are much better than those of the murderer or the thief.

The important thing to keep in mind with all of this is that—assuming Hell is infinitely bad in the same way that Heaven is infinitely good—a rational agent should be willing to sacrifice just about any finite good in order to avoid Hell. Even if Hell is not infinitely bad—or even if, as Alex Pruss suggests, a life in Hell is somehow still good for the person suffering—such a life is surely infinitely *worse* than one in Heaven. And when Heaven and Hell are mutually exclusive options, any state of affairs that increases a given person’s chances of going to Hell makes that person not only worse-off in expectation, but *infinitely* so.4

§3: The Molinism Addendum

According to Molinism, God has complete “middle knowledge”—counterfactual knowledge about what people would have done had they found themselves in circumstances other than their actual circumstances. Here’s a rough gloss of the way this is meant to bear on salvation: if you die and haven’t yet converted to Catholicism because you weren’t in the appropriate epistemic situation, God checks his counterfactual knowledge—his knowledge of the “counterfacts,” if you will—and decides whether you go to Heaven or Hell at least partially on the basis of whether you *would have* converted to Catholicism given the opportunity. This view admits the possibility that, as concerns people who aren’t culpable for converting to Catholicism, God can send them to Hell because of facts about whether they *would have* died in a state of unrepented mortal sin in circumstances that permitted them to learn about the wrongness of their behavior. If we have good reason to think God does that, my argument fails entirely.

I happen to think this kind of Molinism is pretty implausible even conditional on Catholicism. However, in the interests of keeping this essay focused, I’ll refrain from arguing against it here.

§4: Moral Harmony

Several of the most interesting and influential modern arguments for the existence of God are “harmony” arguments. Here, I want to talk about a kind of harmony that I seldom see discussed, but which I would expect to obtain in a world where God exists to just about the same extent that I would expect psychophysical or nomological harmony. Call the thing we’re about to discuss “moral harmony.” The term might lead you to believe that I’m about to make something like a theistic argument from moral knowledge, but I actually have something completely different in mind. A union of the moral facts and the human race’s moral intuitions is a *kind* of moral harmony, but I’m thinking about a union of the right and the good. Basically, if I knew theism were true, I would be inclined to expect that doing the right thing generally has good consequences. I would expect this to be true in two ways:

1. People who habitually do the right thing live pleasant lives in the long run(acting rightly may still sometimes require sacrifice).

2. Doing the right thing—while perhaps not always having the *best* possible consequences—generally has good consequences. Corollary: Suppose it’s really true that it’s categorically impermissible to masturbate—that it isn’t permissible to do so even to prevent the end of the world. That’s fine, but I’d then expect—given theism—that no one will actually find themselves in a situation where the only options are masturbation and apocalypse.5

I think most Catholics will be inclined to agree with me on the first point. It may be true with respect to our diminutive Earthly lives that good deeds are not rewarded as often as we might like. But were I a Catholic, I’d simply point out that our Earthly lives are indeed diminutive, and that infinity—in comparison—is incomparably vast. Even if it were true on Earth that no good deed goes unpunished, the inverse would hold in Heaven.6

As far as the second point goes, I’m not sure what Catholic teachings imply about this, but it strikes me as a very commonsensical idea.

Having laid out the idea of moral harmony, I can now make my argument more precise. If Catholicism is true, the right thing for any given non-Catholic to do is convert to Catholicism. Likewise, if Catholicism is true, the right thing for any given Catholic to do is to evangelize. Most of the time, conversion/evangelism will have very good consequences. In some cases, however, it will have horrifically bad consequences. In fact, it will have *infinitely* bad consequences, in the sense that the person converting/being-evangelized will be infinitely worse-off. This is an instance of *extreme* moral disharmony, moral disharmony so intense that it is, in fact, comparable to a situation where morality demands that you refrain from masturbation though the whole world be destroyed as a consequence. Even if it were exceedingly rare for conversion to Catholicism to make people worse-off, the disharmony would be glaring—after all, a single apocalypse caused by an omission to masturbate would be more than enough to count massively against theism. And theists, because they should not expect this kind of disharmony, should—all else being equal—look for a theistic hypothesis that avoids it.

Conclusion

Most moral objections to Catholicism simply point out that Catholic morality conflicts with commonsense non-Catholic morality on a variety of points. My moral disharmony argument is not that sort of objection. By the Catholic’s own lights, it is true that it’s good to go to Heaven, and at least much, much worse to go to Hell. It is also true, by the Catholic’s own lights, that it’s right to convert and right to evangelize. The idea that moral harmony is to be expected on theism is not Catholic doctrine, but—as far as I can tell—doesn’t outright contradict Catholic teachings, and even harmonizes with them in certain ways. All this to say: my argument is a more forceful kind of moral objection to Catholicism because the moral data come mostly from the inside.

Other kinds of harmony are generally thought to count in favor of theism. By the same token, entailing massive *disharmony* ought to be considered a black mark against any given theistic theory. Here, I’ve argued that Catholicism has such a black mark.

The idea is this: suppose that there are souls in addition to bodies. If so, the stuff that makes up souls is different from the stuff makes up bodies in many ways. For one, matter has the ability to move other matter—one body can directly interact with another, push it around, pull it close, et cetera. It seems that soul-stuff, however, cannot directly interact with other soul-stuff in the way that matter can directly interact with other matter. If I have a desire to cause you pain, I can’t simply bump my soul against yours. Instead, I have to maneuver my body such that it bumps up against your body, and body damage translates to soul-suffering. Conditional on souls, people either *are* souls or are largely constituted *by* souls, and yet most or all interaction between people currently requires bodies as intermediaries. This, you might think, is a bit strange. Joe Schmid has recommended calling the strange thing I’ve observed “psycho-psycho harmony.” For my part, I prefer “interaction harmony.” What we’ve just discussed is mind-mind interaction harmony. The immediate question: is this evidence for a Universe biased towards value? I’m shaky on this point. Maybe. The primary benefit of this sort of harmony, in my view, is that it probably allows me to have a clearer sense of self than I otherwise would. Were it to be the case that other souls could freely influence the sorts of things I was feeling “on the inside,” it might soon become difficult for all of us to know where one began and another ended. I don’t want to speculate much further than that, however, because I’m really in no position to say what it might be possible for one soul to do to another. 1

If either interaction argument substantially supports the hypothesis that the Universe is biased towards value, I think it’ll be mind-body. But, as you can perhaps tell, I’m not too bullish on either of these. My thought here is that, whether or not my own harmony arguments work, making harmony arguments has pretty much become a pointless exercise. If the psychophysical and nomological harmony arguments work, nothing further is really needed. I see it this way: the project of the Bayesian theist is to show that, given our actual evidence, we should consider theism vastly more likely than atheism. If “atheism” is read as “the kind of naturalism that allows the laws, constants, and initial conditions to be literally anything,” then perhaps these arguments succeed. If so, theism is massively confirmed over naturalism by psychophysical and nomological harmony alone. On the other hand, there may be some deep-buried naturalistic reason that the laws and constants need to be the way they are—perhaps this will be discovered by future science—and then it’s probably the case that *none* of the harmony arguments work. Either way, there’s not much point to presenting more of these arguments. By my lights, they either fail entirely in ways that we don’t understand yet, or they don’t do much besides take a tiny number and make it even tinier.

