Bryan Caplan  

The Basics: A Socratic Dialogue

The Hansonian Moralist... Have an opinion about string t...
Glaucon: Guess what?  I'm writing a book about moral philosophy!

Socrates: Cool.  What do you say?

Glaucon: I'm a pragmatist. We should do whatever maximizes human welfare.

Socrates: So the main point of the book is to advocate open borders?

Glaucon: No, no, no, Socrates.  I said I'm a pragmatist.

Socrates: You don't think allowing low-skilled immigrants to get jobs in the First World would sharply raise human welfare, on balance?

Glaucon: I suppose it would.  But voters wouldn't accept it.

Socrates: But if your thesis is correct, they are morally obliged to do so, no?

Glaucon: Well, who's to say?  You think you've got a hotline to moral certainty?

Socrates: I'm struggling to understand your thesis.  You begin by asserting we should maximize human welfare.  But when maximizing human welfare is unpopular, you back off.  Then when I point out the inconsistency, you accuse me of unwarranted confidence in my moral judgments.

Glaucon: I bet you're one of those deontologists, who thinks we should follow moral rules regardless of the consequences.

Socrates: [scowls]

Glaucon: Look, we just have to start with popular moral intuitions, and weigh them against each other.  You may not care what other people think, but I do.

Socrates: Glaucon, I've listened to you long enough.  Frankly, you need to stop writing, learn the basics of moral philosophy, and start over.

Glaucon: [stunned] There's no need to get personal.

Socrates: It is nothing personal.  But you're conflating an array of moral distinctions.  Almost any philosophy professor who teaches ethics would be appalled.

Glaucon: [laughs] Why don't you tell me what you really think, Socrates?

Socrates: [laughs] Happily.  Let's start with the first distinction: moral realism (also known as moral objectivism) versus moral relativism (also known as moral anti-realism).  Moral realism holds that at least some moral claims are literally true.  Moral relativism holds that all moral claims are literally false.

Glaucon: Then no one's a moral relativist, because everyone makes moral claims.

Socrates: Not exactly.  The smart moral relativist will reply, "Relativism is a meta-ethical claim, not an ethical claim.  Sure, I concede that murder is wrong.  But that claim is still literally false."

Glaucon: I don't know anyone who says such things.

Socrates: Look in a mirror.  When I asked you if you favored open borders, you objected that voters wouldn't accept it.

Glaucon: So?

Socrates: Well, if we really should do whatever maximizes human welfare, and open borders genuinely maximize human welfare, then what moral difference does it make if voters disagree?

Glaucon: None?

Socrates: Right.  None.  Unless, perhaps, moral claims are never literally true in the first place.  

Glaucon: Not sure if I follow.

Socrates: Well, suppose I say, "There are no unicorns, but here's what I think about unicorns." 

Glaucon: Kind of silly.

Socrates: Indeed.  But once I say that, you shouldn't be surprised if my claims about unicorns are internally inconsistent, even senseless.  After all, my whole position is internally inconsistent - or, to be more charitable, "poetic."  So is your book on moral philosophy actually a thinly-veiled work of poetry?

Glaucon: God no!  

Socrates: Then you're a moral realist.  That means you need to strive for logical consistency, even if most people don't.

Glaucon: Example?

Socrates: Once you say (a) we should maximize human welfare, and (b) open borders accomplishes this, you are logically bound to either revise your moral principle or embrace open borders.  What people accept is irrelevant.

Glaucon: Oh, now I see where I went wrong.  I don't actually think we should maximize human welfare to the exclusion of all other goods.

Socrates: But you think we're morally obliged to maximize some weighted average of goods?

Glaucon: Right.  That makes me...?
Socrates: ... a consequentialist, but not a utilitarian.  A utilitarian says we should maximize human welfare.  A consequentialist says we should maximize some richer function.

Glaucon: Isn't that just a tautology?

Socrates: Probably not.  Suppose two actions' consequences have equally valuable overall moral consequences.  But one requires you to tell a lie.  Should you tell that lie?

