The Brennan Thought Experiment: Further Explorations
Ross Levatter argues that a thought experiment Jason Brennan uses to test our intuitions about the morality of markets has too many problems to genuinely be helpful.
Back in 2011, Georgetown political theorist Jason Brennan penned a quick piece at the blog Bleeding Heart Libertarians (BHL) consisting of the following thought experiment:
Imagine that a bunch of economists provide compelling evidence that life in a strictly libertarian polity would be a disaster. Imagine that they show conclusively that if people everywhere were to live in a Nozickian minimal state or a Rothbardian anarcho-capitalist civil society, with everyone strictly observing property right rules, that 10% of people would starve, 80% would be near subsistence, and only 10% would prosper…
Here’s Brennan’s Question regarding the thought experiment: “What would you advocate and why?”
Here’s Brennan’s Goal in developing the thought experiment: Clarifying “how your moral commitments interact with your empirical beliefs about how economies actually work.”
Frankly, I found this thought experiment not particularly well developed, and recently wrote an essay explaining why. In summary, I argued that while thought experiments may be, and often are, “unrealistic,” while still being quite helpful, problems develop when they are so strikingly divorced from reality that a) we cannot trust our intuitions as to how to respond, and b) cannot easily translate our response to the thought experiment into real-world applications.
Not surprisingly, Brennan disagreed with my analysis, and jotted off an amazingly quick response at BHL. There he managed to bring forward some interesting claims that deserve response and analysis. Thus this essay.
Perhaps the most revealing passage in Brennan’s response is his saying: “My most charitable guess is that Levatter is just widely misinformed about the state of economics.” We learn from this that Brennan believes people’s reaction to his thought experiment has something to do with their knowledge of “the state of economics.” It follows Brennan must think his thought experiment has something to do with economics. But, as I show below, this is problematic at best.
Reasons Why It Is Not Rational to View the Brennan Thought Experiment as About Economics
1. People with Widely Different Views on Economics May Give the Same Answer to Brennan’s Thought Experiment1
Consider two students to whom Brennan might pose his thought experiment. One is named Milton. He’s a strong free market advocate. The other is Karl. He’s a committed communist. Both Milton and Karl are, however, very sensitive to the consequences of the policies they advocate. Both Milton and Karl give the same answer to the thought experiment, the answer I gave in my earlier essay: “In the … world you posit, no sensible person would any longer advocate a Nozickian minimal state or a Rothbardian anarcho-capitalist civil society.” That is, Milton and Karl, who differ markedly in their economic beliefs, give the same answer to the thought experiment because they share the same sensitivity to hypothetical consequences. Their differing economic beliefs only affect their attitude about how realistic the thought experiment’s scenario is. Only Milton would, as I did, call the thought experiment consequences “unimaginable.”
2. People with Identical Views on Economics May Give Categorically Different Answers to Brennan’s Thought Experiment
Next consider two other students to whom Brennan gives his thought experiment. One is named Ayn. She cares only about her own self-interest and has no concern at all about how her actions affect anyone other than herself. The other is named Adam, and he is highly sensitive to the sufferings of others. Both, however, are strong believers in the free market. When THEY answer the thought experiment, Ayn, who is confident she will be among the prosperous 10%, says she has no problem with continuing to favor a Nozickian minimal state.2 Meanwhile, Adam indicates that in THOSE circumstances, he, like Milton, would abandon his favored free market impulses.3
So in summary, varying economic beliefs does not necessarily lead to different answers to the thought experiment. And holding economic beliefs constant does not guarantee that answers to the thought experiment will always be the same. That is, knowledge of a person’s economic beliefs is neither a necessary nor sufficient condition for determining his answer to the Brennan thought experiment. Therefore, it is not reasonable for Brennan to believe his thought experiment has much to do with “the state of economics.”
3. What Happens When We Change ONE Word?
There are other reasons to be skeptical of the claim Brennan’s thought experiment has anything to do with economics. Consider this variant thought experiment:
Imagine that a bunch of moral philosophers provide compelling evidence that life in a strictly libertarian polity would be a disaster. Imagine that they show conclusively that if people everywhere were to live in a Nozickian minimal state or a Rothbardian anarcho-capitalist civil society, with everyone strictly observing property right rules, that 10% of people would starve, 80% would be near subsistence, and only 10% would prosper."