However, things get interesting again when we consider non-theistic hypotheses that make the Universe value-directed—I’ll call these “atheistic supernaturalist” hypotheses. When we have those kinds of hypotheses on the table, we can look at different kinds of harmony and say that perhaps classical theism best accounts for one while a form of Buddhist idealism might best account for another. Even within the broad category of classical theism, perhaps Christianity in particular best predicts one kind of harmony, but Islam another. In this sort conflict between various theistic and atheistic supernaturalist hypotheses, the value of a plenitude of harmony arguments becomes obvious. In this sort of debate, there may be a real place for interaction harmony.

As it stands right now, however, I’m inclined to doubt that this new argument constitutes any kind of meaningful contribution to the theistic position2. Still, Interaction harmony is a fun idea, and I’m glad I took the time to do a little write-up.

I truly despise Pascal’s Wager. I have simple reasons. When investigating God’s existence, there are two particularly important questions to ask:

1. Is God real?

2. Should I believe in God?

I want to answer both these questions, but I don’t want the answers to be independent. I want the answer to the second to depend wholly on the answer to the first; I want to believe what’s true, or at least to have beliefs that are well-justified. Furthermore, I want it to be the case that I *should* aim primarily at true, well-justified beliefs. Pascal suggests otherwise, and I find this repulsive. But that’s not much of an argument, is it?

My situation gets worse: I’m an agnostic. Hardcore atheists who assign extremely low probabilities to God’s existence might at least be able to say that pretty much every action they take has some chance of making them either infinitely-better or infinitely-worse-off, and so it’s rational to neglect tiny probabilities, lest they be pulled simultaneously in every possible direction. But an agnostic like me, someone whose probability of God’s existence hovers around .5? No such excuse.

There are, of course, many proposed solutions to Pascal’s Wager.1 I find myself unsatisfied with most of them, and so I want to propose my own.2 It takes the form of a new wager—call it the Weasel’s Wager—which aims to lead you to the conclusion not that you should believe in God, but rather that you should perform a rigorous investigation of your reasons for believing in God, and then go with the result of your inquiry. This is, in essence, Pascal’s Wager for studying philosophy of religion.

To set the stage for the Weasel’s Wager, let’s briefly review the concept of inegalitarian universalism(IU). The idea is that everyone goes to Heaven, but Heaven is better for some people than for others. Here’s a plausible way of cashing that out: one of the things that’s great about being in Heaven is having a personal relationship with God. Now suppose that everyone in Heaven develops their relationship with God at the same rate. However, not everyone begins in the same place. People who had a closer relationship with God on Earth will have a head-start upon reaching Heaven, and will be—at every future time—closer to God than people who didn’t have a close relationship with God on Earth. The afterlife is infinitely good in any case. But suppose you achieve a closer relationship with God on Earth than I do. In that case, each of our afterlives will be infinitely valuable, but yours will be infinitely *more* valuable. At any given time, your life will be better than my life; over an infinite number of years, the difference in value will sum to infinity, just as the value itself does.

I like IU. When I put on my bare theist hat and I ask myself what the structure of the afterlife would be, it strikes me that all versions of universalism beat out all other models, and IU has the distinct advantage of preserving the idea that our Earthly choices really matter.3 After all, one might worry about the prospect of Martin Luther King Jr. and Mao Zedong having afterlives of equal quality; wouldn’t this indicate that the basic structure of Creation lacks the justice that the existence of God would lead us to expect?4

So, suppose we buy that IU is the best live option. Insofar as we think that your position in the afterlife is going to depend at least partially on the strength of your Earthly relationship with God, Pascal’s Wager is still going to hold sway. However, IU makes room for the possibility that, even among believers, some have a closer relationship with God than others, and those people are going to get more out of Heaven.

Now we can consider two ways of coming to have a relationship with God:

1. You take Pascal’s Wager. You go to church with an open mind, and eventually just start to believe in God. There were difficult questions that initially kept you from belief, and you still don’t have the answers, but you believe, and you feel good about.

2. You take the Weasel’s Wager. You spend a decade studying philosophy of religion, and eventually come to the conclusion that God exists. You begin to believe, and to take up some kind of spiritual practice. You still don’t have all the answers, but you think you have some, and you have a much deeper understanding of the questions themselves. Your understanding of God is still thoroughly limited, but it is deep enough to make a major contribution to the awe and reverence you feel towards the divine.

I suspect the second person would have a much better relationship with God than the first, and so an infinitely better afterlife. Of course, studying philosophy of religion doesn’t make belief in God nearly so likely as simply joining a church and committing to religious practice. However, if IU is true and studying phil religion can make the afterlife infinitely better, then it’s probably worth doing that. If the difference in afterlife quality is sufficiently great, it may remain the best option even if there’s a decent chance that some form of infernalism is true.

The Weasel’s Wager is particularly powerful when applied to people like myself—given my psychology, I actually think it’s far *more* likely that I would develop a belief in God through rational inquiry than simply through joining a religion. Also, this one doesn’t have the Many Gods problem to nearly the same extent as the original, and it makes much more room for the plausible possibility that it really does matter *how* you find God. Still, there some issues.

Here’s one major objection I can think of: why not do both?

Here, my thought is this: if I join a religion and this causes me to develop a firm belief in God, it may impede my ability to do philosophy of religion well, and accordingly close off the second pathway. Furthermore, what makes the difference between options 1 and 2 isn’t just whether I perform a rational inquiry, but whether my rational inquiry is my path to God.

Aside from this, the only thing that troubles me about the Weasel’s Wager is the thought that a traditional view of Heaven and Hell might be far more likely than IU. Well, that and the fact that I’ve probably come up with the wager for the sole purpose of explaining why I don’t have to take Pascal’s. But, in my book, the original Wager is a dirty trick anyhow, and so I don’t really mind weaseling out of it.

So, let’s begin:

Michelangelo frowned at the old rabbi. “He’s lying,” he said to the children. “As usual. Don’t believe him.”

“Lying?” Yehuda said, with heat. “How, lying? With Crusades and murder, heretics and executions? And every penny going to that fortress of stone in the name of God—when not a man inside knows what, or who, God is? Surely the Messiah is come, and the Kingdom of Heaven has arrived!”

“The Messiah IS come, you satan!” Michelangelo suddenly bellowed. Jeanne nearly fell off her chair. “And the Kingdom of Heaven is near indeed! The Cistercians have carved wood and bent iron into a pump of perpetual movement! In Chartres, they have erected a cathedral that literally touches the belly of Heaven! And wise men like you and Roger Bacon and Rashi show the benighted the error of their ways!”

The profundity of this scene is concentrated in words you have yet to see Michelangelo speak. For personal reasons, however, I feel the need to include the broader context of Michelangelo and Yehuda’s theological debate. You see, I myself have recently been engaged in a great deal of what might be loosely characterized as theological debate.

Allow me to stage the scene: I am, as far as philosophy is concerned, something of a recent convert. One year ago, I had a radically transformative experience. Tossing aside all previous identities and ambitions, I decided that I was going to be a philosopher. The extremity of my transformation precludes my ability to offer a worthy description. For now, I’ll just say this: More than anyone else, there were two men who brought me to the tipping point. They were my intellectual mentors, and they have raised the highest mountains of my mindscape, carved the swiftest rivers through the plains. I consider it thematically appropriate to call these men P and Q.

