For many philosophers these four rules are unassailable. In reality, each can be contested, and their overall coherence placed in doubt.
There exists an abundant literature in which R1 (“one cannot derive an ought from an is”) is rejected in the name of R2 (“ought implies can”).
1R3 (“like cases must be treated alike”) is a purely formal rule that could oblige us to endorse a whole series of morally repugnant judgments. It shares certain characteristics with a more than dubious argument, the one known as the slippery slope. These are two reasons, among others, to use it with caution.
Finally R4 (“It is pointless to oblige people to do what they will necessarily do of their own accord” and “It is pointless to prohibit people from doing what they will not do voluntarily in any case”) seems to contradict certain habits, such as those that consist in encouraging those who are already in the act of doing what they have been asked to do (“Go on, continue!”).
2Let us consider this more closely.
QUESTIONS WITH REGARD TO R1: “ONE CANNOT DERIVE AN OUGHT FROM AN IS”
Despite its superior position as the mother of all rules of moral reasoning, “one cannot derive an ought from an is” is by no means a principle that must be respected at all costs if we aspire to produce a coherent overall philosophical conception. Valid theories can be constructed without respecting it.
1. Those philosophers who are the most radically subjectivist where morality is concerned are certainly prepared to concede that what is good morally, or what it is fitting to do, is nothing else than what everyone approves, everywhere or else in a particular society.
3 From what is (what everyone approves), they derive what ought to be (what is good or what we must do).
2. Bertrand Russell denounces utilitarianism because, according to him, it is a doctrine that derives the desirable (pleasure must be maximized) from what is desired (we naturally seek to maximize pleasure), that is to say, the norm from the fact.
4 But that has not stopped utilitarianism from flourishing.
3. The naturalists inspired by Aristotle have proposed all manner of examples tending to show that the rule “one cannot derive an ought from an is” is a modern invention that distances us more and more each day from true morality, which would be founded upon the nature of each being and their needs. When we say, “you must water the green plants,” we spontaneously draw a normative conclusion (“you must”) from a fact relating to the natural needs of plants (absorbing water, among other things). It is, admittedly, a hypothetical “you must,” a conditional duty, which depends upon a desire for preservation on the part of these plants, wherever it may come from. But the idea that there could be other moral duties, “categorical imperatives,” unconditional, absolute, completely detached from the desires, needs, or interests of concrete beings—is it not completely harebrained, a philosophical fantasy of no real importance?
54. For almost the whole of the last half-century, normative research in law, in ethics, and in politics has found quite solid epistemological foundations, with, among others, the development of John Rawls’s so-called method of reflective equilibrium.
6 This is a method that renounces the attempt to “found” ethics. It merely proposes to bring face to face the spontaneous moral beliefs of competent judges, our moral principles, and our systematic philosophical theories in order to construct, little by little, a sufficiently coherent overall moral conception.
In certain of its versions, this method rejects the idea that our moral beliefs and our nonmoral beliefs could be justified separately, as if there were absolutely no relation between the two. One then speaks of a “wide” reflective equilibrium.
75. Experimental moral and political philosophy is a new area of study that mobilizes very high-level researchers nearly everywhere in the world.
8 The idea informing this program is that a normative quest completely independent of the empirical facts generally culminates in unrealistic, weak, or pointless conclusions. They are unrealistic due to their not taking into account certain empirical data relating to mental architecture or to social organization. They are weak, given that they consist in bringing back certain vague and general principles like “human dignity” and that they disregard the particular maxims followed by people when going about their business. Finally, they are of very limited interest as regards the orientation of the agents in question, for they take no account of their opinions and of their concrete concerns.
The subjectivists, the utilitarians, the naturalists, the “coherentists” in the broad sense, and the experimental moral philosophers do not at all seem to think that it is completely illegitimate to establish links of all kinds between what is and what ought to be.
That amounts to many philosophers all told, and they cannot all be bad!
QUESTIONS REGARDING R2: “OUGHT IMPLIES CAN”
Certain thought experiments could lead us to doubt the clarity of this rule.
Romeo lives thirty kilometers from the center of Rome. He arranges a date with Juliet, one Thursday, at 8:00 P.M., in Rome, in the Piazza Navona. By arranging this date, Romeo has made Juliet a promise, placing him under an obligation or making it his duty to be in Rome, on that Thursday, at 8:00 P.M., and in the Piazza Navona.
But on the day of the date, Romeo dines too lavishly. He decides to have a siesta. He does not hear the alarm clock, needless to say. He wakes suddenly at 7:59
P.M. Too late! He
cannot be in the Piazza Navona at 8:00
P.M.Does it follow that his duty is annulled? No, from the fact that he cannot keep the date, it does not follow that he ought not to. Would it be contradictory to say that Romeo has a duty to keep his date even if he does not have the capacity? No.
This is indeed why Juliet does not seem to be behaving wholly irrationally when she reproaches Romeo for having failed (as usual, we should say) in his obligations.
