Saturday, May 12, 2007

Paul Churchland: Rawls' Theory of Justice

Paul Churchland, professor of philosophy at University of California, San Diego, rose to address the Beyond Belief 2006 conference to present the moral/political theory of John Rawls.

Churchland began his presentation by reporting that he was asked to speak about the moral and political philosophy of John Rawls probably because the conference organizers wanted to show the audience a system of morality that did not rely in any way on scripture. Rawls’ theory of justice certainly met that objective.

Rawls: A Theory of Justice

Briefly, Rawls suggested that we can best intuit justice by asking what a rational person would choose for a political system if he were placed behind a ‘veil of ignorance’. This veil left the agent ignorant of any of the particular facts of his life – his intelligence or access to education, his economic standing, who his friends were, his physical appearance, his race, etc.

According to Rawls, the position that a rational person would choose is one that provided the highest standard of living for those who were the bottom of the social and economic ladder. Such a system, for example, would allow for great inequalities of wealth if the system that produced that wealth left the least-well-off people better than they would have been in a system without such inequalities.

The idea was that the option which raised the position of the least well off was a uniquely rational best option.

Churchland’s Criticism

Churchland then threatened pulled the rug out from under the conference organizers by claiming that this theory of justice does not work. Of course, what good does it do to assert that we can have a system of ethics that makes no use of God if there is no system of ethics that works.

Churchland’s objection was that Rawls’ theory of justice employed the same type of reasoning that Rene Descartes used to defend particular facts of nature, including the existence of God.

Descartes argued that if he can form a clear and distinct idea of a natural principle than this demonstrated that the principle was a natural law. This meant that all that scientists needed to do was to sit around and measure the clearness and distinctness of their ideas in order to understand nature. Descartes also asserted that one of these clear and distinct ideas was the idea of God; therefore, God exists.

Rawls is using his veil of ignorance to form “clear and distinct ideas” of moral principles. It promises to be no more useful in helping us come to a theory of justice than Descartes plan.

As a matter of fact, science has not made the progress we have seen by testing ideas for clearness and distinctness. It would, in fact, be quite nice if all a scientist has to do is sit back in his office, put his feet up on his desk, and test his ideas for clearness and distinctness.

Yet, science requires a lot of hard work. The scientist has to get up off of his desk, go to the lab (which often is a very uncomfortable place such as a hot desert where he looks for dinosaur fossils, a damp and sweltering jungle, or an arctic research station where he is picking up ice cores). He has to carefully and meticulously take and record precise measurements. Then he needs to fit his favorite theory to those measurements or – GASP! – discover that his theory does not fit.

Churchland’s argument is that morality and justice require the same type of experimentation. He described human history substantially as a social science experiment, with an emphasis on science. We set up a society. We observe the results. From our observations we evaluate whether the assumptions on which our society are validated or falsified. We make adjustments. We then establish a new society, or modify the existing society, to take into consideration what we have learned. We make a new set of observations. And so on.

Churchland used Prohibition as an example of a social experiment. People devised a theory, they set up an experiment, they drew some observations that suggested that prohibition was a bad idea. They tossed out prohibition. I suppose we are to think of inquisitions, slavery, monarchy, Nazi Germany, and Soviet Russia as failed experiments.

This may be a salient point against Rawls’ methods for arriving at moral truth, but it is strangely hollow. Imagine somebody responding to a theory of planetary formation by saying that current theories are only the most recent component of millennia of scientific study and hard work that can be traced back to the first civilizations. This may be true, but it does not help us to decide which theory of planetary formation is correct. It suggests that certain methods should not be used to judge theories of planetary formation, but it says nothing about the methods that should be used.

I have encountered a similar issue with respect to the desire utilitarianism that I have defended here. One of the more frequent objections that I receive says that desire utilitarianism must be rejected because, if desire utilitarianism were true, some moral questions would be difficult to answer. The objector makes the completely unfounded assumption that a moral theory would make all moral questions easy to answer, and that desire utilitarianism must be rejected for its failure to do so.

I would argue for rejecting any theory that distributes answers to moral questions like answers to scientific questions. Some of them are easy to answer. Some are difficult. Some may even remain outside of our ability to answer forcing us to live in a universe with some measure of moral uncertainty. Yet, over time, we have the ability to make moral progress as we make scientific progress, never quite arriving at perfect moral knowledge (just as we will always lack perfect scientific knowledge), but getting closer over time – as long as religion doesn’t muck things up by insisting on teaching moral (scientific) myth.