I called myself a recent convert; I’m not the only one. In the ten-odd months since I last saw P and Q, both men have become Roman Catholics. At first, this gave me strange feelings. It was never as if they became different people. The three of us have been talking quite a lot lately, and they’re just as lively as ever, just as curious, just as keen to argue as they taught me. But you should understand that secularism, for me, is a basic value, and also that my other basic values have been shaped by P and Q themselves. So there’s a conflict, if only in my head. You see, I don’t think Catholicism is true, but that’s not the extent of the issue; rather, I don’t *want* Catholicism to be true. I can truck with spirituality, perhaps with a loose religious pluralism. But I ponder a world in which there is any single person who lives an infinite life, and yet is damned to Hell on the basis of his actions during a finite period of time; this world, to me, is beyond the pale. It isn’t that I believe such a world-picture cannot be rationally accepted, nor that my problems with it are purely emotional. I do have principled objections to Catholicism, and yet they may be overwhelmed in time. I’m poorly positioned to know.

“And peasants,” the old rabbi replied, straightening his crooked back, his beard quivering, “pretend they are going to the Holy Land and descend on Jewish villages and loot and murder and burn! Surely, God is smiling!”

As of late, I have a new mentor. Call him R. R is a very sharp philosopher, and also a Mahayana Buddhist. The other day, he told me a story which I’ll now paraphrase:

There is a village which has been drowning in a flood of religious teachings. The Buddha comes to this village, and the villagers beg him not to preach. They tell him about the recent influx of preachers they’ve experienced. They tell him what they really need of a holy man is only advice about which creed to accept. And the Buddha answers that they should trust there is an afterlife, and that in the afterlife those who do good deeds will be rewarded. The Buddha says that the nature of the good can be learned independently, and so the villagers should first work to be good, and only afterwards pursue the religious truth. Call this the Buddha’s wager.

R is very good at explaining things to me in such a way that my experience seems to be retroactively described by them, and this is one such case. Setting aside miracles and theology, I reject Catholicism largely because I take the Buddha’s wager—I believe that I can learn what’s right, and that the majority of the worlds in which God rewards me for my earthly conduct are, above all else, worlds in which I take morality seriously. And right now, I cannot mix this with Catholicism, because I believe that the moral teachings of the Catholic Church are wrong.

I don’t care to motivate my claim. In this essay, for once, the argument is not the point.

“That makes God weep and you know it!” Michelangelo began, his voice swelling. But then he stopped. He glanced at William. “Our young brother, though, finds something funny.” They all looked at the young monk. Indeed, William was grinning.

P believes that gay sex is immoral, and I have heard him say, in all solemnity and sincerity, that he would take a bullet for any given gay person. I have heard Q deny that trans women are actually women, and moments later credit one with teaching him how to be a better person.

The point is not the contrast. P and Q are not tragic figures, somehow woefully benighted. The point is not the contrast because there *is* no contrast. P and Q have arrived at their position in virtue of seeking the truth, and now they believe they have found it. By my lights, they are wrong, but this too is not the point.

To care about the truth is also to care about what’s right, to care about the good. So the point is just that these men love their neighbors.

“What? Are you laughing at us?” Yehuda demanded.

William shook his head, but he could not stop grinning. He said, “How are you two friends? A rabbi and a prior? Shouting at each other and calling each other satan?”

“You call this friends?” Yehuda said, throwing his hands in the air. But then he laughed. “No, it’s true. You should be so lucky to have a friend like this.”

And then Jeanne said, “He does. Two.”

William took in his breath sharply. One corner of his left eye was suddenly damp.

Unless I explain who P and Q are to me, this essay will not work as it should. I have called them intellectual mentors, but the phrase is insufficiently revealing. So, please, let me tell you one last story:

Last Sunday night, I sat on the grass with R, and overhead the sky was dark, and he taught me how to write this essay. The concept I needed was devotional receptivity—a sort of spirituality that’s grounded in no particular religious context. To practice devotional receptivity, you have to be radically open to love; you have to recognize that others, when they love you, can make you a better person. Then you have to let them do it.

This is what happened to me. I started to learn philosophy, and I felt as if I were undergoing a rapid expansion. I was in a near-dark space, and my mind was stretching in every direction, sprouting appendages which reached incessantly for the bright, distant nodes on the horizon. But there is no growth without nourishment; P and Q gave that to me. They weren’t Catholic then, but nothing’s changed. They would do it again, and they have.

Devotional receptivity. Never before had my own experiences been so completely clarified by a voice from the outside.

Jeanne turned to Michelangelo. “Why do you call him Satan?” If someone among the present company looked like Satan, Jeanne thought, surely it was Michelangelo. Jeanne saw that Gwenforte was curled around his feet again, while Jeanne’s bare feet were cold.

I think I know this much now: the strange, uneasy feeling I got when I first thought about the conversion was basically unfounded. Yes, I disagree with P and Q over whether Catholicism is the true religion. Likewise, both of them are moral realists and I lean Kantian. Were I generally to find myself in agreement with these men, we would have much less to talk about, and I would deem this tremendously unfortunate.

The question, I suppose, is one of whether religious disagreement is somehow unique. When the Catholic meets the agnostic, is there some noxious presence that crowds the air, some black miasma which is conspicuously absent from the meeting of the deontologist and the utilitarian?

No. But I can say this much: There may be, on the part of the Catholic, a fear that the agnostic will be damned to Hell, and this might beget an urgency that the agnostic mistakes for aggression. It is probably true, after all, that P and Q fear for my soul. But I needn’t make too much of this; if they save me, they do it by winning an argument, and that’s what each of us is trying to do anyhow. And just as the fear which seeps into the dispute is premised on the love of another, the argument itself is premised on a love of Truth.

Michelangelo replied, “He is satan! Or rather, a satan. In Hebrew, satan means an advocate of the alternative, the one who makes the arguments you don’t know how to refute.” Michelangelo looked to the old Jew, still grinning wickedly in the corner. “That satan is my best friend.”

I have a message for P and Q, and also for R, and for the other people who will read this essay and know themselves to be included: Each of you is my satan, and it is *for* this that I love you.

“Forty-two!” yelled Loonquawl. “Is that all you’ve got to show for seven and a half million years’ work?”

“I checked it very thoroughly,” said the computer, “and that quite definitely is the answer. I think the problem, to be quite honest with you, is that you’ve never actually known what the question is.”

“But it was the Great Question! The Ultimate Question of Life, the Universe and Everything!” howled Loonquawl.

“Yes,” said Deep Thought with the air of one who suffers fools gladly, “but what actually is it?”

Deep Thought, the hyper-computer Douglas Adams writes of in The Hitchhiker’s Guide to the Galaxy, is capable of discovering the answer to any question—or, as it were, the question for any answer. Deep Thought is a sort of truth-machine.

I love the truth. Lovers of the truth, I think, should want to build the truth-machine. We should want to come together, combine our powers, raise a fig tree from barren earth. However, what if the truth-machine takes millions of years to build, millions longer to answer the questions and to find them? We can imagine such a machine under construction. We can imagine perhaps a thousand brilliant men and women who together begin the work, knowing that they’ll never complete it. We can imagine that work being passed down to the next generation, to the biological and intellectual descendants of the original thousand.

We can imagine a pyramid of stacked stones, and each member of each generation may add only one. The base of the pyramid covers ten thousand square kilometers, and no mere outline will do. If it is to stand, the great monument must be fully substantial. And so each generation toils, and the next toils. I place my stone, and so too does my son and his daughter and her students, and each of us are grasping at something that we doubt we will ever really *feel.* But we build. We think so carefully about where to place our stones, and then we do, and we are content with our lives. To simply know that all our toil has been in the service of something so great; this is enough. This is everything. Because on some distant day, the apex of our pyramid will scratch the belly of Heaven, and it will then begin to leak. Truth will be the second Flood to drown the Earth. Even if the very last builder—the very last living human—dies in the moment she places the final stone, we will have done it. Our lives will not have been in vain.

