The Definition of Keren
Translated by David Silverberg
As we have seen, the mishna lists the four "avot nezikin" – general categories of damages for which one bears liability, each of which is defined by its unique properties and characteristics. On the basis of these properties, we extend the given category to include liability for its "toldot" – other situations with similar properties. We must therefore carefully define each av by identifying its unique properties. In this shiur, we will discuss the precise definition of "keren," one of the three avot associated with damages caused by animals.
The sugya on 2b indicates that any damage an animal intentionally causes falls under the av of keren. ("Keren," which literally means "horn," refers to damage caused by an ox's goring with its horns.) The Gemara remarks, "What is unique about keren – that it [the animal] intends to cause damage, it is your property and you are responsible to guard against it [this type of damage]. These [other damages caused by an animal's body, such as through kicking], too, [are cases where] it intended to damage, it is your property and you are responsible to guard against it." Similarly, the Gemara on 4a distinguishes between keren and shein (damage caused by an animal eating) on the basis of the fact that keren refers to situations where the ox intended to cause damage, whereas shein describes cases where the animal had no specific intention to cause damage. This concept of "kavanato le-hazik" (the animal specifically intends to cause damage) appears again on 5b. All these sugyot clearly point to "kavanato le-hazik" as the defining characteristic of keren.
This conclusion, however, appears to directly contradict a later sugya dealing with "chatzi nezek" – the 50% compensation payment required for keren damage by a "shor tam" (ox that has yet to commit three keren offenses). The Gemara (15a) brings a debate between Rav Papa and Rav Huna Berei de-Rav Yehoshua as to whether this payment constitutes "mamon" – a standard compensation payment incurred by virtue of the owner's liability – or "kenas" – a punitive fine. Ultimately, the Gemara concludes that the chatzi nezek payment is a kenas, and it adds, "Now that you have established that chatzi nezek is a fine, if a dog ate a sheep, or a cat ate a chicken – this is a deviation [from the animal's standard conduct], and it is not collected in Bavel." Meaning, if a dog eats a sheep, since it has deviated from the standard conduct of dogs, this situation falls under the category of keren, and thus the laws of keren – including chatzi nezek – apply. And since the chatzi nezek payment constitutes a kenas, a punitive measure which the Babylonian Batei-Din did not have the authority to implement, the dog's owner will not be ordered to compensate for the sheep. Even though the dog committed this act not with the intent to cause damage, but rather for its own gratification, the Gemara nevertheless considers this a situation of keren, since the dog deviated from the standard conduct of its species. This clearly suggests that the defining characteristic of keren is not specific intention to cause damage, but rather deviation from standard conduct, seemingly in direct contradiction to the aforementioned sugyot.
A similar problem arises from Rashi's comments. On 2b, Rashi repeatedly emphasizes the point that the essential definition keren lies in the element of the animal's intent. The Gemara there identifies several violent acts of an ox as toledot of keren – pushing, biting, crouching and kicking. Rashi explains the precise meaning of each act mentioned in the Gemara:
"Negifa – it pushed with its body and intentionally caused damaged; revitza – it saw utensils on the road and it went ahead and crouched on them in order to break them; be'ita – it kicked with its legs and broke the utensils. Each of these is a toleda of keren because [in each case] it intended to cause damage, like in keren, and no benefit is derived by causing the damage, just like keren, and this sort of damage does not commonly occur."
Clearly, then, Rashi considers intention to damage a necessary condition to define a given act of damage as keren or a derivative thereof. Later, however, commenting on the Gemara's discussion of the dog who eats a sheep (15b), Rashi writes, "This is a deviation – for they are not accustomed to such [conduct], and it is thus a toleda of keren, requiring that he pay chatzi nezek." Meaning, the factor of deviation renders this situation a toleda of keren, despite the absence of specific intent to cause damage.
