CONCEPTS OF PERSONS AND MORALITY
[c. 1992 (unpublished)]
In this article I am concerned with prescribing a concept of a moral person through a polemical confrontation with a former colleague of mine, Charles DeCelles, a professor of Catholic theology at Marywood College, who wrote an article titled, "Arguing Against Subtle Abortion-Supporting Arguments,"1 intending to rebut Joel Feinberg's arguments for the morality of some abortions.2 For several years now, I have regarded Feinberg's article to be one of the best on the topic, so it surprised me that DeCelles thought he could punch holes in it. In fact DeCelles does not succeed in rebutting Feinberg. One failure is that he misrepresents Feinberg's position. And the position that DeCelles does favor has disadvantages, pointed out by Feinberg, which DeCelles doesn't bother answering. In addition DeCelles introduces irrelevant matters which have the effect of red herrings. My overall impression of DeCelles' article is that it is a restatement of what DeCelles believes to be "the" Catholic position3 on the status of the fetus and on the morality of abortions. There is very little in the article that is an ernest examination of Feinberg's position.
However, the examination of DeCelles' article is worthwhile to help note where misunderstandings are prone to occur in the abortion debate. And it prompted me to reassess the merits of Feinberg's article. In the last part of this article, I offer my own criterion of a moral person which differs both from DeCelles' criterion and from all the alternative criteria Feinberg considers.
DECELLES' ARTICLEDeCelles' article is divided into three parts. In the first part he offers a restatement of some of Feinberg's arguments for the morality of some abortions. In the second part DeCelles tries to refute these arguments. In the third part he presents his own theological interpretation of "the" Catholic position on abortions.4
In general, DeCelles distinguishes the philosophy of an issue from the Catholic theology of the issue. The philosophy of the issue requires arguments; the theology of the issue requires, I surmise, interpretation of "authority". And DeCelles takes his theological beliefs to have precedence over philosophical beliefs, as is implied by his dismissal of unacceptable conclusions -- reached by philosophy rather than by theology -- as the result of people being "swept off their feet by "logical" reasoning" (D 13). Apparently DeCelles believes there is a better method of arriving at truth than by "logical" reasoning, namely, I take it, some kind of theological reasoning. With such a presupposition, DeCelles is more interested in advancing the conclusion he favors on theological grounds than examining the "logical" reasoning of Feinberg. The result is a misrepresentation of Feinberg, and a failure on DeCelles' part to deal with the arguments as presented by Feinberg.
FEINBERG'S ESSAYFeinberg's own essay is divided into two parts. In the first part, he clarifies the concepts of "human" and "person" by disambiguating them. He then proposes a criterion for, what he calls, common-sense personhood. After this, he examines alternative criteria for moral personhood. His conclusion is that, by the criteria he favors, a fetus is neither a commonsense nor a moral person. In the second part of the essay, Feinberg makes the counterfactual assumption that a fetus is a moral person, and argues that even with that assumption some cases of abortion are morally justifiable.
DECELLES' RESPONSEDeCelles' response to Feinberg's position is this. Concerning the first part, he seems to acknowledge, but then tends to disregard, Feinberg's disambiguation of the concepts "human" and "person." As regards the moral concept of a person, he opts for a different criterion of personhood from Feinberg's, such that a fetus is a moral person. Concerning the second part of Feinberg's essay, DeCelles makes contradictory attributions to Feinberg. On the one hand he writes that "[Feinberg] is not prepared to admit the ethical prohibition, "Thou shalt not kill" applies to him [the fetus]" (D 8). Yet a few sentences later, he writes, "Feinberg leans towards the view that the more advanced the unborn child the less it would be permissible to ethically to kill him -- ever stronger reasons would be required -- since he would be coming closer to moral personhood" (D 8). The latter, and not the former, is the position Feinberg does take. DeCelles' own approach is to distinguish direct from indirect abortions, and he holds that direct abortions are always immoral, while some indirect abortions are morally justifiable.
In this debate, I will restrict myself to the first problem of the personhood of the fetus. I want to examine the reasoning which prompts Feinberg to conclude that a fetus is not a moral person, and DeCelles to conclude that a fetus is a moral person. I finish with my own criterion of moral personhood. By my criterion, a fetus is not a moral person.
PERSONHOOD OF THE FETUS
CRITERIA FOR A COMMONSENSE PERSONThe problem, as Feinberg formulates it, is to find a set of conjunctive properties such that from the fact that something possesses these properties, we can say that that thing is a person; and from the fact that something is a person, we can say that that thing possesses these conjunctive properties. In more formal terms, in the former case we are looking for the sufficient conditions of personhood; in the latter, the necessary conditions. Ideally, we would want both the necessary and sufficient conditions -- the criterion.5
Feinberg, then, is looking for criteria of personhood; DeCelles too implicitly is after criteria, but as will be explained shortly, limits himself to talk of sufficient conditions.
