What Counts as an Abortion, and Does It Matter?

In the U.S. Supreme Court’s recent case of Burwell v. Hobby Lobby, the respondents challenged their federal obligation, pursuant to regulations promulgated under the Patient Protection and Affordable Care Act (“ACA”), to include coverage in their employees’ health insurance package for four forms of birth control that the respondents consider abortifacients.  According to the respondents, the Religious Freedom Restoration Act (“RFRA”) protects them from having to violate their religious commitments by funding abortions through employer contributions.

By a 5-4 majority, the Supreme Court held for the respondents, for-profit closely-held corporations, ruling that requiring coverage for contraception methods that the corporations’ owners regard as abortions imposes a substantial burden on the respondents’ exercise of their religion, and the requirement at issue is not the least restrictive means available for serving the government’s (assumed) compelling interest in ensuring women’s access to contraception.

One of the underdeveloped factual issues that arose in Hobby Lobby is the question of whether in fact the two types of IUD and two forms of the morning-after pill, FDA-approved methods of contraception to which the respondents demanded coverage exemptions, do factually qualify as abortifacients.  The Court deals with this question in the opinion by saying of one set of respondents, for example, that “the Hahns believe that ‘human life begins at conception.’  It is therefore ‘against [their] moral conviction to be involved in the termination of human life’ after conception, which they believe is a ‘sin against God to which they are held accountable.’ . . .  The Hahns have accordingly excluded from the group-health-insurance plan they offer to their employees certain contraceptive methods that they consider to be abortifacients.”  The Court adds that “HHS [the U.S. Department of Health and Human Services] acknowledges [that the challenged methods of contraception] may result in the destruction of an embryo.”

Since the Supreme Court handed down its decision in Hobby Lobby, some commentators (such as this one) have objected to the Court’s deference to what they consider to be a factually inaccurate characterization of the challenged birth control methods.  Because the morning-after pill is not actually an abortifacient, according to this objection, the Supreme Court should not have simply accepted the claim that it is.  My colleague and fellow columnist, Professor Michael C. Dorf, here offers a very compelling response to this “you have the right to your own opinion but not to your own facts” analysis.

Dorf contends that courts properly defer to sincerely held factual beliefs of religious claimants, inasmuch as religious belief encompasses not only moral but factual teachings as well, and courts ought no more to be rejecting the “truth” of the latter than they ought to be rejecting the truth of the former, in considering whether complying with federal law does or does not substantially burden religious exercise.

Regardless of how one resolves this issue—of whether sincerely held religious beliefs receiving protection under RFRA include only value judgments or factual assessments as well—it remains useful to examine whether the Hobby Lobby respondents’ claim that IUDs and morning-after pills are abortifacients is really a factual claim, a values claim, some combination of the two, or none of the above.  Considering this question may also have implications for how we think about abortion more generally.

The Respondents’ Claim About Four FDA-Approved Methods of Contraception

Let us begin by looking at what the respondents claim about two types of IUDs and two forms of the morning-after pill, four contraceptive methods covered by regulations under the ACA to which the respondents sought an exemption.  According to the Supreme Court’s majority opinion, the respondents believe that life begins at conception.  This belief is common to those who consider themselves part of the pro-life movement:  they believe that the value we ascribe to a human life attaches to that life at the moment that a sperm cell fertilizes an egg cell.

Once conception occurs, on this view, terminating the life of a zygote, an embryo, a fetus, or a born individual has the same moral weight, regardless of chronological age.  Stated differently, just as most of us consider it no better to murder a one-year-old baby than it is to murder a three-year-old toddler, a person who is pro-life considers it no better to terminate the life of a fertilized egg than it is to terminate the life of a third-trimester fetus or a newborn baby.

The religious belief that life begins at conception is not specifically a factual belief about IUDs or other methods of birth control.  It is a moral belief about the zygote or embryo.  Nonetheless, this belief does provide a standard for evaluating whether participating in a particular method of birth control does or does not violate one’s religious commitments.

The standard is this:  if a method of birth control can operate by killing a developing human after the point of its conception, then using that method of birth control qualifies as an abortion and accordingly violates the religious commitments of the pro-life individual.  If method X, for example, can operate by killing a fertilized egg or embryo, then a person who believes that life begins at conception would consider method X to be an abortifacient.  The remaining factual question (on the basis of which some commentators have criticized the Court for improper deference) is whether the methods challenged by the respondents do in fact work by killing a zygote or an embryo.