What has changed? Only that the term “economists” was replaced by the term “moral philosophers.” Notice I made NO changes to the economic arguments or rationales in the thought experiment, since there ARE NO economic arguments or rationales provided in Jason’s thought experiment, merely an assertion by “a bunch of economists,” just as, in this variant, there is an assertion by a bunch of moral philosophers. Moral philosophy is, like economics, at least in part an empirical subject. It is not like logic, after all. So, as Brennan might put it, “[t]here’s some possibility, and perhaps even some non-zero probability” that a bunch of moral philosophers might provide compelling evidence for this claim.
So the question of interest is: Is this a completely different thought experiment, or a slight variation of the original thought experiment? People can make up there own minds, but it seems obvious to me that this is but a minor variant of the original. Granted, per arguendo, it is easy for Brennan to imagine markets stink, but is it easy to imagine lots of people giving one answer to his original query, and yet a significantly different answer to my variant? Is it easy to imagine lots of libertarians saying, “Oh, well if a bunch of ECONOMISTS asserted 90% of the world would be immiserated by a minimal state, I would no longer favor it, but if a bunch of MORAL PHILOSOPHERS asserted 90% of the world would be immiserated by a minimal state, I’d still strongly favor it.”?
Now, if you agree that my variant thought experiment is really not significantly different from Brennan’s original, that is of course further evidence his is not a thought experiment about economics. It is a thought experiment about how one’s sensitivity to consequences–adverse, unexpected, highly unlikely, no-reason-provided-to-believe-accurate, stipulated consequences–impact on one’s advocacy of an otherwise-favored society.
4. The Borkean Problem with Brennan’s Hypothetical
In reason #3, we explored the implications of replacing one word, “economists,” with another. Now we’ll explore what happens if we just blank out the word altogether. Many readers no doubt recall the problem the late jurist Robert Bork had with the 9th amendment. As he said in his Supreme Court confirmation hearing: “if you had an amendment that says ‘Congress shall make no’ and then there is an ink blot and you cannot read the rest of it and that is the only copy you have, I do not think the court can make up what might be under the ink blot if you cannot read it.”
Let’s apply the Borkean test to the Brennan hypothesis. We get:
Imagine that a bunch of INKBLOT provide compelling evidence that life in a strictly libertarian polity would be a disaster. Imagine that they show conclusively that if people everywhere were to live in a Nozickian minimal state or a Rothbardian anarcho-capitalist civil society, with everyone strictly observing property right rules, that 10% of people would starve, 80% would be near subsistence, and only 10% would prosper…
Like the 9th amendment for Bork, Brennan’s now modified thought experiment is epistemically incomplete. It involves dealing with a bunch of….something, for sure. But we’re not clear as to what. Can we figure out what’s under the inkblot? Let’s stipulate, correctly as it happens, ONLY ONE WORD IS COVERED. Does the context of the remaining structure of the thought experiment help? We see the “bunch of…” somethings provide compelling evidence. So it’s reasonable to assume they are a “bunch of…” experts, perhaps, of some type. But what type? The thought experiment deals with a “strictly libertarian polity,” so perhaps the experts are political scientists. That would go along with the line about strict property rights. But Nozick is mentioned, and property rights come up in philosophy as well. Maybe they’re a “bunch of philosophers”. Then again, maybe the problem with the libertarian polity is, as some argue, that it cannot defend itself against attack from abroad. Maybe the horrible results Brennan describes are due to defeat in war. So maybe the experts are military in nature. It’s hard to say. Rothbard is mentioned, and he’s formally an economist, but the reference is to his writings on anarcho-capitalism, more political philosophy than economics. But perhaps we can’t rule out “economists” being under the inkblot…
Why do I use this inkblot test? It’s just one more piece of evidence that it is foolish to view Brennan’s test as a pellucid thought experiment about economics. By removing only one word, it’s very challenging to even determine one is dealing with economic concepts at all. By removing only one word, it’s very challenging to know one is dealing with the opinion of a bunch of experts in ECONOMICS.