Rabbenu Tam's Solution
We might resolve this contradiction in light of the position of Rabbenu Tam. The Gemara on 5b, to which we briefly referred earlier, alludes to a dispute as to whether the liability for keren damages is more intuitive or less intuitive than that of the other avot nezikin. One view considers liability for keren a novelty of sorts, at least in relation to the other avot, because an animal is not initially "mu'ad" for this type of damage; meaning, the owner cannot be expected to anticipate this kind of behavior unless it repeats itself three times. The three other categories, by contrast, represent standard conduct that can potentially cause damage, requiring the owner to protect against these kinds of damages. The second view in the Gemara maintains that to the contrary, keren liability is more intuitively compelling than the other categories, since in instances of keren the animal specifically intends to cause harm. Rashi comments that nowhere does the Gemara ever specify who authors this second position, viewing keren as the more intuitive category of damage liability. Rabbenu Tam, by contrast, claims that the Gemara here refers to the aforementioned debate (on 15a) between Rav Papa and Rav Huna Berei de-Rav Yehoshua as to the nature of the chatzi nezek payment. Rav Papa considers this payment mamon – bona fide damage compensation, because the owner must always beware of the ox's possible tendency to violent behavior. In principle, the owner should pay in full for all the damages, but the Torah exempted him from half the damages until three occurrences of violent conduct. Rav Huna Berei de-Rav Yehoshua, by contrast, maintains that we do not expect an owner to anticipate this type of conduct and thus restrain the ox. In principle, then, he should be exempt altogether from payment until the ox establishes a pattern of violent behavior. But the Torah requires him to pay a fine as a means of encouraging owners to restrain their animals.
According to Rabbenu Tam, then, Rav Papa and Rav Huna Berei de-Rav Yehoshua debate the issue of whether we consider an ox's goring an unusual occurrence. Rav Huna maintains that such conduct cannot be anticipated, whereas Rav Papa holds that the owner must, in fact, take precautions against such behavior, which Halakha deems as standard conduct for oxen. Thus, when the Gemara on 5b mentions the view that keren liability is less intuitive than the other avot nezikin, it refers to Rav Huna Berei de-Rav Yehoshua, who considers goring a deviation from the standard conduct of oxen. The second view mentioned in the Gemara is that of Rav Papa, who considers this type of behavior standard. In his view, then, keren liability is in fact more compelling than the other avot, since the ox specifically intends to cause damage.
Thus, only according to Rav Huna, who holds "palga niska kenasa," can we consider incidents of keren damage "meshuna" (unusual). Rav Papa, who holds "palga nizka mamona," considers this behavior normal. Indeed, Tosefot (15b s.v. ve-hashta) claim that according to Rav Papa, if a dog eats a sheep, the owner need not pay at all, since there is no source in the Torah – in his view – for an obligation to for damages caused by an animal's unusual behavior. Only Rav Huna, who holds "palgnizka kenasa,"would require the dog's owner to pay chatzi nezek, for this ca– like all cases of damage resulting from an animal's unusual conduct – constitutes a toleda of the category of keren. Despite the fact that in standard cases of keren the ox specifically intends to cause damage, whereas here the dog ate the sheep for its own gratification, nevertheless, the defining characteristic of keren, according to this view, is the element of meshuna – the animal's deviation from expectable conduct.
Similarly, the mishna (15b) establishes that we consider biting expectable behavior for snakes. Tosefot (16a s.v. ve-ha-nachash) argue that a snake's bite cannot fall under the category of shein, since unlike cases of shein, the snake's intention is to cause damage, rather than its own enjoyment. Instead, Tosefot classify snakes' biting under the category of regel (damage caused by an animal walking), since the snake bites as part of its usual routine ("darko be-khakh"). Thus, Tosefot explicitly hold that a type of damage caused with purposeful intent is nevertheless not considered keren if it does not involve a deviation from the animal's standard conduct. Tosefot's position becomes perfectly clear in light of Rav Huna Berei de-Rav Yehoshua's view, which is the accepted view, defining keren on the basis of deviation from normal conduct, rather than specific intention to cause damage.
In light of Rabbenu Tam's approach, we can easily resolve the contradiction between the various sugyot. After all, the sugya on 15b that classifies under keren all damages caused by unusual behavior, even without specific intention to cause damage, is working within the view of "palga nizka kenasa." The sugya on 2b, by contrast, which defines keren as damages caused with purposeful intent, does so in the context of its discussion of Rav Papa's statement that "toldoteihen lav ka-yotzei ba-hen" (certain toladot do not follow the same rules governing their av). Understandably, then, the Gemara here works within Rav Papa's position on 15b, that we do not consider an ox's goring unusual conduct, and thus we define keren on the basis of purposeful intent, rather than deviant behavior.
It would thus turn out that according to the accepted view, that "palga niska kenasa," a given act's classification as keren depends not on the factor of the animal's specific intent to cause damage, but rather on the element of "meshuna" – unexpected conduct. An act that does not deviate from the animal's normal conduct would not fall under the category of keren, regardless of the animal's intent to cause damage, whereas deviant conduct would qualify as keren, even if the animal had no specific intention to cause harm.
