On ‘Midrashic Conservatism’ and the Status of Default Rules (Column 570)

The motivation for writing this column arose from a discussion I conducted with A’ here. The background is my claim about two types of conservatism, which I also presented in the series of columns 475 – 480. I will open with a brief review of this distinction and then move on to discuss A’s question. From there I will enter the question of the status of default rules that arose in our discussion.

The distinction: two types of conservatism

Imagine a group of people who have walked for generations in hot Africa, they and their forefathers, dressed in swimsuits. That is their tradition. At some point they arrive in cold Scandinavia, and the question arises whether to change dress. I distinguish between four positions:

1. One should keep wearing a swimsuit, since this is the tradition of our ancestors—to go in a swimsuit.
2. One should switch to warm clothing, simply because it’s cold.
3. One should switch to warm clothing, since the tradition is not “wear a swimsuit” but “wear clothing suitable for the weather.”
4. One should keep wearing a swimsuit simply because it’s an interesting challenge (life is richer that way).

Positions 1 and 3 are conservative, i.e., committed to tradition. Positions 2 and 4 are not committed to it. There is a difference between the two conservative positions and also between the two reformist ones in their reasoning. From this you can learn that there is no dependence between actual behavior and commitment to halakha. You can be committed to halakha and wear a swimsuit (position 1) or a coat (position 3), and you can be uncommitted to halakha and wear a swimsuit (position 4) or not (position 2). Moreover, we see that it is not necessarily true that conservatism is always stringent (continuing to wear a swimsuit even in the cold), nor is it necessarily true that lack of commitment is always lenient (putting on a coat when it’s cold).[1] Note that position 1, by definition, is leniently conservative, for while living in hot Africa they advocated going in a swimsuit, and there it fit the climate. Their stringency appeared only with the move to Scandinavia. Contrary to the natural inclination of many of us, the rule “It’s too good to be kosher” is not correct.

What interests us here is the distinction between positions 1 and 3, i.e., between two types of conservatism: position 1 is peshat-conservatism (plain/literal), whereas position 3 is midrashic conservatism. The peshat-conservative preserves the tradition as-is, literally, without cleverness and without interpretations. The midrashic conservative preserves the tradition with the same devotion and commitment, but in his view the tradition does not say “wear a swimsuit.” That is the literal surface. In his view, the tradition is the principle learned from a midrashic reading of the surface: the fact that they wore a swimsuit is incidental. The actual duty is to wear clothing appropriate to the climate. The midrashic conservative can be absolutely and fanatically conservative, just like the peshat-conservative, but he preserves the traditional principle as it emerges from midrash, not the literal principle.

What makes the midrashic conservative a conservative? What differentiates him from the heretic or reformer (positions 2 and 4)? The existence of the conservative midrash. The midrashic conservative justifies the change he proposes by a midrash that explains why this change preserves the tradition (the swimsuit is an expression of the duty to wear clothing suitable to the climate). By contrast, the heretic or reformer changes the tradition without offering a justification that this is the correct implementation of the tradition under the new circumstances. The conservative midrash is what determines whether you are a conservative or not. In addition, note that in the eyes of the midrashic conservative the tradition requires us to wear a coat. That is not a mere leniency but an obligation. In his view, the peshat-conservative who continues to wear a swimsuit is a transgressor, for he is not fulfilling the requirements of the tradition (perhaps under duress, because he thinks that is what is proper).

Let us now move to the halakhic plane. For this purpose, consider an example of an argument that could be raised in favor of qualifying women for testimony. The Talmud disqualifies women from testifying (Shevuot 30a). One could raise the following conservative midrash: women once lacked education and were not involved in economic life, and therefore were disqualified. Today they are not so, and therefore today they should be qualified. On the face of it, this looks like a reformist argument, but note that there is here a conservative midrash that justifies it. The claim is that the Talmud did not disqualify “women” per se but rather groups of people who were, by and large, uneducated and not involved in economic life.[2] This is a conservative midrash, and therefore this proposal is conservative (i.e., committed to halakha) and not reformist. The reformist, by contrast, will argue that they should be qualified because disqualifying them is unequal and harms them. He does not offer a conservative midrash but simply says he dislikes the rule (see elaboration in column 478). The peshat-conservative, in contrast to both, argues that if the Talmud disqualified women, then they are always disqualified; it does not depend on circumstances. It would seem that in his view the reason for this law is not lack of education or economic involvement but some hidden reason (and he also assumes, for some reason, that this hidden reason is necessarily eternal and not circumstance-dependent). Note that unlike the heretic or reformer, both kinds of conservatives assume that the Talmud’s rulings have authority and validity and are committed to them. The dispute is only about what exactly the Talmud instructed us: that women are disqualified, or that uneducated groups are disqualified.[3]

It is important to note that in practice you will not find pure peshat-conservatives. There are conservatives who tend more toward peshat-conservatism (Haredim) and those who do so less (modern religious). But everyone uses conservative midrashim from time to time. The distinction between peshat– and midrashic conservatism does not speak about people but about arguments. There are arguments of peshat-conservatism and arguments of midrashic conservatism. When you look at people, they bring both kinds of arguments in varying proportions.

