to life even before it has been subjected to the process of injection and transformation. The possibility of transforming kittens into persons will not make it any more wrong to kill newborn kittens than it is now.
Thirdly, in view of the symmetry principle, if it is not seriously wrong to refrain from initiating such a causal process, neither is it seriously wrong to interfere with such a process. Suppose a kitten is accidentally injected with the chemical. As long as it has not yet developed those properties that in themselves endow something with a right to life, there cannot be anything wrong with interfering with the causal process and preventing the development of the properties in question. Such interference might be accomplished either by injecting the kitten with some “neutralizing” chemical or simply by killing it.
But if it is not seriously wrong to destroy an injected kitten which will naturally develop the properties that bestow a right to life, neither can it be seriously wrong to destroy a member of Homo sapiens which lacks such properties, but will naturally come to have them. The potentialities are the same in both cases. The only difference is that in the case of a human fetus the potentialities have been present from the beginning of the organism’s development, while in the case of the kitten they have been present only from the time it was injected with the special chemical. This difference in the time at which the potentialities were acquired is a morally irrelevant difference.
It should be emphasized that I am not here assuming that a human fetus does not possess properties which in themselves, and irrespective of their causal relationships to other properties, provide grounds for ascribing a right to life to whatever possesses them. The point is merely that if it is seriously wrong to kill something, the reason cannot be that the thing will later acquire properties that in themselves provide something with a right to life.
Finally, it is reasonable to believe that there are properties possessed by adult members of Homo sapiens which establish their right to life, and also that any normal human fetus will come to possess those properties shared by adult humans. But it has just been shown that if it is wrong to kill a human fetus, it cannot be because of its potentialities. One is therefore forced to conclude that the conservative’s potentiality principle is false.
In short, anyone who wants to defend the potentiality principle must either argue against the moral symmetry principle or hold that in a world in which kittens could be transformed into “rational animals” it would be seriously wrong to kill newborn kittens. It is hard to believe there is much to be said for the latter moral claim. Consequently one expects the conservative’s rejoinder to be directed against the symmetry principle. While I have not attempted to provide a thorough defense of that principle, I have tried to show that what seems to be the most important objection to it – the one that appeals to a distinction between positive and negative duties – is based on a superficial analysis of our moral intuitions. I believe that a more thorough examination of the symmetry principle would show it to be sound. If so, we should reject the potentiality principle, and the conservative position on abortion as well.
VI Summary and Conclusions
Let us return now to my basic claim, the self‐consciousness requirement: An organism possesses a serious right to life only if it possesses the concept of a self as a continuing subject of experiences and other mental states, and believes that it is itself such a continuing entity. My defense of this claim has been twofold. I have offered a direct argument in support of it, and I have tried to show that traditional conservative and liberal views on abortion and infanticide, which involve a rejection of it, are unsound. I now want to mention one final reason why my claim should be accepted. Consider the example mentioned in section II – that of killing, as opposed to torturing, newborn kittens. I suggested there that while in the case of adult humans most people would consider it worse to kill an individual than to torture him for an hour, we do not usually view the killing of a newborn kitten as morally outrageous, although we would regard someone who tortured a newborn kitten for an hour as heinously evil. I pointed out that a possible conclusion that might be drawn from this is that newborn kittens have a right not to be tortured, but do not have a serious right to life. If this is the correct conclusion, how is one to explain it? One merit of the self‐consciousness requirement is that it provides an explanation of this situation. The reason a newborn kitten does not have a right to life is explained by the fact that it does not possess the concept of a self. But how is one to explain the kitten’s having a right not to be tortured? The answer is that a desire not to suffer pain can be ascribed to something without assuming that it has any concept of a continuing self. For while something that lacks the concept of a self cannot desire that a self not suffer, it can desire that a given sensation not exist. The state desired – the absence of a particular sensation, or of sensations of a certain sort – can be described in a purely phenomenalistic language, and hence without the concept of a continuing self. So long as the newborn kitten possesses the relevant phenomenal concepts, it can truly be said to desire that a certain sensation not exist. So we can ascribe to it a right not to be tortured even though, since it lacks the concept of a continuing self, we cannot ascribe to it a right to life.
This completes my discussion of the basic moral principles involved in the issue of abortion and infanticide. But I want to comment upon an important factual question, namely, at what point an organism comes to possess the concept of a self as a continuing subject of experiences and other mental states, together with the belief that it is itself such a continuing entity. This is obviously a matter for detailed psychological investigation, but everyday observation makes it perfectly clear, I believe, that a newborn baby does not possess the concept of a continuing self, any more than a newborn kitten possesses such a concept. If so, infanticide during a time interval shortly after birth must be morally acceptable.
But where is the line to be drawn? What is the cutoff point? If one maintained, as some philosophers have, that an individual possesses concepts only if he can express these concepts in language, it would be a matter of everyday observation whether or not a given organism possessed the concept of a continuing self. Infanticide would then be permissible up to the time an organism learned how to use certain expressions. However, I think the claim that acquisition of concepts is dependent on acquisition of language is mistaken. For example, one wants to ascribe mental states of a conceptual sort – such as beliefs and desires – to organisms that are incapable of learning a language. This issue of prelinguistic understanding is clearly outside the scope of this discussion. My point is simply that if an organism can acquire concepts without thereby acquiring a way of expressing those concepts linguistically, the question of whether a given organism possesses the concept of a self as a continuing subject of experiences and other mental states, together with the belief that it is itself such a continuing entity, may be a question that requires fairly subtle experimental techniques to answer.
If this view of the matter is roughly correct, there are two worries one is left with at the level of practical moral decisions, one of which may turn out to be deeply disturbing. The lesser worry is where the line is to be drawn in the case of infanticide. It is not troubling because there is no serious need to know the exact point at which a human infant acquires a right to life. For in the vast majority of cases in which infanticide is desirable, its desirability will be apparent within a short time after birth. Since it is virtually certain that an infant at such a stage of its development does not possess the concept of a continuing self, and thus does not possess a serious right to life, there is excellent reason to believe that infanticide is morally permissible in most cases where it is otherwise desirable. The practical moral problem can thus be satisfactorily handled by choosing some period of time, such as a week after birth, as the interval during which infanticide will be permitted. This interval could then be modified once psychologists have established the point at which a human organism comes to believe that it is a continuing subject of experiences and other mental states.
The troubling worry is whether adult animals belonging to species other than Homo sapiens may not also possess a serious right to life. For once one says that an organism can possess the concept of a continuing self, together with the belief that it is itself such an entity, without having any way of expressing that concept and that belief linguistically, one has to face up to the question of whether animals may not possess