Article Text

Download PDFPDF

Cursed lamp: the problem of spontaneous abortion
  1. William Simkulet
  1. Philosophy, Mid Michigan Community College, Harrison, MI 48625, USA
  1. Correspondence to Dr William Simkulet, 311 Sunnyslope Ct. S, Andover, Kansas, LS 67002, USA; Simkuletwm{at}


Many people believe human fetuses have the same moral status as adult human persons, that it is wrong to allow harm to befall things with this moral status, and thus voluntary, induced abortion is seriously morally wrong. Recently, many prochoice theorists have argued that this antiabortion stance is inconsistent; approximately 60% of human fetuses die from spontaneous abortion, far more than die from induced abortion, so if antiabortion theorists really believe that human fetuses have significant moral status, they have strong moral obligations to oppose spontaneous abortion. Yet, few antiabortion theorists devote any effort to doing so. Many prochoice theorists argue that to resolve this inconsistency, antiabortion theorists should abandon their opposition to induced abortion. Here, I argue that those who do not abandon their opposition to induced abortion but continue to neglect spontaneous abortion act immorally. Aristotle argues that moral responsibility requires both control and awareness; I argue that once an antiabortion theorist becomes aware of the frequency of spontaneous abortion, they have a strong moral obligation to redirect their efforts towards combating spontaneous abortion; failure to do so is morally monstrous.

  • abortion
  • ethics
  • killing
  • embryos and fetuses
View Full Text

Statistics from

Cursed Lamp: Allison comes across an ancient lamp and, on a whim, she rubs it. From the lamp emerges a djinn who says ‘I will grant you unlimited wishes. However, I can only grant wishes intended to benefit only you. Rest assured that the wishes I grant will not harm anyone’. Allison, believing events to be an elaborate prank, makes a wish, ‘I wish for a box of diamonds’, she says. Granted, a box filled with diamonds appears.

Allison spends her newfound fortune, and the next few years researching the diamonds and finds no evidence her wish harmed anyone else. Satisfied that the djinn was truthful, Allison makes several more wishes. The djinn grants the wishes intended to benefit her, but when a wish was intended to benefit another—Allison’s friends, family or community—he does not grant them.

One day Allison is bitten by a rabid animal, heads to the hospital and is unexpectedly told there is no cure. Flustered, she summons the djinn and wishes to be cured. Granted. She wishes to know why there was no rabies vaccine, and the djinn tells her ‘Whenever I grant a wish, I also remove a cure for a disease from the world – only you remember that a cure ever existed; for everyone else there was never a cure’. Allison is furious, ‘You told me that the wishes you grant would not harm anyone’. to which the djinn replies ‘Ah, but I haven’t harmed anyone, I’ve merely acted in such a way that allows harm to befall many more than would have been harmed had I not acted’.

According to Aristotle, moral responsibility requires both control and awareness. In other words, ought implies can and know-how.1 James Rachels argues that morality is an effort to guide one’s conduct by reasons—the right thing to do in any given situation is that which one has the best reasons to do, and the wrong thing is anything else.2 In Cursed Lamp, after her research, Allison believes she has many reasons to make wishes and no reasons not to; as such it makes sense to say that she ought to make wishes when doing so will benefit her because she does not believe that there is any drawback to doing so. However, after discovering that her wishes carry with them unforeseen, horrible consequences, it is clear that Allison ought to change her behaviour. If she were to continue to use the lamp for personal gain knowing that doing so would allow terrible harm to befall others, she would be a moral monster, even if no one else would ever know that her actions allowed harm to befall anyone.

Many people believe that human fetuses, from the moment of conception (or soon afterwards), are morally comparable with normal, adult human persons like you or I and because of this, voluntary abortion is prima facie seriously morally wrong. A variety of antiabortion theories purport to be consistent with these beliefs–notably the substance view and deprivation views. Proponents of the substance view contend what makes it wrong to kill us is that we are essentially rational substances and that human fetuses are rational substances from the moment of conception.3–10 Don Marquis contends that what makes it wrong to kill us is that killing us deprives us of a future of value and argues that human fetuses are numerically identical to the adult human organisms they will become approximately 2 weeks after fertilisation—after cell totipotency ends and cells begin to specialise.11–13

Recently, a number of philosophers have argued that the frequency with which spontaneous abortion occurs in human beings presents a serious problem for these views.14–19 Roughly 60%—3 out of every 5 fertilised of human eggs—end in spontaneous abortion.20 21 If these numbers are anywhere close to being accurate, then it seems these views are committed to the belief that an overwhelming number of beings of significant moral worth die on a regular basis; spontaneous abortion would be the greatest killer of human beings, killing more human beings than have ever been born! Toby Ord summarises the argument as follows:

  1. (Moral claim:) The (human fetus) has the same moral status as an adult human (person).