On the other hand, it does indeed seem that Romeo’s duty would have been annulled if he had been kidnapped during his siesta or if he had died of a stroke due to his lavish meal. Finally, this thought experiment risks leading us to two contradictory conclusions:
1. “Ought implies can” is true: Romeo ought to keep his date with Juliet if and only if he can be there.
2. “Ought implies can” is false: Romeo
ought to keep his date with Juliet even if he
cannot be there.
9
Thus, valid philosophical theories can be constructed without respecting the rule R1, “one cannot derive an ought from an is” And, to the question of knowing whether the rule R2, “ought implies can,” is true, there is no definite answer.
Furthermore, R1 and R2 could contradict each other.
CONFLICTS BETWEEN R1 (“ONE CANNOT DERIVE AN OUGHT FROM AN IS”) AND R2 (“OUGHT IMPLIES CAN”)
The rule R2 (“ought implies can”) can be specified as follows: “Charlie ought only to go to the moon if he can go there.” From this formula we can logically derive another:
“If Charlie cannot go to the moon, he cannot then be obliged to go there: he ought not to go there.”
But this new formulation poses a problem for all those who admit R1 in the logical version “we cannot derive a normative conclusion from nonnormative premises.”
Indeed, “if Charlie
cannot go to the moon, then he
ought not to go there” well and truly violates R1 in its logical version, since a normative conclusion (“Charlie
ought not to go there”) is derived from a nonnormative premise (“Charlie
cannot go there”)!
10Thus R2 seems to contradict R1.
If this conflict is real, we must choose between the two rules, or amend both of them so as to render them compatible.
The subjectivists, the utilitarians, the naturalists, the “coherentists” (in the broad sense), and the experimental moral philosophers take the liberty of disregarding the rule R1: “one cannot derive an ought from an is”
Are there arguments for doing without R2, “ought implies can,” or for amending it?
Yes. It is a rule that seems to have absurd implications. It should lead us to think the following:
1. A ruined debtor does not have a duty to pay his debts, since he does not have the capacity to do so.
2. A kleptomaniac does not have a duty not to steal, since he cannot not steal.
3. A sadist and a psychopath do not have a duty not to massacre their victims, since they do not have the capacity to do otherwise.
11
Besides, the rule “ought implies can” seems to exclude moral conflicts. If we reckon that moral conflicts do not exist, the rule is not threatened. But if we believe in moral conflicts, it is.
I ought to jump into the water in order to try to save a child to my right who is drowning in a lake.
But this child’s twin is drowning in the same lake, to my left, and at a distance such that I cannot save both twins at once.
By virtue of the principle of impartiality, which demands that like cases be treated alike, if it is true that I should try to save the one, it is also true that I should try to save the other. In other words, I
ought to try to save both. But it is, obviously, a thing that I
cannot do. This kind of case enables us to envisage the possibility that “ought” does not necessarily imply “can.”
12What these examples show is simply that there is no logical link between ought and can, and not that the rule “ought implies can” has no justification.
On the basis of these examples, it does not seem illegitimate, indeed, to conclude that, in our world at any rate, it is
logically possible to demand the impossible.
13But, supposing that it is, indeed, logically possible to demand the impossible, the question would then be that of knowing why we should.
What indeed would a world in which we could demand the impossible be like? Would we like to live in such a world?
QUESTIONS REGARDING R3
The rule R3 “treat like cases alike” poses a very general problem. It is a rule of formal consistency that demands that we persist in our moral judgments without explaining what justifies them.
But let us suppose that, through completely failing to understand what the moral vocabulary signifies, we have sincerely judged Hitler to be a fine fellow. According to the rule “treat like cases alike,” we ought to assert that all those who behave like Hitler are fine people!
The principal moral question is that of knowing why we have been so stupid as to formulate the initial judgment “Hitler is a fine fellow”
In reality, the rule R3, “treat like cases alike,” only has a secondary moral importance: its moral value depends upon the quality of the initial judgment.
In this particular case, it would be better not to follow the rule.
An identical problem is posed by the moral requirement of impartiality, of which the rule R3 is one expression. We can be perfectly impartial by treating everyone equally badly.
Besides, the rule R3, “treat like cases alike,” presents certain characteristics that it has in common with an argument that is often employed in moral debate but whose validity is doubtful: the slippery slope argument.
14It is a less embarrassing objection, for it is possible in part to refute it.
THE SLIPPERY SLOPE ARGUMENT
It is not hard to illustrate this argument, so frequently deployed in public debate.
15Those who are more liberal say: “We begin by limiting late abortion, and we will end up banning contraception. Why will sexual relationships whose purpose is not procreation not then be banned?”
The more conservative will reply: “We begin by permitting abortion, and we will end up allowing infanticide. Why will we not decriminalize voluntary homicide?”
The more liberal say: “We begin by banning actively assisting those who beg to die, and we will end up prohibiting suicide, as in former times.”
The more conservative retort: “We begin by permitting assisted suicide, and we will end up allowing the elimination of the old, the poor, and the handicapped. Why not then allow it for any other category within the general population whose life is deemed not worth living?”
