Okimtas – A Mini-Lecture and Exercises (Column 632)

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.

The phenomenon of okimtas (restrictive readings) causes discomfort for many learners. In my article “A Platonic Look at Okimtas,” I tried to explain why this difficulty is only apparent, and argued that okimtas are a necessary feature of Talmudic discourse. The explanation I offered makes claims about Talmudic sugyot and the proper way to approach them, and these claims can be tested across different sugyot (in principle, this is a falsifiable thesis). Indeed, several people have already presented me with sugyot in which, ostensibly, my thesis fails the Talmudic test, and I think that at least so far I have always managed to show that this is not the case.

A few weeks ago two more sugyot were sent to me, and I thought to use the opportunity to briefly present the general picture from that article and then bring several examples in order to practice applying it. If the article about okimtas is the lecture, these applications are the exercises. As in university, such practice usually helps sharpen the general picture presented in the lecture and its implications. Beyond that, following the discussion a few more nuances will be added here to my picture of okimtas.

Background: An explanation for the okimta phenomenon

To keep it short, I’ll do this through a well-known example. The Gemara in Gittin 78b brings a statement of Rava:

Rava said: If he wrote her a get and placed it in the hand of her slave who is asleep and guarding it—this is a valid get; if he is awake, it is not a get, for this is like a courtyard that is guarded not with her knowledge. [If the slave is] asleep and guarding it—this is a get. Why? Is it not a moving courtyard, and a moving courtyard does not acquire! And if you would say “asleep is different,” did not Rava say: Anything that would not acquire while walking does not acquire while standing or sitting?—and the halacha is [that it is valid] when [the slave is] bound.

Rava says that you can give a woman a get by placing it in the hand of her slave when he is asleep. The Gemara challenges: this is a moving courtyard, and even when he is asleep it is still deemed moving. It concludes that the case is a slave who is asleep and bound.

The difficulty is obvious. Rava does not so much as hint that the slave is bound. It is not plausible that he intended to introduce a ruling specifically about a bound slave, since a crucial detail would be missing. So how can the Gemara “set” (do an okimta of) his words in a way he did not originally intend? Seemingly, Rava said you can give the get via any slave, bound or unbound, and the Gemara removes his words from their plain sense. True, this is an okimta the Gemara does to an Amoraic statement, whereas usually okimtas are made to Mishnayot or Tannaitic sources, but the problem is the same.

In my article on okimtas I brought several common explanations and showed that they are implausible. I argued there that the phenomenon can be understood by two assumptions: (1) When the Mishnah presents a specific ruling, it always intends to establish a general principle, and the case cited serves only as an illustration of that principle. (2) A general law never manifests in its pure form within a concrete real-world situation. To see it, one must examine it in a Platonic, abstract situation (illustrated by a laboratory case).

To illustrate assumption (2), think of Newton’s first law: a body upon which no force acts moves at constant velocity in a straight line (no change in speed and no change in direction). No one has ever seen this law occur as stated. There is no situation in our world in which a body moves with literally no forces acting on it. In order to test this law we must imagine a world with no air and no friction, no other bodies besides the one in question, temperature at absolute zero, and perhaps other impractical features. This is an imaginary situation that never actually occurs in our reality. The same goes for Newton’s law of gravitation: it never acts “alone,” because there are always other forces acting on the bodies in question. Here too we speak of a fictional reality. Practically, this is why science needs laboratories: the lab tries to create a state that approximates, as closely as possible, that imaginary-Platonic situation. If we evacuate air, lower the temperature to zero, remove all other bodies, and then measure the motion of the body, we will get a result very close to the general scientific law (never exactly, because the state we created is still part of reality and can never be perfectly “clean”).