  2. (Empirical claim:) Medical studies show that more than 60% of all (human fetuses) are killed by spontaneous abortion.

  3. (Conclusion:) Therefore, spontaneous abortion is one of the most serious problems facing humanity, and we must do our utmost to investigate ways of preventing this death—even if this is to the detriment of other pressing issues.18

Ord sees this argument as a reductio ad absurdum; he begins by assuming the antiabortion moral claim and shows that assumption would lead to a conclusion he believes to be absurd. Rob Lovering and Timothy Murphy take a similar stance, noting that the moral claim leads to counterintuitive implications.16 17 To avoid this, Ord and others claim antiabortion theorists have to reject the moral claim that human fetuses are comparable with adult human persons. To do so, however, would undermine most contemporary antiabortion arguments; as Don Marquis said, ‘Many of the most insightful and careful writers on the ethics of abortion… believe that whether or not abortion is morally permissible stands or falls on whether or not a fetus is the sort of being whose life it is seriously wrong to end’.11

Here I argue that when antiabortion theorists learn of the frequency of spontaneous abortion, they face a dilemma:  Either they should (1) abandon the moral claim that fetuses are morally comparable with adult human persons (and in doing so, abandon their antiabortion stance), or (2) re-evaluate their priorities and work to prevent both spontaneous and voluntary, induced abortions. Either horn of the dilemma seems to show that contemporary opposition to abortion is misguided. If (1), then either abortion is not wrong, or if it is wrong, it is likely far less wrong than killing an adult human person, and thus not as pressing an issue as contemporary antiabortion theorists believe it to be. If (2), then it seems antiabortion theorists have strong moral obligations they are failing to address. If one has a strong moral obligation to oppose voluntary abortion because voluntary abortion results in the loss of life of something comparable with you or I, then it seems one would have a comparable moral obligation to oppose spontaneous abortion for the same reason, where many spontaneous abortions can be prevented with education and current technology, and many more can be prevented with future research, education and resource allocation. Consider the following case:

Rich’s Ass: Richard loves his wife very much and wants her to live. She is currently suffering from an incurable disease and will die without expensive medication. He believes he has a strong moral obligation to care for his wife. Rich spends his savings on his wife’s medication, but knows that he will soon run out of funding.

On one windy day, Richard finds himself equidistant between two stacks of abandoned bills—on his left, he sees a stack of $1 bills, and on his right, he sees a stack of $100 bills—and believes he only has time to grab one, before the other stack is lost forever in the ocean.

On a whim, Richard chooses the stack on his left. The other stack is blown away, lost to the ocean. He spends the stack of $1 bills on enough medication to keep his wife alive for another day.

Richard believes that he has far more reasons to choose the stack of $100 bills than the stack of $1 bills, as such his behaviour is inexplicable and inconsistent with his moral beliefs. I think it makes sense to say that his behaviour is morally monstrous.

Charles C. Camosy contends that the high rate of spontaneous abortions is a ‘relatively well-known fact’.22 Yet, given that many antiabortion advocates devote substantive time, resources and thought to preventing and mourning voluntary, induced human abortions and little to no time, resources, or thought to preventing and mourning the much larger number of spontaneous abortions that result in what they would, by assumption, view as the same kind of loss, I genuinely doubt the average antiabortion advocate is aware of this medical fact. However, I contend that those that are aware and freely choose to disregard it act as moral monsters.

Here I will address six responses to the problem of spontaneous abortion.

  1. The numbers response: Robert George and Christopher Tollefsen argue that many of spontaneously aborted fetuses are not true human substances, and thus the number of regrettable human deaths caused by spontaneous abortion is significantly smaller than it initially appears.10

  2. The control principle response: Henrik Friberg-Fernros and Marianne Burda argue that most cases of spontaneous abortion are unavoidable, and thus we cannot have a moral obligation to prevent spontaneous abortions.4 23

  3. The visibility response: Sarah-Vaughan Brakman argues that we are more likely to be upset by the death of those we know than those we don’t, so our lack of response to spontaneous abortion—the deaths of beings we’ve never met—is not morally problematic.24

  4. The causing/allowing distinction response: some philosophers argue that because it is worse to intentionally cause harm than to allow it, and spontaneous abortions are not intentionally caused, there problem of spontaneous abortion is not as dire as it initially appears.25

  5. Weak obligations response: other philosophers argue that our moral obligations with regards to spontaneous abortion are more manageable than critics make them out to be.26 27

  6. The asymmetrical response: finally, some philosophers contend that there is an asymmetry between human fetuses and normal, adult human persons that often makes the loss of many human fetuses a lesser loss than the loss of one normal, adult human person like you or I.25

Here I argue that each of these responses fails to solve the problem that spontaneous abortion poses to antiabortion theorists. Indeed, many of these responses are inconsistent with the very moral intuitions antiabortion theorists claim consistency with—(1 , 3 , 4) and (6)—and many would allow, or in the last response even obligate voluntary, induced human abortion—(3, 4) and (6).