Glaucon: Uh... no.

Socrates: Then you're not a consequentialist.  You're a deontologist.

Glaucon: Aren't deontologists those crazy people who think consequences are morally irrelevant?  We shouldn't tell a lie even if it will save the world?

Socrates: Nope.  Consequentialism is the polar position that NOTHING BUT consequences are morally relevant.  Deontology is the rest of logical space.  This is just standard usage among philosophers.  Most deontologists hold that consequences are morally relevant, but aren't the whole story. 

Glaucon:  So where does my pragmatism fit in?

Socrates: Unclear.  Some people treat "pragmatism" as a synonym for utilitarianism.  Others treat "pragmatism" as deference to popular moral views.  Sometimes "pragmatism" is a sugar-coated word for moral relativism. 

Still, I suppose you could equate "pragmatism" with the moderate view that consequences and other stuff morally matters.  But when a word is this muddled, it's probably best to avoid it.

Glaucon: [sigh] So you've got absolute moral truth?

Socrates: I wish!  Sadly, noting that the word "pragmatism" is confusing fails to make me morally infallible.

Glaucon: I've listened to you, but I feel less able to write my book than before our conversation started.

Socrates: That's because you see the state of your knowledge more clearly.  Now you know that you're not ready to write your book.

Glaucon: You're a piece of work, Socrates.  What would your great treatise on moral philosophy say?

Socrates: It's complicated.  But let me tell you what my book on moral philosophy would not say.  It wouldn't equate moral realism, moral certainty, and moral indifference to consequences.  It wouldn't equate utilitarianism, social acceptability, "pragmatism," and open-mindedness.  And before proposing any general moral theory, I would test it against a wide range of hypotheticals. 

Glaucon: Doesn't sound very pragmatic.

Socrates: If "pragmatic" means "appealing to intellectually lazy readers," you're right.  You're a talented writer.  If you just kick a bunch of half-baked claims toward the goalie of public opinion, you'll score.  But Glaucon, you're better than that. 

If you were writing a book about pool, you'd start by getting a clear grip on the basic concepts.  You wouldn't equivocate between pocket billiards and swimming pools.  You wouldn't assume that people who don't play 8-ball don't play pool at all.  You'd be open to the possibility that shooting hard is a bad strategy, even though almost every novice does it.  Moral philosophy isn't easy, dear Glaucon.  But if you don't take the time to get a clear grip on its basic concepts, it's impossible.

Comments and Sharing

COMMENTS (28 to date)
pliny writes:

[Comment removed pending confirmation of email address and for policy violations. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

john hare writes:

Most of the open borders posts that Bryan does move me the opposite way. I.e. Bryan moves my opinion toward con most of the time. Open Borders>One World Government>Nowhere to Run. This one was entertaining in an ivory tower humor sort of way.

I am pro toward most immigration, just not in an anarchistic sort of 'free for all' sort of way.

Andrew_FL writes:

@john hare-Probably because this post takes for granted the claim (defended elsewhere, to be fair) that open borders would be a move toward maximizing welfare. Bryan could have replaced it with anything other than his favorite issue that both allegedly moves towards maximizing welfare and would be resisted by a "pragmatist" as politically unfeasible, and the philosophical discussion would still be the same.

He probably should have, too, because the philosophical discussion is more widely applicable and useful on its own.

Andrew_FL writes:

@Andrew_FL-Put another way, if Glaucon answers "No, I don't think it would." instead, the actually interesting part of the discussion never happens. The interesting part comes because Glaucon agrees with the premise.

Ed Hanson writes:


Feel free to state that open borders would maximize human welfare, but I believe that a better expression for your case would be 'improve' human welfare, maximize is an high bar, for there is certainly other known and unknown out there which may maximize.

But that is not my criticism at this time, which is, like the fanatic, you give only one part of the argument. As a trained economist, you intimately are aware of trade offs. Open borders come with trade offs. Rather than list all the trade offs, which are unknown, I will just give examples.