My claim there was that an okimta is essentially a laboratory setup whose purpose is to demonstrate the general principle in the Mishnah or statement in its purity, without interference from additional principles. Returning to the sleeping slave: the Mishnah’s purpose is to say that a slave is like the master’s courtyard. This principle is always true, for every slave, bound or unbound, asleep or awake. It can have ramifications in the laws of acquisitions, divorce documents, chametz, and so on. It is a general principle, and therefore the statement speaks of “a slave” without specifying that he is bound.[1] However, the example Rava chose to illustrate this principle is acquisition of a get through a chatzer (courtyard). But in the laws of kinyan chatzer there is a requirement that it be “guarded with the owner’s knowledge,” and it must not be “moving.” Therefore, the ruling is set in the case of a slave who is bound and asleep. Note that the general principle—that a slave is his master’s courtyard—is true for every slave. That is what Rava is saying; he is certainly not speaking only of a bound, sleeping slave. Rather, he presents the principle through a specific example of acquiring a get. The need to make an okimta does not arise from the principle Rava is teaching (since, as stated, it applies to every slave), but from the laws of kinyan chatzer, which are merely the incidental example chosen to express that principle. For the example to work, the slave must be bound and asleep. My claim is that an okimta never touches the conceptual innovation of the Mishnah or statement. That is always presented precisely. The okimta is always offered to solve a side problem that relates only to the incidental example through which the conceptual innovation was expressed.

Consider Newton’s law of gravitation, according to which any two masses attract each other with a force inversely proportional to the square of the distance between them. No one states this explicitly in a world with no other bodies and no friction, etc. Otherwise, other forces would act on the body. The reason is that all those are technical constraints that do not touch the general law itself. The general law is always true—with or without friction, with or without other bodies. Its actual manifestation in the world is enabled only in Platonic laboratory situations, i.e., in a fictional reality without friction and without other interfering bodies. Framed differently to sharpen the point: one can express the law of gravitation as follows—when a ball is subject to the influence of some planet, it will accelerate at such-and-such a rate. This is an instance of the general law about the force exerted between any two masses, but it is expressed through one specific consequence (the acceleration of a ball under the influence of a planet). The general law is always true, but this specific consequence is not true in the presence of friction or other bodies. The force exerted by the mass is indeed well-described by the general law, regardless of friction and other masses; but its practical expression as acceleration requires an okimta, i.e., a Platonic description of a situation in which this occurs (in which the acceleration is indeed as stated).

In psychology, too, one can speak of a law that aggression stems from frustration. This does not mean that frustration will always cause aggression. Education and social pressure may lead a person to restrain himself and not act aggressively. Therefore, the actual manifestation of the general law occurs only if that person received no education and is under no social pressure—something that of course never happens. The general law is entirely true: frustration does cause aggression. But the concrete emergence of aggression depends on other circumstances and constraints, so it will be seen only in a laboratory setup that eliminates them.

I brought a Talmudic example as well (attributed to R. Elchonon Wasserman). The Gemara says that “one who recites Havdalah over a cup will have male children” (see Shevuot 18b; of course there are many other ways to merit sons—see the sugya there, Berakhot 5b and 57a, Bava Batra 10b, Niddah 70b). Yet our eyes see that many who recited Havdalah over a cup did not have sons, or had only daughters. Does this contradict the Talmudic statement? Not necessarily. There are other considerations that influence whether a person will have male children: that he has a wife and is intimate with her at the right time; that he has no sins of vows (for due to the sin of vows a person’s sons die—see Shabbat 32b; and there are other causes there, as well as in Moed Katan 25a), or other sins that could render either spouse liable to death (which vows can also effect). Therefore, the actual manifestation of the Talmudic assertion requires an okimta: we are speaking of a person with no issues of vows, who married and was intimate at the right times, with no physiological impediments for either spouse, and so on.

In that article I presented further Talmudic sugyot through which I showed applications of this explanation. One can see there that a proper understanding of okimta does more than solve the “okimta problem”; it also sheds light on the sugya as a whole and resolves additional difficulties that arise in it.