Other Objections to Rawls

I would like to take this opportunity to throw in a few additional objections to Rawls’ theory of justice.

(1) Implications of the Veil of Ignorance

Rawls asks us to imagine the choices that one will make from behind a veil of ignorance. He then says that these conclusions apply to the real world in which we live. But why would implications drawn from obviously false premises have any relevance in the real world?

Imagine, as you read this, that a fire breaks out nearby filling the room with smoke. What would you do? No doubt, you can come up with a plan. You can pick an escape route and imagine yourself using it. However, it does not follow from the fact that you would perform a particular set of acts in an imaginary world in which there is a nearby fire, that you should – in the real world where there is no fire – perform the same actions.

Similarly, the mere fact that you would choose a particular political system when behind a veil of ignorance certainly does not imply that you should choose that same political system in the real world where you suffer from no such ignorance.

(2) The Extent of Ignorance

Another problem arises from the question, “How ignorant am I supposed to be?” Are we supposed to be ignorant of our beliefs?

For example, am I to assume that I do not know from behind a veil of ignorance whether I am a Christian Scientist? It would seem that I must be ignorant of this if I am going to vote for a system that treats Christian Scientists justly. However, am I also supposed to be ignorant as to the cause of diseases? If I am not to be ignorant of disease, then I would likely support a system that helped Christian Scientists realize their error and would force doctors (rather than priests) to attend to my illness.

Consider the choices that a person who believes in an arrogant and megalomaniacal God who condemns to eternal damnation any who do not believe in him. What would this person choose behind a veil of ignorance?

Or the person who believes that homosexuality is fundamentally wrong? To such a person would likely vote for a society where homosexuals are compelled to seek some sort of treatment that he would want even for himself.

In short, if people are not allowed to take behind the veil of ignorance their beliefs about the nature of the world – the causes of disease, the existence of a God, the intrinsic value of various states of affairs – then we will be too ignorant to make rational decisions. On the other hand, if we can carry our beliefs about such things behind the veil of ignorance, people will end up promoting substantially the same principles behind the veil of ignorance that they promoted in front of the veil.


In short, Rawls’ theory of justice is a secular theory, but it does not work. As such, it is a poor candidate for proving that it is possible to have a system of morality that does not depend on God. This job must be given over to theories that actually work.


Anonymous said...

You’ve probably answered this before, but I’ve not read all your posts, so I’ve not yet seen it. If we have an isolated place, where the only people are a thousand child molesters and one child, we might predict that molesting the child would fulfill the more and stronger desires of nearly all the people there at the cost of only thwarting the desires of the one child.
In this event Rawls’ theory would protect the child because any of us would not know if he or she might be the child. How do we use desire utilitarianism to protect the child?

Alonzo Fyfe said...

atheist observer

I have answered the question. However, it is inevitably going to be one of the most frequent objections raised no matter how many times I answer it.

The answer hinges on the distinction between desire utilitarianism and a closely related theory - desire fulfillment act utilitarianism.

Desire utilitarianism measures desires by their consequences and measures acts by whether they are acts that good desires will cause.

Desire fulfillment act utilitarianism does not measure desires at at all, but measures only acts by their consequences (in terms of desire fulfillment).

Desire fulfillment act utilitarianism has the consequences you mention.

If the desires of the child molesters did not exist - no desires would be thwarted. Not the child's. Not the people's. With the existence of these desires, some desires are thwarted. Thus, there is more reason to create a society without these desires than to create a society with these desires.

Now, as to Rawls, how would Rawls' theory protect the child? Because, in your example, the odds are 1001:1 against being the child.

Anonymous said...

It protects by the magnitude of the consequences. I would chose not to buy a car that has a thousand to one chance of exploding and killing me, because the consequences are horrific enough that even this probablility is not one I would chose.
I think Rawls' approach is to say we could be in the worst possible situation in the society, and we should act to make even this situation no more horrible than necessary.

Alonzo Fyfe said...

Atheist Observer

You are forgetting the magnitude of the consequences for the child molesters. The car with 1 in 1001 chance of blowing up is the best option if you have to have a car, and the only other option blows up 1000 out of 1001 times.

Chad Van Schoelandt said...

"Now, as to Rawls, how would Rawls' theory protect the child? Because, in your example, the odds are 1001:1 against being the child."

Rawls is very explicit that the make up of the society is also hidden from the Original Position agents.

"But why would implications drawn from obviously false premises have any relevance in the real world?"