Ukraine. The country has a large Russian speaking populous due to open borders in the past, who retain a larger allegiance to Russia than Ukraine. The economic gains, the freedoms by leaving behind the constricting state of Russia, have been completely undone by civil war. This is true for the native Ukrainians and the immigrant. The deaths, the destruction of property and wealth, the ruin of the economy, and the distrust of neighbor by neighbor far into the future, have resulted in no gain in welfare, in fact a reduction in welfare.

Ukraine is not the lone example. They are too numerous to recite. But here are some examples of open borders where human welfare can be questioned, from the past that still effect today. Tibet by the Chinese, Korea by the Japanese, Africa by the Europeans and Muslim countries. Need I emphasize that the list is huge.

Like all economic events and processes, open borders have trade offs.


Dan S writes:

For the life of me, I'll never understand why you find that moral realism essay so convincing.

The universe just is. It doesn't have opinions about what we do. Moral realists are projecting their mental states onto the cosmos.

Andy Hallman writes:

That's interesting Dan, because I find Huemer's treatment of moral realism thoroughly convincing.

One typical argument against moral realism is that moral facts seem unlike other facts we claim to know, and that this queerness somehow makes them suspect. Moral properties are things we perceive with our intellect, the same way we perceive light with our eyes and sound with our ears. It's true that moral facts cannot be perceived with the eyes, but neither can the eyes perceive sound, and yet that does not make sound any less real, nor does it matter that sound is nothing like light.

Matt Skene writes:

Moral realism isn't the view that the universe has opinions about what we do. In fact, if it was possible for the universe to have opinions, and moral truth depended on those opinions, then what you would have is an odd sort of subjectivism comparable to divine command theory, and not a form of moral realism at all. This is exactly the sort of thing Caplan is complaining about.

StrangeLoop writes:

What leads us to believe that the dominant strand of pragmatism inflexibly promotes "whatever maximizes human welfare"?

Anon. writes:

@Andy Hallman

Someone unable to perceive sound (say, someone deaf) can still detect it though. When you say "perceive sound" what you really mean is "transform mechanical waves propagated through the air into qualia". The waves are observable in a way that moral facts are not.

People in general tend to agree about observations regarding sound, and we can construct implements to verify these observations and experiments to test hypotheses regarding them.

In contrast, there is extremely widespread disagreement regarding moral facts. How could they be the result of perceiving something real if everyone's perceptions are so radically different? Why have we been unable to construct tools to aid or confirm our perceptions in this realm?

Dan S writes:

Andy and Matt,

Thanks for the replies, I will turn the snark dial down a little bit now :)

Actually Matt, yes, I do think moral realism is in practice a sort of confused divine command theory, even if its practitioners would never describe it as such.

Let me approach my point from a different angle. One of the revolutionary conclusions of the modern scientific era is that everything in the universe, from "the heavens" to the earth to our own minds and bodies, are not part of separate spheres with their own laws, but are in fact parts of one consolidated whole, the behavior of which is reducible to a single body of physical law. To put it more crudely, it turns out that the universe is a bunch of atoms bumping into each other, perhaps with some cosmic dice thrown in (depending on how you want to interpret quantum mechanics).

What I've just described is a universe that has no room for unconditional "oughts," hence the is-ought gap.

And yet, moral realism is the proposition that there are in fact unconditional oughts out there, and that you "should" do A, regardless of whether you actually want to do A or whether A is in conflict with some other goal that you have. But these "shoulds" have no teeth, and they are incompatible with an otherwise physicalist universe.

That's why I describe moral realism as a confused divine command theory. It's the insistence that there are other rules on top of the ordinary physical ones that govern the collisions of atoms. But there are no such extra rules, as far as we can tell.

Reginald writes:

I could use a similar didactic approach but in the interest of time, I wont. I'll just say that when those low skilled foreigners come here and start voting against your interests, you will regret your deontological approach to this issue. Remember we live in the most free of unfree places. Those who want to leave centrist states see us as a richer trough to feed from. It isn't that I care what voters HERE and NOW think but rather what voters in the future will think about every other good we offer. And before you say we are a nation built on the wiles of eager immigrants, recall that we had no social welfare state when that was true. Immigration greatly changed the politics of this nation. Certain identity groups are almost monolithic in their preference for less liberty. They aren't fleeing their statist environment as much as they are bringing it with them.