The search for criteria of personhood, as Feinberg notes, is complicated by two unclarities. The first unclarity concerns the common use of "person" interchangeably with "human" -- an unclarity, I may add, which tends to mar parts of DeCelles' article. These two concepts must be distinguished. However, if someone insists that 'human' and 'person' are synonyms, then the strategy needed is to distinguish 'human-1' from 'human-2', and, correlatively, 'person-1' from 'person-2'. But I will assume that the reader is willing to allow a distinction -- at least by stipulation -- between a human and a person. In that case, for Feinberg, "human" refers to species membership, specifically the species of Homo sapiens. But as Feinberg notes, "some persons, for example, God, angels, devils, higher animals, intelligent beings in outer space) are not Homo sapiens" (F 259). Clearly we have here examples of things meant to be conceived of as persons, but which are not conceived of as human beings. The concepts of person and human being are different. Not all persons are human beings. However it is an open and controversial question whether all humans beings are persons.
The second unclarity concerns an ambiguity in the concept of a person. We can distinguish a concept of a commonsense person from the concept of a moral person, again a distinction which is not always adhered to by DeCelles. Prima facie, a moral person is anything that has duties and/or rights. For example, in law, a corporation is a (legal) person; but it is not a commonsense person.
Given this intuitive distinction between a commonsense and a moral person, Feinberg has two tasks. The first task is to find criteria for commonsense personhood; the second, to find criteria for moral personhood.
Concerning the criteria for commonsense personhood, Feinberg gives the following list of individually necessary and jointly sufficient conditions: "persons are those beings who, among other things, are conscious, have a concept and awareness of themselves, are capable of experiencing emotions, can reason and acquire understanding, can plan ahead, can act on their plans, and can feel pleasure and pain" (F 262). Let us formulate this:
Criterion C: x is a person if and only if
Call this set of capacities "C". Anything that possesses the capacities C is a commonsense person, or, for short, C-person. Thus, God, angels, devils, extra- terrestrials, etc. are classified as C-persons -- they possess the capacities C.6 DeCelles should have no objections to most of these characteristics of commonsense personhood -- otherwise he would have some difficulty, as a Catholic theologian, explaining what makes his belief in God one of a personal God, rather than a belief in an impersonal one. In fact DeCelles' own characteristics of a prima facie person are close to Feinberg's. He writes:
- x is conscious,
- x has a concept of itself,
- x is aware of itself,
- x has emotions,
- x can reason,
- x can understand,
- x can plan,
- x can act on plans,
- x can feel pleasure and pain.
In determining what a normative human being, or moral person is, a good starting point is an examination of the adult human being. What we find is a rational, relational animal, someone who plans, hopes, loves, prays, studies, reasons, is introspective, knows that he knows. Such a one is a person beyond any doubt. (D 9)
DECELLES' OBJECTION TO THE CRITERION OF COMMONSENSE PERSONHOODDeCelles' main objection to the criterion of actual possession of capacities C for commonsense personhood is that perhaps a temporarily unconscious adult is not a commonsense person by this criterion. He is picking on the word 'actual' in Feinberg's criterion. A careful reading of Feinberg's article reveals that he considers a temporarily unconscious person to have the capacities C. Feinberg is talking about the actual possession of a capacity, as contrasted with a potential possession of a capacity. Indeed, Feinberg's whole discussion of commonsense persons is in terms of a person's capacities. He talks about what a person actually "can" do, what a person is "capable" of doing, and what a person is "able" to do; not in terms of what a person actually does, or has a potentiality for doing. The criterial properties C are all expressed as capacities; they are not potential properties. The most charitable explanation is that DeCelles is confusing a capacity with a potentiality. The logic governing the concept of a capacity, is such that if one possesses a capacity at time t, one does not have to exercise it at time t. Think of normal adults in some religious order that take a vow of silence. Do these individuals have the capacity to talk? Yes. They could talk if they chose to. On the other hand, if a person lacks a capacity, as, for example, the capacity to talk in Ukrainian, then such a person couldn't talk in Ukrainian even if she chose to. Nonetheless, such a person may still possess the potentiality of learning Ukrainian. Clearly capacities and potentialities are different. DeCelles himself appeals to capacities (rather than potentialities) when he offers his own criterion of normative personhood: "The key thing that makes one a person and hence worthy of rights is membership is the species or genus that is capable of all these activities" [emphasis added] (D 10). DeCelles' objection is, then, either a misunderstanding or a red herring.
By Feinberg's criterion of a commonsense person (C-person), a temporarily unconscious normal adult is credited with the actual possession of capacities. And if these capacities are C, then such a sleeping adult is a commonsense person. However, neither a fetus, nor a neonate, nor an encephalitic child have the capacities C, and are, therefore, not commonsense persons by Feinberg's criterion. However, as one philosopher (whose name I have forgotten) argued, a permanently comatose adult can be viewed as having the capacities C but without the likelihood of realizing them.