The numbers response

Ord contends that if fetuses have the same moral status as adult humans, spontaneous abortion is ‘the greatest killer of all’, and ‘leads to more regrettable human death than all other causes put together’.18 One strategy of responding to the problem of spontaneous abortion is to argue that the number of morally problematic spontaneous abortions are far less than what they initially appear to be.

Substance view theorists Robert George, Christopher Tollefsen and Henrick Friberg-Fernros argue that many early spontaneous abortions result from chromosomal defects ‘so significant that a human embryo probably failed to form’.4 10 According to the substance view, humans are essentially rational substances; if a biologically human embryo possesses severe chromosomal defects, then it is unviable, and lacks the capacity to become a fully rational substance, and thus while such things may be biologically human, they are not human substances.

Rob Lovering contends this response is a red herring; citing Carlson that at best 50% of spontaneous abortions are caused by such defects.16 28 He says ‘that still leaves roughly 3,000,000 spontaneously aborted human fetuses in the United States alone, more than 200 times the number of people who die each year from AIDS in the United States’.

If we take George and Tollefsen’s response seriously, a couple of things become apparent: first, a significant number of human biologically fetuses—roughly 30%, or about 1 in 3—are not truly human. This is surely at odds with the common-sense moral intuitions held by most antiabortion theorists. Second, this would still mean that 42% of all true human beings die from spontaneous abortion—a staggering, morally significant amount. Third, as I’ve argued in an earlier work, for proponents of the substance view there would be no reason to bother saving the lives of chromosomally deficient biologically human embryos.19 According to the substance view, whether or not something possesses rationality is an essential trait, so any intervention that would correct the chromosomal defects of these embryos would destroy them and replace them with a different substance—a full rational substance with comparable moral status with you or I. Thus, even if there was an inexpensive pill with no medical risk that could easily correct these chromosomal defects, according to the substance view, using such a pill would be optional and non-obligatory; this is to say that a woman with a chromosomally deficient embryo is morally permitted to refrain from taking such a pill, while a woman with a non-chromosomally deficient embryo facing a different medical risk would not be morally permitted to refrain from taking steps to save that embryo’s life. This consclusion is certainly  inconsistent with the intuitions of most antiabortion advocates.

A more compelling set of numbers comes from Don Marquis.12 He argues that our moral intuitions about early pregnancy are unreliable and contends that a human organism does not come into existence until after cell totipotency ends at about 2 weeks. Most spontaneous abortions occur before 2 weeks—approximately 5 out of every 6.18 If human embryos do not become human organisms until the end of this 2-week period, then only 20% of human organisms die from spontaneous abortion, compared with the substance view’s 42% of human substances. This number is still quite large and substantively higher than other deaths by a wide margin, and thus, on its own, is insufficient to address the problem of spontaneous abortion, but the margin is much smaller than competing anti-abortion views.

The problem with Marquis’s view is that it shifts the debate; it is no longer a question of if abortion is morally acceptable, but when. Abortion is always morally acceptable within the first 2 weeks of pregnancy. Rather than oppose abortion, if Marquis is correct, we ought to advocate that those considering abortions act swiftly to secure an abortion in the first 2 weeks. Education, vigilance and easy access to abortion would prevent the vast majority of unwanted pregnancies.

The control principle response

In response to the problem of spontaneous abortion, Friberg-Fernros says, ‘… the existence of obligations is dependent on our capacity to fulfill these obligations.4 We cannot reasonably be obligated to do something which we cannot do – as the famous Kantian maxim ‘ought implies can’ stresses. … it is not immoral per se to fail to do something about the embryo loss if we were subject to conditions which precluded attempts to prevent spontaneous abortions’.

Roughly the argument is as follows:

  1. The control principle—moral responsibility requires control.

  2. At present, we have no control over whether many spontaneous abortions occur.

  3. Therefore, we are not morally responsible for failing to prevent many spontaneous abortions.

There are three substantive problems with this argument. First, just because we cannot prevent all spontaneous abortions does not mean that we should not prevent any. Friberg-Fernros notes that even with current medical technology, we do have the possibility to ‘do something about embryo loss’ (214). However, preventing such losses would be difficult, he says, because the majority of spontaneous abortions occur before the woman is even aware that she is pregnant. He says this is ‘a fact which also makes it hard to see how one ‘can’ do anything about it’.