Michael Stack writes:

I suppose Socrates is standing in for Tyler Cowen? If so I think this is a bit unfair, it sounds to me like somebody who doesn't quite understand the points Tyler is trying to make.

If this has nothing to do with Tyler then please ignore me.

Justin writes:

From the perspective of a purely physical universe, it goes beyond "shoulds" not having any teeth - if everything is reducible to purely aimless interactions between atoms bumping into each other, then all human actions are determined by physical forces leaving no room for agency and therefore morality doesn't make sense as a concept. In fact, no action would even be "about" or "directed towards" any purpose at all, regardless of moral content. If you reach for the TV remote, it's not because you "want" to change the channel, it's entirely the result of unthinking physical forces. Ultimately, visible light (or perhaps sound) from the TV causes you to reach for the remote, but that "you" "decided" to do so "for" a particular reason would necessarily be illusory.

StrangeLoop writes:

Dan S wrote, "What I've just described is a universe that has no room for unconditional 'oughts,' hence the is-ought gap. . . . That's why I describe moral realism as a confused divine command theory. It's the insistence that there are other rules on top of the ordinary physical ones that govern the collisions of atoms. But there are no such extra rules, as far as we can tell."

In fact, you are fully comfortable with many non-spatiotemporal "oughts," like the rules of logic. I would, likewise, argue that mathematical sets exist (in the ontological sense), but are not reducible to atoms.

Philo writes:

The main point of a book on moral philosophy cannot be to advocate open borders. Moral philosphy distinguishes right from wrong *actions*. *Open borders* is a state of affairs, not an action. *Advocating open borders* is an action; whether it is obligatory, impermissible or neither will depend on the circumstances.

Andy Hallman writes:


I suggest moral facts can be perceived by considering normative propositions of the form "murder is wrong," for example.

I think there is in fact fairly widespread agreement about morality. For instance, nearly everyone who has heard of the Holocaust thinks it was wrong. Furthermore, as Steven Pinker has pointed out, human societies have become more liberal over time, and I think this is evidence in favor of moral realism that is hard to explain purely on evolutionary grounds.

ted writes:

I think that "maximising human welfare" is at least a very subjective term, if not downright nonsensical.

It certainly doesn't follow at all that open borders give any sort of such maximisation unless you're also willing to believe that killing the top 1% richest people and distributing their wealth to the remaining 99% is a valid strategy for maximising human welfare.

Sure, letting 100 million people from Pakistan and Bangladesh into Switzerland might make their lives better, at the expense of the destruction of the existing 8 million Swiss. Prof. Caplan's Socrates might describe it as maximisation of human welfare.

I am not convinced this is a moral position. I am not convinced at all that some people should accept the injection of poverty, crime and unsavoury habits into their lives just to benefit others, who are otherwise totally free to maximise their human welfare on their own, in their own countries.

Unless Prof. Caplan's Socrates assumes these people incapable of doing so, because of their culture/IQ/etc, in which case the open border policy makes even less sense, as rather seems like a negative sum game.

Matt Skene writes:

Well, a number of moral realists nowadays are naturalists, who would disagree with the claim that there is no room in a purely physical universe for morality. I'm not one myself, but the move from physicalism to moral anti-realism is hardly uncontroversial.

As for physicalism, I really can't think of any good reason to accept it. It appears to be clearly at odds with any number of things we know, including the existence of moral facts. It's not part of science, and it's not necessary for it. It appears to be a fairly speculative and unsupported ontology to me.

