Kai Nielsen’s Support of Consequentialism and Rejection of Deontology
Adrian Pritchett

Kai Nielsen defended consequentialism and showed how it can still agree with commonsense, deontological convictions in his article “Traditional Morality and Utilitarianism.” His article focused on closing the gulf between consequentialism and deontology by showing how closely they can agree, and he further evaluated the systems and found that consequentialism as he sees it should be practiced is morally superior to traditional deontology. First, this essay will explain his argument that consequentialism squares with the commonsense convictions of deontology, and second, it will show how Nielsen arrived at the conclusion that consequentialism is a good moral system while deontology is faulty.

To show that consequentialism squares with the commonsense moral rules used by deontology or “moral absolutism,” Nielsen assumes, as many do, that outside of cases where one may has to choose the lesser of two evils, consequentialists generally make the same moral decisions as deontologists. He alluded to this general understanding when he wrote that “a consequentialist has very good utilitarian grounds for being so appalled” at acts like the “judicial execution” or “punishment, torture, and killing of the innocent,” but we should consider this assumption to be a premise in his argument so that the argument will be represented here with its full force.

Nielsen’s next major premise is that if a consequentialist is faced with a decision from which the overall value of the consequences is unclear, then consequentialism should yield to the relevant deontological rule. That is to say, if it is possible that violating a deontological rule to bring about greater good may lead to bad consequences in the future, or if that violation may not with certainty bring about greater good, then the agent should follow the deontological rule. This is a case where consequentialism does not have enough empirical evidence to make a proper decision. Nielsen’s example here was the classic case of whether a justice system should frame and execute an innocent person in order to appease threatening rioters that are upset about a crime for which the actual perpetrator cannot be found. Although the classic consequentialist viewpoint may say that framing an innocent person is acceptable because the mob will cause greater loss of life, Nielsen contends that in such a situation the overall consequences from framing an innocent person are usually unpredictable and therefore an innocent person should not be framed. In such a situation, appeasing the mob by finding a scapegoat could backfire: people might learn of the miscarriage of justice and riot again, or respect for the justice system might be undermined, which would lead to rampant crime in the future. Such uncertainty is a consequentialist reason to follow the deontological principle of not harming the innocent instead of making what looks like a shortsighted consequentialist move.

We can now derive that consequentialism generally follows deontology because a third major premise of Nielsen’s is that this would follow if, firstly, consequentialism often agrees with deontology, and, secondly, consequentialism must sometimes yield to deontological rules. These conditions have been met, and it is now clear how these two theories frequently yield the same results. However, having similar results is not enough to show that these two theories square with each other.

Next, Nielsen posits that the deontological exceptions prescribed by consequentialism can actually be universalized. To support this, Nielsen claims that deontological types of rules should be regarded as universal only up to the point that certain conditions exist; when other conditions exist, the rules are inapplicable and are always inapplicable when those other conditions exist. Deontological rules should not stand alone as blanket prohibitions but should be reformulated as conditionals that maintain prohibitions until certain extenuating circumstances exist.

The fifth major premise is that if consequentialist exceptions can be universalized, then consequentialism is an a priori theory. Deontologists generally claim that only their theory is an a priori theory because the proscriptions and duties they espouse actuall exist as universal moral facts and, conversely, consequentialism is strictly a posteriori due to its emphasis on empirical facts. However, Nielsen has made the case that consequentialism can be universalized, so we can now derive that it is indeed an a priori theory as well.

Finally, our last premise is that consequentialism squares well with deontology if two conditions are met: (1) that consequentialism generally follows deontology, and (2) consequentialism can be construed as an a priori theory, which deontology is. Consequentialism does generally follow deontology, which was derived from the first three premises, and consequentialism is a priori, which we derived from the fourth and fifth premises. So, from this argument, it follows that consequentialism does indeed square well with deontology.

What does it mean that consequentialism squares with deontology? Nielsen set out to show that consequentialism agrees with “commonsense moral convictions,” which are the sort of rules that deontologists support. Consequentialism can satisfy what many believe are two important requirements of a moral theory since it can be universalized and generally match agents’ intuitions. This is an important conclusion since consequentialism is often regarded as an untenable theory with extreme, unacceptable implications.

Throughout the article, Nielsen concurrently argues that deontology should be rejected but that consequentialism is viable. We may reconstruct his argument as follows: Deontology, as a morally absolute theory, makes mistakes. Likewise, an absolutist form of consequentialism also makes mistakes. So absolutism is wrong. Unfortunately, deontology can only be formulated as some type of “moral absolutism,” while consequentialism can be flexible. Therefore, deontology should be rejected, and by rejecting deontology we are left with consequentialism as a viable theory.

Nielsen relied heavily on examples to support his first premise that deontology makes mistakes. He discussed warfare to show how it is not the case that one is necessarily morally corrupt if he or she knowingly kills the innocent while making moves to kill combatants, but this point would not have been salient without having seen the movie he referred to, The Battle of Algiers. Nielsen did present an effective example, though, with the “case of the innocent fat man.” In this thought experiment, a fat man is leading a group of people out of a cave when he gets hopelessly stuck in the opening. There is a rising tide that will cause everyone inside the cave to drown unless they can get out. The only option for removing the fat man is to blast him out with dynamite that someone happens to have. Nielsen explains that the deontologist would hold that the fat man must not be blasted and killed because this would violate the prohibition against killing and it is only nature responsible for everyone else drowning. Nielsen challenges this principle by declaring that anyone in such a situation, including the fat man, should understand that the right thing to do is blast the fat man out in order to save the many lives in the cave. Furthermore, the deontologist exhibits “moral evasion” whenever he stands idly by and allows a greater tragedy than is necessary to occur. Nielsen explains that this is the kind of example that highlights the corrupt nature of deontology.

The two major arguments presented here could have been construed differently, either by formulating the premises differently or by organizing the claims into one argument or even more than two. Nielsen clearly describes his two main points — essentially, that consequentialism is viable and deontology is not — but the structure of the article comprises several short arguments that rely on the reader’s careful attention to see how they support those two main points.

The major question that Nielsen left unanswered is exactly what kind of consequentialism is viable. He clearly rejects an absolute form since he advocates avoiding consequentialist-style actions when empirical evidence is scant, but one is left wondering where to go from there. Nielsen seems to support a sophisticated blend of rule-consequentialism and act-consequentialism: a moral agent should generally follow the same rules that deontologists use except that rules can be broken when it is empirically clear that the agent will achieve better consequences by doing so (even though it would just be the lesser of two evils).

A common objection to consequentialism, that agents are burdened with duties to help others at the expense of their own happiness, was not even addressed. This in itself seems to be one form of absolutism that riddles consequentialism in general. Nielsen made it clear that one should not be absolute about insisting on weighing consequences when they are barely known, but would he reject this notion as well? It is not clear that this absolutism, of always valuing the good of others over the agent’s own self, is separable from the concept of consequentialism; so it is not clear that consequentialism can escape absolutism as Nielsen concluded in the second argument recounted here.