Article Text


Can the Catholic Church agree to condom use by HIV-discordant couples?
  1. L Bovens
  1. Correspondence to Luc Bovens, London School of Economics and Political Science, Dept. of Philosophy, Logic and Scientific Method, Houghton Street, London WC2A2AE, UK; L.Bovens{at}


Does the position of the Roman Catholic Church on contraception also imply that the usage of condoms by HIV-discordant couples is illicit? A standard argument is to appeal to the doctrine of double effect to condone such usage, but this meets with the objection that there exists an alternative action that brings about the good effect—namely, abstinence. I argue against this objection, because an HIV-discordant couple does not bring about any bad outcome through condom usage—there is no disrespect displayed for the generative function of sex. One might retort that the badness of condom usage consists in thwarting the unitive function of sex. I argue that also this objection cannot be upheld. In conclusion, if there are no in-principle objections against condom usage for HIV-discordant couples, then policies that deny access to condoms to such couples are indefensible. HIV-discordant couples have a right to continue consummating their marriage in a manner that is minimally risky and this right cannot be trumped by utilitarian concerns that the distribution of condoms might increase promiscuity and along with it the HIV infection rate.

Statistics from

Pope Benedict XVI recently provoked some public health dignitaries, researchers and aid workers by proclaiming that promoting condom usage risked increasing the rate of HIV infections in the developing world.1 The argument is that the distribution of condoms increases promiscuity and condoms do not eliminate but can only reduce the risk of HIV transmission. This is not a moral claim—it is a projection of a scientific nature. The Vatican appeals to empirical studies that assess the actual risk of HIV transmission with condom usage and that track the number of expected infections given alternative public health campaigns (abstinence, condom usage, combined campaigns, …) to support its case.2 But, in the end, the Vatican’s opposition to condom usage does not hinge on an exercise in risk analysis. Moral questions, for the Vatican, are not decided on utilitarian grounds.

What, then, are the grounds of its opposition to condom usage? The Vatican opposes extramarital sex in all forms and shapes, and so if the spread of condoms facilitates extramarital sex we can understand its opposition. But what does it say about condom usage within the bounds of marriage? Suppose that a couple is HIV-discordant, that is, one and only one of the partners is infected with the HIV virus. Would condom usage not be permissible for them? The Vatican has not taken an official stand, but, to say the least, it is also not forthcoming in condoning condom usage for HIV-discordant partners. So how might it defend its opposition in a way that is consistent with Church doctrine?

Thwarting the natural purpose of sex

The following reasoning underlies the Vatican’s opposition to contraception. Human activities have particular natural purposes inscribed in them by the Creator. Such a natural purpose imposes a normative constraint. That is, it is immoral to engage in the action while not respecting this natural purpose—it is not respectful of the intentions of the Creator and hence of the Creator himself. What does this mean with regard to sex? One such natural purpose of sex is procreation. We do not respect this natural purpose when we engage in sex and at the same time thwart its natural purpose. This is what contraception does and this is what makes it impermissible.

But then what about natural family planning, of which the Vatican approves? Is this not a sexual regimen designed to thwart procreation? Certainly, but the question is not whether we thwart procreation by choosing when to have sex. The question is whether we thwart procreation while engaging in sex. What is objectionable is to engage in sex in a particular manner that minimises the chances of procreation. What is not objectionable is to time when to have sex so as to minimise the chances of procreation. There is disrespect in using the Creator’s design in a way that contravenes his intentions. There is no disrespect in timing when to refrain from using the Creator’s design.

Here is another objection that will become particularly relevant to us. So what if the couple is (knowingly) infertile or what if the woman is post-menopausal? Is it not permissible for them to have sex? Certainly—they can have sex because they do not thwart the natural purpose of sex. They know that sex will not lead to procreation, but this is different from saying that they are somehow thwarting procreation when having sex.