Rawls uses the veil of ignorance to eliminate irrelevant information from the decision procedure. One may similarly not put an applicant's race on an application, or ask a hiring committee who they would hire if they did not know the races of a set of applicants. I assume that we would not take as a response, "But I know their races and I want to hire the people of my race."

As for your fire example, I'll point out that there is not a question of fairness involved in the attempt to escape from the fire. The specific reason Rawls says to rule out the information is to ensure fairness, which it is hard to do if you appeal to your known race, gender, religion or the like in justification of distributing the benefits of society.

"Are we supposed to be ignorant of our beliefs?"

Yes. Agents would know general, uncontroversial facts of science, sociology, economics and the like, but not much particular. Rawls is attempting to set up a standard in which particular facts can be appealed to, but the particular facts are not needed from the start. This is why Rawls proposes the difference principle, and leaves it to economists to put forth the particular facts about how to bring about the principle.

This may actually be a way of bringing back your fire example. What we would do in the relevant analogy is come up with principle of escape from behind the veil (escape soon, saving one's life as priority over saving one's entertainment equipment...) and then consider the facts, comparing proposed escape plans to the standard generated from behind the veil. This is still an extreme stretch, as fairness is not involved, but I hope it is clear enough. The original position will not spit out a tax code, but will create a general standard by which to judge the tax codes put forward.

Anyhoo, there is much in your post which I take to be oversimplifying Rawls, but this is why his work has generated a huge literature over the past decades. Will I am doing my thesis on (technically against) Rawls, and have studied his work for many years, I will not attempt to endlessly expand this comment.

Alonzo Fyfe said...


Some oversimplification is necessary in a 4-page blog post. Yet, please assume that just as Rawls' arguments are oversimplified, so are my answers.

(1) The Makeup of Society

Rawls' claim that the makeup of society is hidden runs into two problems.

First, Rawls' claims about what we would know about the composition of society behind a veil of ignorance is actually ambiguous. He allows us to know that we are humans, and (as you point out) we are allowed knowledge of certain psychological facts. The line between what we are permitted to know and what we are not permitted to know about human nature is not entirely clear.

Second, and most important, the composition of society is relevant to certain policy decisions. Behind the veil of ignorance we will adopt rules that will take the composition of society into affect where it is relevant. For example, our policies would be quite different if 99.999% of the population had a fatal allergy to peanuts and 0.001% liked peanuts. If we do not know these distributions behind the veil of ignorance, we are still going to want to adopt rules that will allow us to make use of them when the veil is lifted.

It is on the virtue of this second fact that I asked the question: How is it the case that Rawls' theory would give the answer that Atheist Observer wanted?

(2) The Fire Example

The fact that there is no fairness involved in escaping a fire is not relevant.

The fact is, no argument is sound if it contains false premises.

There is no rule of logic that says, "An argument is sound if and only if its premises are true and the argument form is valid, except when we talk about fairness, in which case we have to use false premises, but don't worry about it, because our argments would still be sound."

I use the fire example to illustrate this general point - that no conclusion drawn from false premises is necessarily true in a different world where those premises are no longer false.

(3) "Agents would know general, uncontroversial facts of science, sociology, economics and the like."

My point is, there are no such things. Practically everything is contoversial. The very example that I used was one in which there is a controversy over the causes of disease - whether they are associated with bacteria or with alienation from God.

All facts are "uncontroversial" only within a particular world view, so we must select a world view before we select a set of "uncontroversial" casts - and people with different world views, behind a veil of ignorance, will never agree on what counts as an "uncontroversial" fact.

In other words, we will never reach an agreement from behind a world of ignorance unless we first prohibit participation by those who do not share our world view.

Anonymous said...

I am new to your blog, and possibly more ignorant than most of your readers. Allow me to say nevertheless how useful I have found your thinking. Thank you.

Alonzo Fyfe said...


Everybody was new to this blog at least once.

Even me.

Chad Van Schoelandt said...

I fully recognize that oversimplification is necessary, and am sure my own response included an excessive amount of it. My replies should never be taken as anything but an honest attempt to dialog, and not in any way insulting, cynical, or any sort of writing off. On that note:

(1) "If we do not know these distributions behind the veil of ignorance, we are still going to want to adopt rules that will allow us to make use of them when the veil is lifted."

This could be taken in at least two ways, one of which I have struggled with. It does seem that the agents could decide on principles which make reference to some possible probabilities, such as Principle X will be just if the population contains a super majority, while Principle Y will be just if there is a divided populous. This of course will not protect the children in Atheist Observer's case. I suspect that the strongest case against Rawls on these sorts of appeals will be in favor of form of preference utilitarianism. John Harsanyi has some work arguing for utilitarianism from the Original Position.