As for moral realism, it hasn't been a confused sort of divine command theory since Plato believed in it. Platonism still strikes me as by far the most plausible view. Moral facts appear to describe certain necessary relationships between non-moral states of affairs and the moral properties that supervene on them. If you have an instance of one person killing another and there are no other morally relevant factors involved, then wrongness has to be there as well. This isn't a matter of how we feel or what anyone, including God, has said, this is just a fact about how certain types of properties must be related to one another. You can't have an instance of willful killing with no redeeming characteristics and not have wrongness any more than you can have a triangle and not have a shape. Certain properties can only be instantiated together, and supervenience captures one type of necessary co-instantiation fact.

We know about moral facts the way we know about other necessary relationships between properties: by thinking about them. Exactly what the relationship between the mind and the underlying facts are that allows us to think about evaluative properties is an interesting question. However, it's not really a different question than the one we get in mathematics, logic, or with other sorts of general metaphysical intuitions in a variety of areas in philosophy and elsewhere. There is nothing distinctively problematic about evaluative facts in terms of explaining how we know about them.

As for the disagreement issue that has come up, if you actually look at moral disagreement, it seems to confirm the existence of a fairly sophisticated and impressive grasp of these dependency relationships. Moral discussion almost always takes the form of describing certain general relationships between states of affairs and supervening moral properties and then considering counter-examples that require refinement of our application of this general knowledge in a variety of circumstances. We instantly grasp both the accuracy of then general claim in most cases and the inapplicability of it to certain situations, which is what allows us to come up with counter-examples to exceptionless moral claims. We also can consider and even devise novel situations and generate widely embraced conclusions about the ethics of those situations. All of this suggests a broad starting point of fairly secure and impressive moral knowledge. Obviously in many circumstances situations become complex, and disagreements arise, but they aren't anything we shouldn't expect from any discipline of this sort. We don't just disagree, we debate and offer evidence, and if we didn't have the sort of knowledge I've been describing to start with, we couldn't do this as effectively as we have been able to. The idea that apparently widespread and easily accessed knowledge that has made possible intelligible, complex, and sophisticated conversation between parties throughout the world and throughout history is all grounded in some sort of shared and magically consistent illusion is a bizarre and purely speculative position, and I can't see much to say on its behalf.

Sieben writes:

@Dan S

The universe just is. It doesn't have opinions about what we do. Moral realists are projecting their mental states onto the cosmos.

are in fact parts of one consolidated whole, the behavior of which is reducible to a single body of physical law.

Why should I think that the "universe" is reality? It is something I perceive with my senses. I know nothing about the origin or nature of my senses. An obvious retort is that evolution provides an explanation, but this returns us to the general problem of qualifying empirics with empirics. But trusting your eyesight is simply intuitionism, and intuitionism can also be used to derive ethical realism.

To me, it's not obvious that we don't live in the matrix. I don't have a good way to estimate the probability that there's something "funny" going on in this life, but that doesn't mean I default to the "safe and responsible" view that science = real. Simply, no one has a good way to estimate the correctness of their assumptions.

One question I often ask is: "what is the minimum it would be willing to take to get you to think something weird might be going on with the universe?", like it having a divine origin or being like the matrix. I have never gotten a good answer. Keyword here is: "minimum".

God is talking to you through the acronyms in license plates.

Dan S writes:


If I may summarize your argument, I am saying that a physicalist description of the universe is causally complete and doesn't require any extra secret sauce of moral properties to explain it, and you are saying that actually there are moral properties that supervene on top of physical ones, that certain arrangements of matter can be inherently "good" or "bad" in addition to the mere observation that they just "are."

So what I would say do you is, what is your basis for making that claim? On what grounds can you claim that one arrangement of matter is "better" than another? Are there moral particles that pop into existence when a child eats an ice cream cone? "You can't have an instance of willful killing with no redeeming characteristics and not have wrongness any more than you can have a triangle and not have a shape." Why? You are merely asserting that without evidence.

Here is another way to put it. According to the moral realism view, a perfectly amoral person (one who acts solely according to his desires without any regard for "right" and "wrong") must necessarily be irrational at some level. But I don't think he is. You can assert that certain arrangements of matter are desirable, but if he chooses not to find that opinion compelling, there are no good-ons and bad-ons to measure to persuade him. Compare that with the person who fails to run away from the tiger chasing after him.