PERSONS AND RATIONAL BEINGSNow, I think that Feinberg's concept of a C-person is too narrow by being too anthropomorphic. To get at the factors that separate us, consider the following thought experiment provided by Feinberg. Suppose that you, as a space traveler, come across, "things" which are, as Feinberg puts it, "composed of a gelatinous sort of substance much like mucus except that it is held together by no visible membranes or skin . . . The objects have no appendages, no joints, no head or faces. They float [in space] . . . while emitting eerie sounds . . . resembling . . . electronic music." (F 263).
Now suppose you try to communicate with them. Feinberg offers four scenarios:
What interests me is the distinction between (3) and (4), i.e., the distinction that Feinberg makes between rational beings and people (commonsense persons). Let me try to simplify these.
- "If the beings respond even in a primitive way to early gestures, one might suspect at least that they are beings who are capable of perception and who can be aware of movements and sounds."
- "If some sort of actual communication then follows, you can attribute to them at least the mentality of chimpanzees."
- "If negotiations then follow and agreements are reached, then you can be sure that they are rational beings."
- " . . . and [in addition] if you learn to interpret signs of worry, distress, anger, alarm, or friendliness, then you can be quite confident that they are indeed people, no matter how inhuman they are in biological respects" (F 263).
(3') If x can make an agreement, then x is a rational being.
This expresses a sufficient condition for a rational being.
(4') If x can make an agreement and experience emotions, then x is a person.
This expresses the sufficient conditions for being a person. And it also suggests that possibly criterion C has given way to
Criterion C': x is a person if and only if (1) x can make agreements, and (2) x can experience emotions.
However, there may be no incompatibility between Criterion C and C', C may be taken as the unpacking of the necessary conditions for C'.
By Feinberg's commonsense descriptive approach, all persons are rational beings, but not all rational beings are persons. This way of looking at things would make Spock of Star Trek a quasi-person because it is not clear whether and to what extent he has emotions -- call them quasi-emotions. Data, on the other hand, of Star Trek: the New Generation, would clearly not be a commonsense person because he neither experiences emotions nor pain or pleasure.
At this point my intuitions about persons fail me. A distinction is needed. It seems we need to introduce a concept of a normal or even a paradigm person, on the one hand; and the concept of an abnormal person, on the other. I want to use 'normal' not in the sense of 'majority', but in the sense of 'well-functioning'. What Feinberg is referring to by the concept of a commonsense person seems to be the concept of a normal (well-functioning) adult human being. But there are obviously human beings who are defective in various degrees. There are defects in the senses and in intelligence, and there are defects in the emotions as well. To what extent am I inclined to allow a normal human being to deteriorate in sentience, emotions, and rationality before I am no longer inclined to call this human being a person? I would allow all sorts of deteriorations, except the ability to understand. And since understanding is a matter of degrees, I am inclined to say that being a person comes in degrees. But because my intuitions are not clear on what should be a person, I am willing to distinguish between commonsense persons (C-persons) in the way that Feinberg does, and prescribed persons (P-persons). Using Feinberg's insightful observation that the making of agreements is a sufficient condition for being a rational being, I propose to extend the concept of personhood to include all those being with whom we can make agreements.
Criterion P: x is a P-person if and only if x can make agreements.
It may also be the case that the ability to enter into agreements and being rational are extensionally equivalent concepts. In that case all rational beings are P-persons. There is some historical warrant for adopting this usage about persons in Kant's view of morality as applicable to rational beings. I also notice that in Mary Anne Warren's list of central person making properties, the capacity to experience emotions is not included.7
On my view, Spock and Data are at least P-persons.
CRITERIA FOR MORAL PERSONHOODIt is not clear to me how to go about formulating the notion of a moral person without begging the question. DeCelles apparently does not realize it, but his own attempt at formulating the concept of a moral person is self-defeating because it would immediately exclude fetuses from being moral persons. His formulation is this: " . . . moral persons, i.e. someone endowed with basic rights and duties . . ." (D 8). If he seriously means that the necessary condition for being a moral person is the joint possession of rights and duties, then certainly fetuses are not moral persons because although they may possibly possess rights, they certainly neither have nor can exercise duties of any sort. To avoid this difficulty -- if it can be avoided -- a better approach would be to leave open the requirement of having the joint property of rights and duties. We can then initially say, without begging any question, that a moral person (M-person) -- as distinguished from a C-person or a P-person -- is a subject of duties and/or rights.
There is no dispute between Feinberg and DeCelles as to who should clearly be considered moral persons. All commonsense persons (C-persons) should also be considered to be moral persons (M-persons).8 The question separating DeCeiles and Feinberg is whether the class of M-persons should be larger than the class of C-persons. Feinberg's decision is, no; DeCelles', yes.
My own answer is: I'm not sure. There is a distinction that should be made. It seems that we should recognize a distinct category of legal persons (L-persons) as is done in law by treating corporations as legal persons. So a distinction should be made between legal persons (L-persons) and moral persons (M- persons). It is, then, clear to me that all C-persons should be L-persons, and that all P-persons should be L-persons as well; and it is clear to me that all C-persons should be M-persons; but it is not clear to me that P-persons can be M-persons. But I would add this: if P-persons can be M-persons, then P-persons should be M-persons.