How do we decide the conservative dispute?

How can one decide the dispute between the peshat-conservative and the midrashic conservative? Were women disqualified due to lack of education, or is this a decree with some hidden reason (or without a reason)? One can ask this of the midrashic conservative himself: how does he arrive at his midrashic conclusion? From where does he know that this is indeed the reason for the law at hand?

In principle one can seek evidence in the Talmud or in the writings of the Rishonim and Acharonim. For example, there is a broad discussion among the commentators about whether the disqualification of a wicked person (rasha) from testimony is because he is suspected of lying, or whether it is a formal disqualification of the person for some hidden reason (a decree). The commentators debate this and bring proofs either way. For example, some argue (an argument that can certainly be disputed) that if we find a Rishon who rules that a wicked person is valid for testimony in the case of a woman (to permit an agunah), then clearly in his view there is no suspicion of lying in a rasha and it is only a formal disqualification; otherwise we would not rely on him even in testimony about a woman.[4] Likewise, one can try to seek evidence from the Talmud and the Rishonim regarding the nature of the disqualification of women: is it a matter of unreliability or a formal disqualification? For example, whether women are valid to testify in matters of a woman’s status, in prohibitions, and so on.

If we have found evidence, the dispute is decided. In such a case even the staunch peshat-conservative will likely accept the ruling and agree that women should be qualified to testify. As noted, he certainly accepts the Talmud’s authority. The question is what we do if we have not found decisive evidence. Beyond the question of whether the midrashic conservative will convince the peshat-conservative (likely not), the more important question is what he will say to himself. After all, he too is a conservative, and preserving the tradition of his ancestors (halakha and the Talmud) is important to him. How will he decide for himself whether to accept the conservative midrash and qualify women for testimony, or, in the absence of evidence, to preserve the law as it is (like the peshat-conservative)? Here we arrive at our topic.

“Sit and do not act” is preferable?

If you think about it, you will discover that conservatism resides at this level. We saw above that if evidence is brought for the conservative midrash, then even the greatest peshat-conservative will accept its conclusion. Hence the dispute between peshat– and midrashic conservatism emerges specifically in situations where we have not found evidence for the proposed midrash: the peshat-conservative will say that in such a case one must continue the status quo, whereas the midrashic conservative will nonetheless consider change.

We must admit that the first instinct of a person committed to halakha is that in such a case “whoever changes bears the burden.” Whoever comes to change the existing halakhic situation bears the burden of proof. If you like: “sit and do not act is preferable.” Therefore, if we have not found evidence for the proposed conservative midrash—he should be rejected and we should continue to wear swimsuits (or disqualify women from testimony). How, then, can a midrashic conservative, who is also entirely committed to halakhic observance, act otherwise? Does he not accept these simple rules? After all, halakha itself states them. Here there are two further important points to consider: the question of cost, and the question of reason.

The question of cost

There is a sense that continuing the status quo is the safe route, and therefore anyone who changes must bring evidence for his proposal. Continuing the status quo has no cost, whereas change can carry heavy costs. But this is a big mistake. Consider the disqualification of women from testimony. You have women’s testimony about murder, theft, damage, or a loan. If you disqualify women’s testimony, then the murderer will be acquitted, and the money will remain with the robber and not reach the owner or the one to whom it is due. That is, if the midrashic conservative is right, then halakha says that women today are valid witnesses. According to him, the peshat-conservative is a transgressor, acting contrary to halakha. On that possibility, continuing the status quo carries a very heavy cost: a perverted judgment.

The conclusion is that continuing the status quo is not costless. On the contrary, usually when a change is proposed there are good reasons for it, and these are acknowledged even by peshat-conservatives; nevertheless, in their view we must still continue the status quo. This argument leads us to the conclusion that both change and preserving the status quo are steps with costs. Therefore, it is not clear that the status quo has any advantage. A policy of “sit and do not act” can certainly depend on the expected costs of change versus continuation. If we reach the conclusion that the cost is heavy, then perhaps the “sit and do not act” is precisely to change in order to minimize cost.

