Introduction
Many bioethicists have a neat conjuring trick up their sleeve, whereby they can instantly convert a philosophical conclusion into a policy prescription. In its simplest form, a philosophical argument is presented that some activity A is morally acceptable or unacceptable, and while the philosopher waves his or her magic wand to distract the reader, there is then an immediate elision in which A is seen as permissible or impermissible, and the argument proceeds from there to claim that A should be permitted or prohibited.
In this article I show that this piece of magic is, in most cases, an illusion, and that there is no valid and sound simple, easy, or immediate conversion from philosophical conclusion to policy prescription. If the trick could be performed, it would truly be “one of the finest spectacles in Nature,” on par with the grand leap of the whale described by Benjamin Franklin in 1765: “Whales, when they have a Mind to eat Cod, pursue them wherever they fly; and that the grand Leap of the Whale in that Chace up the Fall of Niagara is esteemed by all who have seen it, as one of the finest Spectacles in Nature!”Footnote 1 For, when a philosopher has a mind to make policy, he or she pursues it wherever it may fly, even leaping through bad argument.
In order to provide a focus for the analysis, I look at a particularly controversial philosophical conclusion that follows from the—among bioethicists—quite commonly held view that only persons have full moral status: that is, the view that infanticide of a healthy infant is not intrinsically or inherently wrong. I explicate why this philosophical conclusion, in which many philosophers seem to have absolute confidence, cannot easily be converted into a policy prescription, except by the magic of underhand elision.
Persons, Abortion, and Infanticide
If we take the view that only persons have full moral status and a right to life and combine this with an account of personhood that requires some higher mental functions, it follows straightforwardly that infanticide of a healthy infant is not intrinsically or inherently wrong. This was argued as long ago as 1972 by Michael Tooley, who, in his seminal article “Abortion and Infanticide,” showed that this conclusion follows from a particular version of rights theory;Footnote 2 he later expanded the argument in his 1983 book of the same name.Footnote 3
From a consequentialist perspective, the same philosophical conclusion has been reached by Jonathan Glover, John Harris, James Rachels, and Peter Singer—to mention just a few prominent scholars—in the 1970s and 1980s.Footnote 4 Some of these philosophers do discuss policy, but in a fairly modest way. Glover, for instance, writes at the end of his article on infanticide, “I have nothing to contribute to the emergence of this policy, beyond the attempt in this book to undermine some of the general views (the sanctity of life, the acts and omissions doctrine) that at present obstruct it. The policy must be worked out by those in a better position to do so.”Footnote 5 And Singer has, in a joint-authored book with Helga Kuhse, tried to develop a more detailed policy prescription for the infanticide of infants with severe disabilities, taking into account the many practical issues that such a policy would have to take into consideration.Footnote 6 Singer and Kuhse have been rightly criticized by Simo Vehmas for being discriminatory, because the fundamental personhood-based argument seems to apply to all infanticide, and not only to the infanticide of the disabled.Footnote 7 Nevertheless, Singer and Kuhse’s book, although imperfect, exemplifies the amount of detailed work that has to be done if a controversial philosophical conclusion is to be turned into an, at least in principle, implementable policy prescription.
But not all authors are as meticulous, and in a now-infamous 2012 article that essentially rehashes the arguments from the 1970s and 1980s (often without proper attribution!) we can read the following nice example of the grand leap of the whale: “On these grounds, the fact that a fetus has the potential to become a person who will have an (at least) acceptable life is no reason for prohibiting abortion. Therefore, we argue that, when circumstances occur after birth such that they would have justified abortion, what we call after-birth abortion should be permissible.”Footnote 8 We move with breakneck speed and no argument from “no reason” and “justified”—that is, the philosophical conclusion—to “permissible,” the policy prescription.
Why Is the Grand Leap an Underhand Conjuring Trick?
Why is the grand leap from philosophical conclusion to policy prescription an underhand and illegitimate conjuring trick? The general issues have been stated clearly by Tom Beauchamp and James Childress:
Public policy is often formulated in contexts that are marked by profound social disagreements, uncertainties, and different interpretations of history. No body of abstract moral principles and rules can determine policy in such circumstances, because it cannot contain enough specific information or provide direct and discerning guidance. The specification and implementation of moral principles and rules must take account of problems of feasibility, efficiency, cultural pluralism, political procedures, uncertainty about risk, noncompliance by patients, and the like.Footnote 9
Let us take this as our starting point and look at four specific issues, moving from the theoretical to the practical:
1. Reasonable disagreement about values and theories
2. General jurisprudential arguments
3. The differences between policymaking and philosophy
4. The messy world of implementation
Reasonable Disagreement about Values and Theories
Philosophical conclusions are unassailable when the argument has a valid logical form and the premises are true, that is, when the argument is sound. But in ethical argument there is often reasonable disagreement concerning the truth of certain premises.