The Medical Definition of Abortion

As a general matter, the medical profession defines an abortion as the termination of a pregnancy, and doctors in turn typically define a pregnancy as a condition in which an embryo has implanted itself in a woman’s uterus, where it continues to grow and develop.  This definition has some unremarkable implications, including that when a fetus dies inside a pregnant woman’s womb, then the woman in question is no longer pregnant, and the removal of the dead fetus from her body does not qualify as an abortion.

More relevant to our purposes is the further implication of this medical definition that even after fertilization has occurred, there is no pregnancy—and there can therefore be no abortion—before the embryo has implanted in the woman’s uterus.  This means that even if a method of birth control works some of the time by killing a fertilized egg that has not yet begun the process of implanting itself in the woman’s uterus, then that method of birth control, under the standard medical definition of abortion, falls outside the “abortifacient” category.

The consequence of this definitional divide between a pro-life individual’s thinking and the standard medical thinking is that a method of birth control that works by killing a fertilized egg prior to implantation is an abortifacient from the pro-life individual’s perspective but is not an abortifacient from the standard medical perspective.  In such a case, there could be an apparent dispute between a pro-life individual and a typical doctor about whether an abortion has taken place, but the dispute would not, at bottom, be factual at all.  This is because both disputants would agree that the (hypothetical) method kills the fertilized egg.

At most, the disagreement might involve the moral status of killing an unimplanted zygote or embryo, and calling the act an “abortion” might be a way of saying, “I regard the act as morally equivalent to any other killing of a human who might be much older than a zygote or an embryo.”  Even if one were unwilling, under RFRA, to defer to inaccurate factual claims, then, the pro-life view about the hypothetical method of birth control would plainly be a normatively and religiously controversial view, rather than a factually controversial view.

The medical definition of abortion arguably does not itself even embrace a moral view about the status of the fertilized egg.  A particular gynecologist might consider herself pro-life and might therefore regard a method of birth control that operates by killing a fertilized egg to be a form of murder.  Yet the gynecologist would nonetheless understand that the particular murder does not fall within the medical definition of abortion (much as shooting a teller at a bank is a form of murder that would not fall within the medical definition of abortion).

The disagreement, in this case, is semantic rather than being either factual or normative.  That is, the pro-life individual and the doctor use the same word, “abortion,” to mean different things for different purposes, but everyone has in mind the same empirical facts (about whether or not a fertilized egg has been killed) and perhaps even the same normative assessment of those facts.

Consider, by analogy, the question of how many weeks pregnant a woman is.  Under the medical approach to counting the weeks of pregnancy, one begins the clock on the first day of the woman’s last menstrual period.  On the definition that proceeds from the medical approach, pregnancy typically lasts 40 weeks.

Pro-life advocates, by contrast, commonly count the number of weeks into which a pregnancy has progressed from the date of conception, which typically occurs approximately two weeks after the first day of the woman’s last menstrual period.  On this definition, pregnancy lasts approximately 38 weeks.  Notwithstanding the apparent disagreement (40 weeks versus 38 weeks), however, a pro-life advocate and a doctor would not differ here on the amount of time that typically passes between conception and birth.  The “disagreement” would, again, be semantic.

Potential Normative Implications of Different Definitions of Abortion

Though the disputes above can be characterized as disputes about definitions rather than disputes about facts or morality, there are subtle moral issues that emerge upon closer scrutiny.  Consider the meaning of the Hobby Lobby respondents’ definition of abortion.  According to the respondents, the challenged contraceptive methods are objectionable because—as HHS apparently acknowledges—these methods could potentially have their intended effect after an egg has been fertilized.

One way in which this might happen, if one accepts the factual claim that it sometimes does happen (a claim that many dispute), is that fertilization would occur before the morning-after pill is taken, and the pill could then prevent the already-developing embryo from implanting in the woman’s uterus, thereby making it impossible for the embryo to continue to develop into a baby.  The morning-after pill, if it in fact operated post-conception, could thus prevent implantation, perhaps by hormonally making the uterine lining a less hospitable environment for implantation than it would otherwise have been.

For purposes of our earlier examination of the competing definitions of abortion, we stipulated that a hypothetical birth control method “killed” the fertilized egg prior to implantation and that therefore, whether or not the medical definition of “abortion” applied, the factual claim that the contraception method killed a post-conception human entity would be accurate.  For someone with a pro-life set of values, the killing in such a case would be morally indistinguishable from the killing that would occur if an already-implanted embryo were killed with a saline solution or terminated by some other uncontroversially classified method of abortion.  But if, as some claim, the morning-after pill operates after conception by making the uterus inhospitable to implantation, then this method is arguably morally distinct from our hypothesized method that directly kills an embryo.