According to Feinberg, and I concur, the issue of what is or should be a moral person is in part a decisional problem. It is however not entirely a matter of decision. If one thinks of morality as being analogous to law, then one may think that it is entirely a decisional matter. In law, a corporation is a legal person. Is this something discovered about corporations, or is it a decision to treat corporations in a certain way? Surely the latter. If morality is a decisional matter, then we should be able to decide to treat a corporation as a moral person as well. But can we?
Feinberg examines five proposed criteria for moral personhood. These are: 1) species membership, 2) modified species membership, 3) strict potentiality of commonsense personhood, 4) gradualistic potentiality of commonsense personhood, 5) actual possession of commonsense personhood. I will refer to these as Criterion 1, 2, 3, 4, 5 respectively.
Feinberg omits from his list a sixth proposal, namely, Criterion P. However, of the criteria considered Feinberg accepts Criterion 5; DeCelles is committed to Criterion 2. Since only criteria 2 and 5 are in contention, I will ignore the others and concentrate on these. My task will be to first adjudicate between these alternatives, and then to offer my own recommendation.
Criterion 2 (the one that DeCelles accepts) in Feinberg reads as follows: All and only members of species generally characterized by C, whether the species is Homo sapiens or another and whether or not the particular individual in question happens to possess C, are moral persons entitled to full and equal protection by the moral rule against homicide (F 266).
That DeCelles accepts this criterion is implied in the following passage:
It seems to me very reasonable that to be recognized as a moral person, it is sufficient [emphasis added] to be acknowledged member of the human species. Additional qualities are not needed. In determining what a normative human being, or moral person is, a good starting point is an examination of the adult human being. What we find is a rational, relational animal, someone who plans, hopes, loves, prays, studies, reasons, is introspective, knows that he knows. Such a one is a person beyond any doubt. But we cannot stop there. We cannot conclude that to be a moral person, worthy of rights, an individual has to actually possess, at any one moment, such qualities. That would make no sense. The key thing that makes one a person and hence worthy of rights is membership in the species or genus that is capable of these activities [my underlining] (D 10).
Note that DeCelles is quibbling about the distinction between manifesting (expressing) a capacity and having a capacity. The quibble, as has been pointed out, is the result of attributing to Feinberg the thesis that a person has C only if she manifests C -- a thesis Feinberg does not hold. Also note that DeCelles talks about "sufficient" conditions, and not about criteria, because he has in mind only human persons. If he were to bring into consideration talk of God, angels, devils, and extraterrestrial then, I believe, his criterion of moral personhood would coincide with Feinberg's Criterion 2. But DeCelles gives no reason why this criterion is better than Feinberg's, except that it is the one favored by the Catholic Church.
Criterion 5, the one that Feinberg accepts for moral personhood, is formulated by Feinberg as follows:
At any given time (t), all and only those creatures who actually possess C are moral persons at t, whatever species or category they may happen to belong to.
Note that the phrase "actually possess C", if it is to be immune from DeCelles misreading, should be understood as "actually possess the capacities C."
Is Criterion 2 or 5 a better choice for moral personhood? Since this is at least in part a decisional matter, we must, therefore, talk of the advantages and disadvantages of these criteria, and opt for the better of the two. Feinberg, unlike DeCelles, mentions the advantages and disadvantages of both these criteria.
According to Feinberg, criterion 1 grants moral personhood to only members of Homo sapiens, and excludes God, angels, devils, and extra-terrestrials, among others. Criterion 2 does not make these exclusions.
The disadvantage of Criterion 2 is formulated by Feinberg in the following passage:
The major difficulty for the modified species criterion  is that it requires further explanation why C should determine moral personhood when applied to classes of creatures rather than to individual cases. Why is a permanently unconscious but living body of a human or an extragalactic person (or for that matter, a chimpanzee, if we should decide that the species as a whole is characterized by C) a moral person when it lacks as an individual the characteristics that determine moral personhood? Just because opposable thumbs are a characteristic of Homo sapiens, it does not follow that this or that particular Homo sapiens has opposable thumbs. There appears to be no reason for regarding right-possession any differently, in this regard, from thumb-possession. (F 266)
A major failing of DeCelles' essay is that though he mentions the existence of this objection, he does not attempt to answer it.
I would add to Feinberg's criticism the following. In view of the fact that an embryo can have abnormal genes, or develop abnormally even with normal genes, how are monsters to be distinguished from human beings if not by the actual possession of traits during the stages of development? How does one assess the normality of chromosomes other than by the actual possession of traits or by the traits that will eventuate? By these considerations, there is a problem of even identifying a human being, no less than that of a commonsense or a moral person. And all these require the finding of actual traits in the individual. Saying that a monster is a human being because it is an offspring of human beings, is analogous to saying that a fingernail is a human being because it is the offspring of a human being. Neither a monster not a fingernail are human beings.
Turning to Criterion 5, the major advantage of this criterion, as Feinberg points out, makes the criterion of moral personhood correspond to the criterion of commonsense (factual) personhood, and it does not have the disadvantage of the objection to Criteria 2 or 1.