A note on the Talmud’s Platonism

The topic of okimtas came up in my series of lectures on Platonism in the Talmud and more generally. There I explained that, based on this understanding of okimtas, the case an okimta deals with need not be practically realizable. Thus the Talmud allows itself to speak of “an elephant that swallowed an Egyptian wicker basket,” “wheat that descended in the clouds,” “one who became stuck in his yevama,” “a warning issued by a demon,” “a flying camel,” and other imaginary cases. This is merely an expansion of “a bound, sleeping slave,” which itself is not exactly a practical case. The Gemara is not bringing those cases to discuss them for their own sake; they are Platonic laboratory cases intended to present a situation (even a fictional one) in which a general halachic principle is expressed. Sometimes one needs, or at least it is convenient, to resort to a fictional situation in order to eliminate interferences and create the laboratory state in question. On this point the questioner raised an additional query, this time from a different sugya in the same chapter. We will return to this momentarily.

First example: The rule of “proximate” in monetary law and in prohibitions

Consider another example that came up in a Q&A on the site a few weeks ago:

A small question, if possible. In the Mishnah in Shekalim (7:1): “Coins found between the [chest for] shekels and the [chest for] donations—if closer to the shekels, they fall to shekels; if to donations, they fall to donations.” Likewise, Rambam rules in Hil. Ma’aser Sheni (6:12): “If one found produce between produce of ma’aser and produce of terumah—they fall to whichever is closer.” Many asked: but when “majority” and “proximity” conflict, we follow the majority and not the proximity (Bava Batra 23b). Why needn’t we check which side contains the majority?

Bartenura writes that the piles are equal. Several later authorities (Radbaz on Rambam; Tiferet Yisrael on the Mishnah in Shekalim; Sha’arei Yehudah on Rambam in Hil. Gezeilah, and more) wrote that it is difficult to fit this into Rambam, who simply stated the law without noting that the piles are equal. There are other explanations in the Acharonim; none satisfied me. If you have a bright idea, I’d love to hear it (this has practical implications for a ruling; otherwise I would not trouble you). Thank you!

I answered him as follows:

I don’t think this is a “bright idea,” but it does seem to me a reasonable and correct explanation. The Mishnah—and Rambam following it—here present the rule of karov (proximity). The novelty is that the rule of proximity applies both in issur-veheter and in monetary law. They are not entering into practical questions such as what happens if there is a conflicting majority to one side, or if the coin resembles those in one pile and differs from those in the other (maybe one pile is shekels and the other dollars), etc. That is the Talmud’s way—and hence the need for okimtas. I do not see any real difficulty here.

By the way, in light of what I explained in my article on okimtas, you can understand that even those who set the case to equal piles do not truly mean that this is what the Mishnah itself is “about.” Their intent is that the Mishnah comes to teach the rule of proximity. And when you ask, “what about a case with a conflicting majority (which prevents the manifestation of the rule of proximity)?,” an okimta will be made that the piles are equal. But the underlying novelty is always true—even when the piles are not equal: that the rule of proximity is relevant in those situations as well. See my article there.

This is a simple demonstration of the okimta phenomenon, and of how understanding it removes quite a few difficulties one might otherwise get entangled in.

“Filling the public domain with barrels”

A few weeks ago, someone in the series on Platonism asked me how my approach squares with the sugya at the start of the chapter “One who set down the jug” in Bava Kamma. The claim was that the okimta there seemingly contradicts the Mishnah. I presented my answer in lecture 19 in that series, which is devoted entirely to this sugya. Let’s look at it.

The Mishnah in Bava Kamma 27a deals with one who sets down a jug in the public domain:

One who set down a jug in the public domain, and another came and stumbled over it and broke it—he is exempt; and if he was injured by it, the owner of the barrel is liable for his damages.

The Mishnah rules that everything is the fault and responsibility of the one who set it down (therefore the one who broke his jug is exempt, and therefore he is also liable for injury caused by his jug).