The Riva's Approach
There appears, however, to be another approach to this sugya among the Rishonim. The Shita Mekubetzet (15b s.v. ve-Riva) cites the position of the Riva, that even according to Rav Papa, who views chatzi nezek as bona fide compensation, a person would bear liability for his dog's consumption of a sheep. Meaning, even Rav Papa, who considers goring standard behavior for ox that the owner must anticipate and thus guard against, keren includes instances of deviant conduct:
"The Riva says that even according to the view that palga niska mamona, keren is deemed unusual, only not unusual enough that we would consider the payment a fine [rather than actual compensation]. Even according to his position, then, it turns out that one pays in cases of unusual behavior."
The Riva holds that even Rav Papa would consider an unusual act as keren, regardless of specific intent to cause damage. Therefore, even our sugya on 2b, which, as mentioned, works within Rav Papa's view, accepts "meshuna" as a defining characteristic of keren. Our original question thus returns: why does our sugya point to intent as the defining characteristic of keren, whereas the sugya on 15b considers abnormality the critical factor?
The answer, perhaps, emerges from the Gemara's discussion on 19b concerning a case of "kishkesha ba-amata" (damage caused by the thrusting of a male animal's organ). On the one hand, the Gemara suggests, one might classify this situation under keren, which likewise involves an act caused by an animal's physical impulses and aggression. Alternatively, since in this instance the animal does not act with the specific intent to cause damage, it perhaps does not qualify as keren, and should thus be considered regel, which would mean that the owner bears no liability in a reshut ha-rabim (public domain). This sugya, too, appears to define keren based on the animal's purposeful intent. For if the element of deviation defines the category of keren, then the issue of whether an action occurred due to physical impulse or with the intention to cause harm is irrelevant. The Riva, cited in the Shita Mekubetzet, indeed raises this question:
"The Riva asked, what does [this sugya] hold? If this is normal conduct, then it is regel, and why does it matter that it has no intention to cause damage? And if this is not normal conduct, then it is keren, and even if it has no intention to damage [the owner] bears liability, as we said regarding the chicken that thrust its head into a glass utensil [thereby breaking it] – since this is abnormal, [the owner] must pay chatzi nezek, even though it did not intend to cause damage."
This question works off the assumption that we define keren based on abnormal conduct. In answering this question, the Riva dismisses this assumption:
"It considers 'kishkesha ba-amata' standard conduct as a result of its physical impulse. This is what [the Gemara means when] it says that [on the one hand,] this resembles keren mu'edet [an instance of keren after a pattern had been established] in which case he must pay for the damages in full, since this is standard conduct as a result of its physical impulse. On the other hand… here, even though it is standard conduct as a result of its physical impulse, this does not resemble keren since it does not intend to cause damage. But it is certainly self-understood that it does not resemble regel, since it is standard conduct only as a result of its physical impulse."
In other words, even if the animal acts normally, we will consider the given act keren if it was motivated by specific intention to cause damage, and the owner would thus bear liability even in a reshut ha-rabim. But since the animal acted normally, the owner must pay nezek shalem (complete compensation) like a case of shor mu'ad (an ox that had established a pattern of violent conduct), rather than chatzi nezek. The Riva would presumably disagree with Tosefot's view, that a snake's bite falls under the category of regel. According to the Riva, since the snake acts normally by biting and does so with the specific intention to cause damage, it would be classified under keren. The practical ramification between these views would arise in a case of a snakebite in a reshut ha-rabim. Since an owner is not liable for regel damages caused in a reshut ha-rabim, Tosefot would exempt the snake's owner in such a case, whereas according to the Riva, who considers snakebites keren, the owner would pay full compensation, even though the damage occurred in a reshut ha-rabim.
Thus, according to the Riva, intent to cause damage suffices to define a given act as keren, regardless of whether or not it marked a deviation from standard conduct, and therefore even normal behavior with destructive intent will yield liability in a reshut ha-rabim. Why, then, does the Gemara require only chatzi nezek when a dog eats a sheep? Since it acted for its own benefit, rather than with the intent to cause damage, how can we consider it keren and demand only chatzi nezek? The answer is that the Riva distinguishes between liability in reshut ha-rabim, which depends on the definition of a given act as keren or shein/regel, and chatzi nezek, which depends solely on the act's abnormality. Indeed, when a dog eats a sheep, we do not consider this act of keren. Nevertheless, the owner will pay only chatzi nezek because of the abnormality of the act. The definition of keren, aour sugya indicates, depends opurposeful intent to cause damage. The issue of chatzi , however, depends not on a given act's classification as keren, but rather on its unusual nature. Therefore, as the Gemara on 15b states, when a dog eats a sheep the owner pays only chatzi nezek, even though the dog acted for its own benefit.