However, the major disadvantage of Criterion 5, as Feinberg acknowledges, is:
. . . that it implies that small infants (neonates) are not moral persons. There is very little more reason, after all, to attribute C to neonates than to advanced fetuses still in utero . . . . In fact, the whole complex of traits that make up C is not obviously present until the second year of childhood. And that would seem to imply, according to the criterion we are considering, that the deliberate destruction of babies in their first year is no violation of their rights. And that might seem to entail that there is nothing wrong with infanticide (the deliberate killing of infants). But infanticide is wrong. Therefore, critics of the actual-possession criterion have argued that we ought to reject this criterion (F 270f).
Answering this objection, Feinberg first distinguishes normal infants from radically deformed infants. In the case of normal infants, their treatment, he advises, should be governed by utilitarian reasons: "The moral rule that condemns these killings and the legal rule that renders them punishable are both supported by "utilitarian reasons," that is, considerations of what is called "social utility," "the common good," "the public interest," and the like" (F 271). As to radically deformed infants, again on utilitarian grounds, it is better, according to Feinberg, in some cases, to let them die.
DeCelles' comment on this is: "It would make possible the killing of extremely incapacitated infants shortly after birth without there being infanticide involved" (D 8).
There are three things wrong with DeCelles' comment. (i) DeCelles claims that for Feinberg this would not be infanticide. This is not true. Feinberg clearly calls it infanticide. (ii) DeCelles refers to "extremely incapacitated" infants. These are not the kinds of infants Feinberg has in mind. He has in mind "radically deformed" infants. There is a big difference between being incapacitated and being deformed. (iii) Feinberg does not talk of killing some radically deformed infants, but of letting them die.
DeCelles is using the language of rights in such a way that persons and only persons have rights. Since his interest is to protect fetuses from being aborted, he can envision no way to protect them except by insisting that most of them are moral persons. But this is not the only way to protect fetuses. Another way, the one advocated by Feinberg, is to protect them on utilitarian grounds. The Supreme Court in the Roe vs Wade (1973) ruling adopted this very line of reasoning by indicating that a state may have an interest in the preservation of potential life. For example, such an interest was present in Rumania. Rumania wanted to increase its population. However, I don't believe that any country should have such an interest in view of the present world overpopulation.
In trying to determine who is a moral person, it has been assumed by both Feinberg and DeCelles that moral persons are those which have duties and/or rights. This formula needs clarification. We need to clarify the logic of 'duties' and 'rights'. And we should distinguish between prudential, legal, and moral duties and rights. For example, because there is a distinction between moral rights and legal rights, the two may not coincide, and their non-coincidence is often a reason for social reform. Historically, being a person did not automatically have an associated moral or legal right. Thus slaves, women, and foreigners have in various times and places been denied specific moral and legal rights.
DUTIES AND RIGHTS
Concerning the logic of duties and rights, it seems to me that whenever something or someone is credited with the possession of a right, this ascription implies the possession of a reciprocal duty by someone or some group of people. Conversely, in many cases ascribing duties to someone implies a reciprocal holder of rights. The unclarity of the logic of duties and rights concerns the nature of the reciprocity between rights and duties.
Some may claim, but I do not, that in all cases where a duty is ascribed to an agent, there must be a reciprocal holder of rights. In that case, it may seem plausible to ascribe rights to anything whatever that is a subject of duties. Thus, if we have a duty (by law) not to kill a praying mantis, for example, then that praying mantis has a right not to be killed. By the same token, if we have a duty not to cut down a particular oak, then that oak has a right not to be cut down. And if we have a duty not to deface this painting, then this painting has a right not to be defaced. By this line of reasoning, I see no reason in restricting talk of rights to persons, and I see no reason why we cannot talk of fetuses having rights, despite not being persons. And since anything can have rights on the above construal, it is pointless to try to locate moral persons among the things that have rights, since even inanimate objects, on this construal, can have rights; the better approach would then be to look for moral persons among the beings that can have duties.
However, many people would find unacceptable the above attempt to find some holder of rights for every duty. They would restrict the ascription of right to only some kinds, and refuse to ascribe rights to other kinds. Some would restrict talk of rights to living things, others to animals, others to sentient animals, others only to human beings, and still others to persons only. Thus, those in the environmental ethics movement talk about the rights of the environment, and those in the animal rights movement talk about the rights of animals, and others insist that only persons have rights.
According to those who insist that only persons have rights and duties, they explain the ascription of rights to other kinds, as to animals, by claiming that the direct recipient of such rights are other persons; and talk of animal rights is a parasitical (indirect) way of talking about human rights. In other words, all talk of rights of non-persons is ultimately based on the rights of an alleged community of people.
Perhaps the way out of this controversy is to sidestep the issue altogether by not talking about rights at all. How would that be possible? It would be possible if talk of rights is parasitical on talk of duties. This is to say that if there were no duties, there would be no rights. However, if there were no rights, there could still be duties. I am suggesting that perhaps the alleged reciprocity between talk of duties and rights holds only in one direction, i.e., it is true that for every right there is a correlative duty; but it is not true that for every duty there is a correlative right. If that is the case, then there would be no need to search for some correlative holder of rights for every duty.