Furthermore, no information is provided as to WHY the experts think as they do. The thought experiment assumes wide deference to the authority of “a bunch of economists.” Wide deference is what one gives to experts when one lacks understanding of a field. The more one knows about a field, the less likely one gives deference to conclusions one believes far-fetched. So rejection of the Brennan thought experiment is not reasonably viewed as eo ipso evidence of a severe lack of understanding of the state of economics.4
5. Following Brennan’s Own Suggestion…
At the beginning of Brennan’s 2011 essay, as I noted above, he indicates the experiment “is directed at libertarians who 1) advocate a purely negative rights-based theory of justice, but 2) who also believe that under ideal libertarian institutions, markets would work really well at making most people better off.” But at the end of that essay, Jason casually adds “By the way, if you aren’t a libertarian, try changing this thought experiment around to deal with your views.” So Brennan himself recognizes that the structure of his thought experiment works independently of people’s actual beliefs about economics and politics.
Reasons Why The Thought Experiment Doesn’t Test The Connection Between One’s Moral Commitments and One’s Empirical Views of How Economies Work
6. I’d Like Some Desert With That…
Let’s re-consider once more Brennan’s thought experiment:
Imagine that a bunch of economists provide compelling evidence that life in a strictly libertarian polity would be a disaster. Imagine that they show conclusively that if people everywhere were to live in a Nozickian minimal state or a Rothbardian anarcho-capitalist civil society, with everyone strictly observing property right rules, that 10% of people would starve, 80% would be near subsistence, and only 10% would prosper.
Let’s ask the “what would you advocate and why” question now of a bright student named Robert, a real wünderkind. He is exquisitely sensitive to consequences. Nonetheless, he says he can’t give an answer to the thought experiment. Queried as to his inability, he says, “Well, you provided only a time-slice of the data. Issues of justice may require historical information. Maybe one year ago, 15% of the people were starving, 84% were near subsistence, and only 1% were prospering. That is, maybe the libertarian polity is moving things in a desired direction.”
“Furthermore,” Robert notices, “we don’t know WHY 10% are starving. We DO know it isn’t because their rights are being violated, since the thought experiment posits a libertarian polity. Maybe 10% of the populace composes a large religious cult that promotes starvation as religious penance. Perhaps many of the prosperous 10% volunteer their time and money to urge those starving to re-consider their position, and bring them free and fine foods with which to end their devastating starvation fasts. What else can the prosperous 10% do? The starving 10% are autonomous adults, just like the prosperous 10%, equal in their rights, who can’t be forcibly religiously re-programmed, after all. Surely Professor Brennan wouldn’t want to force feed 10% of the population like they were Guantanamo prisoners?!”
“And maybe,” Robert goes on, “most of the 80% who are near subsistence are very limited in intelligence, or have extremely poor work habits, or are unremittingly vicious.” Maybe those who are prosperous work mightily to try and educate them as to certain bourgeoisie values, only to be rebuffed by most, yet successfully educating and rehabilitating a small minority of the 80%, who year after year as a result become part of the growing prosperous percentage.” Maybe the prosperous 10% are doing well by helping the 80% not starve–opening PayDay Loan operations, putting in grocery stores in dangerous parts of town, generating job opportunities for low-skilled non-punctual workers, etc. In such circumstances–completely consistent with the thought experiment–Robert suggests it would be unjust to force the prosperous to pay for the choices of those who remain at a subsistence level. Robert notes the possible unintended consequences of taxing the productive to assist the indolent, raising concern it may lead to a decrease in the former and an increase in the latter. And Robert says all this because he is quite familiar with the current state of economics.
Brennan anticipates that the more sensitive one is to consequences, the more likely one will be willing to violate one’s moral commitment to the NAP in the face of horrific consequences. But here is a perfectly permissible construction of Brennan’s thought experiment, one that allows all capitalist acts among consenting adults, such that moving away from the libertarian polity may make the horrific consequences even worse. Thus, Brennan’s thought experiment cannot be a sensitive measure of the interaction between devotion to deontological principles and empirical beliefs about economics.