Not all people accept personhood theory, and, among those who do, there are differences concerning exactly what cognitive function or set of functions is necessary for personhood. Those who reject personhood theory, or those who accept it but have another account of the wrongness of killing that makes it wrong to kill nonpersons,Footnote 10 have good reasons to reject some of the premises in the long chain of arguments leading to the philosophical conclusion that infanticide of healthy infants is not intrinsically or inherently wrong.
The argument is thus only conditionally sound (i.e., sound for those who accept the premises) and can be rejected without any inconsistency by anyone who has good reason to reject one or more of the premises. Or, to put it differently, the philosophical conclusion is not even binding on all or even most philosophers of good faith. Those who hold other views are not perverse, deluded, or obviously wrong. They just disagree!
So, we ought to make room for at least a modicum of epistemic humility. I think I am right, but I have to at least entertain the possibility that my interlocutor is right when she disagrees with me. The penalty for getting it wrong philosophically is often small; few people are directly harmed by fallacious philosophical arguments. But the penalty for getting it wrong policy-wise can be big. The philosophical fallacies in Marxism-Leninism are many and egregious, but it was the implementation of the philosophy that caused the real harm.
Here one may complain that the prior acceptance of a particular philosophical or policy conclusion can, nevertheless, compel acceptance of the argument on pain of inconsistency. Arguments by analogy and parity-of-reasoning arguments abound in bioethics. In this particular instance, the claim would be that if someone thinks that elective abortion of a healthy fetus is acceptable, then that person is committed by parity of reasoning to find the infanticide of a healthy infant permissible.
I have previously analyzed the pitfalls of parity-of-reasoning arguments in an article in the Cambridge Quarterly of Healthcare Ethics Footnote 11 and will not rehearse the full argument again. However, the possibility of a reasonable plurality of values and ethical theories, which is the focus of this section, also has implications for parity-of-reasoning arguments.
There are many people who are committed to the view that women should have a legal claim right to publicly funded abortion on demand for early abortions and a legal claim right to later abortion in specific circumstances. But those who are so committed are not all committed for the same reasons. It is perfectly possible to be committed to such a set of legal rights and to still be committed to the view that many specific instances of abortion to which women have a legal right are profoundly wrong. Or, it is possible—although probably unlikely—that there are people who have been convinced by Judith Jarvis Thomson’s famous violinist argument and have come to the view that, even if fetuses have a full moral status, abortion is still justified.Footnote 12 According to Thomson, we can draw an analogy between a woman who has, against her will, been hooked up to a violinist who needs to use her organs as support for nine months and a pregnant woman. Thomson argues that, even though the violinist has a right to life, the woman can legitimately cut him off, even if that means that he will die. And, on the basis of the claimed analogy, she further argues that this shows that abortion is permissible, even if fetuses have a right to life. A convinced Thomsonian would be able to make a clear distinction between cutting off, and thereby killing, the violinist during the period in which he is attached (i.e., abortion) and killing the violinist after he has been disconnected (infanticide).
So we may be committed to the same conclusions, but for very different reasons. And because parity-of-reasoning arguments are parity-of-reasoning arguments, it is the reasons and the structure of the argument—and not the conclusion—that matter when we decide what consistency requires. So, to torture the metaphor even more, we don’t even know whether our whales are swimming in the same rivers before we try to make them jump. When the rights whale jumps, it may end up in a very different place from the consequentialist whale, and the Thomsonian whale may not be willing to jump at all.
Jurisprudence and the Conversion of Ethics into Law
It is a jurisprudential platitude that the mere fact that something is ethically acceptable or unacceptable has no direct implications as to whether it should be permitted or proscribed by law. There are, for instance, many different instances of theft that are philosophically permissible or perhaps even mandatory, but it does not follow that all or any of these should be made explicitly permissible in law. And there are many cases of sexual unfaithfulness to a life partner that are profoundly ethically unacceptable and wrong, but it does not follow that sexual infidelity should be legally penalized. The relation between ethics and law is much more complicated than just a one-to-one translation, and the lawmaker has other legitimate concerns than ethical concerns.