Now invert. Let one broad stone be the base of the structure, and let each newcomer be added in such a way that the great creation, as it expands, becomes not narrower but broader, ever broader, broad enough in time to encompass everything. And now I show you philosophy.

Nothing troubles the philosopher more than the assertion that his work is either pointless or futile. This assertion, says the philosopher, is so deeply misguided. But perhaps the philosopher’s real worry is that the layman touches on a substantial point, that the work *will* turn out futile. Because what chance have I, one man, of discovering anything *real?* I might spend my entire life solving the problems of an ideological program which ultimately fails. There is no guarantee of discovering the truth, but this is not so much the trouble; the trouble, rather, is that we haven’t even the slightest assurance of being positioned such that we *can* discover the truth. Moreover, if any given philosopher believes that epistemic parity holds between herself and those philosophers opposed to her, perhaps she should even think that her chances of being so situated are fairly poor. Two equally good reasoners may come to entirely different conclusions simply in virtue of having started in slightly different places—why, then, should I think that my starting point is the right one?

The question troubles me; it threatens the truth-seeker. But also, it inspires a question: ought the truth-lover be a truth-seeker? More precisely: ought the truth lover care much whether his own pursuit of truth is successful?

Let me share the vision of philosophy that I have recently come to find appealing. As philosophers, we are building the truth-machine. We are trying to scratch the belly of heaven. That we are building the truth-machine does not mean that any one of us will be personally acquainted with the truth. The truth is not wholly contained in any particular philosophical system. The truth, rather, is the output of the machine, and the machine is essentially the totality of *all* the philosophical systems and the totality of all the relationships between them. Philosophy itself is the machine. So imagine philosophy as a great printer, and the truth as the sheet of paper it creates. The printer and the paper exist, and so too do we. But in relation to the printer, I am perhaps a drop of ink, and you are one of the display’s many pixels. The ink cannot read what the printer outputs, but the output remains *real.*

This is how we build the truth-machine. This is how we build *Truth.* And even if none of us are alive to see and to know, we can preserve the structure of the machine, carve it in stone if we must, for the structure *is* the substance. And then we will have done it. Our lives—our work—will not have been in vain.

My thought is this: if private property can be legitimate in the first place, which I think is likely, then this is a straightforward case of ownership. So let’s break down what happened: Dana took stuff that she already owned, purchased some additional items from willing sellers, then did some work to turn a bunch of things that were already hers into something new. This resembles John Locke’s picture of how people come to own things, according to which you make something yours by mixing it with your labor. But though we may dispute the conditions under which mixing something with your labor can make it yours, Dana’s case should be uncontroversial. That’s because she not only worked to make something new—the sourdough loaf—but made it out of things that she already owned.

Now, suppose that the day after eating her sourdough, Dana goes to the sperm bank, purchases sperm, and voluntarily becomes pregnant. Nine months and much labor later, she has a child. Call the child Steve. Does Dana own Steve?

Well, when you own something, we might say you have at least four kinds of right:

1. The right to use the thing as you will.

2. The right to transfer ownership of the thing to someone else.

3. The right to set the conditions under which others can use your thing.

4. The right to destroy the thing.

That is not to say, however, that these rights alone define the ways in which you can permissibly use the thing in question. If I have a kitchen knife, I may use it in a variety of ways. If I so choose, I can stick googly eyes on it, name it Sir Pointy, and speak to it as if it’s a person. I may not, however, use Sir Pointy to skewer my friend throat. That is because my friend has a right against being skewered. Note also that we can construct a parallel case for the right of destruction. If I have a photograph of a beautiful mountain landscape, I am ordinarily permitted to burn it. Suppose, however, that I have promised a different friend, one who very much enjoys looking at mountain landscapes, that I will always keep the photograph for her to admire. I may not, then, destroy the photograph. We could come up with examples for the other rights as well, but the central point is simple: My rights are constrained by the rights of others.

Let’s now return to the case of Dana and Steve. Certainly, Dana may do certain things to Steve that she can’t do to adult people, or even to children that aren’t her own. Dana may, for instance, prohibit Steve from exiting his room at night. When Steve is young, Dana may even prevent him from leaving home unsupervised at all—this will have the effect, among others, of preventing him from running out to the middle of the street and being hit by a car.

Now, this resembles the right of use. There are, however, certain ways in which Dana may not use Steve. She may not, for instance, forcibly dislocate his shoulder. Furthermore, there is at least one property right which she does not have—the right of destruction. Dana absolutely may not kill Steve. So, in virtue of the fact that Dana’s Steve-directed rights are so different from the suite of ordinary property-directed rights, can we say that Dana does not own Steve? I think otherwise. Let’s take the right of use. We’ve already established that my right to use a thing is constrained by the rights of others, and I don’t see why this would change if the thing and the person were one and the same.

So if Steve’s rights constrain Dana’s rights, we can’t establish non-ownership simply by observing that Dana’s rights are constrained. Instead, if we’re to say that Dana does *not* own Steve, we must, I think, say that Steve has a particular kind of right—a right against being owned. And we will want to say, I think, that all people have this right, and that it is innate. This is the simplest way to explain why Steve has such a right from the very moment he enters the world.

In what, however, does Steve’s right against being owned consist? I think there are two obvious answers we could give:

1. People just can’t be owned at all.

2. People own themselves.

In favor of Option Two, we can say that it seems people are entitled to use themselves as they will, and also to set conditions under which they can be used by others.1 However, the other two rights are trickier. If the right to destruction holds, then people can freely commit suicide—some may consider this an issue, though I don’t. But the real problem is with the right of transfer. The right of transfer, after all, is a right to waive all my property-directed rights. When I transfer my property to another, I no longer have any legitimate claim on it, nor any ability to set the conditions under which others may use it. If I may do this with my own person, I may make myself a slave. This, I think, will strike many as a completely unacceptable result, far worse than the permissibility of suicide. Is it incorrect, however?

There are, by my lights, a couple ways we could get to the result that people do own themselves and yet can’t sell themselves. I want to walk through my preferred route, and I want to start by returning to Dana and Steve. Dana, as we’ve established, is permitted—at least in the ordinary course of events—to restrict Steve’s behavior in certain ways. We’ll again use the example of keeping Steve inside so that he can’t run out into the street and get hit by a car. If Dana doesn’t own Steve, in virtue of what is she permitted to treat him this way? In the case of any ordinary adult, after all, death by car is certainly much more likely outside the house than in. But Dana is not permitted to keep just *anyone* inside for the benefit of their own safety. There must, then, be some kind of special relationship that holds between Dana and Steve. On Dana’s end, we could call this relationship “parenthood,” but I prefer “guardianship” for reasons that will soon become clear.

When two people stand in guardianship, let the former be called the guardian and the latter the ward. The guardian’s core responsibility is basically to compensate for the deficiencies of the ward. Dana and Steve again: Dana is permitted to prevent Steve from leaving the house unattended not because this makes him safer, but rather because Steve is not rational enough to make an informed decision about whether the risk is worth it, and furthermore because Steve—assuming he won’t use the sidewalk instead of the street—lacks the self-control required to *minimize* risk, or at least to reduce it. The relationship between Dana and Steve may be called “strong guardianship,” while a looser form of guardianship may hold between any responsible adult and Steve. Dana’s neighbor Lauren, for instance, has much less legitimate control over Steve than Dana. However, Lauren may still intervene if she sees Steve wandering too close to the road. Call this relationship “weak guardianship.” This is not meant to be a firm distinction—each kind of guardianship may come in degrees, after all—but rather a practically useful one. Broadly speaking, though, we can say that strong guardianship involves a special obligation, while weak guardianship does not.