7. The Black Box Problem
Let’s expand and abstract Robert’s insight in reason #6. We can say that Brennan’s thought experiment has a Black Box character. The Black Box is the economy. You place some input–people, resources–into the black box. The black box does SOMETHING. The output is a measure of prosperity. In Jason’s thought experiment, his Black Box is a libertarian polity, and his output is something we all agree is horrific. He seems to think that is sufficient for him to determine, from people’s answers of what to do and why, how one’s moral commitments interact with one’s empirical beliefs about how economies actually work. But what about the inputs? As noted above in raising the question of desert, inputs obviously matter. Ignoring the people and their character, intelligence, industriousness, preference for leisure, time-preference, etc., what if Jason’s thought experiment occurred on a world whose only land mass was a small barren island rock? The point should be clear: If one REALLY wanted to test how one’s moral commitments interact with one’s empirical beliefs about how economies actually work, one would give us the Black Box AND the input and offer the hypothesis about the output. But Brennan doesn’t do that. He completely ignores the input. This makes it more difficult to take Brennan’s thought experiment with the seriousness he seems to think it deserves.
Let’s be clear about this point, as it’s important. Let’s elaborate on the Black Box problem. Imagine two societies, one of which dominates from a freedom perspective. Society A is a libertarian polity, however conceived in terms of government organization, tax levels, border openness, etc. Society B is identical to society A except that there is a 0.25% tax (above any other taxes) on all exchanges. That collected money is burned, availing no one. Thus Society A clearly is the preferred society from a libertarian perspective. The Black Box for Society A is labeled “Libertarian Polity” and for Society B the Black Box is labeled “Not As Good A Libertarian Polity”
But also imagine that on a variety of important and relevant parameters, say virtue and intelligence, among others, Society A has a standard Bell curve distribution while Society B is composed of people only in the “upper” (right) half of Society A’s Bell curve. Thus, the input to Society A’s Black Box is labeled “Normal people” while the input to Society B’s Black box is labeled “Better people.”
It would seem to follow easily that it would be more likely society A than society B suffers a higher percentage of those who are starving or at subsistence. The minor marginal barrier of the 0.25% tax on society’s prosperity would seem swamped by the opportunities that arise from the application of intelligence and virtue. That is to say, the output from Society A’s Black Box is more likely to be worse than the output from Society B’s Black Box.
What does this prove? It’s a mistake to confuse a society’s structure–the rule of law; property rights; adherence to the NAP–with the combination of structure PLUS input–the natural talents of the populace–when viewing output–the degree of prosperity and poverty. Jason’s thought experiment goes straight from structure to output without assessing input at all. Jason simply assumes his described deplorable output must be the result of society’s structure, not in any way related to the population’s input. But we see that’s wrong. This is but one more reason I don’t think the thought experiment accomplishes what Brennan believes it does.
8. The “Are You Smarter Than A 5th Grader” Variant
Here’s another simple way to see Brennan’s thought experiment doesn’t really test the interaction between libertarian deontological moral principles and empirical economic beliefs. Let’s make two simple assumptions:
5th graders, even smart ones, know relatively little about the state of economics
5th graders have SOME reasonable moral intuitions and preferences. In particular, they understand and usually agree with David Boaz’s three-part summary of libertarianism for children: Don't hit other people. Don't take their stuff. Keep your promises.
So now let’s ask, “What would happen if you gave a suitably modified version of the Brennan thought experiment to 5th graders?” It would look like this (italics denoting change):
Imagine that a bunch of very smart people provide compelling evidence that life in a free country would be a disaster. Imagine that they show conclusively that if people everywhere were to follow the three rules: Don't hit other people. Don't take their stuff. Keep your promises, that 10% of people would starve, 80% would be very poor, and only 10% would be OK or do well…
Imagine we ask this of 100 or so smart 5th graders. My guess: Almost all of them would say, “We should stop following those three rules. Because we don’t want 10% of the people to starve and 80% to be very poor.” If I’m right–and I encourage Brennan or some other political theorist to run this test–does Brennan think that means he’s learned something about how these 5th graders’ commitments to those three simple rules interact with their empirical beliefs about how economies actually work? We already speculated that 5th graders have no deep understanding of how economies actually work. Yet I submit–Jason is free to explain why I’m wrong–I made no substantive change to his thought experiment as it impacts on “how economies actually work.” So, again, here is one more example of why Brennan is wrong about what his own thought experiment actually tests. This speculation about 5th graders shows you don’t really need to know much of anything about economics to meaningfully answer Brennan’s question. It immediately follows that Brennan’s thought experiment cannot be testing the interaction between one’s moral commitments and one’s empirical views of economics.