This can itself be formulated in two ways:

• “Sit and do not act” is not a directive to refrain from action but a directive to play it safe, or to adopt a policy that will yield minimal cost. Therefore, when the costs of continuing the status quo are heavy, the directive “sit and do not act” can be interpreted precisely as a recommendation to change it.
• In another formulation one can say that indeed “sit and do not act” is always a directive to refrain, but cost is a legitimate consideration, and sometimes it overrides the “sit and do not act” consideration.

In a different formulation: the directive “sit and do not act” applies only when we have no criterion whatsoever to decide between the options—when they are evenly balanced. But if one side entails a much heavier cost, we will decide based on cost; and now that we have a decision, there is no weight to the directive “sit and do not act.” In my article on leniency in halakha, I explained in a similar way the guideline to follow a lenient opinion in a time of duress. And in my article on autonomy in halakhic ruling I similarly explained the directive to rule according to what seems right to me and not to follow the majority or the rules of decision.

If we consider the cost factor, it is hard to decide our dilemma on that basis alone. I do not know how to measure whether the cost of continuing the status quo—assuming the midrashic conservative is correct—is heavier than the cost of changing it—assuming he is wrong. As noted, even if the disqualification of women is based on a different principle, it presumably has logic and a rationale. If so, qualifying women for testimony assuming they are in fact disqualified also has a cost (though it is hard for me to understand what it is). At first glance, the first cost is very clear (perversion of justice). There the damage is heavy and visible, whereas here we are dealing with a doubtful, hidden harm. And yet my assumption is that that harm exists as well. Even according to the peshat-conservative, the Torah did not disqualify women for nothing. There is some reason, even if hidden; hence accepting women’s testimony has a cost. In any case, it seems cost considerations lead us to the conclusion that continuing the status quo is not necessarily preferable, but still this is not a decision in the other direction. We have brought the issue to parity, but we still have no decision. This brings me to the question of reason.

The question of reason

The reason proposed by the midrashic conservative is reasonable. If women lack education and are not involved in social and economic life, there is logic in disqualifying them from testimony. Conversely, if we already see that they were disqualified, it is quite reasonable to attribute this to lack of education or involvement in life. Therefore, even the peshat-conservative must admit that this midrash has logic and is possible. However, in his view it is doubtful that this is the correct reason. He claims that perhaps there is some hidden reason that we do not understand, and therefore believes there is insufficient basis to change the law in question.

But if we have a reasonable explanation for some law, I do not see how a skeptical argument can dislodge us from it. In clarifying halakha or Talmudic analysis we always use our reasoning to explain things and reach conclusions. The claim that perhaps our explanation is not correct is not an argument. It is mere skepticism, and in any other context we would dismiss it with the dictum “a judge has only what his eyes see.” I am indeed not certain that I am right, but if I have found a reasonable explanation, why dismiss it in favor of an ungrounded supposition that perhaps it is not correct and there is another explanation? In other words: and perhaps it is correct? After all, even the peshat-conservative is not convinced that this reason is incorrect. So why is he not troubled by the same skeptical doubt?

In my article on the fifth principle I explained why the common interpretation of the rule that we do not seek the “reason of the verse” (ta‘ama de-kra) is not logical. It is customary to explain this by saying we are unsure whether our reason is the true one. Beyond the fact that I showed there that we do not seek ta‘ama de-kra even when the reason is explicitly written in the Torah (and then it is certainly correct), I explained that such a consideration cannot lead to the conclusion that we should not seek the reason of the verse. After all, there is also the possibility that it is correct. So if I have a reasonable explanation, there is no reason to prefer the possibility that perhaps it is not correct and there is another explanation that I did not think of. That is of course possible, but the first option is preferable; hence we should apply the rule that a judge has only what his eyes see.

In several Talmudic sugyot we find the statement “Shall we act merely because we reason so?!” (Gittin 19a and 37a), which simply means exactly this: we have a sevara (reason), but reason alone is insufficient to take action. In other words: even if you have a logical explanation, “sit and do not act is preferable.” But clearly this speaks only of a situation in which we have significant doubt regarding the reason and there is much logic in continuing the status quo (and heavy cost to changing it). If this were the rule in every situation, I would expect that we would never interpret anything based on reasoning, for there is always the concern that perhaps we erred.

The conclusion from all this is that we now have a principle that can decide the dilemma: if there is a reasonable explanation for some law, it is always preferable to the supposition that perhaps it is not correct and that there is some other hidden explanation that we do not know. Therefore, if the conservative midrash proposes a reasonable explanation for the law in question (the disqualification of women from testimony), it is preferable to the conservative supposition that perhaps this explanation is not correct and there is another explanation. Unlike cost considerations, this consideration gives clear priority to midrashic conservatism over peshat-conservatism. Here we already have a decision. This explains why the midrashic conservative is willing to rely on his reasonable reasoning to change the law and does not adopt the rule “sit and do not act is preferable.” That rule applies only when we have two equally balanced options, but not when one of them has a clear advantage.