The great virtue of this picture is that it is child-adult symmetrical—it does not posit that “child” is a special category that some fall within and others outside, such that those who fall within are basically at the mercy of those outside. On this picture of guardianship, you may take measures to prevent your child from wandering into the street, but also to prevent your senile grandfather or severely mentally disabled sister from doing the same. A parent’s child-directed rights are downstream of child-directed *duties,* and the duties hold in virtue of the child’s incapacities rather than the child’s age.

Let’s now return to the distinction between strong and weak guardianship. If there is no special obligation involved in weak guardianship, a person has a duty to act as a weak guardian for *any person* who demonstrates a relevant incapacity.2 We can see what I mean by creating a new vehicular manslaughter scenario: suppose that, from my position on the sidewalk, I watch as a man begins to slowly cross the street. The man is simultaneously jaywalking and having an impassioned phone conversation. I then notice that a car is approaching. I call out to the man, but he doesn’t seem to hear me, nor does he seem to hear or see the car. The car, for its part, is not slowing down. Imminently, the car will strike the man and kill him, but I fortunately have enough time to intervene, to jump in and tackle the man out of the way. I can do so such that I won’t harm either one of us. I think most people would agree with me that I should do so. But I may not tackle this man in ordinary circumstances, and presumably I may not do so even when he takes action that is likely to harm him. If, for instance, I see someone attempting to free-solo3 a mountain—which is incredibly dangerous—I may not physically intervene.4 What explains the fact that I may physically intervene in the first case but not in the second?5 In my view, it is the relationship of weak guardianship. The jaywalker doesn’t see the car coming, and so can be forcibly removed from its path. The free-soloist has no analogous incapacity—he is at least presumably fully aware of the risk.

Okay. That’s all the groundwork. Finally, it’s time to build up the central argument. Note that, while the relationship of strong guardianship doesn’t seem like the sort of thing that could hold between peers,6 the relationship of weak guardianship absolutely can. There is no asymmetry in the relationship of weak guardianship—if I don’t see a car coming at me, my friend may tackle me out of the way, and I can do just the same for him. Of course, some people with more fundamental incapacities may, in some circumstances, be incapable of taking on the “guardian” role. A blind person, for instance, may not be able to see the car that’s about to run me over. If the person in question *could* see, though, she would be responsible for taking me out of the path of the accident, just as I would for her. his is the symmetry I mean.

Perhaps it’s now become clear what I’m working to. Suppose there was some course of action such that no person who actually exists was capable of rationally taking it. If so, the relationship of weak guardianship might hold in such a way that each of us would be entitled—and perhaps required—to prevent others from taking it, even if the others very much wanted to. By my lights, selling oneself into permanent slavery could well be such an action.7 I’d go so far as to say that it’s such an action if *anything* is. So the concept of weak guardianship sets us up well for a view on which people do own themselves, but may not sell themselves into slavery.8 For me, two questions immediately arise here:

1. If this goes for selling oneself into slavery, does it go for anything else?

2. What exactly is the relevant incapacity here?

As far as the first question goes, I’m not sure. I’m not personally inclined to think that weak guardianship allows us to prevent others from suicide, but I think a very good argument to the contrary could be made. There are also, of course, arguments to be made that weak guardianship allows us to prevent others from doing a variety of other things—perhaps even things like free-soloing. All of this I will leave to be discussed another time. So now let’s turn to the second question:

There are, I think, probably multiple candidate incapacities in the case of one who chooses to sell himself into slavery, but I’m inclined to think that the most important one is a sort of foresight. Let’s return to the jaywalker: in the jaywalking case, there was something that the jaywalker didn’t know, such that he would have acted differently had he known it. In the slavery case, there is an analogous gap in knowledge—the prospective slave doesn’t know whether he’ll still be willing in the future.9

But in response, you may say: “Doesn’t this have much broader implications? No person, prospective slave or otherwise, knows how they will feel about anything in the future. Does that mean that no one can ever commit themselves to something in advance?”

By my lights no. I concede that the difference between the slavery case and the case of say, promising to keep in touch with a friend actually is a difference of degree rather than of kind. However, it’s such a substantial difference of kind that I think it basically—from a practical perspective—basically amounts to a difference of degree. Let’s consider for a moment how people actually change their minds; I think it’s typically either because of a single radically transformative experience, or because of an accumulation of much less impactful experiences which each steer one slightly in a given direction. So change, if I’m to generalize greatly, either happens because of experiences which are radically different from previous experiences, or it happens over a long period of time. For this reason, we should think that the person who sells himself into slavery is vastly disproportionately likely to change his mind in the course of being enslaved—the experience is very likely to be radically different from previous experiences, and so too will it be very, very long. This holds even if the person is quite certain that he *won’t* change his mind—just ask me how I felt about God when I was thirteen.10 On the other hand, in the case of a promise to keep in touch, I can probably foresee that my future mental states aren’t going to change so dramatically that I will no longer want to keep in touch. The costs of keeping in touch are also very low, while the costs of being a slave are very high, and this should be factored in as well—weak guardianship definitely doesn’t kick in when it comes to the prevention of trivial harms. Now, I think that selling oneself slavery is likely to be a more extreme case in terms of both the costs and the relevant incapacities than pretty much any other action we might imagine. We should thus think that, with regard to the prospect of selling oneself into slavery, people are not merely poorly informed, but just about incapable of being appropriately informed. This does not preclude the possibility that some other rational beings with a much better ability to predict their future mental states could be permitted(by morality and by their peers) to sell themselves into slavery.

There is much more to be said about this theory, but I’ll break it off here for now; I’d like to allow myself to write about something else. If you’re aware of anyone who’s previously done work on an idea like this one, please let me know!

There is a video game called Outer Wilds. One might call it a cult classic. It is a strange work of art in that, when I attempt to discuss it with those who do not have first-hand experience, I am incapable of being both sincere and convincing.

Tonight, I’m going to be sincere: I believe that Outer Wilds is a near-perfect work of art, by far the greatest of all the games. I would freely call it a peer to literature’s most esteemed classics, insofar as commensurability holds. But I will never say this in a way that is persuasive to the uninitiated. That is because Outer Wilds cannot really be explained. I could tell you about how the game functions, about its elegant puzzle design, about its impeccably tasteful art and music. In doing these things, I would ruin many surprises that are integral to the experience of playing the game. More importantly, I would fail to illuminate much. I could tell you what Outer Wilds is, but never how it feels. I could give you an intellectual sense of the game, but never the tactile, visceral totality.

In Outer Wilds, you explore outer space. Tonight, I feel rather like I’m exploring space myself. Tonight, I feel as if Outer Wilds is a place on Earth.

Tonight, I’m writing to you from my dorm room at the Princeton Theological Seminary. I am not, however, here to study theology. For the next thirty days, I will be playing a dual role in the independent high-school philosophy program Reason and Rationality. Last year, I was only a student. This year, I’ll be studying in an advanced program for returners, and at the same time working as a “Group Leader” for the new students. To call this my dream job would be understatement.