Brennan’s Burden: The Extreme Nature of His Thought Experiment
9. A Radiologist’s Point of View
Brennan’s thought experiment is stated in such an extreme fashion as to lose most all discriminative power. What do I mean by this claim? Well, I’m a radiologist by trade, so let me explain that by describing a standard curve used in radiographic physics and applying it to Jason’s problem.
In learning how to expose radiographs, radiology technologists (RTs) are introduced to what is called a Hurter–Driffield curve, or H & D curve, also known as a Characteristic Curve because it has a characteristic appearance for a given film/screen combination. Here’s the point: When you plot the optical density (the gray scale) of a radiographic film as a function of (the log of) the exposure (the percentage of photons transmitted through the patient), you get a sigmoid shaped curve like this:
Looking at the curve, you see a low horizontal portion on the left, near Dmin. Known as the toe of the curve, little information is to be had there. Doubling the exposure leads to no significant change in density. It will still all look white (underexposed). Similarly, on the upper right of the graph, near Dmax, is another horizontal portion, known as the shoulder. Here, too, there is little information to be had. Doubling the exposure will have no effect on the density. The image will still appear all black (overexposed).
The helpful information, we teach the RTs, is to be had in the linear upward sloping area, the body of the curve. There, making a small change in the exposure will lead to a visible change in the density, large or small depending on the slope of the body of the characteristic curve.
What does this aside have to do with Brennan’s thought experiment? It creates an analogy to explain why his thought experiment doesn’t work. It is way up on the shoulder. It cannot detect how marginal changes in economic consequences impact on one’s moral deontologic intuitions, because it’s already maxed out.5
To make this even more clear, quickly consider these Brennan variants:
Shoulder thought experiment: “Imagine if a purely libertarian polity with a fully free market economy caused EVERYONE TO DIE. Would you still desire a purely libertarian polity?”
Toe thought experiment: “Imagine if a purely libertarian polity with a fully free market economy caused everyone to be fabulously wealthy and have all their significant needs met, but also resulted in an occasional street corner that was not wheelchair accessible. Would you still desire a purely libertarian polity?”
Neither the toe nor shoulder thought experiments allow you to learn anything about how a respondent’s moral commitments interact with his empirical beliefs about how economies actually work. They test instead as to whether the respondent is sane or has a firm grasp of English. What, seriously, would you think of a respondent who answered “Yes” to the shoulder thought experiment or “No” to the toe thought experiment?
So, to re-state my position here: Brennan’s thought experiment, to anyone who has even basic familiarity with economics, is much closer to the shoulder thought experiment than to anything that might credibly create a tension between a preference for a libertarian polity and one’s empirical understanding of how economies work.
If Brennan REALLY wanted to test what he claims he desires to test, he would need to position his thought experiment in the body of the curve, so to speak. He would need, to stress it again, a more reasonable, less absurd, albeit still negative, economic consequence, one that is sufficiently realistic that it creates a serious tension between economic consequences and one’s moral beliefs. The fact that it is easy to create such a thought experiment adds further intellectual discredit to Brennan’s apparent lack of interest in doing so.
Would It Help Brennan If His Thought Experiment WERE About Economics?
10. The Anti-historical Nature of Brennan’s Thought Experiment
A recent WSJ review of Abigail Carroll’s book “Three Squares,” quotes Dr. Alexander Hamilton as he
cataloged his journey from Maryland to Maine in 1744. Invited to dine with a ferryman and his family, he declined. He described the meal: "They had no cloth upon the table, and their mess was in a dirty, deep, wooden dish which they evacuated with their hands, cramming down skins, scales, and all. They used neither knife, fork, spoon, plate, or napkin because, I suppose, they had none to use."
By the standards of the age, the ferryman's repast was ordered: "Only about a third of the families in seventeenth-century Virginia had chairs or benches, and only one in seven had both," writes Ms. Carroll. Only about a quarter of the early Virginian houses had tables.”