The Gemara there on 27b challenges the reisha of the Mishnah:

“Another came and stumbled over it and broke it—he is exempt.” Why is he exempt? He should have looked carefully as he walked!

Why is one who stumbled over the jug exempt? A person (unlike an animal) is required to pay attention when walking on public roads; if he failed to do so, he injured himself and has no claim for compensation. The Gemara in its answer offers three okimtas, and finally brings a fourth opinion.

First okimta:

The academy of Rav said in Rav’s name: [The case is] where he filled the entire public domain with barrels.

Even if the person had been careful, he would have had no alternative. The whole public domain was filled with barrels/jugs, and there was no way to pass without breaking one.

Second okimta:

Shmuel said: They taught [the Mishnah] regarding [a case of] darkness.

The road was dark and he could not have taken care.

Third okimta:

R. Yochanan said: [It was] at a corner.

The jug was placed around a bend, so it could not be seen in time to avoid breaking it.

The Gemara infers from the Mishnah against Rav:

Rav Pappa said: The Mishnah is precise only like Shmuel or like R. Yochanan; for if it were like Rav, why specify “stumbled”—even if he intentionally broke it, he would be exempt!

If Rav is correct that the case is where the entire public domain was filled with barrels, the Mishnah would not need to speak about breaking unintentionally; in such a situation one may break a barrel intentionally in order to pass.

Rav Zevid resolves this difficulty and explains the Mishnah according to Rav:

Rav Zevid said in Rava’s name: The same would apply even if he intentionally broke it; and the reason the Mishnah taught “stumbled” is only because it wished to teach in the seifa “and if the owner of the barrel was injured by it, [the one who stumbled] is liable for his damages”—specifically when he stumbled, but if he intentionally broke it, then “he injured himself” and the owner is of course exempt; therefore the reisha also taught “stumbled.”

Indeed, he may even break one intentionally to pass. And the reason the Mishnah spoke of “stumbling” and not of one who broke intentionally is only because in the seifa—where it speaks of a case in which the owner was injured by the jug—it is only when he stumbled. If he acted intentionally, he injured himself and the owner is certainly exempt.

Finally, as noted, the Gemara raises a fourth possibility:

R. Abba said to Rav Ashi: Thus they say in the West in the name of R. Ulla: Because it is not people’s way to pay attention in the roads.

Here the presupposition in the original challenge is rejected: a person does not typically take care in the road, nor is he required to. If so, there is no need for any okimtas.

From here on we will deal with Rav’s okimta (the questioner wondered about it). Seemingly, this law does not appear in the Mishnah, which speaks of one who placed a jug, not one who filled the public domain with barrels. This okimta seems forced in the Mishnah’s language.

Explaining the okimta and the Mishnah’s novelty

In light of what we saw above, there is a systematic way to look for the solution. The starting point is that indeed, the Mishnah comes to teach a general principle that is true even regarding a single jug. The okimta of “filling the public domain with barrels” is offered only for side reasons, in order to allow the general principle to manifest in practice. Filling the public domain with barrels is supposed to be a laboratory setup that eliminates other constraints and enables the general principle to appear in its pure form.

So what might be the principle that exempts one who stumbled over a single jug placed in the public domain from paying for the damage? The situation is a person walking in the public domain who encounters a jug someone else placed there. A person who causes damage is liable even in cases of accident, so ostensibly the one who stumbled should be liable. On the other hand, everyone is permitted to use the public domain, including placing a jug there. There is a clash between the walker’s use and the setter’s use. That is, one might have thought that in such a situation the walker will be liable for damage, and the owner of the jug will be exempt if the jug injures the walker. Therefore, the Mishnah teaches that the walker is entitled to do as he pleases, and—as the Gemara extends—it is even permitted for him to break the barrel intentionally. You may use the public domain, but you may not use it in a way that obstructs others.