Marianne Burda takes a similar stance:

Since 60% of early pregnancy losses occur in the first 2 weeks before a woman may realise she is pregnant, treatment directed towards preventing these losses must be initiated prior to the woman possibly being able to detect that she is pregnant.23

Elsehwere I have argued if one was serious about preventing spontaneous abortions, one ought to recommend that women that have engaged in sexual activities err on the side of caution and assume that they are pregnant until medical testing shows otherwise.19

Burda notes the difficulties and expense that such vigilance would require:

Therapies to prevent these possible losses may require a woman and her partner to undergo testing before she is pregnant to identify any factors that may possibly contribute to pregnancy loss. Testing and prophylactic treatment may be required each month before or around the time of conception. This type of testing and treatment is unnecessary for women who will have no problems with their pregnancies and subjects them to risks and side effects from the testing and treatment. This approach also wastes limited resources that can be directed to other areas of medical care and research, including helping women who have a history of recurrent losses.23

However, this response is unsatisfying for two reasons:

  1. The burden to undergo regular, rigorous testing after sexual activity is not all that different from the burden the antiabortion position argues women have to carry fetuses to term.

  2. Burda contends that such an approach is a waste of limited resources,not because it will not save lives, but because it is a less efficient means of saving lives than had we allocated resources differently. This is an empirical matter that antiabortion advocates need to stay abreast on, but if this empirical fact requires us to allow fetuses to spontaneously abort, it seems it would just as easily obligate allowing voluntary induced abortions if we expect that we can save more lives by other methods. Since there is compelling reason to think we can do just this, the antiabortion theorist’s focus on induced abortion is misguided at best.

Second, even if we cannot currently prevent most spontaneous abortions and ought to refrain from preventing some of the preventable early spontaneous abortions because it would be an inefficient use of scarce medical resources, we might still have a strong moral obligation to dedicate resources to medical research on spontaneous abortion, doing so would likely result in more efficient practices and treatments that could be used to significantly reduce the number of spontaneous abortions in the future.

Third, and I think most importantly, the control principle response seems to miss the point: spontaneous abortions result in an overwhelming number of lost lives. Even if we do not have control over whether such loss of life occurs, we certainly have control over how we respond to this loss of life. Contemporary antiabortion theorists mourn the loss of human fetuses killed in voluntary, induced abortions that they could not prevent, yet refrain from mourning the far greater loss of human fetuses killed by spontaneous abortions they could not prevent. In other words, they show more reverence—more respect—for some human fetus lives than others.

The visibility response

Thomas A. Marino notes that many pregnancies end in spontaneous abortion before anyone suspects there was even a pregnancy.27 In some such cases, had a woman received currently available medical treatment, she might have been able to prevent the spontaneous abortion, but if she was not aware of the need to secure such treatment, it would not make sense to hold her morally accountable for failing to secure it. Furthermore, if she remains unaware of her spontaneous abortion, it does not make sense to hold her blameworthy for failing to mourn the loss of her fetus’s life.

Marino says the spontaneous abortion problem can be understood as a ‘missed opportunity’ to educate women about prenatal health and that such education ‘ought to be further promoted’. Marino undersells the obligations here—those that believe fetuses, from conception, are comparable with adult human beings have a strong moral obligation to educate about the fact that women might be pregnant before they realise it and promote vigilance regarding the moral obligation to secure prenatal care immediately following any kind of behaviour that one believes might lead to pregnancy. However, these obligations do not end there; one would also be morally obligated to educate people about the rate of spontaneous abortion and about the potential value of medical research that might lead to less abortions—either spontaneous or induced. There is good reason to think that this should include education about contraception and alternative sexual behaviour that would not result in pregnancy. Indeed, given the high rate of spontaneous abortion, would-be parents ought to be educated about adoption, as a successful adoption satisfies their goal to become pregnant without bringing into existence a human fetus that is more likely to die of spontaneous abortion than be born.

It does not make sense to hold a woman morally accountable for failing to mourn the loss of a child she never knew she had, but this lack of awareness does not stop her from mourning the faceless masses lost because of spontaneous abortion. It is not uncommon to pay one’s respect to groups of deceased people—those who died in war, in terrorist attacks, in school shooting, of disease or the like. Spontaneous abortion culls more living things than all of these things combined, yet few antiabortion theorists express any regret or concern.

Sarah-Vaughan Brakman offers an explanation for the antiabortion theorist’s lack of grief towards spontaneously aborted fetuses: ‘it is a normal human response to feel more deeply the loss of life of those one knows or with whom one is in immediate community than a loss of human life that is unknown, or more distant or otherwise not as present or visible’.24 She continues, ‘Therefore, the fact that there is NOT presently a concerted effort to limit [spontaneous abortion] as there is to limit deaths by cars, cancer and cyclones speaks in part more to the power of visibility and degree of personal connection to others than to any moral status claim or to a weighing of number of lives lost’.