The slippery slope concern

At this stage slippery-slope concerns always arise. For in this way one could abolish all of halakha. One can always propose midrashic suggestions to remove halakhic directives from their relevance and implement them differently or cancel them. If we accept conservative midrashim as part of the halakhic method, this could empty all our halakhic commitment of content.

To this I will note several points. First, behind this claim lies a common suspicion—as if conservative midrashim are a post-facto excuse. You are actually a reformer and only de facto present a “conservative” argument to reconcile your decisions and conduct. This is a suspicion, and suspicions are hard to argue with. But it is important to understand that a conservative midrash can also lead to stringent conclusions. If the tradition were to wear warm clothing in a hot place, the conservative midrash could conclude that one must wear light/cool clothing in a cold place—and that is a stringency. A conservative midrash can lead to many conclusions in stringent directions, not only lenient ones. Several examples of this can be seen in the critiques raised against my article in Makor Rishon: “A Gentile Whom Halakha Did Not Recognize.” Dear Kanai and Rabbi Chaim Burgansky raised there several such conclusions and wondered whether I would adopt them as well. My answer was of course yes. Contrary to the suspicions underlying their challenges, the midrashic conservatism I proposed is not intended to achieve particular goals—convenience, leniency, or alignment with fashionable values. In the eyes of the midrashic conservative, this is the truth and the will of the Torah under the given circumstances. Therefore, he does this both toward leniency and toward stringency. As we saw, in his view the peshat-conservative is a transgressor because he acts against the will of the Torah. It is important to understand that midrashic conservatism does not find a halakhic permission to act this or that way; rather, it interprets the halakhic obligation under the given circumstances.

I will also add the biased assumptions underlying the labels of leniency and stringency. In my article on leniency and stringency mentioned above, I explained that these concepts are often interpreted rashly. People tend to think that a leniency is what is comfortable for me and a stringency is what is uncomfortable. I showed there that this is not the correct interpretation. Here I will suffice with the well-known story about the woman who came to R. Chaim of Brisk and told him that her young son had been drafted into the Russian army, and asked whether she could travel on Shabbat to St. Petersburg to try to secure his release. R. Chaim permitted and urged her, emphatically, to travel immediately. When asked how he could be so lenient in the laws of Shabbat, he replied that he was being stringent in the laws of saving life. Likewise, qualifying women for testimony is not only a leniency but also a stringency: it is a stringency regarding the duty to judge justly and not to pervert judgment. From this perspective, the peshat-conservative is the lenient one. Also note that there is here an interpretive act whose aim also—and perhaps mainly—is to realize a halakhic value (to judge justly) and not only to adapt halakha to a modern value fashion (women’s equality). Again, this is a common accusation that is not necessarily correct.

But even if the midrashic conservative does not necessarily deviate from halakha and even if his intentions are pure, one cannot deny that this method could serve as cover for people whose aims are less pure. One can always present an argument that will permit whatever one wants; thus, de facto, it is a tool that can also serve transgressors (those who wish to shirk halakhic obligation) and can empty halakhic commitment of content.

To this I say two things: 1) Despite slippery slopes, truth stands in its place. 2) Not every midrashic argument is legitimate. There is common sense. I will now expand on these two points.

What about the truth?!

Even if these concerns are justified, and let us even suppose that the suspicions about the midrashic conservative’s dark motives were correct, it is still important to distinguish them from the truth. First acknowledge that he is correct on the principled level, and only thereafter discuss whether perhaps it should be prohibited. One cannot say that because of slippery slopes this is not halakha. In other words, one cannot, out of such a fear, negate the validity of the conservative midrash. Fine—but one can now still demand that we refrain from using such midrashim due to slippery-slope considerations. In my opinion, absolutely not.

As a result of the distinction between the essence of the law and concerns (slippery slopes), the following situation arises. We assume that in principle the midrashic conservative is correct, i.e., that on the principled plane women are indeed valid for testimony. But we are troubled by the possible consequences, and therefore think that it is not proper to use such midrashim because it could lead to the destruction of halakha. Yet even if there is a concern that this will lead to such destruction, what can we do?! If there were a Sanhedrin today, it could enact a decree not to use conservative midrash, and since it has the authority to forbid the permitted and permit the forbidden, this would be halakhically valid. But when there is no authorized institution, then if this is the halakhic truth—this is the truth—even if it may lead to problematic results. In our time, no one has the ability to enact a decree not to use such inferences. If the truth is that a woman is now valid for testimony, then she is valid. If we have concerns, we must address them, but at least as long as there is no Sanhedrin, this does not change the halakha itself. At most one can warn people and recommend that they be stringent (or lenient) nonetheless. But the halakha stands in its place.