The appeal to the doctrine of double effect

In normal circumstances, condom usage does thwart the natural purpose of sex. However, a marital relation with HIV-discordant partners is not a normal circumstance. It is inviting to appeal to the doctrine of double effect (DDE) to justify the use of condoms in such circumstances. The standard appeal to the DDE in Church doctrine is the following. A pregnant woman is diagnosed with cancer of the uterus. If left untreated, she will die before the fetus is viable. This is the simplest variant of the case. A hysterectomy—that is, a removal of the uterus—is the only available treatment. In this case, according to the DDE, it is permissible to remove the uterus even though this will entail the death of the fetus. Would this not be tantamount to abortion? The answer is no, and here is where the DDE comes in. If the good outcome of the procedure—the survival of the woman—is sufficiently weighty in comparison with the bad outcome—the premature death of a fetus that would not have come to term anyway—and we do not intend the bad outcome, neither as a means nor as an end, then the procedure is permissible.

Applying this to our case, the good outcome of condom usage is that the healthy partner will not become infected. This is clearly a sufficiently weighty reason relative to the absence of procreation. So long as one intends only not to infect one’s partner, and one does not intend to thwart procreation, neither as a means nor as an end, condom usage seems to be condoned by the DDE.

Does the option of abstinence void the DDE?

What does the Vatican say about this appeal to the DDE? There is no official statement, but considering its opposition to the distribution of condoms in developing countries with a high HIV infection rate, it is clearly reluctant to endorse this argument. Why so? What would they recommend to a couple in this predicament? When condoms are ruled out, abstinence seems like the only reasonable route. And indeed, this brings us to a qualification that we left out in our earlier presentation of the DDE.

Let’s go back to the hysterectomy case. Suppose that there is an alternative and equally effective treatment that does not involve the removal of the uterus (and the fetus inside it). Then, clearly, the hysterectomy would no longer be permissible. And what if that treatment is equally effective, but will cause more discomfort to the patient than the hysterectomy? The hysterectomy still would not be permissible. On the assumption that fetuses are persons with a right to life, this seems entirely reasonable. Appeal to the DDE is precluded when there are alternative courses of action (even if they involve discomfort) that do not yield the bad outcome.

So let us return to condom usage. The appeal to the DDE, one might say, proceeds too hastily, because there exists an alternative course of action that secures the good outcome of keeping the non-infected partner healthy—namely, abstinence. Granted, it may cause discomfort within the marital relation. However, the discomfort of the alternative treatment did not justify resorting to a hysterectomy. Neither does it justify resorting to condom usage.

Restoring the import of the DDE

Does this reasoning hold? Let us carefully consider the analogy to a hysterectomy when there is an alternative treatment. What was the bad outcome in the case of the hysterectomy? Well, clearly, the premature death of the fetus. And this bad outcome is present even if the sole intention was to take out the cancerous uterus and to save the pregnant woman. This is the reason why it is imperative to choose an alternative treatment if it is available. But what is the bad outcome in our special case of condom usage? It is presumably that the couple thwarts the natural purpose of sex. But do they? And if they do, would this really be a bad outcome?

Let’s take the easy case first. Suppose that the couple is practicing natural family planning. Then, clearly, they are not putting on that condom to thwart the natural purpose of sex. There is no procreation to thwart here, since they are choosing to have sex only when no pregnancy will ensue. What could they possibly be thwarting?

Once we grant that there is nothing objectionable to the usage of condoms by a couple in a marital relation with HIV-discordant partners practicing natural family planning, the public policy issue is decided. Condoms should be made accessible on public health grounds so that HIV-discordant couples—and there are many of them—have access to them. The couple can then secure a great good by continuing to consummate their marriage.

Now, the Vatican might worry that condoms might fall into the wrong hands and encourage promiscuity, which, given the failure rate of condoms in HIV protection, may leave a population worse off than they would have been with a complete ban on condoms. Fair enough, but these are utilitarian considerations. If an HIV-discordant couple can secure a great good in this manner, then we should not thwart their chances of doing so on the grounds that condoms become instruments of vice when they fall into the wrong hands.

The situation is analogous to the availability of morphine in the developing world. People are dying in excruciating pain because of the lack of morphine. This lack is due to the fear that morphine would fall into the wrong hands and enter the drug trade to the developed world in the form of heroin. Be that as it may, this is no reason to deny people the right to die with dignity. Similarly, whatever condoms may do to public morality and whatever the overall effect may be on the transmission rate of the HIV virus, this is no reason to deny a couple the right to secure a great good through enabling them to continue consummating their marriage in a licit manner.