I think that is a very interesting discussion to have, but you're being inconsistent with your demands for proof. I am skeptical about the existence of moral facts. I am not obligated to also prove that the universe is real and that we don't all secretly live in the matrix in order to make my point.

Matt Skene writes:

The evidence I offered in my previous post was that the existence of moral knowledge is the best explanation for the abilities people demonstrate in discussing moral issues. What was wrong with that evidence?

As for statements such as "willful killing with no redeeming features is wrong" it's not clear to me why I need to offer evidence for this. It seems obvious on its face, and pretty much everyone has always agreed with it. It seems that in a situation like this the person denying the claim has the burden of proof, not the person presenting it.

If you think these claims are incompatible with physicalism, then physicalism is in an even worse position. Why should I accept a general position about metaphysics which seems far from obvious and which you've given no evidence in support of rather than seemingly trivial facts that people have known for thousands of years? You're the one asserting the speculative philosophical position and saying its incompatibility with common sense should get us to reject common sense. Given this, the burden is on you, not me, and it's quite a heavy one at that.

Dan S writes:


I suspect we're not going to agree. Heh, internet philosophy debates have a tendency to be unconvincing for most people involved. Until next time, I'm sure.


Christopher Chang writes:
I am not convinced this is a moral position. I am not convinced at all that some people should accept the injection of poverty, crime and unsavoury habits into their lives just to benefit others, who are otherwise totally free to maximise their human welfare on their own, in their own countries.

This is an unnecessarily weak statement. Caplan's "what moral difference does it make if voters disagree" position when the US's place premium is inextricably intertwined with institutions requiring a high level of voter consent, and there already are countries like Sweden that are seriously trying and largely failing to capture the welfare gains Caplan takes for granted is unambiguously immoral.

With that said, I still condemn use of violence in self-defense at this stage. The current situation mirrors the constitutional crisis of 1937 (where the FDR and Congress faithfully represented the people but the Supreme Court threatened to singlehandedly obstruct everything)[1]; given that precedent, there is no good reason to deny the judicial and legislative branches their chance to check the executive and Fourth Estate. Only when US government as a whole is acting against the interests of US citizens at large scale, and there no longer are any practical nonviolent ways to correct that, does violence become advisable. I think we are very unlikely to reach that point.

[1] I'm really not a fan of the tactic FDR employed, but he did nonviolently resolve the problem of governmental failure to represent citizen interests, and ultimately I have to give him a lot of credit for that. Those who argue that the median citizen is dumb/lazy and unworthy of representation can already peacefully demonstrate the superiority of their position in other, more autocratic countries; if they instead work to undermine the US, they're admitting that they know their ideas cannot build a society better than the US--instead they have to bring the US down to their level.

Sieben writes:

@Dan S,

I am skeptical about the existence of moral facts. I am not obligated to also prove that the universe is real and that we don't all secretly live in the matrix in order to make my point.

So do we agree? I'm as certain about the existence of moral facts as I am about the existence of the world I see.

Liberty writes:

Does this dialogue have an actual argument or is it just making fun of people who throw around words like 'pragmatist'?...

Marty writes:

What if I don't care about human welfare writ large, but only about the human welfare of those already living here by legal right, and as a matter of fact that opening the border would decrease their welfare?

What if I further am more interested in the welfare of those nearer the bottom of the scale, who are more likely to see their wages depressed by an influx of low-skill competition?

I mean, this is an interesting piece if it's about moral philosophy, but is almost totally beside the point of the economics of the actual immigration debate in this country at this time.

emerich writes:

Thanks Bryan, I enjoyed this post quite apart from the specific question at issue--open borders. It's rare to get clear discussion on philosophical first principles like this. I think it was from one of your earlier posts that I learned the word deontology, though I'm going to have to re-read the definition a few times before I remember what it means.

Comments for this entry have been closed
Return to top