Although both Marino and Brakman contend the fetus’s lack of visibility helps to explain why one has diminished responsibility to prevent its death, for Marino, one’s diminished responsibility is a matter of ignorance that can, and should, be overcome through education; in contrast, Brakman seems to suggest visibility is tied to personal connection and the greater our personal connection to the subject, the greater our moral obligations towards them.

For much of human history, our moral obligations have largely been local because there was little we could do for those both literally far away, such as those on the opposite side of the world—and figuratively far away—in this case, undetectable fetuses. We did not have moral obligations towards far away people because there would be no way of satisfying such moral obligations. Today, things are different; we can help those both literally and figuratively far away. Many philosophers have argued that distance is not, by itself, morally relevant.29 30

Although many philosophers agree that we may have stronger moral obligations towards our family, friends and community than towards strangers, this does not mean that we have no obligations towards strangers. In light of the sheer number of spontaneous abortions, then, our relative lack of interest in prevention spontaneous abortions compared with deaths by car accident is morally wrong.

Still, Brakman’s contention that personal connection is morally relevant is somewhat compelling, but it seems to be a reverse-edged sword for the antiabortion theorist. This stance seems to be inconsistent with those that oppose abortion on the grounds of the fetus’s moral status. If our moral obligations are tempered by visibility and personal connection, then the belief that fetuses are morally comparable with adult human persons is not sufficient to show that abortion is morally wrong. Either one can have a personal connection with a fetus that does not yet exhibit signs of personhood or one cannot. If one cannot, then all abortions are morally acceptable on this view regardless of the fetus’s moral status. However, if one can have a personal connection with a fetus, then causing or allowing abortion is only seriously morally wrong for those with a sufficiently close personal connection to the fetus regardless of the fetus’s moral status. This is to say that individuals can have a strong moral obligation to refrain from causing or allowing abortions to occur to their fetuses, but they cannot have a strong moral obligation to act to prevent others from causing or allowing abortions to occur to the fetuses of others.

Consider the antiabortion theorist that opposes abortion but contingently lacks any personal connection to a fetus at risk of induced abortion. Such a theorist seems to stand in the same (lack of) relationship to fetuses at risk of induced abortion as they do to fetuses at risk of spontaneous abortion; this is to say that they are strangers, largely invisible and most likely thought of as abstract statistics. On Brakman’s view, either one can have a moral obligation to such things or one cannot. If one cannot, then for most people, there is no moral obligation to oppose either induced or spontaneous abortion. This is inconsistent with contemporary antiabortion views. If one can, then the lack of a personal connection to spontaneously aborted fetuses does not excuse the antiabortion theorist’s relative lack of interest in preserving their lives.

The causing/allowing distinction response

Many philosophers believe there is a morally relevant difference between acting and refraining, such that in some cases acting to bring about a harm can be morally worse than merely allowing that harm to occur. Christopher Dodsworth, Tihamer Toth-Fejel and Zach Stangebye take this stance:

Ord ignores the distinction between doing evil and failing to do good. We are always responsible for the former (absent excusing conditions), but we are not always responsible for the latter. Moreover, the former is usually morally worse than the latter.

It is wrong to ignore drowning children, but it is an entirely different type of evil to drown them yourself and consider it a ‘right’ to hold them underwater as they thrash frantically for air, until their young collapsing lungs bleed into the water.25

Is the difference between acting and refraining—in this case killing and letting die—morally relevant? James Rachels employs a thought experiment to gauge our moral intuitions on the matter; we consider two parallel cases that are identical except for one thing—the first involves a killing, the second a letting die.31 Consider the following two cases:

Smith Kills: Smith stands to make a significant amount of money from future fertility treatments if something should happen to his patient’s fetus during pregnancy. During the birth, he wraps the umbilical cord around the fetus’s neck, causing it to suffocate to death and be stillborn. He acts to bring about the fetus’s death for financial gain.

Jones Lets Die: Jones stands to make a significant amount of money from future fertility treatments if something should happen to his patient’s fetus during pregnancy. During the birth, he notices that the umbilical cord is wrapped around the fetus’s neck and does nothing, allowing the fetus to die and be stillborn. He refrains from acting to bring about the fetus’s death for financial gain.

If you have the moral intuition that Smith and Jones have acted equally wrongfully, this is evidence that there is no morally relevant difference between killing and letting die. Rachels contends the rightness or wrongness of an action depends on the reasons for or against that action; in these cases, Smith and Jones have the same reasons to do what they do: they act or refrain from acting to bring about the fetus’s death for financial gain despite compelling reasons to do otherwise.2

Suppose that you have the intuition that Smith acts more wrongly than Jones, this would be evidence of what Dodsworth and colleagues believe: that killing is worse than letting die. Even if this is the case, it strikes me that most of us would conclude that what Jones does is still seriously morally wrong. Similarly, a person that freely ignores drowning children they could otherwise easily save is a moral monster. Even if acting to cause a death is more morally wrong than merely allowing death to befall someone, the latter is still uncontroversially morally significant.