Now comes the hard case. Would it then become permissible for condom users in this predicament to abandon natural family planning and to have sex all through the cycle? Let me repeat: this does not affect the policy issue of whether condoms should or should not be made accessible. That issue is already decided by my earlier argument. But it is an issue of sexual morality and the Church aims to provide guidance in this matter. So what can be said?

We stipulate that the couple chooses to use a condom in order not to infect the healthy partner. Of course they acknowledge that procreation will not occur, and they may either welcome this or not. We need to make a distinction that we have ignored so far in our discussion, namely, that between thwarting procreation and intending to thwart procreation. Clearly the couple need not intend to thwart procreation, though it cannot be denied that by putting on a condom during the fertile period they do thwart procreation. But is it a bad outcome to thwart procreation without intending to thwart procreation when having sex?

What would constitute a case of thwarting procreation without intending to thwart procreation? Imagine a couple who are fully open to procreation. Unbeknown to them, one of the partners is taking a medication or adheres to some diet, that, in the particular case, makes conception impossible. Or, suppose that, unbeknown to them, they favour a particular position (strictly involving vaginal sex) that, due to their particular physiologies, makes conception impossible. (I admit that these are fanciful examples, but just bear with me.) Did they do anything wrong? Would we say that what they did was wrong (though of course they are not culpable, because, in Aristotle’s terms, they did not know the particulars of their actions)? Or would we say that what they did was not wrong, because the wrongness consists only in intending to thwart procreation?

I am tempted to say that what they did was not wrong. As long as one does not intend to thwart procreation, there is nothing wrong with thwarting procreation. 3 4 It has the same status as having sex when conception is impossible due to infertility or due to the post-menopausal status of the woman. Note that the Vatican approves of the use of contraceptive pills for medical reasons—this is precisely a case in which one thwarts procreation without intending to do so.

In this respect, the case is radically different from that of a hysterectomy. In a hysterectomy, there is a bad outcome to one’s actions, namely, the violation of the fetus’s right to life, whether it was intended or not. In the case of condom usage, there is no bad outcome whatsoever, as long as thwarting procreation was not intended. This is the reason why you are obliged to choose the less comfortable alternative treatment in the hysterectomy, since you are obliged to avoid the bad outcome. But you are not obliged to choose the less comfortable route of abstinence, since there is no bad outcome to be avoided as long as condom usage is not intended to thwart procreation.

So there are two types of permissibility that the DDE can grant. In one type of case, permissibility is gained because the bad outcome is not intended, but it remains present. In this case, we do continue to bring about the bad outcome. Hence we are under the constraint of due care. If there is an alternative action that can avoid this bad outcome, even at some cost to ourselves, then we should choose the alternative action. This is the case of the hysterectomy. In the other type of case, permissibility is gained because what would have been a bad outcome if it were intended is no longer a bad outcome if it is not intended—its badness dissipates. Now there is no longer a constraint of due care. There may be alternative actions that avoid the outcome that would have been bad had they been intended. But why would you move in that direction? There is no reason to do so and there is even less of a reason to do so if they come at a cost. That is the case of condom usage in the special circumstance that we were considering.

There is a standard objection to this position. Janet Smith (2006, p52–9)5 argues that one simply cannot use a condom without intending to thwart procreation, because the end of thwarting procreation is inherent in (embedded in, intrinsic to) the action. Similarly, she writes, you cannot hit the brakes to signal danger without intending to slow down, cut off a leg to stop gangrene without intending to make the patient lame, or eat a gazillion hamburgers to win a contest without intending to absorb calories.

I find this position problematic. First, I repeat, how can natural family planners intend to thwart procreation when having sex with a condom during the infertile period—they are not even thwarting procreation when doing so. Second, Smith’s reading of “intending” is excessively broad and does away with the DDE even in canonical cases. How could one remove a cancerous uterus with a fetus inside it to save a woman’s life without intending the death of the fetus, on Smith’s reading? Is the death of the fetus not inherent to the act of removing the uterus that it inhabits? How could one administer morphine to relieve the pain without intending the life-shortening effect, on Smith’s reading? Is the life-shortening effect not inherent in the administration of morphine? (This is another standard appeal to the DDE to make the use of morphine as a painkiller consistent with the Church’s condemnation of euthanasia.)