Of course, as we go further back in history, further from the development of a global market economy, we find, as historian Ralph Raico has commented, that King Louis XIV’s wine and chamber pots froze in his castle during winter. So, brushing with broad strokes, we find history replete with evidence that people we consider poor in contemporary America literally live better than kings of centuries past by most reasonable parameters. And this doesn’t even get to the issue that “subsistence” and “prosper” are societally-relative terms. Most Americans today wouldn’t interpret the presence of a table in one’s home as evidence of prospering.
See also Angus Deaton’s new book, The Great Escape, which describes, as economist and blogger Don Boudreaux put it, “the liberation of much (although not yet all) of humanity, over roughly the past 250 years, from the grinding and to us moderns almost-unimaginably harsh and dreary poverty that was the lot of nearly every person in all generations until this Great Escape.”
These historical anecdotes merely give flesh to the bare bones of this kind of graph, no doubt familiar to most libertarians. This one is from economist Brad DeLong.
A fair reading of this graph is that everything we know from history indicates that moving toward a market economy moves us away from the entire prior experience of mankind, much worse than 90% of the populace immiserated and only 10% prospering. As we move to a world with comprehensive markets, we move to a world where these proportions are actually viewed as a bad thing rather than a blessing, where we routinely expect and receive much more. This is why I previously described Brennan’s thought experiment, where he imagines markets leading to the very result they historically eliminated as akin to a thought experiment where you imagine demand curves slope upward rather than down. Both are something no one has ever experienced. And this is why, above, I describe Brennan’s thought experiment as too severe to achieve his desired goals.
Is the Brennan Thought Experiment Salvageable?
11. A Framework For Thought Experiments: Now Up to 7 Billion!
I want to be charitable. Is there, as Justice Roberts might put it, a saving construction to Brennan’s thought experiment? Let me now construct 7 billion thought experiments, a scaffold, or framework, if you will, of thought experiments.
The goal, you recall, is to test one’s commitment to deontological principles, like the NAP, when stressed by one’s sensitivity to consequences. Brennan’s thought experiment suffers from some ambiguities. How prosperous are the 10% who prosper? Is the distribution in the libertarian polity Pareto optimized (that is, are there yet trades to be made such that at least one person is better off and no person worse off)? So I begin by slightly simplifying Brennan’s thought experiment:
Imagine that a bunch of economists provide compelling evidence that life in a strictly libertarian polity would be make one person worse off and otherwise not affect anyone. Imagine that they show conclusively that if people everywhere were to live in a Nozickian minimal state or a Rothbardian anarcho-capitalist civil society, with everyone strictly observing property right rules, the only consequence would be that one person would be worse off.
And now we iterate: Let i vary from 2 to 7,000,000,000, such that…
Imagine that a bunch of economists provide compelling evidence that life in a strictly libertarian polity would be make i persons worse off and otherwise not affect anyone. Imagine that they show conclusively that if people everywhere were to live in a Nozickian minimal state or a Rothbardian anarcho-capitalist civil society, with everyone strictly observing property right rules, the only consequence would be that i persons would be worse off.
Now of course we need to tighten this up. What does it mean to say no one else would be affected? Does it mean they would live as well as they do now or that they would live as well as we expect people to live in a fully free society? How should we deal with interpersonal utility comparisons, given the switch from the original thought experiment where some do poorly and others well to the modified experiment where some do poorly and others are unchanged? I give Brennan free reign to fill in such details so as to maximize the pedagogical value of the thought experiment.
Presumably even those with the most tenuous dedication to the NAP, the guiding deontological principle of the libertarian polity, will be comfortable with the thought experiment results when i is very low. And presumably even the most devoted adherent to the NAP will grow uncomfortable as the sky falls and i approaches 7 billion. And perhaps this framework of thought experiments will allow us to learn something, as Brennan desires, of “how your moral commitments interact with your empirical beliefs about how economies actually work.” The higher i goes before you switch, the stronger your moral commitments to a libertarian polity relative to your sensitivity to consequences. I offer this modification to Professor Brennan as a charitable effort to improve and focus his thought experiment.