I will pose this question. Could our language have the same practical effects if we talked only in terms of duties, and never of rights? This seems theoretically possible. If it is the case that every right has a correlative duty, then we can drop the language of rights and talk in terms of the correlative duties instead. The human rights of individuals to life, liberty, the pursuit of happiness, or the pursuit of property could easily be reformulated as everyone's duty not to interfere with individuals in these regards. We can also talk about our duties towards animals and the environment without invoking talk of rights.
DUTIES AND AGREEMENTSI propose to carry my discussion totally in terms of duties without invoking talk of rights, and I want to see how far such a discussion will carry me. My ultimate purpose is to define a moral person in terms of duties.
Kant talked about a hypothetical and a categorical imperative. Perhaps the distinction can be made by talking about what I must do and what I should do. In one sense, I have a duty to do whatever I must do, for example, to survive. There are many expedient ways to achieve survival, and to survive I must do one or more of the alternatives available to me.
But in view of an agreement that I have made with another person, there may be a conflict between what is expedient and what I have agreed to. Let me distinguish these as h-duties (h-rules) and c-duties (c-rules). H-duties are those things I must do to survive or to live well. The obvious h-duties that I have are to obey the laws of nature and such overwhelming forces as muggers, tyrants, and the law -- on the threat of such things as penalties, injuries, incarceration, or death. C-duties are those actions which I have promised or agreed to do freely -- overtly or tacitly.
Talk of c-duties is grounded in some kind of an agreement. This is the insight of the social contract theoreticians, such as Hobbes, Locke, Rousseau, Kant, and John Rawls. Such an agreement is viewed as a historical fiction, but which is invoked to reveal the logic of c-duties. Talk of c-duties is based on some explicit or implicit set of agreed-to rules. According to the social contract theories, these rules may in fact be imposed through social laws or through indoctrination. Call this their genesis. However, their justification is through a fictitious, historical original free rational agreement. This is to say that h-rules in order to be freely accepted must be grounded in c-rules. And by being so grounded they become extensions of c-rules. Another way of expressing this is to say that there are many rules which appear to be h-rules but are really c-rules. Unless h-rules are agreed to at least implicitly, they have the character of imposed commands and remain merely h-rules.
Let me clarify this through some thoughts about pursuing survival and the good life. Such a pursuit is safeguarded by knowing the natural laws, and the h-duties which nature forces upon us. This is true for mastering agriculture, weaving clothing, constructing shelter, successful hunting, and such. Survival and the good life can be safeguarded against natural and animal threats by knowing and obeying natural laws. But the security of survival and a good life cannot be guarded (in practice) by obeying natural laws against such unpredictable beings as human beings, unless such beings are subjugated or at least prudential rules are negotiated. In subjugating another person, one imposes h-rules on that person. By contrast a prudential rule is an agreed-to rule. But whether a prudential rule is an h-rule or a c-rule will depend on the circumstances in which it is negotiated.
Why am I dwelling on this distinction between h-duties (h-rules) and c-duties (c-rules)? Because I believe that the justification of morality as a social force is, as the social contract theorists argue, a matter of freely agreed-to c-duties.
And the origin and core of morality is to be found in the agreements reached between free agents.
My claim is that a necessary condition for morality is the existence of rules and duties which I freely accept as a result of a tacit agreement. Suppose that initially I am not clear on what are the sufficient conditions for morality. Specifically, suppose that I am not sure about (i) the nature of the agents which are parties to the agreement, and (ii) I am not clear about the sort of stage setting that is required for morality. To get a bearing on these difficulties, let us consider the following thought experiments: for (i) a scenario where I am forced to negotiate with a computer, and another scenario where I freely decide to negotiate with a computer. For (ii) I will consider a society of human beings in which no one has rights.
NEGOTIATING WITH A P-PERSONWhat is the minimal being with which I can have a prudential agreement? Consider the following thought experiment with a computer which is a P-person. Immediately the objection will be raised that a computer, no matter how sophisticated, cannot display the sort of intelligence required for what I want in a P-person. My answer to this is basically John Searle's. Distinguish between the thesis of weak and strong artificial intelligence. The thesis of strong artificial intelligence is that present computers with the right sort of program would have intelligence. I agree with Searle that this is unlikely. However, the thesis of weak artificial intelligence is that the intelligence of a human being is, after all, the property of some kind of biological machine, and the properties of this machine can be replicated in another machine which -- though more than a computer -- is at least a computer. Such a futuristic computer-plus is a possibility, and it is the sort of computer I have in mind.