This can be likened to the law of a rodef (pursuer). The novelty of the Mishnah is that the barrel is “pursuing” the walker in the public domain, since it prevents him from using that domain, and therefore it is subject to being broken (it “has no monetary value”). However, when the walker has a way around and can avoid breaking it, he must take the detour; there is no permission to break the barrel. This is akin to a rodef whom one can neutralize by injuring a limb (ein lo damim—the barrel is forfeit—but only if it is necessary to “kill” it; if one can go around, there is no license to break it). Therefore, the Gemara made okimtas: that the case is where the entire public domain is filled with barrels. In such a case the walker has no other option (apart from not passing at all, in which case the barrel is indeed a pursuer). The same is true when he cannot take care because it is a corner or dark. But now you can see that these three okimtas merely remove a side problem and do not touch the Mishnah’s core novelty. With a single jug one can pass without breaking, and thus there is a prohibition to break it. There is a side factor here that neutralizes the “rodef” law regarding the barrel (since one can “save” himself by injuring a “limb”).

If so, the Mishnah comes to teach the very “rodef-like” status of the barrel (that it has no monetary value), and this is true even with a single jug. Therefore, the Mishnah does not state “when one filled the public domain with barrels,” but speaks of a single jug. The side constraint—that the walker can bypass it without breaking—forces us to resort to okimtas, but that is not the Mishnah’s subject. On the face of it, there is no principled dispute between the Amoraim; these are merely alternative portraits that can explain the Mishnah’s application (they remove the side obstruction to the Mishnah’s principle). Indeed, the commentators debate whether the different okimtas disagree with each other or are merely alternative suggestions. Still, it seems that only in Rav’s portrait is one permitted to break l’chatchila, because the barrel has no value; in the other two okimtas there is permission to walk normally without extra care, but not to break intentionally. Yet these are merely different portraits and not necessarily a dispute. In any case, the explanation of okimta that I proposed can work according to all views.

Explaining the Mishnah’s wording

We saw above that the Gemara adds that a person is even permitted to break it, not only exempt if he stumbled. So why does the reisha use the word “stumbled”? The Gemara explained that this is because in the seifa the correct wording is “stumbled,” but in the reisha it is not precise and even misleading. Why didn’t the Mishnah say “broke” in the reisha and “stumbled” in the seifa? According to our approach this is readily understood. The reisha used “stumbled” rather than “broke” because it wished to convey that the barrel has no monetary value (it has the law of a pursuer). When there is no alternative, one may also break; but that is a law stated only about a public domain filled with barrels. When it is not filled, then in principle it has no value, but since one can pass without breaking, one is obligated to do so and may not break. Thus, the Mishnah could not state that one may break, since it is not dealing with a case of a public domain filled with barrels but with a single jug, and there, certainly, there is no license to break.

Incidentally, R. Abba at the end—who does not make an okimta—takes the same principle one step further: one need not take special care at all, because the barrel has no value. In his view one need not be careful, but of course even according to him the walker cannot break the jug intentionally. The term “stumbled” fits precisely as we explained. If so, he too learned the Mishnah’s core novelty just as the makers of okimtas did, except that in his view there is no need for an okimta because the side difficulty raised by the Gemara isn’t a difficulty.

It is now also clear that the commentators who wrote that the exemption—when one filled the public domain with barrels—is because the damage is a case of total accident (ones gamur) were not correct (see Nachal”D here: “Therefore it resolves that we are dealing with filling [the street], or darkness, or a corner—this is a great accident.”). According to our approach, this is not an exemption of “great accident” versus “small” (especially since a person who causes damage is liable even in accident; see Nachal”D here). After all, we saw that one may even break l’chatchila, because the barrel has no value. Filling the public domain with barrels is not about creating a case of accident but about neutralizing the option to bypass the barrels; once there is no way to “save with a limb,” the license to “kill the pursuer” applies. This is not teaching the Mishnah’s core law—namely, the basic exemption concerning the jug.[2]

The law of “pursuer” with property

We saw above that the status of the barrel is akin to that of a pursuer. It is permitted to break it because it “pursues” our use of the public domain. To sharpen this, consider Rambam at the end of Hil. Chovel u’Mazik (8:15):

If a ship was on the verge of breaking due to the weight of its cargo, and one of them arose and lightened its load and threw [cargo] into the sea—he is exempt, for the cargo in it is like a pursuer after them to kill them, and he performed a great mitzvah by throwing [it] and saving them.