The unitive function

So far, I have not mentioned the unitive function of sex. The unitive function is indeed another natural purpose of sex in Catholic sexual morality. The Aristophanes–Nozick line on sex and love may be helpful in this respect. Through loving, according to Nozick, we come to invest our identities into a shared we. We soften the boundaries of our private selves to construct an extended self with our beloved. Lovers no longer conceive of themselves independently of the union with the beloved.6 Now, a natural purpose of sex within a marriage is that it procures this extended self—it fuses two into one and it continues to fuse two into one. This was also Aristophanes’ view of sex in Plato’s Symposium, though he thought that the fusion was short-lived and could only provide temporary solace. We should respect this natural purpose as well every time we have sex. So the question is, when HIV-discordant partners use a condom, do they thereby thwart the unitive function of sex?

I see two lines of argument. The first is that the unitive function and the generative function—that is, what we have so far called the natural purpose of procreation—are intertwined. We cannot properly respect the unitive function without respecting the generative function. Certainly this is a controversial claim, but let us give it a charitable reading. If every time that marital partners have sex, they intend to close off any chance of procreation, then they will fail to genuinely form a we. Clearly this is different from saying that an infertile couple cannot form a we—they may know that procreation will not ensue, but they do not intend to thwart procreation. Similarly, let us return to the couple who unbeknown to them were thwarting procreation. Clearly this cannot stand in the way of forming a we, since they did not even know that they were thwarting procreation. So it is necessary for the unitive function to fail that the couple intends to thwart procreation. But if this is the case, then an HIV-discordant couple is not thwarting the unitive function of sex, since, as I argued before, they need not intend to thwart procreation.

Here is the second line of argument. One might just take the unitive function by itself and argue that this little piece of latex stands in the way of a genuine fusion. Proponents of this view write that, in sex with a condom, “husband and wife do not become “one flesh””7—it “is simply two bodies rubbing against each other or, in fact, rubbing against latex” (Smith, 2006, p48).5 What is essential in sex is that semen is actually deposited into the vagina.8 To use an analogy from a different spiritual tradition, think of New Age gardeners who feel that they do not become one with the earth unless they can take off their gloves and garden with bare hands, unless they can mix some of their sweat with the moist soil.

This position raises multiple concerns. Do condom users report a frustration in failing to form a we? More so than users of other forms of contraception? And where do we draw the line—would the use of lubricants also stand in the way of fusion? Furthermore, suppose that not condoms, but spermicidal gels, diaphragms or even contraceptive pills were to protect against HIV-transmission. With these methods, semen is properly deposited into the vagina. So would these methods then not be subject to the objection that they thwart the unitive function of sex? And would the Vatican then consider it to be permissible for HIV-discordant couples to use these methods? I find it hard to believe that this difference could make a moral difference.

But even so, suppose that we could work up some sympathy for the position that a couple really cannot become truly one unless genitals actually touch and seminal fluids make contact with vaginal secretions. Becoming a we is still not a binary issue—that is, there is a continuum between forming a shared we and fully remaining separate selves. And so we may not fully succeed in fusing with one another now that there is a piece of latex between us. We may lose some of the we in lovemaking. But then the question remains, would our HIV-discordant couple retain more of a sense of a we through committing themselves to perpetual abstinence? Is this little bit of unitive function that is thwarted by latex really more worrisome for the we than a cold bed would be?


The DDE can condone the usage of condoms by HIV-discordant couples in a marital relation when the intended outcome is to prevent the transmission of the HIV virus and not to thwart procreation. The availability of an alternative course of action—abstinence—that also brings about the good outcome is of no import. Furthermore, an appeal to the unitive function of sex is also unable to block the permissibility for condom usage for HIV-discordant couples. A public policy that prevents access to condoms to married couple with HIV-discordant partners would deprive them of a great good and it is no less worrisome than a public policy that keeps morphine out of the hands of patients in the Third World who are dying in excruciating pain. The fact that the accessibility of condoms and morphine may increase vice, be it promiscuous sex or heroin usage, may be a matter of concern, but not a sufficient reason to keep condoms and morphine out of the hands of those for whom they would constitute great goods.


I am grateful for comments from Foad Dizadji-Bahmani, Sara Fovargue, Ann Furedi, Alice Obrecht, Michael Otsuka, Alexander Pruss and Alex Voorhoeve.


View Abstract


  • Competing interests None.

  • 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

  • The concise argument
    Søren Holm