I’m still not convinced, for reasons given above, it tests one’s “empirical beliefs about how economies actually work,”. But I can say this with confidence: My framework must test “how economies actually work” at least as well as Brennan’s original thought experiment, since that part of the thought experiment was left untouched. In particular, there must be an i such that for that i, that thought experiment is as close as desired to the original thought experiment. Which is to say Jason’s original thought experiment is part of my larger framework.
12. Looking at the Matter Cartoonishly…
Alan Moore is frequently praised as the author of what many call the greatest graphic novel ever written, Watchmen. And while those not fully appreciative of graphic novels disdainfully refer to such work as cartoonish, I think we can learn much about the tensions between moral commitments and sensitivity to consequences from studying this work.
To briefly summarize a very complex and layered plot, in a world like ours save that costumed heroes and even super-heroes exist, the planet’s smartest man, who in costume goes by Ozymandias, faces a moral dilemma. Engaged in costumed heroics for more than a decade, never failing to best the villains he confronts while developing a reputation for never killing his foes, he is convinced, correctly, that the world’s two super-powers, the USA and the Soviet Union,7 are irrevocably moving toward World War III, a nuclear holocaust that will destroy all human life on the planet. To save the day, he develops and carries out a plan that gets the two super-powers to put aside all past enmities and form a peaceful alliance, albeit at the cost of Ozymandias having to kill half of the population of in New York City. That is to say, the reader is forced to evaluate the moral propriety of violating, in grave fashion, the Non-Aggression Principle and killing about 3.5 million people to potentially save the lives of 7 billion. Note the ratio here is 0.05%.
Here is a thought experiment–do you agree with the protagonist’s actions?–that creates some tension between one’s moral commitments (do not kill) and one’s sensitivity to consequences (nuclear holocaust is horrific). I suggest there are at least three significant ways the Moore thought experiment improves on the Brennan thought experiment, achieving what Brennan himself indicates was his desired goal:
Moore gets the proportions right. Ozymandias kills off 0.05% of the population to save 99.95%. One’s sensitivity to consequences makes it an easy choice to save the 99.95%, but one’s commitment to the NAP rebels at killing MILLIONS of innocents. Thus, tension. If Moore had Ozymandias kill off 49.99% of the population to save 50.01%, it may STILL have made sense using consequentialist logic. But the tension is gone. Either way, the NAP is violated and half or more of the planet would die. So no tension develops between favored consequences and moral principles. Similarly, in the Brennan thought experiment, the suggested consequences of complying with the NAP are so horrific, no reasonable person would choose to comply with the NAP at the cost of the death of 10% of the globe and immiseration of 80%. So there is really no tension here in making a choice, despite Brennan’s professed desire to create it.8
Moore demands a personal moral choice. Brennan describes a horrific society and asks what you would do. Although vague and not fully developed, the impression is that he anticipates answers like, “I’d agree to taxation on the prosperous 10% to help feed the starving 10%” rather than “I’d break into the houses of the prosperous 10% and use the loot obtained to feed the starving 10%.” That is, Brennan’s thought experiment is set up to see what it takes to get you to assent to government intervention, not to personally engage in immoral behavior. Nowhere are you asked to personally violate the NAP. To the extent that morality tests YOUR ethical choices, you are not clearly asked in the Brennan thought experiment to make a moral decision. This also has the effect of lessening the tension between one’s moral commitments and one’s alleged understanding of how economies work. This is very different from the Moore thought experiment, where you are in essence asked to imagine yourself facing the decision Ozymandias faced, asked if YOU would yourself kill 3.5 million people to save the planet. That, again, enhances the tension between one’s moral commitments and one’s sensitivity to consequences.9
Moore provides needed detail. Brennan and I agree there is a sense in which thought experiments need not be “realistic.” But there is a strong sense in which Brennan’s thought experiment is not merely unrealistic, but grossly unformed. He describes the end state of following a libertarian polity–10% starving to death, 80% eking out only a subsistence living, only 10% prospering–with NO explanation of HOW that could POSSIBLY result from a libertarian polity, not even a bare outline of an unrealistic mechanism. In the alternative, Moore provides extensive details as to how (albeit unrealistically) Ozymandias puts together his decades-long plan to save the world, how he kills half of NYC. The addition of this level of detail adds to the moral tension by creating verisimilitude, making the question Brennan asks–what would you do and why?–more compelling, difficult, challenging. Here, again, Moore, who makes a living producing cartoonish characters that communicate using word balloons, outshines Brennan, for whom discussing cartoonish characters is only an avocation. Moore creates the more “realistic,” more intellectually demanding, thought experiment.