There are two significant problems with this reply to the problem of spontaneous abortion. First, even if allowing abortions is less bad than causing them, the antiabortion theorist would have to hold that there is a wide gulf between the two—that causing the abortion of a human fetus is morally equivalent to causing the death of an adult human person, while allowing the abortion of a human fetus is not morally problematic at all—such that allowing spontaneous abortions to occur is not problematic at all. Note, of course, this would mean that allowing others to perform induced human abortions, too, would not be morally problematic at all, completely undermining most contemporary antiabortion efforts to restrict access to abortion.

Amy Berg takes a similar stance, contending that contemporary ‘person at conception’ opposition to abortion is not typically restricted to individuals pursing their own abortions but includes opposition to ‘other people’s abortions’ as well.14

Second, in all likelihood many spontaneous abortions may be inadvertently caused by the actions of others—whether the pregnant woman, or those around her; in other words, many spontaneous abortions are probably accidentally induced abortions as well. Thus, the contrary to Dodsworth and colleague’s assumption, the distinction between acting and refraining cannot explain the different stances antiabortion theorists take towards intentional induced human abortion and spontaneous human abortion.

Finally, the appeal to a distinction between acting and refraining seems to miss the point; for most antiabortion theorists, what makes abortion seriously morally wrong is the fetus’s loss of life, not the means that it loses its life. Don Marquis says ‘The loss of one’s life is one of the greatest losses one can suffer. The loss of one’s life deprives one of all the experiences, activities, projects, and enjoyments that would otherwise have constituted one’s future’.10 The reason most antiabortion theorists oppose abortion is to prevent this kind of loss, and this loss is experienced by both killed fetuses and fetuses merely allowed to die. Cursed Lamp demonstrates this clearly: the djinn explains that his actions do not cause harm to others, they merely allow harm to befall others; but I suspect most of us would find this distinction completely irrelevant to determining whether or not Allison should make more wishes.

The weak obligations response

According to Ord, if human fetuses are morally comparable with adult human beings, we are currently failing our moral obligation to do something about spontaneous abortion because spontaneous abortion kills far more human beings than induced abortion, cancer, AIDS and so on.18 Similarly, Rob Lovering says of spontaneous abortion ‘Now, if a natural epidemic were killing the same number of, say, standard adult human beings annually, we would at least be talking about if not doing all that we could to eliminate the epidemic’.16

One response to the problem of spontaneous abortion is to argue that our obligations in the face of spontaneous abortion are significantly weaker than Ord believes they would be. Thomas Marino says:

Ord also suggests no one is doing anything about pregnancy loss. There are three points to be made here. First, for women who do not have repeated pregnancy loss, there is nothing to do. They try again and get pregnant. Second, there are numerous scientific and clinical studies designed to help those women who do have documented pregnancy loss that is affecting their ability to get pregnant. Third, if anything, this is a missed opportunity because more needs to be done to educate women about the importance of prenatal health, which ought to be promoted further.27

Marino’s response is rather unsatisfying. First, it is not clear why Marino would think that women who have yet to become pregnant have nothing to do; after all, his third point just is that there ought to be more done to educate women, so surely women have a moral obligation to educate themselves long before pregnancy loss—reported or otherwise. Furthermore, if women can take steps to reduce the chance of future spontaneous abortions, it seems they would have a strong obligation to do so. Second, although there are some studies to this effect, Ord’s contention is that given the sheer size of the problem of spontaneous abortion, if fetus lives are morally significant, there should be many more.

Robert Card proposes a different set of obligations one has in face of the possible scourge of spontaneous abortion, contending that those who believe fetuses are morally comparable with adult human persons ought to ‘change their sexual behaviours to avoid undesired procreation and/or alter their methods of contraception when engaging in sexual intercourse’.26 He goes on to say, ‘The cost of an educational campaign that promoted non-procreative sexual behaviors or advocated, for example, combining methods of contraception to substantially reduce the chance of unwanted fertilizations would be relatively low, and it certainly would not entail eliminating the funding of cancer research’.

There are three substantive problems with this. First, it seems that even relatively minor advances concerning conception and spontaneous abortion would lead to far more lives saved than major advances in other areas that we currently devote substantive scarce medical resources to investigating, such as cancer research. More importantly, if our goal is to save fetal lives, we ought to devote far more resources to fighting spontaneous abortion than fighting induced abortion.

