Policy Forum
May 2007

The Principle of Double Effect and Proportionate Reason

Nicholas J. Kockler, MS, PhD
Virtual Mentor. 2007;9(5):369-374. doi: 10.1001/virtualmentor.2007.9.5.pfor2-0705.


Mrs. Wilson recently discovered that she had an ectopic pregnancy—the embryo was implanted in her fallopian tube. Her physician admitted her to a Catholic medical center for treatment and scheduled a salpingostomy (a surgery that makes an incision in the fallopian tube through which the embryo is removed). But an ethics-savvy surgery center nurse questioned whether that procedure was morally permissible given the Catholic identity of the hospital. The nurse called the bioethics committee to inquire whether the doctor should perform a salpingectomy (surgical removal of the fallopian tube) instead, fearing that the Catholic Church considers a salpingostomy to be a direct abortion.

Mr. Jones has advanced metastatic liver cancer with neoplasms in his bones that cause excruciating pain. He has built up tolerance for virtually all pain medications; his doctors believe that one of the few remaining ways to alleviate his pain is to sedate him. Mr. Jones has said that he no longer wants any curative treatments. The Catholic hospital in which Mr. Jones is receiving care has a strong stance against euthanasia and physician-assisted suicide. Members of Mr. Jones's family approach his doctors and unanimously request, "Please end his life so he can stop suffering."

How does one begin to make sense of these cases? What morally viable options are available for the patients and health care professionals? The principle of double effect enables bioethicists and Catholic moralists to navigate various actions that may or may not be morally justifiable in some circumstances. The questions in this essay are the following: What are the principle of double effect and its proportionate reason condition? How do they function in clinical situations—at both the beginning of life and at the end of life? Despite historical and contemporary debates on the interpretation and application of these concepts, the principle of double effect developed within the long history of Catholic moral theology as a conceptual tool for determining the moral permissibility or justification of actions that have both good and bad (evil) effects. Proportionate reason is one of four conditions of the principle of double effect. In various ways, the principle of double effect and proportionate reason assist decision makers in moral analysis in both Catholic and non-Catholic health care settings.

The Principle of Double Effect

The history of the principle of double effect dates at least as far back as the work of St. Thomas Aquinas. Although St. Thomas did not use the term "double effect" or refer to the principle, he used the concept in justifying killing in self-defense [1]. In so doing, he recognized the bad effect (death of the assailant) and the good effect (preservation of the victim's life). Can one justifiably kill an attacker to save his or her life? St. Thomas answered in the affirmative. Likewise those who use the principle of double effect today attempt to discern the rightness or wrongness of actions that will have both good and bad (evil) effects.

To make such a determination, one must analyze an action on the basis of four conditions; all of which must be met for the action to be morally justifiable. The conditions of the principle of double effect are the following [2]:

  1. The act-in-itself cannot be morally wrong or intrinsically evil [3].
  2. The bad effect cannot cause the good effect.
  3. The agent cannot intend the bad effect.
  4. The bad effect cannot outweigh the good effect; there is a proportionate reason to tolerate the bad effect.

In analyzing acts within the framework of the four conditions, one considers that, if the act satisfies the four conditions, then the act is indirect and, therefore, morally licit. If, however, the act does not fulfill these four conditions (or, according to some interpretations, just the first two conditions) the act is direct and, therefore, the act is not morally licit [4]. Some theologians argue that application of the four conditions depends on several factors, not the least of which are how one formulates the conditions and how one describes the act. In fact, some argue that the first three conditions are three statements of the same moral proposition: the act cannot be intrinsically evil.

By analyzing our example cases we can appreciate why certain acts are permissible and others are not. In Mrs. Wilson's case, a traditional application of the principle indicates that salpingostomies are directabortions whereas salpingectomies are indirect abortions. This conclusion is not without controversy, especially given the development of salpingostomy as the standard of care for ectopic pregnancy [5]. Salpingostomy "directly" attacks the developing embryo, so it does not satisfy the first condition. A surgeon performing a salpingectomy, however, removes the pathological tissue (fallopian tube), which does fulfill the first condition. The death of the embryo does not cause, in and of itself, the good effect—preservation of the mother's life; it is the removal of the pathological tissue that causes the good effect, thus fulfilling condition 2. The agent (physician or mother who consents to the procedure) does not intend the death of the embryo, but rather intends the cure of the ailment, thus fulfilling condition 3. The last condition, whether there is a proportionate reason to tolerate the unintended bad effect, asks if the good effect (preserving the life of the mother) outweighs the bad effects—death of the embryo, and, incidentally with salpingectomy, reduction or elimination of the mother's fertility. I will examine proportionate reason more closely below.

In Mr. Jones's case, a traditional application of the principle of double effect indicates that one can administer pain medicine even if the patient's death is a foreseen, unintended consequence. How is this justifiable? Is this not euthanasia? The key for Catholic moralists in distinguishing palliative sedation (or allowing to die) from euthanasia (or killing) is the way in which the Catholic tradition understands intentionality. Here, the administration of pain or sedative medicine is not, in and of itself, morally wrong (fulfilling condition 1). The death of Mr. Jones, were it to happen, does not cause his relief of pain (fulfilling condition 2)—the sedative medicine accomplishes this. The agent, the physician or Mr. Jones's surrogate decision maker, does not intend on the death of Mr. Jones (fulfilling condition 3). This last statement may seem to contradict the statement provided by the patient's family. Nevertheless, a close examination of the intent behind their statement is his relief from suffering caused by his pain. Arguably, Mr. Jones's family sees his death as the only means to achieve this end or is unable to distinguish between pain relief and death. Again, is there a proportionate reason for tolerating the bad outcome that would permit sedating Mr. Jones? I now turn to that question.

Proportionate Reason

As mentioned above, proportionate reason grounds the fourth condition of the principle of double effect. How does one determine whether the good effect outweighs the bad effect? The phrasing of this question is immediately problematic. One of the main critiques of proportionate reason is its mathematical connotation: how can a good effect outweigh a bad effect, especially in end-of-life decisions where the bad effect is often death? Proportionate reason is a moral principle that one may employ to determine objectively and concretely the rightness or wrongness of actions [6]. Given the other conceptual problems with the principle of double effect, many Catholic theologians and moralists have appealed to proportionate reason in an attempt to delineate a more useful interpretation of the principle or to replace it entirely [7]. Thus, proportionalism developed in response to the more problematic approaches to the principle. One should note, however, that even in the traditional formulations of the principle, proportionate reason is a central feature of the four conditions, so traditional interpretations require a concept of proportionality [8].

One should not understand proportionate reason in purely mathematical terms, but rather as a balance between values and disvalues in determining whether the means (an act) is proportionate to the intended end or reason. The "reason" (ratio) here is not "some serious reason" that an agent identifies to justify the evil effect of the act; alternatively, what many commentators "mean by 'reason' [is] a concrete value which is at stake in the act of an agent" [9]. The term "proportionate" means a formal relation between the reason for the act and the premoral values and disvalues in the act [10]. "More specifically, the term signifies a proper structural relation (debita proportio) of the means to the end or of the end to further ends" [11]. Thus, the proper understanding of proportionate reason contains these two dimensions: the reason (ratio) and the proper structural relation (debita proportio) of the premoral values and disvalues involved in the action. Proportionalism is the general analytic structure for determining the rightness or wrongness of actions within which one appeals to proportionate reason [12]. Scholarship on the principle of double effect, proportionalism and proportionate reason is immense and complex. Unfortunately, I can only treat it cursorily here.

One can easily imagine the problems that emerge in the analysis of moral dilemmas using proportionate reason. Nevertheless, several thinkers have offered criteria for whether proportionate reason obtains. Walter describes some candidates:

(1) a non-contradiction between the means and the end or between the end and further ends, (2) the means do not undermine the end, (3) the means do not cause more harm than is necessary, (4) in the action as a whole the good outweighs the evil, (5) the means are in a necessary causal relation to the ends, and (6) the means possess the inherent ability to effect the end [13].

Considering first the definition of proportionate reason and second the criteria that establish it, one should recognize that there are various ways of knowing whether proportionate reason obtains. Walter suggests that there are two general ways of knowing: pre-discursive and discursive knowing. His discussion relates to moral epistemology (i.e., the study of moral knowledge), which need not be discussed in detail here. It suffices to say that several modes of knowing exist, from the intuitivist modes to those of discursive reasoning (i.e., analysis and argument) [14], all of which give one insight as to whether the criteria for proportionate reason have been fulfilled.

Pope John Paul II's encyclical, Veritatis Splendor, explicitly condemned proportionalism as a normative ethical theory [15]. But some Catholic theologians suggest that the pope's understanding of proportionalism may not have been entirely accurate [16]. The pope categorized proportionalism as a species of consequentialism, which the church condemns because, using consequentialist reasoning, a desirable end can justify any means. No Catholic moralist or theologian would agree with this extreme position. Like consequentialism, proportionalism is teleological, but one can distinguish it from consequentialism precisely because proportionalism accounts for both means (the debita proportio) and ends (the ratio). Moreover, proponents of proportionalism, so-called proportionalists, did not develop proportionalism explicitly as a normative ethical theory. Rather, it was an attempt to expand the fourth condition of the principle of double effect. Whether proportionalism evolved into a normative ethical theory is subject to further theological and philosophical inquiry beyond the scope of this essay.

In terms of our cases, one sees that proportionate reason exists in both. In Mrs. Wilson's case, one may claim that a salpingectomy fulfills the fourth condition because the good effect (preservation of her life) outweighs the bad effect (death of the embryo). Because the means (removal of pathological tissue) is indirectly ending the early life of the embryo, such means are proportionate to the intended end; there is a non-contradiction between the means and the end. What about a salpingostomy? Or administering methotrexate? The permissibility of salpingostomy requires a re-interpretation of the act in question and a determination of whether it passes the first two conditions. Is a salpingostomy a direct abortion?

In Mr. Jones's case, one may argue that terminal sedation fulfills the fourth condition because the good effect (relief of pain) outweighs the bad effect (death of Mr. Jones). Here, the means (palliative sedation) is proportionate to the end (relief from pain) insofar as it is the last remaining option. The question of alternatives can help physicians and surrogate decision makers discern what the true intentions behind certain requests are. Thus, a physician might ask Mr. Jones's decision maker, "If there were any other way to relieve Mr. Jones of his pain, would you want to pursue that option?" If he or she answers yes, then one can claim that his or her intent is not in the death of Mr. Jones, but relief of Mr. Jones's pain. The agent cannot intend both to cause the patient's death and relieve his pain. In this hypothetical case, if there are no alternatives to relieving his pain except for sedating him, there is a proportionate reason to do so, and such an act is not euthanasia (direct killing of Mr. Jones).


In both example cases, one finds justification for certain actions by applying the principle of double effect, which relies upon specific criteria to establish proportionate reason. Though controversial and subject to various interpretations, the principle of double effect and proportionate reason allow sensitivity to various moral issues in health care, especially from a Catholic perspective; they inform the moral reasoning behind several moral norms in Catholic teaching (e.g., in the Ethical and Religious Directives for Catholic Health Care Services); and they represent useful analytical tools for resolving complex moral dilemmas confronted by providers in a variety of health care contexts, Catholic or non-Catholic.

Read More


  1. Thomas Aquinas. Summa Theologiae. Fathers of the English Dominican Province, trans-ed. New York, NY: Benzinger Brothers; 1947:II, Q. 64, Art. 7.

  2. Kelly DF. Contemporary Catholic Health Care Ethics. Washington, DC: Georgetown University Press; 2004: 108-123.

  3. The "act-in-itself" is distinguished from the totality of the action, which includes the agent's intention (end sought), the act itself, and the circumstances.

  4. Kelly, 111-112.

  5. Kelly, 113-116.

  6. Walter JJ. The foundation and formulation of norms. In: Curran CE, ed. Moral Theology: Challenges for the Future, Essays in Honor of Richard A. McCormick, S.J. New York, NY: Paulist Press; 1990:132.

  7. See, for example, Hoose B. Proportionalism: The American Debate and Its European Roots. Washington, DC: Georgetown University Press; 1987; Kaczor C, ed. Proportionalism: For and Against.Milwaukee, WI: Marquette University Press; 2000.

  8. Kelly, 111.

  9. Walter JJ. Proportionate reason and its three levels of inquiry: structuring the ongoing debate. In: Kaczor CK, ed. Proportionalism: For and Against. Milwaukee, WI: Marquette University Press; 2000:394.

  10. Following Walter, I use the term "premoral" here to signify real value or disvalue that is present prior to an agent's moral judgment or decision; such premoral values and disvalues are relevant for moral judgment and decisions but do not determine the morality of an action.

  11. Walter, Proportionate reason, 394-395.

  12. Walter, The foundation, 131.

  13. Walter, Proportionate reason, 397.

  14. Walter, Proportionate reason, 399-403.

  15. See John Paul II. Veritatis Splendor. Vatican City: Vatican Press; August 6, 1993. http://www.vatican.va/holy_father/john_paul_ii/encyclicals/documents/hf_jp-ii_enc_06081993_veritatis-splendor_en.html. Accessed April 13, 2007.

  16. For examples, see McCormick RA. Some early reactions to Veritatis Splendor. Theological Studies. 1994;55:481-506; Selling JA, Jans J, eds. The Splendor of Accuracy: An Examination of the Assertions Made by Veritatis Splendor. Grand Rapids, MI: Eerdmans Publishing Company; 1994.


Virtual Mentor. 2007;9(5):369-374.



The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.