We could have a long philosophical discourse here about the differences between “killing” someone and allowing someone to die, but suffice it to say that our legal system (and most systems of morality) do draw a distinction between the two.  There is, in other words, a moral (and legal) consensus that directly killing someone, all things being equal, is morally more culpable than allowing someone to die without intervening.  This is why the people who watched Kitty Genovese being murdered in Queens, New York, in the 1960’s, and (allegedly) did not call the police or otherwise attempt to stop the crime were much less responsible, from a moral standpoint (and not at all responsible, from a legal standpoint), for the murder than the man who repeatedly stabbed Genovese until she died of her wounds.

There have long been debates about whether abortion should be considered direct killing or a failure to intervene.  (Judith Jarvis Thomson developed the original “violinist” analogy elaborating the insight that abortion might best be understood as a failure to help rather than as direct violence).  I have offered my own, somewhat different, thinking on the subject in chapter 7 of my book, Mind If I Order the Cheeseburger? And Other Questions People Ask Vegans, a chapter entitled “Are You Against Abortion?.”

Even among those who believe that (1) life begins at conception and (2) abortion generally qualifies as “killing” rather than “letting die,” it is still possible to take a different position from that of the respondents on a method of birth control that makes the uterus less hospitable than it would otherwise have been to an embryo “attempting” implantation.  One who believes that killing is distinct from letting die might view the ingestion of hormones that prevent the further thickening of the uterine lining as something very different from actively killing the embryo (for example, by ingesting a poison that targets embryos).  One could view the disrupted thickening of the uterine lining to make it less hospitable to implantation as akin to shutting the door to one’s house when someone outside is running toward the door seeking refuge from dangerously low temperatures outside.  The person will die of exposure if not admitted into the house, but refusing to admit the freezing man is not, by most assessments, the moral equivalent of shooting the man dead with a gun or even of taking a man who is already warming up inside one’s home and throwing him out into the cold.

For a closer analogy to making the uterine lining inhospitable to implantation, consider the plight of an embryo that has been produced through in vitro fertilization and is now ready for implantation.  Assume that the couple that produced the embryo has decided to split up and agrees that the embryo should not be implanted after all.  The woman in this case, like the woman who uses a contraceptive method that disrupts the thickening of her uterine lining, is making her uterus “inhospitable” to implantation by refusing the embryo’s insertion (and by thus preventing its implantation).  Without a place to go, the embryo dies.

Does this mean that the woman who (with her partner) refuses to have the embryo inserted for implantation is directly killing the embryo?  Would anyone argue that she is having an abortion, by refusing to have the embryo inserted?  There is an embryo involved, and for those who believe that life begins at conception, there is therefore an entity with moral weight equivalent to that of a newborn baby at issue.  Yet there has been no pregnancy, and it would thus seem a stretch to say that there has been an abortion.  If there is a case of “letting die” rather than “killing” in the context of embryos, this would appear to be such a case.

The in vitro fertilization situation, as I explained in an earlier column, illustrates an often-overlooked difficulty one faces when attempting to argue that the use of post-fertilization but pre-implantation birth control should qualify as an abortion.  The difficulty is not about whether the zygote is or is not worthy of moral consideration.  If an embryo is worthy of moral consideration after implantation has occurred, then there is no reason to deny the same embryo equivalent moral consideration just prior to implantation.  It has not changed, after all; only its environment has changed.

At the same time, however, prior to implantation, the person who refuses entry into her uterus is not killing the embryo so much as refusing to rescue it from the death it will inevitably face at its stage of development without a uterus in which to implant.  Just as it would be peculiar to say that the woman who refuses to have a test-tube embryo implanted is “killing” the embryo, it is strange to say that preventing implantation is tantamount to killing.

One can, of course, take the position that whenever one is in a position to provide refuge to a life that would otherwise be lost “but for” the intervention, and one fails to intervene, one is “killing” the one to whom one fails to provide refuge.  But that is not ordinarily how our law or our moral systems work.  And it is for that reason that one might legitimately have qualms about the respondents’ claim that their religious belief that life begins at conception entails the proposition that the morning-after pill is an abortifacient that literally kills embryonic life.  The qualms are moral rather than factual, though—they revolve around the morality of refusing to rescue, and the qualms arguably implicate the respondents in moral hypocrisy rather than (or rather than only) in empirically implausible contentions about the nature of four methods of contraception.  It may be, then, that in assessing the sincerity of the respondents’ moral beliefs, such hypocrisy could provide an(other) reason for rejecting the ultimate outcome in Hobby Lobby, if the respondents are among the vast majority of people who distinguish between direct violence and a failure to rescue.