Second, suppose that one agrees with Marino or Card, that our moral obligations with regards to spontaneous abortion are relatively minor, involving mostly inexpensive, minor changes in education and sexual behaviour. It is important to note that contemporary antiabortion theorists fail to emphasise even this! Whereas the control principle, visibility and causing/allowing distinction responses have attempted to show that one’s moral responsibility for allowing spontaneous abortion is not as morally significant as critics say it is, the weak obligations response takes it for granted that spontaneous abortion is a serious moral problem, but one that is easily addressed. Thus, failure of contemporary antiabortion theorists to address the issue of spontaneous abortion seems all the more egregious. Even if you believe that it you have a strong moral obligation to save the lives of others, it is somewhat understandable to refrain from donating a kidney to save the life of a stranger, but it is far less understandable to refrain from donating an aspirin to save the life of a stranger when one has hundreds of aspirin at hand.

Finally, even if one disagrees with Ord regarding the extent of our moral obligations towards preventing spontaneous abortion, Lovering’s point still stands: at the very least, if you believe that fetuses are morally comparable to adult human persons, you ought to talk about the threat of spontaneous abortion. Yet, it seems that contemporary discussion with regards to noticing, let alone caring about, spontaneous abortion seems to come exclusively from the prochoice side of the debate, chastising antiabortion theorists for treating like things differently, and spending their efforts to prevent what they believe is a relatively minor harm, while leaving a far greater harm unmentioned. If Marino and Card are correct, and our moral obligations with regards to preventing spontaneous abortion are relatively minor and easily satisfied, this is all the worse for those antiabortion theorists. Imagine a parent chooses to spend her life’s savings—let’s say a million dollars—treating one of her children, while letting her other 11 children die of starvation. It is hard to imagine one making such a bizarre choice, yet it seems this is what many antiabortion theorists choose when they focus on the comparably small evils of induced human abortion, rather than the overwhelming evils of spontaneous abortions.

The asymmetrical response

Roughly, the problem of spontaneous abortion is that many antiabortion theorists treat like things differently, as they contend that all human fetuses, from conception, are comparable with adult human persons, yet while they take active steps to prevent the deaths of human fetuses from induced abortion, they do not take similar steps to prevent the death of human fetuses from spontaneous abortion, where spontaneous abortion kills more human fetuses than induced human abortion, and prevention of deaths from spontaneous abortion may very well be far easier to achieve in both the short and long term with relatively little investment—whether the major shift of priority Ord suggests, or the relatively minor shift of resources suggested by Marino and Card.2 6 18 27 Dodsworth and colleagues and Friberg-Fernros offer another possible response to the problem of spontaneous abortion, contending that there is an asymmetry between human fetuses and adult human persons that makes the loss of many human fetuses morally insignificant compared with loss of even one adult human person.4 25

Dodsworth and colleagues say, ‘In several ways, embryos may have much less to lose by death than adult humans do: they do not have life plans, desires, responsibilities, or intentions that are thwarted; early embryos do not feel pain, and they are not yet participating members of any personal relationships’.25 Friberg-Fernros takes a similar stance, contending that the life of one adult, human person might outweigh the lives of many human fetuses because while normal adult human persons have strong time-relative interests, or a psychological connection to their future and past selves, human fetuses lack time-relative interests.4

The problem with this response is that it rejects the common antiabortion stance that human fetuses, from conception, are morally comparable with adult human persons; on this view, they are significantly less valuable, and thus abortion is nowhere near morally equivalent to the killing of an adult human person. This is, of course, the very thing Ord sees himself arguing for.

Furthermore, elsewhere I’ve argued that if Friberg-Fernros’s appeal to time-relative interests is correct, his view is best described as a proabortion stance, as in many situations a woman’s time-relative interests make it morally obligatory to perform an abortion for relatively minor gain, even in cases where that woman takes a strong time-relative interest in having children.19 Consider the following case:

Vacation: Janine is young and has strong time-relative interests in both her career and having children. She becomes pregnant with donor sperm, but shortly thereafter she learns of a tremendous job opportunity that will require her to travel for a year, but will pay extremely well and will allow her to take an extended period of time off following that year of travel.

For Friberg-Fernros, it seems that Janine’s strong time-relative interests in her career advancement obligate her to have an induced abortion, as she can have a replacement child using donor sperm in the future.

Although the asymmetry response is able to explain why spontaneous abortion is not comparable to the deaths of adult human persons, it seems this response throws the baby out with the bathwater, rejecting the commonsense moral claim that so many antiabortion theorists believe—that human fetuses have a moral status comparable with that of adult human persons—the belief that most contemporary opposition to abortion rests on in both academic and non-academic circles.