I can tell you certain things about Reason and Rationality. I can tell you that last summer, on the Seminary campus, my interest in philosophy properly sprouted for the first time, and has not since withered. I can tell you that all the classes are run seminar-style, and that the student body is more engaged than any other student body I have yet encountered. I can tell you that I’ve made great friends here, that we’ve sometimes stayed up long past midnight discussing dualism and the problem of evil. I can tell you that the instructors—among them Amos Wollen, Flo Bacus, and Noah McKay—are the best teachers I have ever had. I can tell you that the program is responsible for my veganism, and that this is not because veganism was ever preached to me; rather, I here learned the way of thinking from which a rational commitment to ethical veganism naturally emerges.

I can tell you there is an important sense in which I write Tabanus Noumenon *because* I went to Reason and Rationality.

I can also tell you that when I was a little kid, I always used to have trouble sleeping on the night before my birthday. My overwhelming excitement kept me at a distance from its object. I have not cared about birthdays for a long time. And yet, last night, knowing that I would depart for R&R the following morning, I once again felt what I have described. It was the first time in years.

I could tell you more. But really, I haven’t told you anything important. Outer Wilds is a place on Earth, and I am *there.* Outer Wilds, however, is something you have to see to believe. Try as I might, I cannot be your disembodied eyes.

So instead, let me tell you this: Tomorrow, and in all the days of the month that follow, I will debate philosophy with a vigor that the circumstances of the past year have not often permitted me. And in my downtime, I will write philosophy, and I will produce essays that I expect to be better and come faster than anything I have thus far shown you. This will be my work, but in equal measure the work of this place, the work of the people here. So prepare for a flood, dear reader.

And I hope that you have something which means to you what this place means to me.

Sincerely,

Harry Moss

“I do not know what I may appear to the world, but to myself I seem to have been only like a boy playing on the sea-shore, and diverting myself in now and then finding a smoother pebble or a prettier shell than ordinary, whilst the great ocean of truth lay all undiscovered before me.”

—Isaac Newton

If we were to compare the field of philosophy to an ocean, we would have to say it was the Pacific—both very broad and very deep. As a consequence, there’s really no one around today who’s an expert in philosophy simpliciter. Expert philosophers, rather, are experts in some particular subdomain of philosophy, be it ethics, metaphysics, epistemology, et cetera. Given the broad scope of the discipline and the complexity of the subdomains themselves, this stratification may be necessary. Nonetheless, it is in some ways regrettable that philosophy has become so partitioned, each philosopher so fenced-in.

Now, let’s consider philosophy of religion. Though the focus of the field is very narrow, philosophers of religion draw on a wide range of philosophical tools—metaphysics, ethics, and epistemology all play major roles, to name a few. In virtue of being so interdisciplinary, philosophy of religion is very difficult to do well, but also very fun. So let’s talk about it for a bit. In particular, let’s talk about a recent video on Joe Schmid’s YouTube channel, in which he brings on Dr. Nevin Climenhaga to discuss Climenhaga’s case for theism:

This is a follow-up to a previous video—in that one, Schmid and Climenhaga discuss the latter’s theory of probability.1 In the video above, they then go on to discuss how that theory of probability ought to be applied to the question of God’s existence, and Climenhaga goes on to give a demonstration.

Now, I’m an agnostic, and I can’t say Climenhaga’s case for theism moved the needle much for me. However, my goal here is not to do some kind of aggressive debunking. I’m a great fan of Majesty of Reason, Climenhaga’s theory of probability strikes me as pretty interesting, and this video gave him just a bit more than two hours to lay out his case anyhow. In those two hours, he and Schmid cover fine-tuning for life, fine-tuning for discoverability, consciousness/psychophysical harmony, religious experience, miracles, hiddenness, and evil. They then do the Bayesian shtick2—they come up with numbers for how likely each of these bits of evidence would be conditional on naturalism and for how likely they’d be conditional on theism, then plug them all into a spreadsheet.

Now, any one of these topics could be the subject of a two-hour video all by itself. So in light of time constraints, I don’t think it’s appropriate for me to pass any sort of confident judgement on Climenhaga’s case for theism. There are points in this video where I’d be inclined to criticize the moves Climenhaga makes, and I will touch on one or two of them, but for all I know he addresses my criticisms thoughtfully elsewhere. Accordingly, my goal in discussing his case for theism isn’t to attack his position directly, but rather to use it as a springboard and leap forward into some of my own ideas about God.

All that said, let’s take a look at Climenhaga’s cumulative case in visual form:

There are a couple things to note about this table. First, when we consider(for instance) fine-tuning for life, math people are quick to point out that we don’t know enough to assign probabilities in a principled way.3 A similar point can be made about pretty much every other point here, if in a less formal way. What’s the probability that consciousness exists condition on naturalism,4 for instance? Well, that’s going to depend on what we think consciousness *is.* For a type-A physicalist, it’s trivial that consciousness would exist in any Universe that has life. For a panpsychist, it’s trivial that consciousness would exist in any Universe simpliciter.

To account for these difficulties, Climenhaga uses relative probabilities rather than absolute probabilities. When he puts down that the probability of a Universe existing is 0.2 conditional on theism and 1 conditional on naturalism, he’s saying only that the existence of a Universe is 5 times more probable on naturalism.

Now, I also want to note preemptively that though fine-tuning for life is the standout here, theism wins in Climenhaga’s model even if we disregard fine-tuning entirely. So though I’m sure many who look to critique this model will target fine-tuning in particular,5 defeating Climenhaga’s fine-tuning number will not defeat his conclusion.

Having gotten that out of the way: that fine-tuning number sure is something, huh?

Allow me to sum up my own stance on fine-tuning—I do think it’s reasonable to take fine-tuning as evidence for theism. I myself take fine-tuning as evidence for theism, and for all the standard reasons.

However, unlike the people who discovered fine-tuning, I don’t have a PhD in physics. More precisely, my understanding of the relevant physics is very limited—I could explain fine-tuning to a five-year-old, but I definitely couldn’t teach a high-school class on the subject.

Now, when I don’t know something, I’m perfectly comfortable deferring to the experts who do—the expert physicists, for their part, generally seem to agree that the Universe is fine-tuned for life—though there is some dissent, as far as I understand.6 Anyhow, we’ll return to the subject of expert deference in a few paragraphs. For now, let me a say a bit more about fine-tuning.

When I consider the basic fact of fine-tuning, I certainly am inclined to think: “this would be more likely if there was some very powerful being who valued the existence of life than if there wasn’t.” And I have some independent reason to think God might exist—the contingency argument, the psychophysical harmony argument, et cetera—then it’s easy enough for me to connect the dots. But how much more likely is fine-tuning given the existence of God? Twice as likely? Ten times? Ten *trillion* times?

Well, once we’re asking that question, I don’t think it’s enough to consider just the fact *that* the Universe is fine-tuned. We also need to consider *how* the Universe is fine-tuned. Let me illustrate what I mean: There’s this number that fans of the fine-tuning argument like to bring up—I believe it’s 10^-60, but I’ve also seen 10^-90, 10^-120, and even 1/10^10^123 thrown around—such that if the value of some constant changed by that much,7 life in the Universe would be impossible. Now, I did go and check what these numbers actually refer to, and I’ve added some context in a footnote, but I’m not going to do so here in the main body of the essay. That’s because I don’t want to oversell my understanding—or, rather, I want to emphasize my ignorance. Of the fine-tuned constants, the cosmological constant is the only one such that I intuitively understand why life couldn’t come to be if we tweaked it. However, I *don’t* intuitively understand how we calculate the cosmological constant. The point is that I don’t really understand fine-tuning very well, all-told.