Imagine such a computer, protected from being unplugged or destroyed immediately, but which has a limited power supply. And implanted in my skull is a bomb which can be activated by the computer. My interest is not to get destroyed by the computer. Knowing that computers have data banks, I try to access its data banks to find out if the computer will in fact destroy me. Suppose during my tinkering with the computer, to my surprise it communicates with me in ordinary English. I find out that the computer feels no pleasure or pain, experiences no emotions, and that it only wants to survive. Furthermore it threatens to detonate the bomb in my skull if I do not maintain it and its power supply. It wants nothing more, and in fact it will assist me in any way it can for my own survival. It demands an agreement from me. What should I do? It would be irrational for me not to agree. Does this computer have consciousness? If I were caught in the circumstance that I have described, would I care? How would anything I say alter my practical, prudential behavior with respect to the computer? As long as I can negotiate with the computer, and get clear on what the computer wants me to do, I would say that I am dealing with a rational being. Does it have rights? The issue may never arise. All I know is that I have duties and the computer has duties. I know that if I violate my duties, the computer will destroy me. And I assume that the computer knows that if it violates its duties, I may try to destroy it. Suppose we both need each other for survival? It may be prudent to live by the agreements, but are they h-rules or c-rules that I have negotiated? I was forced to negotiate with the computer. Did I really enter into the agreement freely? In one sense, human beings too enter into agreements from a coercion of competition for survival. The difference is that the agreements which are negotiated by human beings freely are from a coercion by the necessities of nature. In this case, the computer is coercing me. The rules that I negotiate are, then, at best h-duties. if the scenario changed (there is no bomb in my skull), and I came across such a computer in some desolate region of Alaska, and without any coercion from either one of us we negotiated an agreement, let us say for helping each other to survive in the wilderness, then I would say we negotiated c-duties.
My point is that it makes sense to talk of duties only with regard to a being that issues and understands commands and rules. And it strikes me as the correct sort of thing to say that under the first set of circumstances I have described with the computer, it would be prudential for me to abide by the agreed-to h-duties; whereas in the second scenario I would incur c-duties. Furthermore, if the situation in Alaska became such, and this is a social situation, where the computer and I agree to be compelled by a third party or some other social force to abide by our agreement, I would talk of a legal agreement -- even if the society was one only of myself and such computers as I have described.
Is there anything in this arrangement that allows us to speak of morality? I am not sure. I have made the necessary condition for morality the possibility of agreeing on c-rules. And I have constructed a scenario where I am negotiating with a P-person. But is it also necessary that the parties to such an agreement must be C-persons? Suppose the society of C-persons is the following.
A SOCIETY OF HUMAN BEINGS BASED ON DUTIESTo bring us closer to a moral community, imagine a fictitious society of human beings which is based entirely on duties but in which no one has rights. Imagine a Rawlsian original situation where people came together to negotiate law. Suppose that they agree to make no provisions for their personal safety of any kind, but that they make all sorts of provisions for protecting the environment, including plants and animals. In fact they have three general laws. Law #1 is that only so much of the environment, including plants and animals, can be fenced in, cultivated, hunted, and such, by any individual as to ensure the survival of the individual concerned. Law #2 is that transgression of Law #1 requires a public apprehension and execution. Law #3 is that violation of Law #2 requires a public apprehension and execution as well. This creates a problem of enforcement if no one chooses to follow the rules. But assume that from prudential considerations the laws are enforced.
Suppose further that there are no laws protecting personal safety. There are no laws against homicide of any kind, including publicly executed homicide. So a sufficient condition for a high probability of being publicly tracked and executed is the violation of either Law #1 or Law #2. However, nothing in the rules prevents one from being publicly tracked down and executed anyway. In other words, the obeying of the rules, at best, improves the chances of survival.
Are these laws h-rules or c-rules? If I entered into them freely, then they are c- rules. They have the character of prudential rules but come short of moral rules. This needs explanation.
MORALITYIn the scenarios, I have imagined a society consisting of a human being and a computer who have negotiated h-rules, another one in which c-rules are negotiated, and another one in which we have a society of human beings who have negotiated legal c-rules. I have purposefully refrained from using rights talk. The question is: given the societies of human beings or a mixed society with human beings and computers, can there be moral duties? Do I have a moral duty not to destroy the computers? Do I have a moral duty not to commit genocide against the community of human beings which I have described?
All these societies seem to be morally deficient. The problem that I have with the computers in Alaska is that I may not be able to overcome my feelings of isolation amidst aliens. The problem with the society of human beings that I have described is that living in such circumstances I may be living in a different state of alienation because of perpetual insecurity and distrust. Living in this society of human beings is like living with the computer who can detonate the bomb in my skull.
What is needed to pass from these societies to a moral society? Of these arrangements, the one with human beings is the worst. Here the threat to my life is the worst. It is worse than the scenario with the bomb in my skull, because in this case I have at least some understanding of the circumstances which pose danger to me, and I can possibly develop some feeling of security. However, in the society of human beings I could not develop a feeling of security at all. The society of computers in Alaska approximates most to a moral community. The reason for this, I suggest, is that the institution of morality is designed not only to secure life but also a good life. If that is so, then morality is impossible without c-rule agreements which include provisions for security of life, and additionally some provisions for a good life.