(2) Back to the fire example, I think you might be thinking of the argument taking the following form:

1)Under a veil of ignorance we would ensure rights A, B and use economic system C
Therefore, we should ensure rights A, B and use economic system C.

This would use a strange logical rule where you can derive "we should [x]" from "Under a veil of ignorance we would [x]". Your fire example does fine to show why this is not a proper derivation. Rawls's argument takes a different form however.

1) The original position, with the veil of ignorance, eliminates unfair conditions from the decision making process.
2) The original position, with the veil of ignorance, is a situation of fairness for deciding on principles of justice.
3) Principles chosen in a fair situation are fair.
4) [From 2 and 3] The principles chosen in the original position with the veil of ignorance are fair principles.
5) In the real world we should follow fair principles.
6) [From 4 and 5] In the real world we should follow the principles chosen in the original position with the veil of ignorance.

I'll have to read that after sleeping to see if I still agree with it, but the derivation is not from "In the OP we would [x]" to "We should [x]". Instead we go from "In the OP we would [x]" to "[X] is fair." I suppose this would be the case with the fire escape plan to, but no one thinks the best escape is the fair one, while people may think of justice as fairness.

(3)"we will never reach an agreement from behind a world of ignorance unless we first prohibit participation by those who do not share our world view"

Not to give Rawls an out for this, but he does seem to concede some of this in his later work where he puts much emphasis on his principles being for modern, liberal-democratic peoples. His arguments depend on a number of globally controversial premises and thus will only make sense within certain cultural contexts such as our own.

As for those who no not agree with us about the source of disease or otherwise have some strange world view, Rawls would claim that while we exclude them from our considerations of what the principles of justice are, our own world views may provide constraints on what it is just to do to them. So while you may be basically correct, "I would likely support a system that helped Christian Scientists realize their error and would force doctors (rather than priests) to attend to my illness," we may not force the Christian Scientists to see doctors. Their children are another story... And note that to help someone see their error is far less than to use the oppressive force of the state to hunt them down for their views and force conversions upon them. At the front of Rawls's thought was how we are to deal with "reasonable pluralism" and at times unreasonable pluralism, and what some of our (most modern liberal-democratic people) strongly held principles (such as what is and is not fair) commit us to. One thing is that for the sake of fairness we should not appeal to certain sorts of beliefs to justify state action, though of course there is a limit to what we would give up appeals to.

In any case, I read your blog and find it interesting, and have found this post and discussion fruitful.

Alonzo Fyfe said...

Chad Van Schoelandt

Your working of Rawls' argument shows some circular reasoning - captured in your example by the use of the term 'fair' (or a cognate) in every premise, rather than deriving fairness from something else.

We decide what we should not allow behind the veil of ignorance based on principles of fairness. Yet, the decisions we reach behind the veil of ignorance are supposed to give us those principles of fairness.

This seems to blur the distinction between the 'veil of ignorance' and Rawls' meta-position regarding reflective equilibrium. (Reflective equilibrium suggests that we use our intiutions to draw up a set of moral principles, which we then use to modify our moral intuitions, and back and forth until our intutions and principles are in equilibrium.)

Reflective equilibrium is supposed to give us this 'veil of ignorance', which in turn is supposed to give us our moral principles.

Speaking very generally, I have nothing against this idea of 'reflective equilibrium'. I am one of those who holds that this is really nothing more than coherentist epistemology applied to ethics, and I like coherentist epistemology. However, Rawls' tight connection between the 'principles of justice' that determines what we are ignorant of behind the veil of ignorance, and the 'principles of justice' that determine the conclusions that he reachers, represents the tight circularity of vicious (question-begging) circles than the broad and lose virtuous circularity of coherentism.

Ultimately, I think that some of Rawls' later work has merit. One of the arguments I am particularly fond of begins with the assumption that we will never get everybody in society to agree on any specific moral system. We will always have a society in which people with different belief sets must somehow get along. He then derives a set of principles consistent with this assumption - not principles that we derive from a state of ignorance, but principles derived from real-world facts. I can make the same argument in desire-utilitarian terms.

Anonymous said...

"Descartes also asserted that one of these clear and distinct ideas was the idea of God"

Interestingly, "God" is a very unclear and indistinct idea, and if I ever have to discuss the existence of God, my first question is "what do *you* mean by God?"

Descartes was just plain outright wrong about this bit.