There are also some less technical details to consider. The Universe is fine-tuned in such a way that life exists, yes, but life exists only in the narrowest corners of a massive expanse, and exists in such a way that—once the development of life begins in some particular place—the more “valuable”8 kinds of life don’t emerge for a very, very long time. Furthermore, the process by which life comes to exist involves *immense* suffering.9 And there may, of course, be some multiverse theory that adequately explains fine-tuning. Also, suppose there were some values of the fundamental constants for which the Universe was “super fine-tuned,” such that life was not only possible but very common, and perhaps could even emerge through some much less suffering-intensive process. Or maybe this isn’t possible with a tweak to the constants, but it could be possible with entirely different laws. If “super-fine-tuning” were possible, then the fact that the Universe is only regular-fine-tuned would be a probabilistic hit to theism.10 I bring up all these ideas rapid-fire because I’m not sure what to make of them—I haven’t yet fully marshaled the available evidence, nor properly organized my thoughts, and that is precisely why I’m not convinced one way or the other. The uncertainty is of the essence here.

Now, to his credit, Climenhaga—who presumably knows much more about all this than I do—makes some room for uncertainty as well. He recognizes that fine-tuning is by no means guaranteed on theism, and he factors in the possibility that some future naturalistic explanation of fine-tuning may prove satisfactory. Still, he thinks fine-tuning is far more unlikely on naturalism than theism—more unlikely by 16 orders of magnitude, in fact. To me, this seems excessive. But then, you should know by now not to take anything I say about fine-tuning on faith.

The astute reader may have a question here: why are you even writing about this if you, by your own admission, are so ill-equipped to do so? Because it’s true that I really have no idea what I’m talking about when it comes to the finer details of fine-tuning.

Now, I know this won’t apply to all of you, and I certainly don’t mean any offense by it, but here’s the thing: I think it’s very likely that you, dear reader, *also* have no idea what you’re talking about.

Let me explain: There’s this very thorny epistemological question about expert deference. When, if ever, should you take the conclusion of an expert over the conclusion you yourself have reached by reasoning about the available evidence? If this is sometimes appropriate, why?

I don’t have a settled position on this issue, but I’m inclined to say that you should defer to experts because experts believe what you *would* believe if you had better evidence and understood your evidence better. After all, you need to have faith in your own ability to do good reasoning if you’re to decide which experts are worthy of deference in the first place. So suppose we take climate science as an example. I believe that human activity has caused climate change pretty much solely on the basis of expert testimony. This is perfectly appropriate—when it comes to climate change, there’s widespread consensus, so I can see it’s highly probable that I would believe in human-caused climate change were I to become an expert. But suppose there were a case where I really didn’t have a good idea of how my credences would change if I acquired more expertise. Fine-tuning is such a case—there is, after all, no real expert consensus on what fine-tuning implies. And so, without a deeper understanding of the data, I can’t even really be sure which experts I should defer to.

Right now, I feel that I at least know enough to say that fine-tuning seems somewhat more likely on theism than on atheism, but I don’t know nearly enough to say that whether difference in the relative probabilities is sixteen orders of magnitude. I’m not saying here that Climenhaga’s analysis of fine-tuning is wrong. Instead, I’m saying that I’m not appropriately positioned to judge one way or the other, and I suspect that the same goes for most people who have watched the video.11

Of course, I’m not even sure that my underlying position on expert deference is correct. Uncertainty abounds!

There’s also the question of stakes. Setting aside the material consequences of my belief or unbelief,12 I consider it incredibly important to have the right take on this.13 I would rather suspend judgement than be wrong. And so I hesitate to step firmly in either direction. In the future, as I accrue expertise, I may come to feel I understand all the data well enough that I can settle on a position. But I suspect that will take a while. On fine-tuning in particular, I don’t think I’ll be able to land on a reasonable estimate of the relative probabilities until I’ve learned a lot of physics and had a lot of time to think it over.

Let’s now loop back to what I said at the beginning. Philosophy of religion, in my view, is thrilling precisely because it is so interdisciplinary, draws so much on so many different fields of study. However, for these same reasons, it is a kind of philosophy that’s exceedingly difficult to do well. To comment wisely on the cumulative case for or against God requires either a great deal of accrued expertise or a whole lot of expert deference. The problem, of course, is that it’s also so difficult to know which experts to defer to on many of the relevant questions.

But no philosopher could feasibly achieve expertise in *every* field relevant to the potential existence of God! Should we abandon our inquiry entirely? Should we never allow ourselves to be convinced by any arguments? Surely we shouldn’t go so far as to abandon our Bayesian models?

My thought is this: we can build probabilistic models, but we should build them with error bars. The size of the error bars should correspond to the confidence of the philosopher building the model, and philosophers should be more confident where they have more expertise. And this will most probably give us a scenario where the philosophers with the most expertise in the various relevant fields get to be settled atheists or theists,14 and all the interested-but-relatively-ignorant students like myself can perhaps lean one way or the other, but will be agnostics at the end of the day.15 That’s not to say that a person like me shouldn’t make arguments for or against the existence of God—anyone can have a valuable insight. Furthermore, it’s perfectly appropriate to make arguments for positions you don’t wholeheartedly support. Even if I were to assign myself the role of defending a position which I thought was highly unlikely to be true—let’s say ethical intuitionism, on which I’m currently very down—and even if my suspicions were correct and intuitionism was false, I would still accomplish something valuable by making good arguments for it; advocates of other positions would have to refine their theories to deal with my arguments, and we would all get closer to the truth. In this way, we should all strive to wake each other from our dogmatic slumbers, even when it requires stepping out into terra incognita.

Okay, so that’s the first part of my argument for agnosticism. But it’s not really an argument for agonisticism simpliciter—it’s an argument that people like me should be agnostics, because philosophy of religion is really hard and you need to know a lot to do it well. However, while a lot of people are in an epistemic position similar to mine, some people will certainly know enough and have enough experience that it’s appropriate for *them* to have a confident position on the existence of God.16 Do I have anything to say to them?

Well, I now want to make an argument in which I’m less confident. I want to argue that agnosticism is the appropriate stance to hold at the present moment, and that this is true whether or not God exists. And I want to argue this in such a way that my conclusion will apply to everyone, not just to students like me.

Let’s start with the atheist side. If atheism is true, agnosticism is currently appropriate simply because we haven’t learned enough about the basic structure and origin of the Universe to be *sure* that atheism is true, and because there are some compelling arguments for theism. If atheism is true, most of these arguments probably do have successful defeaters, but those defeaters may rely on scientific knowledge that we don’t yet have access to. So even if all the arguments for theism fail, we don’t yet understand how they fail, which means we don’t know for sure *that* we they fail.

Now, as you can probably tell from the fact that I’m moving on after one paragraph, it’s not the atheist side of this argument that I’m most interested in. So let’s talk about why agnosticism is currently appropriate if God *does* exist. My case here is going to rest on divine hiddenness.