Even if we take a jurisprudential thinker like Lon Fuller, who is sympathetic to the view that law should be moral, we find that this position is at a very general level of respecting the rule of law.Footnote 13 None of Fuller’s famous eight desiderata (there must be rules, they must be promulgated, they must be retroactive as seldom as possible, they must be clear, they must not be contradictory, they must not require the impossible, they should not be changed too frequently, and there must be congruence between declared rule and official action) commit the lawmaker to any substantive congruence between what the law says and what is held to be ethically right or wrong by society or by philosophers. Even the point that the rules must not be contradictory is not of much help, because many of the instances in which a philosopher might point out that two sets of legal rules are not consistent are instances in which there is no direct contradiction in the rules (e.g., it is not directly contradictory that in the UK there are different time limits for embryo experimentation and abortion). The claimed inconsistency is at the level of justification of the rules, and not in the rules themselves, which can be implemented without contradiction.
John Stuart Mill, the patron saint of liberalism, is not of great help here either. In general, the no-harm principle tells us what we cannot legislate against if we are liberals— essentially, actions that are purely self-regarding and harm no one else—but not what we should legislate against, for example, the level of harm to others that is sufficient for legislative action or what behaviors the law should promote and support.
We could find help in old or new natural law theory, in which there is a direct link between ethics and good law, but that only works for those whose ethics are of the natural law variety, so converting ethics into policy via natural law theory comes with a lot of inbuilt philosophical commitments, including of course the controversial “master commitment” to the coherence of a concept of natural law.Footnote 14
The Differences between Policymaking and Philosophy
It has long been recognized by reflective philosophers with policy experience that policymaking and philosophical analysis are two quite distinct types of activity with different aims and different criteria for success.Footnote 15 Philosophical success is, broadly speaking, achieved by developing a valid and sound argument, with premises that one’s opponents cannot deny, whereas policy success is achieved by implementing a sensible policy without expending more than the necessary political capital.
Policymaking rarely starts with a blank piece of paper on which to draw up new substantive policies and regulatory structures; it almost always starts with some policies and structures already in place,Footnote 16 and the policy question therefore becomes, “Where can we move to, from where we are now?” Or, to put it differently: policymaking is path dependent, and where we are now and where we can move from here may to a significant degree be determined by previous policy choices.
Policymaking is, in contradistinction to philosophy, almost always a collective process, and it often involves compromise among different groups; this means that the final outcome may not be supported by any particular, recognizable principle. Let us take abortion legislation as an example: in the 1970s, all Scandinavian countries introduced legislation giving women a legal right to “abortion on demand” early in pregnancy, but the gestation time limit below which women have this right is different in each country. This is best explained not as an example of some failure on the part of politicians to discern the “best” argument underlying a determination of the “right” number of weeks, but as an example of legitimate political compromises being struck slightly differently in each country. It is said that the most important ability of a Danish prime minister is to be able to count to 90 (there being 179 members in the Danish parliament), and if 90 members of the Danish parliament supported the Danish 12-week limit, and less than 90 any higher limit, then it is fully legitimate for the policymaker to legislate for 12 weeks. Given that policymaking in a democracy involves different policymakers at different times, an expectation of consistency between policies made at two different points in time is also problematic. Just as it is legitimate as a matter of democratic policy that tax policies may vary over time depending on whether the majority is to the right or to the left, it is prima facie legitimate that reproductive policies vary according to whether the majority is conservative or liberal .
Because policymaking is a collective process, it might also involve bargaining and trading. Two different aspects of a particular policy area may be of different importance to two sets of policymakers, and the members of each set may be willing to “trade” their position on one of these issues in order to get the policy on the other issue closer to what they want. If we again use abortion legislation as an example, we see that strong protections for the conscientious objection of healthcare professionals are often traded against relatively liberal access to abortion for women.
So, whereas consistency may be the master virtue of philosophy, it is not the master virtue of policymaking. Policymaking is, to a large extent, the art of the possible, and what is possible now may not be consistent with what was possible last year, and vice versa.
Here it might be objected that both ethics and policymaking proceed according to a form of “wide reflective equilibrium”:Footnote 17 that both philosophers and policymakers aim at maximal consistency and coherence among intuitions, principles, and empirical facts; and that the two activities are therefore very similar. It may well be true at a very abstract level that both activities aim at coherence and the minimization of cognitive dissonance, but the principles that matter differ, and the range of facts that are relevant to a practical decision is different from (and often much larger than) the range of facts that are relevant to a theoretical analysis. Legal, political, and social facts matter (and should matter) to policymakers in a way that they do not matter to philosophers.