The case is a ship bearing a load so heavy it threatens to break the ship. One may throw cargo into the sea and there is no liability for the damage. Rambam grounds this in the law of a pursuer, for the cargo is “pursuing” them; therefore it is permitted to “kill” it.

Ra’avad objects there:

“I say”: There is neither salt nor seasoning here—for there is no law of pursuer at all, and this is not similar to the case of the wine barrel in the chapter “Ha-gozel,” and this law [of] one who cast into the sea—even if he cast what belonged to one man—one calculates for all according to their loads, as stated in the Gemara.

He cites another sugya from which it emerges that there is no law of pursuer here, and that the losses must be shared. This seems to be in direct conflict with Rambam.

However, as other commentators note (see, e.g., Magid Mishneh there), Rambam himself brings that Gemara as binding law in Hil. Gezeilah 12:14:

If a ship was traveling at sea and a wave rose to sink it, and they lightened its load—we calculate according to the weight and not according to the monetary value, and one should not deviate from the custom of sailors.

The Magid Mishneh and others explain the distinction thus: In Hil. Chovel u’Mazik the load itself threatens to break the ship. In that case the load is a pursuer. In Hil. Gezeilah the storm threatens to sink the ship; casting the load into the sea is in order to save the ship from the storm. There the load is not a pursuer; the storm is the pursuer. In such a case one is saving oneself with another’s property, and then one must compensate him. In the past (see this article, and also Column 438) I explained that this distinction parallels the difference between a case of a pursuer where Reuven himself threatens Shimon (so one may kill Reuven as a pursuer) and a case where Shimon wants to kill Reuven in order to save himself from an external threat by Levi. There Reuven is not a pursuer, and killing him is merely a means of escape; in such a case Reuven certainly does not have the status of pursuer, and killing him is forbidden. This is precisely the distinction between the two cases in Rambam. With property, even if it is not a pursuer, one may cast it into the sea to save oneself—but one must pay.

A similar reasoning regarding a cow in the public domain

We see a similar reasoning in the sugya of “this one benefited and that one did not lose” (zeh neheneh ve’zeh lo chaser; see also Column 431). The Gemara in Bava Kamma 20a addresses a situation where a person benefited from Shimon’s property without Shimon suffering a loss (for example, he lived in a house that was empty anyway). The question is whether, in a case of “this one benefited and that one did not lose,” one is liable or exempt. The Gemara there on 20b: Rami bar Chama tells Rav Chisda that this is explicitly taught in a Mishnah (19b), which rules liable:

How is the tooth (shen) [an animal] forewarned to eat what is fit for it? An animal is forewarned to eat fruits and vegetables. If it ate clothing or vessels—it pays half-damages. When does this apply? In the victim’s domain; but in the public domain it is exempt, and if it benefited it pays according to the benefit.

We see that a cow that ate fruit in the public domain is exempt for the damage (for tooth-damage is exempt in the public domain), but it pays for the benefit. Evidently, “this one benefited and that one did not lose” is liable.

The difficulty, as Rava objects, is immediate:

Rava said: How dull and unfeeling is the man whom Heaven helps! For although it is not comparable to the Mishnah, he accepted it from him. That [Mishnah] is “this one benefited and that one lost,” while this is “this one benefited and that one did not lose.”

In the Mishnah the victim lost the fruit, hence payment; but no proof can be brought to a case where there was no loss.