Considerations of common sense

The fear that one can conjure “conservative midrashim” to permit anything and cancel any law is formalistic and childishly hysterical. Consider the following midrash-on-letters intended to permit wearing a coat in Scandinavia: In Africa, which begins with an aleph, our ancestors went with a swimsuit (beged yam), which begins with a bet. So in Scandinavia, which begins with a samekh, one should go naked (‘erom), because that begins with an ‘ayin. There is here a conservative midrash that interprets the “halakhic” directive thus: always dress in a manner whose descriptive term begins with the letter following the first letter of the place where you are. What would you say about such a “conservative midrash”?

Well, what one should say is that this is nonsense that does not hold water. It is obvious that someone who wants to be a transgressor can invent for himself any “midrash” he wishes. But someone who wants to be a transgressor does not need midrashim. He can simply violate halakha—and that’s that. Here we are dealing with people whose aim is to interpret halakha, and who understand that it is proper to interpret it through such conservative midrashim. Such people will propose reasonable midrashim (like disqualifying women due to lack of education), and as we saw, their logic is accepted as such even by the peshat-conservative; and only about that is there room to discuss. If someone brings the above “midrash of letters,” we will tell him not to talk nonsense.

Theoretically, one can always propose a “conservative midrash” for any purpose and in any direction, but practically in most cases such midrashim will not hold water. This fear is of a formalistic slippery slope. But not every valid argument is equivalent to any other valid argument. As is known, the validity of an argument merely says that the conclusion follows from the premises, not that it is true. The matter seems to me similar to rejecting kal va-chomer (a fortiori) because it can be used in formally invalid ways, such as obligating a four-cornered garment in mezuzah (see column 52): and if a doorpost (mashkof) that is exempt from tzitzit is obligated in mezuzah, is it not a fortiori that a four-cornered garment that is obligated in tzitzit should be obligated in mezuzah?! Formally it looks fine, but we all know it does not hold water (by the way, the explanation is by no means simple). Shall we therefore reject entirely the use of kal va-chomer?!

In short, my discussion is only about midrash that has logic in itself; and there, the opposition is not because it is illogical but because of conservatism as an ideology. This—and only this—is our discussion. In such a case, we have seen that there are very good reasons to prefer change over maintaining the status quo. Conservatism is illogical (see also in columns 217, 263, and others that deal with conservatism, where I also objected to it for precisely the same reasons).

We now come to A’s reflections from the above thread.

Interim summary and a note on “sit and do not act”

A asked there:

Now, I would like to ask: do you accept that there is weight to the rule “sit and do not act”? That is, sometimes people think it is merely a behavioral rule, applicable only in cases of “playing it safe.” However, in my opinion there is an a priori guiding rule here; namely, the burden of proof lies on the one who changes the norm. Do you agree?

It is important to note that even if we accept that this rule has weight, that still does not automatically mean that the preference is always for the peshat-conservative, since in cases where the cost is high the preference goes to the midrashic conservative. Let us suppose that in the case of accepting women’s testimony, non-acceptance is a cost that in my view is unbearable. Therefore, the logic is on the side of the midrashic conservative, even though he is acting in an affirmative manner (kum va’aseh).

I have gone on at length. In short: do you accept that there is an inherent advantage to the peshat-conservative because he passively continues the reality, whereas the midrashic conservative, who demands an active change, bears the burden of proof (unless, as noted, the cost favors him)?

As follows from what I explained above, there is indeed an inherent advantage to the peshat-conservative. But this is only if there is no difference in cost between the two possibilities and if the explanation proposed by the midrashic conservative is not more reasonable than the alternative. In such a case the rule is “sit and do not act is preferable.” On the other hand, in such a case the issue is not pressing. This is not a case of midrashic conservatism but of mere pilpul. The dispute arises only when the midrashic conservative proposes a reasonable suggestion and the peshat-conservative objects without offering an alternative. In such a case, as I showed, the rule “sit and do not act is preferable” does not apply.