The philosophers have sought, of course, to formalize these common arguments in order to evaluate them.
According to Bernard Williams, for there to be a slippery slope, two conditions are necessary:
16
1. The outcome we end up with should be indisputably horrible (legal permission to eliminate children at an early age or the handicapped and so on).
2. What causes us to slip toward this horrible outcome is not a logical or conceptual necessity but a
natural progression due to social, psychological, or biological factors. In other words, the slippery slope argument must not be confused with what are known in logic as “sorites,” those paradoxes that can cause us to doubt the existence of the bald, of heaps, or of dwarfs.
17
Now, it often happens that these two conditions are not met.
It can be asserted that an outcome would be horrible without demonstrating it, as in the following example:
Slippery slope: If cloning for therapeutic purposes is permitted, we will end up legalizing reproductive cloning. Since reproductive cloning is horrible, let us ban cloning for therapeutic purposes.
Objection: We have not proved that reproductive cloning is horrible. We have merely asserted that it is.
It can be asserted that there exists a necessity to pass from the tolerable to the horrible, without specifying the nature of this necessity, as in the following example:
Slippery slope: If cloning for therapeutic purposes is permitted, we will end up with legalizing reproductive cloning, and then the instrumentalization of the body of the clone: it will serve as a storehouse of health products, of limbs and organs for the benefit of the progenitors. It will be a moral catastrophe.
Objection: Why should we necessarily pass from the first stage, controversial but tolerable, to the last, which everyone should find horrible? Why should the usual repressive mechanisms not suffice to avoid the slide from the tolerable to the horrible? If we vote for a total ban on cloning, it is because we reckon that it will be respected. Why do we not have the same confidence in a more limited prohibition, which would only exclude the monstrous treatment of clones?
In short, where do we get the idea that a limited ban or a strict legal framework for this biotechnical innovation would not be respected?
Is it based upon the psychological hypothesis that the desire to dominate or to exploit our fellows has no bounds?
Is it based upon the metaphysical hypothesis that humans have a propensity to always do the worst?
Is it based upon the sociological hypothesis that we are subject to crazed rules of competition, which oblige us to go ever further in technical innovation without heeding the human harm?
In the public debate, this is not spelled out. Nor is it among the philosophers who take an interest in cloning.
CAN WE DEFEND THE RULE “LIKE CASES MUST BE TREATED ALIKE” WITHOUT SUCCUMBING TO THE SLIPPERY SLOPE ARGUMENT?
At first glance, it is logically impossible to defend the rule R3, “like cases must be treated alike,” without conceding the slippery slope argument.
Indeed, defending “like cases must be treated alike” amounts to not taking into account certain differences that are not morally meaningful. It is this that justifies the passage from one case to another. Thus, we can consider that there is no morally meaningful difference between the fact of ridding yourself of an intruder who wishes to immobilize you for nine months and certain forms of voluntary interruption of pregnancy. We will conclude, according to rule R3, that the two cases should, from a moral point of view, be treated in the same way.
18Now, this is also what we seem to do when we deploy the slippery slope argument. We assert that the legislation covering assisted suicide is akin to permission to exterminate the poorest people in our hospitals, and that we will necessarily pass onto the second once the first is accepted.
But, in reality, no defender of “treat like cases alike” suggests that we pass, in fact, necessarily, from one similar case to another. The comparison is purely conceptual. It is never a question of a “natural progression.”
Asserting, for example, that there is no meaningful moral difference between a masseur and a sex worker absolutely does not suggest that, if you start out as a masseur in a physiotherapist’s consulting room, you will necessarily end up doing tricks in Central Park!
The rule “treat like cases alike” is conceptual. It has nothing to do with the slippery slope argument, which envisages a “natural progression.”
QUESTIONS REGARDING R4
When you are ordered to shut the door even though you are in the very act of shutting it, and no one giving you the order knows it, this order is redundant from the conceptual point of view, and intellectually pointless. It is stupid, in a certain sense.
When you are forbidden to leave your bedroom even though you have absolutely no intention of doing so because you have the flu, and no one forbidding you knows it, this prohibition is redundant from the conceptual point of view, and intellectually pointless. It is stupid, in a certain sense.
But that does not mean that this order and this prohibition express nothing, that they have no practical function.
Ordering someone to do what they are in the very act of doing can sometimes serve to mark approbation or express support, as when we yell at a boxer in the act of finishing off his opponent: “Finish him off!”
19Prohibiting someone from doing what he did not wish to do can have the same positive function of support or approbation, as when someone says to you, “don’t go out, you’ve got the flu,” when you had absolutely no intention of going out.
But, in many cases, ordering someone to do what he is in the very act of doing or prohibiting him from doing what he did not wish to do can have a less sympathetic function. It may involve an act that humiliates and thus underlines a certain state of domination or subordination.
Is not ordering someone to do what he was in the very act of doing a way of telling him that his will counts for nothing?
Is not prohibiting someone from doing what he did not wish to do a way of telling him that his will counts for nothing?