So what did Rami bar Chama think? The Gemara explains:

And Rami bar Chama? [He holds that] the default of fruit in the public domain is that their owners declared them ownerless (hefker).

Since one who puts fruit in the public domain has effectively rendered them ownerless, he did not “lose.” But this cannot be literal, for if they are ownerless, why pay for the benefit? Clearly the Gemara does not mean the fruit are legally hefker, but that it is considered “not a loss,” since the owner harmed himself by placing them in the public domain. With respect to the loss element, it is as if they are ownerless; but they remain his legally, and therefore if someone benefited from them he must pay him for the benefit. In modern legal terms we might say that placing the fruit in the public domain constitutes contributory negligence by the owner, which exempts the damager from tort liability (but not from paying for unjust enrichment, which is not based on fault and to which contributory negligence is irrelevant).

Essentially, the Gemara’s point is that the cow need not have “taken care” to avoid eating the fruit, because it has the right to walk in the public domain; and unlike a person, we do not expect a cow to “take care” regarding damage in the public domain. Therefore, in that sugya there is no need for an okimta (e.g., darkness or filling the public domain with fruit) to explain why it could not take care. There the Mishnah’s principle—that one who places something in the public domain is as if he abandoned it and the one who damages it is exempt—appears in its purity. In our case, the same principle applies to a person; but because it is a person, and a person is expected to take care, we needed an okimta in order for that same principle to be expressed.

Incidentally, the Rishonim explain the exemption of tooth and foot damages in the public domain along these lines: the cow is permitted to walk there, and therefore any damage it does in the course of its walking does not obligate its owner to pay. This parallels our sugya: a person is permitted to walk in the public domain and therefore, in principle, would be exempt for damage to the jug (if not for the fact that, as a person, he is expected to go around it when he can—i.e., to “save with a limb”).

Wounding on Shabbat

As noted above, my claim is that the laboratory case need not be practically realizable; its entire purpose is to create a laboratory state in which the general principle emerges.

The Mishnah in Bava Kamma 34b presents the following comparison:

If his ox ignited a stack on Shabbat—he is liable; but if he himself ignited a stack on Shabbat—he is exempt, because he is liable with his life [for capital punishment].

And the Gemara there infers:

It taught “he” analogous to “his ox”: just as “his ox” is a case of “not for his own needs” (lo kav’ei lei), so too “he” is “not for his own needs”—and yet it taught that he is exempt because he is judged with his life.

From the Mishnah’s analogy it follows that the person’s damage is like the ox’s damage, and clearly in the case of the ox it is “not for its needs” (since an ox does not act with intent). Nevertheless, it says that the person is exempt due to kim lei be-derabbah minei (the greater liability swallows the lesser). It emerges that he is liable to the death penalty for that ignition even though it was “not for his needs.” Hence one who “spoils” on Shabbat is liable, like R. Abahu, against R. Yochanan there who held that all spoilers are exempt.

The Gemara answers according to R. Yochanan:

No—“his ox” is analogous to “him”: just as “he” is a case of “for his own needs,” so too “his ox” is “for its needs.” How is the ox case found? Rav Avya said: We are dealing with a clever ox that had a bite on its back and wished to cauterize itself; it rolled in hot ash. And how do we know this? Because after it burned [the stack], it then rolled in the ash.

The case is one where both the person and the ox had a “need” for the ignition. The ox is a clever ox, and since it needed to rub in ash to heal a wound, it kindled a fire. How do we know this was indeed its motive? Because after the burning it went to roll in the ash. The Gemara brings a precedent:

And is there such a thing? Yes—for there was an ox in the house of Rav Pappa that had toothache; it entered and knocked open a jug and drank beer and was healed.

There was an ox in Rav Pappa’s house that had a toothache; it opened a barrel of beer and drank from it (granted, there are differences between the cases, but this is not the place).