Either (1) human fetuses, from conception or soon afterwards, are morally comparable with normal, adult human persons, or (2) they are not. Most prominent antiabortion theories—in both academic and non-academic circles—rest on the belief that (1) is true. As it so happens, there is overwhelming scientific evidence that far more human fetuses die from spontaneous abortion than from induced abortion. Indeed, if human fetuses have the same moral status as human adults, then far more things with that moral status die from spontaneous abortion than all other means combined. If this is the case, and (1) is true, then it seems we would be morally obligated to devote significant time and resources to studying and preventing spontaneous abortion and preventing unwanted pregnancies, rather than devoting time and resources to preventing induced human abortion. However, at the very least, it would be morally inexcusable to devote significant time and resources to preventing induced abortion and completely neglect the comparably far worse problem of spontaneous abortion and intellectually dishonest to discuss the wrongness of induced human abortion without even talking about the overwhelming similar losses caused by spontaneous abortion, especially when these losses could be prevented easily and cheaply with education and vigilance. Meanwhile, if (2), then antiabortion theorists face two challenges: first, they have to explain the wrongness of abortion in some other way, and second, they have to argue that the common-sense rationale for opposition to abortion—the belief that (1) is true—is false.

The problem of spontaneous abortion can be summed up as follows: either (A) spontaneous and induced abortions are not morally problematic at all, because it is the death of something without significant moral worth, or (B) induced abortion is morally problematic, but it—like every other threat to human life—is insignificant compared with the threat of spontaneous abortion. If either (A) or (B) is true, then the current opposition to induced abortion—both academic and non-academic—is morally negligent.

In this paper, I discussed six responses that antiabortion theorists have taken towards the problem of spontaneous abortion. The first—the numbers response—contends that the number of morally relevant spontaneous abortions is not as overwhelming as critics of antiabortion theorists make them out to be; however, even with Don Marquis’s numbers, it seems more fetuses die of spontaneous abortion than any other threat, far outpacing the threat of induced abortion. The second—the control principle response—contends that because we cannot stop many spontaneous abortions, we cannot have a moral obligation to stop those abortions. However, there are many spontaneous abortions we can prevent today, and still more we can prevent with future research. Furthermore, our inability to stop spontaneous abortions do not prevent us from grieving them, yet most antiabortion theorists fail to do even this. The third—the visibility response—contends that fetus’s lack of visibility lessens our moral obligations towards them, but such fetuses need not be invisible; while accidental ignorance absolves, willful ignorance does not. The fourth—the causing/allowing distinction response—turns on there being a morally significant difference between killing and letting die; however, even if there is a difference, it does not seem that the difference can be sufficient to completely ignore the deaths caused by spontaneous abortion. The fifth—the weak obligations response—contends that we only have relatively minor obligations to stop spontaneous abortion, yet it seems that most critics of abortion fail to satisfy even these minor moral obligations, despite the high stakes. Lastly, sixth—the asymmetric response—rejects the widely held belief that human fetuses are comparable with normal adult human persons, and thus the loss of fetal life is not all that morally significant. However, this view is at odds with the common-sense antiabortion stance in question; to take this stance just is to abandon the stance that abortion is comparable to killing an adult human person and to undermine the foundation of contemporary opposition to abortion.

Given the infrequency with which contemporary antiabortion theorists discuss either the frequency of spontaneous abortion or potential methods for preventing spontaneous abortion, I suspect that many opposed to induced abortion—in both academic and non-academic circles—are simply unaware of the implications of their belief that fetuses, from conception or soon thereafter, have the same moral status as adult human persons. Given their beliefs, their focus on preventing voluntary, induced abortion may very well be morally obligatory, in very much the same way that given Allison’s beliefs, postresearch, in Cursed Lamp, she ought to make wishes to benefit her,because there are substantial reasons in favour of it and no reasons against it. However, the moment Allison became aware that making wishes allowed substantive moral harm to befall others, it seems clear that she has a strong moral obligation to refrain from doing what she originally thought was morally acceptable. The same can be said of those opposed to induced abortion because of the belief that human fetuses are morally comparable with adult human persons; as soon as they become aware that spontaneous abortion is a much greater threat to human life than induced abortion, unless they reject this moral belief, as Ord believes they should, then their newfound awareness obligates that they act significantly differently. If Marino or Card are correct, then one might oppose spontaneous abortion relatively effortlessly while continuing to oppose induced abortion; otherwise it seems one’s moral obligations to prevent the loss of human fetuses would require a far more significant redistribution of the resources we are currently putting toward preventing (among other things) induced human abortions.  Contemporary antiabortion theorists almost universally fail to act in either way; suggesting they are either unaware of the overwhelming loss of life caused by spontaneous abortions, or behave monstrously when they oppose voluntary, induced abortions while ignoring the much greater loss of life caused by spontaneous abortions.


View Abstract


  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles