A Look at Leniency and Stringency and Halachic Ruling: C. Minimal-Cost Determinations (Column 743)

With God’s help

Disclaimer: This post was translated from Hebrew using AI (ChatGPT 5 Thinking), so there may be inaccuracies or nuances lost. If something seems unclear, please refer to the Hebrew original or contact us for clarification.

In the first column of this series I defined the concepts of “leniency” (kulah) and “stringency” (chumrah), and we saw that a halachic leniency means opening up more options, regardless of how convenient any given option is for the subject of the law. In the previous column I discussed the implications of this picture for halachic ruling and the role of the decisor, and we saw that a posek is nothing but a halachic expert. His role is to present the different possibilities along with the implications and “price” of each one, while the questioner is the one who should choose among them and decide what to do in practice. I would now like to present a certain expansion of the concept of leniency, via decisions I will call here “minimal-cost determinations.”

Ambiguous situations: An asymmetry between leniency and stringency

I’ll begin with a special type of halachic situations in which it is difficult or impossible to determine what would even count as a lenient or stringent ruling. A first example appears in Ketubot 15b, where we find a discussion about majority status regarding an abandoned infant:

The text itself: If one found in it an abandoned infant—if the majority are idolaters, he is [treated as] an idolater; if the majority are Israelites, he is [treated as] an Israelite; if it is half and half, [he is treated as] an Israelite.

The Gemara rules that when an abandoned infant is found and his origin is unknown, his status is determined by the environment in which he is found: if there is a Jewish majority—he is considered Jewish; if there is a non-Jewish majority—he is considered non-Jewish. If it is half and half—he is in a state of doubt. The sugya then discusses various halachic ramifications in such a case, where we must decide based on the rules for cases of doubt (lenient for rabbinic law and stringent for Torah law). Yet a question may arise about what he should do regarding Torah study or Shabbat observance. In both of these commandments there is a situation in which a Jew is obligated (not merely permitted) to perform them, whereas a non-Jew is prohibited (not merely not obligated). Note that these are two “stringent” options in the sense that each one allows only a single mode of action (the other is forbidden). Regarding these two questions and similar ones, there is no solution that is simply a leniency or a stringency, since every halachic decision here carries a halachic “cost.” There is no decision that selects a “safe” course of action that could be deemed a “stringency.”

What generates the difficulty here? It is commonly thought that a stringent solution is one without a halachic price. For example, with respect to saving this infant’s life, either there is an obligation to save him (even on Shabbat) or there is not. In that case, a stringent decision would be that there is an obligation to save him. Such a decision could be seen as stringent because it carries no halachic price (at worst we saved someone whom we were not obligated to save). But in a situation where each of the two possible decisions entails a prohibition (=a halachic price) if we are mistaken, it is difficult to define what counts as lenient and what as stringent.

In a certain sense, the case of doubtful blessings is similar: on the one hand there is an obligation to recite blessings of benefit, and on the other hand there is a prohibition to recite an unnecessary blessing. Thus, every decision has a price: if we bless—perhaps we have recited an unnecessary blessing; if we do not bless—perhaps we have derived benefit from this world without a blessing. We saw there that the ruling is to be lenient, and it is done as follows: regarding the obligation to bless—this is a rabbinic obligation, and therefore we rule leniently. We saw that this does not mean we decide not to bless; rather, we decide that both courses are legitimate (as we saw, leniency is the absence of a norm, i.e., opening up the maximum number of options), whereas with respect to the prohibition of “You shall not take [God’s Name in vain],” if this is a Torah prohibition we rule stringently, i.e., not to bless. Only the combination of these two directives together yields the final halachic decision. In this case, since the obligation is rabbinic and the prohibition is Torah-level, the practical ruling is not to bless. Conceptually, the final decision is neither lenient nor stringent, since it pertains to what to do in practice, while leniency and stringency concern the existence or non-existence of norms, not the practical decision. The practical decision is a combination of two determinations in the normative plane: leniency in the laws of blessings together with stringency in “You shall not take [the Name in vain].”

What about the abandoned infant? There, a non-Jew is prohibited from observing Shabbat, and a Jew is prohibited from not observing Shabbat. These are two Torah-level prohibitions, and thus we have a clash that has no simple solution within the usual rules for doubt. Nevertheless, according to our definitions the situation differs from what one might have thought. The decision to be lenient is in fact defined even here. We saw that a decision to be lenient means that both possibilities are legitimate, and that will be true here as well. A lenient decision would mean that one may act in either of the two ways before us, e.g., instruct him that he may observe or not observe Shabbat, as he wishes. A stringent decision is a ruling that invalidates one of these options and leaves only one of them as legitimate.[1] Therefore, only the stringent decision is not uniquely defined in this situation, since we could be “stringent” by requiring him to observe Shabbat, or “stringent” by prohibiting him from observing Shabbat. Going stringent on both sides leads us to contradictory results. There are two possible decisions, both of which are “stringent.” The conclusion from this is that leniency and stringency are not always complementary determinations. We see here that there are cases in which the lenient state is well defined, whereas the stringent state is not.

It is true, however, that when we separate the discussion and consider each prohibition on its own, stringency is well defined. With respect to the prohibition of a non-Jew’s observing Shabbat, the stringent directive is: do not observe Shabbat. With respect to the obligation upon a Jew to observe Shabbat, the stringent directive is: observe Shabbat. The problem exists only when both prohibitions operate together in the same situation. But as we saw above, in such a discussion it is not correct to speak about leniency and stringency, since these terms refer to prohibitions (=norms) and not to operative decisions. Just as we saw regarding doubtful blessings, so too for the abandoned infant: the stringent state is well defined for each of the two prohibitions on its own, and only with respect to the overall operative instruction for the situation as a whole is it undefined.

The conclusion is as follows. We have seen that stringency and leniency are always defined with respect to norms and not with respect to practical decisions. We now add that they are uniquely defined only relative to a specific halachic norm when examined in isolation (a prohibition or an obligation). The ambiguity I described characterizes situations in which we must apply multiple halachic norms together. Thus, indirectly, the conclusion remains that the ambiguity characterizes reality, not the halachah. With respect to each isolated halachic norm there is a clear definition of stringency (the norm applies) and leniency (the norm does not apply). The ambiguities arise when multiple different halachic norms apply to the same practical situation.

It is worth recalling here the well-known story about R. Chaim of Brisk. A woman came before him whose son had been conscripted into the Russian army, and she feared for his life (spiritually and physically). R. Chaim instructed her to travel on Shabbat to try to save him from military service. When he was asked why he was being lenient in the laws of Shabbat, he replied that he was being stringent in the laws of saving life. Here too we see that the features of leniency and stringency relate to norms, not to situations. R. Chaim was stringent with respect to the norm of saving life and lenient with respect to the norm of Shabbat. Leniency and stringency characterize determinations relative to norms. As for the operative decision—what to do in practice in the given situation—there is no way to define it as lenient or stringent at all.

“Sit and do not act (shev ve’al ta’aseh) is preferable”

What should we do in situations where the two sides are symmetrical (a positive commandment versus another positive commandment, or a prohibition versus another prohibition)? Seemingly the natural solution for such cases is: “shev ve’al ta’aseh is preferable” (passivity is preferable). The source is in Eruvin 100a, which also deals with a situation of conflicting norms of the same type:

Rav Huna the son of Rav Yehoshua said: It is a dispute between Tannaim. Items that are given with one application that became mixed with items given with one application—apply once; those given with four applications with four—apply four; those given with four with those given with one—Rabbi Eliezer says: apply four; and Rabbi Yehoshua says: apply once. Rabbi Eliezer said to him: But he thereby transgresses “Do not diminish”! He said to him: But he thereby transgresses “Do not add”! Rabbi Eliezer said: They only said so when done by itself. Rabbi Yehoshua said: “Do not diminish” was stated only when it is by itself. And Rabbi Yehoshua further said: When you gave—you transgressed “Do not add” and you performed an action with your own hand; when you did not give—you transgressed “Do not diminish,” but you did not perform a physical action. According to Rabbi Eliezer, who says there: kum va’aseh (active performance) is preferable—here too he descends [to fix it]. According to Rabbi Yehoshua, who says there: shev ve’al ta’aseh is preferable—here too he does not descend. Perhaps not so—up to there Rabbi Eliezer said kum va’aseh is preferable only because he performs a mitzvah; but here, where he does not perform a mitzvah—so too he would not descend. And if you wish, say: Up to there Rabbi Yehoshua said shev ve’al ta’aseh is preferable only when he does not commit a transgression; but here, where he commits a transgression—so too he descends.

Without entering the details, the dispute is what to do when there are two opposing norms, both Torah-level. For Rabbi Yehoshua, passivity is preferable; for Rabbi Eliezer, activity is preferable. From the flow of the Gemara it emerges that the usual solution is that passivity is preferable, though in this case Rabbi Eliezer has special reasons to deviate (since he would transgress “Do not diminish,” which is considered an active transgression). In any event, the Rambam rules like Rabbi Yehoshua to give a single application; in other words, here too passivity is preferable. We may learn from here that where two norms of equal weight conflict and there is no distinction between the sides, the decision is: passivity is preferable.

As a rule, a positive commandment overrides a negative prohibition, but what do we do when there is a clash between one prohibition and another prohibition, or between one positive commandment and another? Here too we have a conflict between two norms of equal weight, and in such cases the rule is: “shev ve’al ta’aseh is preferable.” It would seem that we should proceed this way for the abandoned infant as well. Because we are dealing with two norms of equal weight, the ruling should be that he should not study Torah but should keep Shabbat.[2] Some might label such a ruling a stringency, since we are imposing a prohibition upon him, but as we have seen, this is not a stringency in the common sense, since even such a decision carries a halachic price (on the possibility that he is Jewish), and yet this is the halachic solution. We will now see that this is indeed a kind of stringency.

Rules for doubt as minimal-cost determinations

Why is this the solution in such situations? Let us consider the meaning of an ordinary stringent decision. A stringent decision means applying the norm to the case at hand. For example, if I am in doubt whether a piece of meat is kosher or not, the rule is: a Torah-level doubt is treated stringently. This means we assume that the norm applies to the case, i.e., that the prohibition against eating pork applies here, and therefore the action (eating the piece) is forbidden. Why is this the decision? Because it is the step with the minimal price. If we take it, we do not pay any halachic price—though we do pay an economic price, namely losing the piece of meat. I have previously explained that for rabbinic doubts we are lenient precisely because of this economic price. For a Torah prohibition the economic price carries no weight (since we must give up all our money rather than transgress a prohibition), but for positive commandments there is no such obligation, and thus the economic price takes on greater weight. Therefore, we may be lenient in a rabbinic doubt, for the Torah has pity on Israel’s money.

One implication of this understanding appears in the sugya of “a matter that will become permitted” (davar she-yesh lo matirin; see for example here). The rule is that for a matter that will become permitted—i.e., a time-limited prohibition—there are halachic stringencies: for a rabbinic doubt we rule stringently, and if such a matter fell into a majority of permitted substance it is not nullified (see Beitzah 3b–4a and elsewhere). For example, chametz on Pesach is not nullified by a majority, and some explain this as a case of davar she-yesh lo matirin (see Rambam, Forbidden Foods 15:9; others disagree). Likewise for the doubt about an egg laid on Yom Tov (see Rambam, ibid. 15:10). The common explanation for this rule is: “Rather than eat it in a doubtful manner, eat it permissibly”—that is, wait until after Pesach or after Yom Tov and you can eat it without resorting to bitul b’rov or leniency in rabbinic doubt. As is known, the decisors discuss whether there is value in being stringent regarding something that has technically been permitted, such as an item nullified by a majority or a rabbinic doubt. From here it clearly emerges that there is indeed such value. The permission to be lenient in a rabbinic doubt or by bitul b’rov is an ex post facto permission, and if there is a way to avoid relying on it, that is preferable. So if it is really preferable to be stringent, why does halachah allow leniency in rabbinic doubt or nullification by majority (for a matter that will not become permitted)? Apparently because it has pity on Israel’s money. Think of a prohibited egg that fell into a thousand permitted eggs. If we do not allow nullification, all 1,001 eggs would be thrown away. Therefore, halachah establishes that if there is a majority of permitted substance, there is no obligation to sacrifice it in order to avoid a prohibition (though it seems that one who nevertheless does so—may be blessed).

Thus, both “Torah-level doubt is stringent” and “rabbinic doubt is lenient” are minimal-cost considerations: either the halachic price (for a Torah-level doubt) or the economic price (for a rabbinic doubt).

Back to “shev ve’al ta’aseh is preferable”

This may also explain the rule “shev ve’al ta’aseh is preferable.” Intuitively, one might interpret it thus: to act, you need a reason; if there is no reason, the solution is not to act. But that would explain the advantage of the muchzak (the current possessor) in monetary law. There, the court must intervene and act, and halachah says that if it has no reason (i.e., the plaintiff has not brought proof), it should not act. But for doubts in prohibitions and permissions—such as a positive commandment versus another positive commandment, or a prohibition versus another prohibition—why should action require a reason? Seemingly I may do what I want, since there is no reason to forbid it. This is essentially a reflection of the “principle of legality” (link; see e.g., column 375), which states that governing authorities need special authorization to act, whereas for a citizen the situation is the reverse: one needs a good reason to forbid him from acting.

It therefore seems that the explanation is different. First, let us adopt the assumption that a negative omission (a transgression by passivity) carries a lower price than an act of wrongdoing (an active transgression). In column 416 I explained that this does not overlap with the distinction between a negative prohibition and a positive commandment. We can now understand that the solution of “shev ve’al ta’aseh is preferable” is nothing but a minimal-cost solution. We are told to sit and refrain, since even if we erred the transgression is an omission (passivity), which is lighter than an active transgression. Note that this is said about a clash of prohibition versus prohibition, or positive commandment versus positive commandment—i.e., it assumes that one can transgress a prohibition by passivity and a positive commandment by activity (as we saw in that column).

Interim summary

The conclusion is that the abandoned infant should observe Shabbat but not study Torah. This is not a stringency by definition, but rather a distinct method of decision-making in favor of the step with the minimal price. In this sense, there is a kind of stringency here, since, as we saw, a stringent decision is also a minimal-cost decision.

The question of minimal cost is deeply intertwined with the question of “leniency.” The reason is that in many cases decisors choose to be stringent because of minimal-cost considerations, and therefore they choose the decision that “passivity is preferable.” In light of what we have said here, such rulings should be examined anew, and at times surprising results emerge. I will examine two examples that I have discussed elsewhere in greater detail.

Separating conjoined twins[3]

There are cases in which two infants are born joined to one another in various ways and share an internal organ (such as a heart). In many cases the medical assessment is that if they are not separated, they will die within a short time. When the organ can be ascribed to one of them, the majority of decisors rule to perform the separation surgery and save the one to whom the organ belongs. But in symmetric cases—when the organ cannot clearly be ascribed to either one—I was astonished to discover that all the decisors agree that the separation surgery should not be performed. From my inquiry it emerged that underlying this determination is the notion that “passivity is preferable.”

In my article I argued that when we assess the price of the two options we must consider not only whether I have committed the transgression of murder, but also the outcome, i.e., whether lives were saved. The decisors assume that choosing “passivity is preferable” means not performing the separation surgery. One might understand that this too is a minimal-cost determination in that situation, since such a decision certainly avoids the prohibition of murder. But once we understand that the rule “passivity is preferable” is merely one expression of a minimal-cost consideration, a more plausible option arises in my view: it seems that the opposite decision is the minimal-cost determination, for beyond the consideration of the prohibition of murder there is also the consequentialist consideration. In terms of the outcome, if we perform the separation surgery we lose only one life rather than two. Once we understand that we are dealing with a minimal-cost determination, we must weigh all the costs, not only the question of which transgressions I commit.

Organ donation[4]

In another article I discussed the issue of the obligation to donate organs to save another person, even before the formal moment of death. It is commonly thought that the permissibility of donating life-sustaining organs (such as a heart or lungs) depends on the dispute about the moment of death. If one holds that death is determined by brain death, then after the brain has died one may harvest organs for transplantation, even if that harvesting causes cardiac death, because the person was already dead at the time of harvesting, and we did not kill him. But if one understands that the moment of death is cardiac death (which is later), then harvesting the heart before cardiac death constitutes murder, and one may not kill one person in order to save another.

My claim in that article was that this is not necessary. The reason we do not kill person A to save the life of person B is that B’s blood is not redder than A’s (mi yemer didama didakh sumak tefei). But when A has already suffered brain death, even if we do not see brain death as the moment of death, it renders him “less red.” The claim is that to kill a person who has suffered brain death, even if halachically he is considered alive, is less severe than leaving a fully living person to die (in that article I also addressed why we do not extend a similar consideration to old versus young or healthy versus ill). For our purposes here I will not enter the details of the argument but rather its logic. My principled claim was that even if the moment of death is cardiac death, and harvesting the heart means killing a living person, there is still room to permit it to save another person’s life, since the blood of one who has suffered brain death is “less red.” In other words, the question of organ harvesting does not depend on the dispute about the moment of death, and the permissibility exists according to all views (at least when the heart-owner consented).

Here too we are dealing with a shift from the consideration “passivity is preferable” to the consideration of minimal cost. The initial inclination to forbid killing a person is a ruling of “passivity is preferable.” We have life versus life and no way to decide whom to prefer. That is a “passivity is preferable” consideration. But once we understand that the rule “passivity is preferable” is merely a particular expression of a minimal-cost consideration, we may—and should—introduce the outcomes into the equation: if we harvest the organ, we will lose the life of a brain-dead patient (whose life is so diminished that some decisors consider him already dead). That is certainly preferable to losing (or failing to save) the life of a person living a full life.

There too the sense among decisors is that the prohibition on donating organs before the moment of death is out of concern for the transgression of murder (by the doctor) or self-harm leading to death (by the patient). They focus on the halachic price. And there too my claim is that we must consider the consequential price in terms of saving and losing lives. Therefore, precisely the decision to permit organ donation is the minimal-cost determination in that case.

The meaning: broadening the term “shev ve’al ta’aseh”

A “shev ve’al ta’aseh” decision always means being passive. But as we have seen, in essence such a decision is one that chooses the side with the lower halachic/moral price. Once this is understood, we discover that at times we must choose a step that takes active measures, because that is the determination with the minimal price—even if it is not “passivity” in the physical sense.

In column 416 I broadened the term “shev ve’al ta’aseh.” I argued there that the nullification of a positive commandment is, in essence, always a transgression by passivity, even when it is accomplished by a physical act. The reason is that nullifying a positive commandment means failing to be in a positive state, and in that sense such a transgression is a passive one. Thus, for example, one who performs labor on Shabbat has violated a prohibition but has also nullified the positive commandment to rest on Shabbat. He performed a physical act, but he nullified the positive commandment to rest on Shabbat, and in that sense this is a passive transgression. He was not in the desired state (rather than being in a forbidden state). The same applies to eating on Yom Kippur.

I will now make a similar claim in our present context. Minimal-cost determinations are, in essence, “shev ve’al ta’aseh” determinations in the substantive sense. “Shev ve’al ta’aseh” means that in the absence of another consideration one chooses the minimal price. Sometimes that is achieved by omission (inaction), but sometimes it requires action. And yet it is still a “shev ve’al ta’aseh” determination, since it tells me not to pay a high price without justification. Perhaps this is merely semantics, but because the term “shev ve’al ta’aseh” has become entrenched in halachic discourse, it is important to understand that my proposals here do not fundamentally contradict the rule “shev ve’al ta’aseh is preferable.”

Summary and closing the circle

Decision rules of the “minimal-cost” type relate to practical situations, whereas decisions of leniency or stringency relate to norms. The two types of consideration and decision complement each other, since each pertains to one of the two planes defined in the previous columns. Once this is seen, the ambiguities that accompany these concepts become clearer. When we understand that stringency and leniency relate to norms and not to the practical choice, there is no longer ambiguity. For each norm it is clear what is lenient and what is stringent: the option that recognizes the legitimacy of more courses of conduct is the lenient one, and applying a norm is the stringent one. By contrast, the practical decision must factor in the different normative sides involved in the determination, and here the decision is according to minimal cost, not in terms of leniency and stringency.

This picture sheds new light on novel halachic rulings that many see as lenient (such as the two examples above regarding organ harvesting and separating conjoined twins). At times these are not leniencies at all but minimal-cost determinations. The perception that these are leniencies feeds on a partial view of reality and of halachah. When all considerations are taken into account, this is the decision with the minimal halachic/moral price—and in a certain sense it may even be seen as a stringent determination.

The connection to the question of change

I concluded my article on leniency and stringency in halachah with a discussion of the connection between leniency/stringency and change in halachah. When someone proposes a halachic change (such as improving women’s status, etc.), he is immediately perceived as being lenient. The rule “shev ve’al ta’aseh is preferable” is replaced by the maxim “whoever changes, his hand is on the bottom.”[5] Change is leniency and conservatism is stringency. This is the basis for the tendency to oppose any change and to place the burden of proof on one who calls for change.

But as with the rule “passivity is preferable,” here too we can run a parallel analysis. The rule opposing changes applies only when change and non-change carry the same price. In such a case, conservatism is in fact an implementation of the rule “passivity is preferable.” But in cases where non-change carries a heavy price, anyone who refrains from changing is “on the bottom.” Change is perceived as an act with a high price. But if we are dealing with a change that reduces the halachic and social price, we may see it precisely as the default and place the burden of proof on the one who prevents the change. As we saw above, we may view R. Chaim’s ruling as a leniency in the laws of Shabbat or as a stringency in the laws of saving life. The same applies to changes in women’s status, which may be seen as a halachic leniency or as a stringency in equitable treatment of women.

The treatment of such rulings as leniencies stems from ignoring the prices attached to the decision that appears “stringent,” which actually make that decision the lenient one. Underlying this is the failure to recognize that halachic stagnation also carries heavy prices, and it is an illusion to think that stagnation is a “safe” solution—i.e., always the minimal-cost solution.

[1] To sharpen the point: according to the Rambam, the obligation to be stringent in cases of doubt is only rabbinic; on the Torah level, a doubt in Torah law is treated leniently. According to his view, for an infant found in a half-and-half environment, on the Torah level both options are legitimate, and the “infant” (once grown) may choose between them himself.

[2] One can dispute this and argue that the prohibitions are not of equal strength (the obligation upon a Jew to keep Shabbat is stronger than the prohibition upon a non-Jew to do so). I use this case only as an example and therefore will not enter a more detailed discussion.

[3] See more detail in my article “Separating Conjoined Twins.”

[4] See more detail in my article “Organ Donation.”

[5] This is, of course, a borrowed expression. In its original context (see Mishnah Bava Metzia 7:6 [76a] and parallels) it expresses a monetary rule and not a meta-halachic approach. I use it here as a label for a common meta-halachic outlook, for which this phrasing is the usual description.