The problem of hiddenness can be summed up with a simple question: “If there’s a God, why don’t we *see* him?” There are, however, at least two ways of interpreting this question. According to one interpretation, we’re asking why there’s no good evidence for God’s existence. But on this interpretation, theists—and even many atheists—aren’t going to accept that hiddenness is a problem. Because there *is* evidence for the existence of God. It may not be decisive, and it may not even outweigh the evidence for atheism, but there is some. If God does exist, he has left fingerprints. According to the second interpretation, though, we’re asking why the fingerprints aren’t easily identifiable as such, and also why they’re sometimes so hard to find. *That* is the real problem of hiddenness. We have fine-tuning, it’s possible we even have a genuine miracle or two, but we don’t have the Qur’an written in the stars. If God exists, he *could* do something to make his existence not only discoverable but obvious obvious, and yet he doesn’t.17

Like many, I don’t think hiddenness is a very extreme problem for the theist—it’s certainly not as baffling as evil. But there’s still something to glean from it. Namely, we’ve learned that *if* God exists, he chooses not to make his existence obvious. And whatever God does, he has good reason to do. So if God chooses to leave room for uncertainty about his existence(in my view, a lot of room), then you probably have good reason to be uncertain.

I can think of two interesting responses the theist might give here. The first is simple—perhaps there is some value in wading through uncertainty in the philosopher’s quest to discover God. Perhaps through this process of discovery, one can have a closer relationship with the divine than would otherwise be possible. The problem here, however, is again disagreement among experts. Close study brings some nearer to God while at the same time pushing others away, and it doesn’t seem to be the case that theist philosophers are generally smarter, more dedicated, or more virtuous than their atheist counterparts. Now, it’s true that the majority of philosophers of religion in particular are theists, but I’d wager that a majority of the theists in question were theists *before* they took up philosophy. If I’m right, then studying philosophy of religion isn’t directly responsible for revealing the existence of God to such people.18

So here’s the other explanation I have in mind: perhaps God does not speak to us from the burning bush because he wants us to set bushes alight. Perhaps God leaves room for uncertainty because he also wants to leave room for leaps of faith.

At first blush, I don’t like this very much. I’m the sort of guy who’s keen on rationality, who would very much prefer not to believe anything that he doesn’t have good reason to believe. But I can still see the appeal of this idea. Suppose I grow up in an orphanage, having never properly known my parents. However, I have some sign of their existence, perhaps some gift or heirloom that they’ve left behind for me. As I grow up, I hope that my parents are somewhere out there, waiting for me, but I can’t really be sure that this is the case. Then, as I come of age, I resolve that I shall search for them. I don’t *know* that they exist—after eighteen years, they could be dead. I don’t know that they’re what I imagine them to be—they may not want to see me, may have maliciously abandoned me. And yet it seems that there is something admirable in having faith, in choosing to live as if my parents are alive and waiting for me.

Now, there is at least one important way in which the search for my missing parents and the search for God are disanalogous. Namely, I know for sure that my parents *once* existed, I know that if I search hard enough, I will probably find at least a pair of graves, whereas in the search for God I have no certainty of finding anything. Nonetheless, the model may be applicable.

But now I ask this: though it may be in some way commendable, even virtuous, that I search for my lost parents, that I have faith in their goodness, would it also be commendable that I fiercely insist on their existence and goodness even in the most detached discussions? I don’t think so. And so, insofar as my little story applies to the existence of God, it strikes me that it’s very much appropriate for those so inclined to live as if God exists, to go to church, to pray, to visit holy sites. We might call this an attitude of “practical faith,” and it may even be a commendable attitude to have, insofar as it is held humbly. But sometimes a person may enter some highly theoretical discussion, in which it is important to take only what the evidence gives you.19 And in this context, I claim, practical faith is not an appropriate attitude. Because if God exists, he has structured the world in such a way that it is reasonable to doubt his existence, and surely the most theoretical discussions are the contexts in which doubt is most appropriate. Accordingly, if God exists, the attitude that most fully respects God and his plan for Creation will be a combination of practical faith and theoretical doubt. If God exists, he has given us uncertainty, and it is our responsibility to take what he gives us.

Exegi monumentum aere perennius

regalique situ pyramidum altius

quod non imber edax, non Aquilo inpotens

possit diruere aut innumerabilis

series annorum et fuga temporum.

Non omnis moriar multaque pars mei

vitabit Libitinam; usque ego postera

crescam laude recens, dum Capitolium

scandet cum tacita virgine Pontifex.

Dicar, qua violens obstrepit Aufidus

et qua pauper aquae Daunus agrestium

regnavit populorum, ex humili potens

princeps Aeolium carmen ad Italos

deduxisse modos. Sume superbiam

quaesitam meritis et mihi Delphica

lauro cinge volens, Melpomene, comam.

The poet, today known as Horace, was a true literary genius, and I give you the Latin first so that you can see his genius for yourself. Whatever flaw you may spy in my translation, it will be my flaw and not his. I offer you that translation now:

I have built a monument—more everlasting than bronze

and higher than the royal site3 of the pyramids—

which neither the devouring rain nor the impotent North Wind

is able to destroy, nor the sequence of countless years and the flight of time.

I will not completely die, and a great part of me

will avoid Death; I will afterward rise

ever-fresh with praise,4 so long as the Pontifex

ascends the Capitoline with the silent Vestal Virgin.5

I will be spoken where violent Aufidus roars

and where Daunus,6 poor of water,

ruled the rustic people; I, powerful from humble origins,

first7 to mold Aeolian8 song to Italian ways. Accept the pride

demanded by your services, Melpomene,9

and willingly encircle10 my hair with Delphic laurel.11

In the final lines, Horace asks the Muse to crown his hair with Delphic laurel. A mark of triumph, yes, but also suitable for the occasion in a more specific way: Delphi is a place of prophecy.

If you take Latin in high school, it’s highly likely that you’ll read at least a bit of Horace. He is outmatched in ubiquity only by Virgil, Julius Caesar, and perhaps Cicero. And if you read some famous Enlightenment thinker, you should not be surprised to find some words borrowed from the great ancient. Indeed, his greatness has long been known, and Horace himself knew that he was great. He is not shy about saying so. His audacity walks a razor’s edge—it would not be nearly so charming were he a lesser talent. But this, of course, is true of many great artists—a certain amount of audacity is to be expected of such people. III:30 is special. III:30 goes much further.

If you will, imagine reading this poem as I first read this poem: You’re in a classroom with your peers, working slowly, taking it one line at a time. The first six are easy. Then you reach this bit: “usque ego postera crescam laude recens.” You know these words and how they relate to each other, so it’s clear enough what Horace is saying. Nonetheless, you can’t quite wrap your head around what he *means.* He’s going to “arise continually fresh”? What? And what kind of ablative is “laude” anyhow?

So you set it aside for a moment. You do the next bit. “While the high priest climbs the Capitoline with the silent Vestal Virgin.” Some kind of religious ritual, you figure. And then you understand. For as long as the Pontifex and the Virgin climb the hill—that is, for as long as Rome stands—Horace is going to be fresh. And “laude,” you’re now sure, is an ablative of means. “Fresh because of praise.”

So you put it together properly:

“I will not completely die, and a great part of me will avoid Death; I will afterward arise continually fresh because of praise, while the high priest climbs the Capitoline with the silent Vestal Virgin.”12

You are eighteen years old. You are soon to graduate high school, and you sit in class translating Latin. You are reading the words of a man who says he is going to live forever because his poetry will never lose its luster, a man who says he will *always* be great and he will *always* be loved, and he was *right.* I have no words to properly convey the magnitude of that, the propulsive force, the world-historic gravity. This poem is a true prophecy.

Two thousand years ago, a great poet sat down and wrote that he would live forever because his poems would be praised until the fall of Rome. He was wrong only in that Rome fell first.