(Legal) Reality Intervenes
A policy of legally permitted infanticide of healthy infants will always have to be implemented in a particular jurisdiction with an already-existing network of legal relations attaching to infants, because whereas some philosophers may not see them as “persons,” the law in general does. An infant can own property, be a beneficiary of a will or a trust (e.g., as the first female grandchild of the testator), be heir to the throne, be a research participant, and so on. And infanticide is not something that just happens; it is something that someone decides to ask for, and something that someone has to perform.
A policy of legally permitted infanticide of healthy infants would thus have to take a position on a range of practical issues with legal import. The following is not an exhaustive list but should illustrate the scope of the problem of carving out legal space for infanticide (and each of the bullet point lists could be followed by an “and so on”):
• For how long after birth should infanticide be a possibility—until the infant becomes a person, or for a shorter period?
• Who should be allowed to request infanticide—the parents jointly, the parents severally, healthcare professionals, or others? Should anyone be allowed to veto a request for infanticide? Should there be any involvement of older siblings?
• Who should perform infanticide? If infanticide is to be performed by healthcare professionals, should there be conscientious objection clauses in the law, and what should be the scope and strength of such clauses?
• What legal status should infants have in the period in which infanticide is legally possible?
○ Can an infant own property in this period? And, if the infant owns property, who should inherit after infanticide?
○ What should happen, legally, if an infant dies naturally during the period in which infanticide is an option? Should it matter whether or not that infant was destined for infanticide?
• Are there other possible effects of giving up the idea that legal personality is obtained at birth, and that all legal persons have an inalienable right not to be killed?
The answers to many of these questions do not flow in any straightforward way from the philosophical conclusion that infanticide of a healthy infant is not intrinsically or inherently wrong, and many of the answers cannot be obtained from either abortion legislation or debate. The fetus is inside the woman, whereas the infant is an independently existing entity, and this may, for instance, mean that decisionmaking and veto capabilities can (or should?) be distributed differently in the two cases.
The real problems of implementation can also be shown by considering whether a policy making the infanticide of healthy infants legally permissible is consistent with the UN Convention on the Rights of the Child, a legally binding convention that has been ratified by all UN member states except the United States and Somalia. Of all the UN human rights conventions, the Convention on the Rights of the Child is the one that has been ratified by the most countries. Article 1 of the convention defines “a child,” Article 3.1 articulates the general principle for how official bodies shall act in relation to children (in their best interest), and Article 6 states unequivocally that states recognize that “every child has the inherent right to life”:
Article 1
For the purposes of the present Convention, a child means every human being below the age of 18 years unless under the law applicable to the child, majority is attained earlier. . . .
Article 3
1. In all actions concerning children, whether undertaken by public or private social welfare institutions, courts of law, administrative authorities or legislative bodies, the best interests of the child shall be a primary consideration. . . .
Article 6
1. States Parties recognize that every child has the inherent right to life.
2. States Parties shall ensure to the maximum extent possible the survival and development of the child.Footnote 18
It is blatantly obvious that a policy of legally permissible infanticide of healthy infants is completely incompatible with the convention. This is not a reason for giving up any philosophical exploration of whether infanticide of healthy infants is morally justifiable, but it is a reason against eliding moral justification and policy prescription. Even if the legislature in a particular jurisdiction became convinced that infanticide of healthy infants should be made legally permissible, to implement such a change in the law, it would have to resile from one of the most important, and widely accepted, binding human rights documents. There is no simple or easy way to make such a change.
Conclusion
The preceding detailed argument has focused on the relation between philosophical arguments showing that infanticide of healthy infants is not intrinsically or inherently wrong and a societal policy of allowing infanticide. It has been shown that there is no straightforward way of accomplishing the grand leap of the whale from philosophy to policy, except by elision between different meanings of terms like “justified” and “permissible.” Despite the focus on a specific set of arguments, this conclusion is quite general. Any philosopher who wants to make suggestions for real policy change must do the hard work that is necessary for converting a philosophical conclusion into an implementable policy prescription.
Let me end by quoting a wise old philosopher with extensive experience in policy engagement:
Those who directly participate in the formation of public policy would be irresponsible if they did not focus their concern on how their actions will affect policy and how that policy will in turn affect people. The virtues of academic research and scholarship that consist in an unconstrained search for truth, whatever the consequences, reflect not only the different goals of scholarly work but also the fact that the effects of the scholarly endeavour on the public are less direct, and are mediated more by other institutions and events, than are those of the public policy process. When philosophers move into the policy domain, they must shift their primary commitment from knowledge and truth to the policy consequences of what they do. And if they are not prepared to do this, why did they enter the policy domain? What are they doing there? Footnote 19