In “Imagine Markets Stink. It’s Easy If You Try,” Brennan states “Libertarianism doesn’t follow from fundamental principles or axioms that are not open to question. Rather, to get to libertarianism, if you are at all sensitive to consequences, you need to make a series of empirical arguments about how institutions work. Now, in light of existing social science, it’s pretty clear that large-scale economies need to be run fundamentally on market principles. But it’s highly controversial to what extent government should supplement, enhance, regulate, or supplant markets in order to generate what consequence-sensitive libertarians would themselves agree are better outcomes.” Since he was responding to me, the implication is that I disagree with this claim. That is not true. What IS true is that Brennan’s thought experiment does not in any way suggest or imply or aid in developing any such empirical argument. That’s because it’s stated in such an extreme fashion as to lose most all discriminative power. A person who is even minimally “consequence-sensitive” can easily agree with Brennan that in the confines of his thought experiment he would no longer seek a Nozickian/Rothbardian society while not agreeing AT ALL that “it’s highly controversial to what extent government should supplement, enhance, regulate, or supplant markets.” Brennan’s thought experiment doesn’t track the intensity of one’s devotion to deontologic principles. It doesn’t strongly track the intensity of one’s sensitivity to consequences. It may be good for shocking undergraduate philosophy virgins, but, for all the reasons I’ve provided, I’m frankly surprised he finds it of any pedagogic value at all.
- Jason noted in his original essay that his thought experiment, “is directed at libertarians who 1) advocate a purely negative rights-based theory of justice, but 2) who also believe that under ideal libertarian institutions, markets would work really well at making most people better off.” Brennan is of course free to direct his thought experiment to whomever he desires, but the structure and coherence of his thought experiment stand independent of this directional preference on Jason’s part. Certainly this directional preference has no bearing as regards the claim that an inability to appreciate the thought experiment is evidence of ideological bias or being “widely misinformed about the state of economics.” Therefore, in some of the following reasons I offer to reject Brennan’s view, I relax this “restriction.” While relaxing the restriction may make the thought experiment less useful or pointed, it shouldn’t make it incoherent. ↩
- Granted, for entirely unrelated reasons, she is aghast at the notion of a Rothbardian anarcho-capitalist civil society… ↩
- I should note as an aside that I view this as a humorous caricature of Ayn Rand’s view of self-interest, and think most thoughtful analysts of her work, favorably inclined or not, would agree. ↩
- It IS true that Brennan, in his thought experiment, indicates the “bunch of economists” “show conclusively” that all Hell breaks lose when you have a libertarian polity. It is, however, far from clear how much heavy lifting “conclusively” can do in a thought experiment. ↩
- To carry the radiographic analogy just slightly further, you might say Brennan is philosophically over-exposed. ↩
- It what immediately follows, crucial plot elements will be revealed. If you’ve not yet had the pleasure of enjoying this wonderful piece of literature, listed as one of the top 100 novels of the 20th century by Time magazine, you may wish to skip this section. ↩
- The story spans many decades, but climaxes in the mid-1980s, when the Soviet Union still existed. The actual comic series also appeared in the mid 1980s, so the potential of nuclear exchange was real to the reader. ↩
- Personally, I think this is Brennan’s fundamental error in creating his thought experiment. He thought he was heightening the tension between following one’s moral commitments and supporting free markets by making the results of following free markets EXTREMELY disastrous. In fact, he was vitiating the tension by going so severely to extremes. To make a medical analogy: the choice to undergo surgery and potentially lose a limb or not undergo surgery and DIE is an upsetting but not difficult decision. The choice to undergo surgery and potentially lose a limb or not undergo surgery and live in constant pain is a difficult decision. ↩
- As does the brief conversation between Ozymandias and Dr. Manhattan, where Ozymandias states he has forced himself to imagine every individual death for which he is responsible. Here Ozymandias does not avail himself of the ethical evasion that one man’s death is a tragedy but a million deaths are but a statistic. ↩