Note that in his words above he wondered whether the rule “sit and do not act is preferable” has weight, or whether it is merely a behavioral rule. His intention is apparently the question whether in any case where there is a slight advantage in favor of the midrashic conservative we ignore the advantage of “sit and do not act,” or whether a certain threshold of weight is required in order to override it. Underlying his words is the intuitive understanding of the rule “sit and do not act is preferable,” namely that it is a default decision rule, which applies only when there is no other substantive way to decide. That is, if I face two perfectly balanced options, I should remain passive. From here it appears that in any other situation—even when there is a slight advantage to one side—that advantage will decide the dilemma, and the rule “sit and do not act is preferable” will be irrelevant.[5]

To this I wrote him:

The rule that “sit and do not act is preferable” has no logical standing in itself, but there is logic to acting in accordance with it. It does not say that this is the truth, but that for the sake of preserving the system it is worthwhile to act in accordance with it. It is not a rule aimed at playing it safe (for as I explained, it does not propose a costless safe route), but I certainly agree that it has weight and that it is proper to act by it in the absence of other considerations. My claim is that when there are other considerations (cost or reason), it is not correct to follow it.

There is ostensibly a problem in my formulation, for I presented the rule “sit and do not act is preferable” as a behavioral rule, that is, a decision rule without logic in itself. On the other hand, I wrote that it has weight, and ostensibly this means that a significant advantage on one side is required to decide against it. Thus, not every advantage decides the dilemma and allows me to act actively. How do these two statements fit together?

Default rules

A sensed this and asked:

Just one clarifying question: indeed, the midrashic conservative proposes a reasoning, and in this lies his advantage; however, his disadvantage lies in that he is making an active change (which requires “broader shoulders,” as opposed to “sit and do not act”). Hence, the less reasonable the reasoning proposed by the midrashic conservative, it is possible that in the final balancing the fact that he has the disadvantage of affirmative action will outweigh it. Is that right? [That is, the mere fact that he throws out a reasoning does not yet mean categorically that this reasoning will outweigh the disadvantage of affirmative action.] Or, in your view, are affirmative action and “sit and do not act” almost meaningless in the face of any reasoning that is proposed?

In other words, he raises the possibility that although we are dealing with a behavioral rule, it may have weight and a minimal advantage is required on one side for us to overcome it. This is exactly what I wrote above. Is there logic in such a conception? Ostensibly a behavioral rule is weightless in itself.

He reinforced his claim with an example he brings in his next message from the Trolley Problem (which has been mentioned here on the site more than once). This is a common dilemma in moral philosophy: a person stands at a railway junction with a lever that can divert a train from its track to another track. He sees a train approaching and down the line five people are lying asleep on the track. If the train continues, it will run them over. He can divert the train to the second track where only one person lies asleep, so that it will run him over; then only one person will lose his life. Should he do it, or remain passive, since “sit and do not act is preferable”?

On this A writes:

By the way, regarding the logical standing of “sit and do not act”—I think that in moral questions—for example, the trolley problem—the intuition is stronger toward “sit and do not act” and affirmative action. That is, there it is quite clear to us that when the decision is through affirmative action, it is automatically less preferable than a decision of “sit and do not act” (although this too is not an absolute rule and can change in practice, depending on the case). And one must consider whether “sit and do not act” and affirmative action in moral domains—where the intuition is clear—indicate that in the legal domain this should also apply, or whether we must distinguish between morality and law. This requires investigation.

Here we apparently see at least a side toward the notion that the rule “sit and do not act is preferable” applies even when there is a difference in cost. After all, there I compare the life of one person with the lives of five, and yet there is a side to remain passive, even at the cost of five people. Here there is ostensibly a difference in the costs, and nonetheless the rule of “sit and do not act is preferable” is applied. If so, it seems this is not a mere rule of conduct but a rule with intrinsic weight.

Comparison to presumptive possession

In my reply there I wrote:

This is an interesting question. In my view, conservatism has no weight in itself; that is, the existing way has no significant advantage over new proposals merely by virtue of being the existing way. Hence any consideration, however small, ought to override it—like presumptive possession (mohzakut), which is only a default rule, and ostensibly any evidence, however small, should override it. However, in presumptive possession this is not so, apparently because we desire legal stability. There is room to argue that the same applies to the rules of “sit and do not act.”

The rule that gives priority to the possessor (the defendant) is also ostensibly a rule of conduct. There is no reason to assume that the possessor is more credible than the claimant. In my pamphlet on migo I explained why we should not apply here the “presumption that what is under a person’s hand is his.” The reason is that while most objects in the world belong to those in whose hands they are found, within the subset of objects that are in legal dispute there is no such majority (especially since there is an opposing presumption that “a person does not make a claim unless he has [a basis] for it”—see Shevuot 40b). This is a subset for which we have no way to check the statistics, but a priori there is no reason to assume that the claimant is more of a liar than the defendant. Therefore, simply, this is a rule of conduct aimed at legal stability.

And nonetheless, not every evidence removes from the possessor. Bari and shema (certainty vs. doubt) does not, in practice, take away from the possessor. Likewise, the halakhic ruling follows Shmuel that in monetary matters we do not follow the majority; that is, a majority does not remove money from the possessor (at least according to some commentators). Migo to extract is not said according to most opinions (except for the Ramban). From all this it emerges that to remove from the possessor, not every evidence suffices. A minimal threshold of evidential strength is required to succeed in removing money from one who holds it. This is ostensibly puzzling, for as we saw, the advantage given to the possessor is a rule of conduct designed for cases of even doubt. If there is some decision, why give priority to the possessor? If the claimant has a migo, that is evidence that he is right, so why should the fact that he bears the status of claimant change the ruling?

As I explained, perhaps this is an extension of the concern for legal stability. By giving advantage to the defendant, we prevent a situation of baseless lawsuits. In other words, if we did not give priority to the possessor, we would face a wave of claims by people trying their luck with a false claim, and at least they would win partially (suppose we would rule that the money is in doubt and should be split). One could say that this very relationship would create a presumption in favor of the possessor, because if many claimants are deceivers, then there is evidence that the possessor is right. Consequently, we have no choice but to give the possessor an advantage—which itself would destroy the presumption in his favor and return us to a balanced state—but still there is sense in leaving him the advantage, because here there is at least a 1/2 chance that the ruling is correct, compared to the alternative that would lead to a ruling that is almost certainly perverted.

I brought a similar consideration in my pamphlet on migo to explain the idea of migo as argument strength (or credibility). I explained that we give an advantage to a person with a migo even in situations where he has no evidential advantage, because refraining from migo would not hurt liars. They would win in any case—either with the stronger claim or with the weaker one. Granting evidential weight to migo at least allows a truthful person to tell the truth without losing.

In any event, for our purposes it emerges that presumptive possession is indeed not a fact-finding rule, but it is a rule that has legal logic behind it (stability),[6] and as such it has non-zero weight. That is, to overcome it, not just any evidential force will do; a minimal strength of evidence is required. But for rules of conduct that have no legal explanation, there is no weight at all, and they are used only when the situation without them is perfectly balanced. This is the case with “sit and do not act is preferable.”

So why use them at all? For example, even when we have two balanced options before us, why prefer passivity? The answer is that apparently one needs a reason in order to act and does not need a reason in order to remain passive. The rule does not say that passivity is more correct, but only that there is no reason to act. Consequently, I remain passive. But unlike the case of presumptive possession, here it is clear that any weight whatsoever will decide in favor of action.[7]

Saving oneself with another’s life

Another example of this is the law of saving oneself with another’s life. A person is forbidden to kill Shimon in order to save himself or Levi. Simply, there is no advantage of Shimon’s blood over his own; rather, there is a determination of parity: neither person’s blood is redder than the other’s. Consequently, when I face the dilemma—me or him—I choose the passive path and do nothing. I do not kill him even if as a result I will be killed. This is a default rule, a rule of conduct applied in a state of parity. The implication of this rule being a default rule appears in the case where I am thrown from a window toward the ground where a baby lies. I can tilt myself to the side and die in order to save him, or continue to fall on him and be saved while killing him. Tosafot (Sanhedrin 74b, s.v. “veha Esther”; Pesachim 25b, s.v. “af na‘arah”) write that in such a case I certainly have no duty to tilt myself, despite the rule “who says your blood is redder,” and indeed because of this very rule. For that rule says that his blood is also not redder than mine, and the parity dictates passivity. In this case, passivity means continuing to fall without tilting, even if the result is that he dies and I am saved.

But here the question arises: what will be the law when there is a difference in “redness of blood” between the two sides? We saw that if this is a mere rule of conduct without intrinsic justification—a default rule—then it is intended only for situations of perfect parity between the two options. But if it is a matter of killing an elderly person to save a young person, or a sick person to save a healthy one, a woman to save a man (who is obligated in more commandments), a fool to save a wise man, etc.—why should the principle of passivity still apply there? Ostensibly there is a consideration, however small, in favor of one side, and consequently the dilemma is decided and the rule “sit and do not act” does not apply.[8]

Perhaps one can say that the value of life as such does not differ among these individuals; therefore this is a dilemma between perfectly balanced alternatives. Wisdom, righteousness, age, gender, health—none of these affects the value of life. It may affect what can be done with life or what is done in practice, but the value of life itself is equal in all—and that is what counts. In another formulation one can say that the value of life is infinite, and therefore any small addition is of no significance.

Back to the Trolley Problem

We are left with the question of why in the Trolley Problem we nonetheless use the rule “sit and do not act.” As we saw, this is a rule of conduct that instructs passivity, i.e., a default rule. If so, why not prefer the lives of five to the life of one? That is a consideration that can decide the dilemma; consequently, the rule “sit and do not act” should not be applied here.