The questioner wondered: why does the Gemara need to show that such oxen indeed exist? According to my approach, the case itself need not be practical. It is a hypothetical case of no intrinsic independent importance, brought only to illustrate a general principle. The need to demonstrate that such an ox exists shows that the Mishnah is not dealing with every ox, but specifically with a “clever ox.” In other words, the okimta here is an interpretation of the Mishnah itself and its principle, not the removal of side interferences in the practical expression of that principle, as I claimed.

I answered that indeed this is a strange sugya. But one can distinguish between a rare case that could occur and an impossible case. For example, if there were an okimta of a person without a heart, or a person made of iron, we would ask how that could be—since a person without a heart is not alive and a person of iron is a robot, not a person. An okimta must portray a coherent case, even if it never actually happens. When you speak of an ox that makes decisions and executes plans, that is like speaking of a talking ox; a talking ox is not an ox but a human (albeit with four legs and horns). Therefore, the Gemara assumes there is no room for such an okimta. For this reason the Gemara wonders how it is possible—i.e., does such an ox really exist—and answers that there are indeed clever oxen that are still oxen (and not humans). That is, it is a coherent case even if rare, or even entirely fictional; we can, via indicators, infer that a certain ox had a plan and an intention.

Moreover, I noted that the example brought is quite far from the case at hand: opening a vessel containing a liquid that the ox “instinctively” “knows” it needs is not such an intelligent act. Just as the ox “knows” to eat, even without conscious understanding, it acts in line with its needs. By contrast, the case of kindling a fire to create ash to rub in is far more extreme—this entails long-term planning and knowledge of the physics that fire produces ash (as well as that ash soothes wounds). It is difficult to accept the inference from an ox that drinks beer to an ox that kindles a fire. But this itself is evidence for my point: the Gemara did not bring a proof that such a case is actually possible, but that it is coherent—that an ox sometimes does engage in planning, even if not very long-term. Therefore, categorically, an ox that ignites a fire need not be deemed to have crossed from the category of “ox” to that of “human.” This does not mean such an ox really exists; only that it is a coherent case. That suffices to justify the okimta.

A note on this very okimta

Until now I dealt only with why the Gemara seeks practical justification for the okimta. But one can question the okimta itself in this Mishnah: does the Mishnah indeed speak of a clever ox, or of any ox? According to the picture I presented, the Mishnah speaks of any ox, and the okimta is there to remove interferences stemming from the example. But here it is not clear what the general principle true of every ox would be, nor what interference is being removed by the okimta that “it is a clever ox.”

On further reflection, it seems to me that this is not an okimta at all. The Mishnah draws an analogy between a person’s ignition and an ox’s ignition. The very comparison shows that it is not speaking of every ox; otherwise there would be no room to compare them. Thus, from the Mishnah itself it is clear that it is not about any ox but about a special case. The “okimta” is not setting the Mishnah to a special case but explaining what the Mishnah means—what case it is addressing. The conclusion is: a clever ox. In this case, the Mishnah indeed discusses only a clever ox and not every ox. Therefore, there is no general principle here that applies to every ox, and the “okimta” is not an okimta but a straightforward explanation of the Mishnah.

Of course, if I am right, the questioner’s initial question falls away. If this is not an okimta, there is no question why the Gemara seeks justification for its practicality. My remarks about the Talmud’s Platonism were said only about okimtas—not about everything that appears in the Talmud. Indeed, in that article (and in more detail in the eleventh volume of the Talmudic Logic Series, which deals with the Talmud’s Platonism; see a summary in my article here) I argued that the Talmud as a whole has a Platonic character, and okimtas are merely a salient example of this. If so, one could indeed conduct the discussion I have done here about the clever ox even if I am right that this is not an okimta.

[1] True, it does specify that he is asleep, and perhaps the general claim is only true for sleeping slaves.

[2] With some forcing one could say that even when the public domain is filled with barrels, he still has permission to pass, and thus even intentional breaking counts as accident.