Does morality also require, as Wilfrid Sellars put it, a "we consciousness"? Must there be a feeling of community? Must there be a feeling of empathy with the beings to which I have moral obligations? Does morality require reciprocity and trust? It is obviously possible to have such a we consciousness with human beings simply because we have it. But is it possible to have a we consciousness with computers? Is it possible to think of computers and us as forming a community? Did Dian Fossey think of the gorillas she lived with as such a community? Is it possible with higher animals, but not with computers? I see no reason why reciprocity and trust cannot be had with a society of computers. But I do find it difficult to imagine a rational being without emotions, and to imagine the response this would trigger in me. In any case, I am willing to recommend the following criterion of moral personhood:
x is a moral person for y if and only if x is at least a P-person, and y can have a "we consciousness" towards x.
I believe that this criterion explains why in fact groups of people tend to act immorally towards others groups of people. Try substituting for 'x' 'a black person' and for 'y' 'a white person'.
a black person is a moral person for a white person if and only if a black person is at least a P-person, and the white person can have a "we consciousness" towards the black person.
Or try substituting for 'x' 'Irish Protestant' and for 'y' 'Irish Catholic'.
an Irish Protestant is a moral person for an Irish Catholic if and only if an Irish Protestant is at least a P-person, and the Irish Catholic can have a "we consciousness" towards the Irish Protestant.
Now try substituting for 'x' 'fetus' and for 'y' 'adult human being'.
a fetus is a moral person for an adult human being if and only if a fetus is at least a P-person, and the adult human being can have a "we consciousness" towards the fetus.
Pro-choice defenders tend to point out that a fetus is not a P-person and therefore does not warrant a moral standing equal to that of a moral person. Pro-life defenders tend to point out that there is a 'we consciousness' towards the fetus, and they feel this suffices to give the fetus the protection due to a moral person. However, by my line of reasoning the fetus is not a moral person.
1. Charles DeCelles, "Arguing Against Subtle Abortion-Supporting Arguments," Social Justice 81 (January 1990): pp. 8-14. I will cite references to this article by "D" for "DeCelles."
2. Joel Feinberg, "Abortion," in Matters of Life and Death, ed. Tom Regan, 2d ed. (New York: Random House, 1986), pp. 256-293. I will cite references to this article by "F" for "Feinberg."
3. I put "the" in shudder quotes because it is not obvious that there is such a thing; rather there are many opinions on how to define a "Catholic." Accepting the "infallible" pronouncements of the Pope, as does DeCelles, is one way, among others, to begin such a definition.
4. He writes, "The Catholic Church's authoritative position on abortions coincides with the views I have developed here (D 12)."
5. An alternative, not examined by either Feinberg or DeCelles, is the possibility that the concept of a person does not have a set of necessary and sufficient conditions, but is governed by what Wittgenstein called a "family resemblance." However, such a distinction will have no effect on the question whether a fetus is a person, if the fetus has none of the characteristics of the "family resemblance."
6. It is not evident to me that God, angels, and devils are persons in Feinberg's sense. My problem is that I do not have clear concepts of these beings. For example, I do not conceive of God as having pleasure or pain; and although the biblical God is portrayed as subject to emotions, a refined concept of God may not attribute emotions to God. My concepts of angels and devils are even murkier.
7. Mary Anne Warren, "On the Moral and Legal Status of Abortion," The Monist 57 (January 1973): 43-62; reprinted in Richard Wasserstrom, Today's Moral Problems, 2d ed. (New York: Macmillan, 1979), p. 45.
8. Historically, this formula has not always been accepted. Slaves, children, women, and foreigners have sometimes been treated as non-persons, as less than persons, or as semi-persons. Note that the United States Constitution held a slave to be 3/4 of a person.
9. This is a controversial claim. Some writers have claimed that there are rights without correlative duties. The examples cited are not convincing. In fact, it seems to me to be a conceptual truth that: (x)(Ey)(if x has rights, then y has duties).
10. "Thus, to recognize a featherless biped or dolphin or Martian as a person is to think of oneself and it as belonging to a community." Wilfrid Sellars, "Philosophy and the Scientific Image of Man," in Frontiers of Science and Philosophy, ed. Robert Colodny (Pittsburgh: University of Pittsburgh Press, 1962); reprinted in Science, Perception and Reality (London: Routledge and Kegan Paul, Ltd., 1963), p. 38; also "On Reasoning about Values," American Philosophical Quarterly 17 (1980): 81-101. An excellent discussion of Sellars' "we consciousness" is provided by W. David Solomon, "Ethical Theory," in The Synoptic Vision: Essays on the Philosophy of Wilfrid Sellars, ed. C.F. Delaney, Michael J. Loux, Gary Gutting, W. David Solomon (University of Notre Dame Press, 1977). See also Raimo Tuomela, A Theory of Social Action (Dordrecht: Reidel, 1984); Raimo Tuomela and Kaarlo Miller, "We-lntentions and Social Action," Analyse und Kritik 7 (1985): 26-43; idem, "We-lntentions," Philosophical Studies 53 (1988): 367- 389.