It is hard to apply here the first formulation from the previous section, since the value of the lives of five people is certainly greater than the value of the life of one person. But if we adopt the second formulation—namely, the conception that the value of life is infinite—then perhaps we can say that the value of one life is entirely equal to the value of five lives, and once again the alternatives are balanced; hence, one should behave passively (“sit and do not act”). It may be that here we should also introduce the prohibition of murder committed by the person who diverts the trolley and include that as well in the equation. That can of course also explain the case of saving oneself with another’s life when the other’s life is “worth more” (discussed in the previous section).

Two levels of “sit and do not act”

In passing I will add that until now I applied the rule “sit and do not act” to the Trolley Problem in order to conclude that it is preferable not to divert the trolley. But there may be a different use of the rule for the trolley case, at a different level of the discussion. Just now I was asked about the Trolley Problem, and the questioner assumed I had a clear stance favoring passivity. I wrote to him that I have no clear stance on the matter. I added there a reflection that perhaps one can apply “sit and do not act” to the dilemma between the arguments (and not to the dilemma between the practical options). The argument in favor of passivity (“sit and do not act”) stands opposite the consequentialist argument (difference between the numbers of people), and in that dilemma itself I will again apply the rule “sit and do not act.” It is not clear to me whether one can apply it to a dilemma in which one of the horns is the rule “sit and do not act” itself.

Further there he asked what the limit is: how many people must be on the first track for me to justify moving the lever to the second track (where one person will be killed). I wrote that I do not know. But for the sake of discussion, let us suppose that in the case of one person it is clear that I will not divert the trolley, and in the case of a million—apparently I will. In intermediate cases I have a dilemma. What will I actually do in the intermediate cases? Perhaps because of the dilemma I will adopt the rule “sit and do not act.” But again this is the application of the rule to a dilemma in which one of the alternatives is the rule itself.

If indeed this rule has no weight in itself and is only a default rule—a rule of conduct for a situation in which the alternatives are perfectly balanced—then if a consideration of weight stands opposite it, it is not reasonable to say that this is a balanced state; and therefore, with respect to it itself, one cannot apply the rule “sit and do not act.” Applying the rule at the second level presupposes that the rule (at the first level) has non-zero weight, and therefore the first-level situation places before us a dilemma between two options of equal weight.

[1] See my reply to the objections to my article in Makor Rishon (many of them pointed to changes toward stringency and mistakenly assumed that I do not support them). Though all this should be discussed in light of my definition of leniency in my article here.

[2] There are always exceptions, but the halakhic disqualification addresses a group. The Torah speaks in terms of the majority (or the rule).

[3] I did not enter here into the question of ta‘ama de-kra (seeking the reason of the verse) for several reasons; I will mention two. First, this is only an example. Second, this is a matter of a derasha (exegesis), and in several places I have shown that in derashot we certainly follow the reason. The rule that we do not seek the reason applies only to laws written in the Torah. I also will not enter the question of authority (who has the authority to make such a consideration and decide on change—Sanhedrin? some halakhic decisor?); this question is not important for our discussion here.

[4] Of course one can wonder from where that Rishon inferred this, but let us leave that here. For the sake of the discussion I will assume that the evidence is from the Talmud itself.

[5] For similar considerations, see the two articles I mentioned above. In my article on leniency in halakha I explained in a similar way the guideline to follow a lenient opinion in a time of duress, and in my article on autonomy in halakhic ruling I similarly explained the directive to rule according to what seems right to me and not to follow the majority or the rules of decision.

[6] In my lectures on doubt and statistics (apparently in lectures 14–17) I discussed three possible types of explanation for rules in the law of evidence: (1) a probabilistic explanation (the evidence is reliable); (2) a legal explanation (there are legal considerations for adopting or rejecting such evidence—for example, the “fruit of the poisonous tree” doctrine rejects evidence obtained illegally even if it is good evidence, in order to prevent motivation to break the law to obtain such evidence); (3) legal intuition (an intuitive sense that this is the proper way to act, when I cannot find a probabilistic or legal justification). My claim above is that the rule giving priority to the possessor (and also migo as argument strength) is of type 2. In those lectures I brought examples of rules of type 3.

[7] It is not clear to me which of the three types of explanation in the previous note this explanation belongs to. It seems very close to type 3.

[8] I will recall that in my article on organ donation I supported taking the life of one who is brain-dead in order to save a sick person from death, even though there is killing involved, because the value of the life of a brain-dead person is less than the value of the life of a sick person who has the potential to recover completely.