The Rambam's Position
Although the Rambam never addresses this issue explicitly, we can determine his position by carefully reading his formulation of the relevant halakhot. The Rambam begins Hilkhot Nizkei Mamon as follows:
"Any animal in a person's possession that caused damage – the owner must pay, since his property caused damage… How much must he pay? If it [the animal] caused damage through actions it normally, always performs, as is common for its species, such as an animal that ate straw or a sheaf, or that caused damage with its leg over the course of walking, he must pay full compensation from his highest quality property, as it says (Shemot 22:4), 'he must pay from the best of his field and his vineyard.' But if it deviated and performed actions that it does not normally, always do, and thereby caused damage, such as an ox that gored or bit, he must pay for half the damage from the body of the animal causing the damage, as it says (Shemot 21:35), 'they shall sell the live ox and divide its price…'"
The Rambam here includes all cases of damage caused by animals, and appears to attribute the liability in all these cases to the fact that "his property caused damage." The difference between damage caused by normal conduct and that resulting from abnormal behavior involves only the payment for the damage. This would indicate that the issue of chatzi nezek or nezek shalem depends not on the formal categories of keren, shein and regel, but rather on the factor of abnormality, just as the Riva seemed to imply.
With regard to the exemption for shein and regel damage in a reshut ha-rabim, the Rambam writes:
"If it damaged in a public domain… if it damaged through shein or regel while acting normally, then he [the owner] is exempt, since it is permitted to walk here and an animal normally walks and eats and breaks over the course of its walking. But if it gored, pushed, crouched, kicked or bit, then if it has yet to be warned, he pays for half the damages, and if it has been warned, he pays full compensation."
In this context, the Rambam distinguishes between the formal categories of keren, shein and regel, which likewise follows the position of the Riva, that the reshut ha-rabim exemption depends on the formal classification of the given act as shein or regel, as opposed to keren. Accordingly, in a situation of an abnormal act of shein or regel, such as a dog who eats a sheep, the owner will pay only chatzi nezek but will be exempt if this occurred in a reshut ha-rabim.
However, the Rambam provides a reason for the reshut ha-rabim exemption: "since it is permitted to walk here and an animal normally walks and eats and breaks over the course of its walking." If, indeed, this is the reason for exempting the owner in cases of shein and regel in a public domain, then presumably he would be liable in cases of shein ve-regel damage resulting from an abnormal act. We are thus uncertain whether the Rambam would exempt cases of "shein meshuna" in a public domain, since, after all, the given incident falls under the category of shein, or if he would obligate the owner to pay in such cases, since the animal acted in an abnormal manner.
Towards the end of the first chapter, the Rambam codifies the Gemara's question on 19b:
"If it shook its tail excessively like it does not always do, causing damage in a public domain, or if it shook its organ in a public domain, causing damage, [the owner] is exempt. But if the person who incurred the damage seized [some property of the owner], he collects half the damages from what he seized. For this matter is subject to a doubt, whether these are toledot of keren, for which one is liable in a public domain, or toledot of regel, for which one is exempt in a public domain, as we have explained."
One might have explained that the Rambam understood the Gemara's question to be whether damage caused by the tail's shaking constitutes a toleda of regel, for which one is exempt in a reshut ha-rabim but otherwise pays nezek shalem, or a toleda of keren, for which one is obligated in a reshut ha-rabim but pays only chatzi nezek. According to this understanding, however, the Rambam should have ruled that had this occurred in the victim's property, if the victim seized the amount of full compensation from the owner's property, we do not force him to return it. If the question surrounds merely the classification of kishkush as either keren or regel, then when the incident occurred in the victim's property, the owner must certainly pay, the only uncertainty being whether he must pay full compensation (if this is regel) or only for half the damages (if this is keren). The Rambam's omission of such a ruling indicates that there is no question that such conduct deviates from standard behavior, and thus the victim certainly has no right to full compensation. The Gemara's question relates specifically to the issue of liability in a reshut ha-rabim.
One might explain that the Gemara's uncertainty revolves around the fundamental issue of whether keren is defined based on destructive intent or deviation from standard conduct. It thus questioned the status of an act that deviated from normal conduct but lacked destructive intent, whether such an act falls under the category of keren – and would thus yield liability even in a reshut ha-rabim – or regel, in which case the owner is exempt in a public domain. Alternatively, we might claim that the Gemara understood that the status of keren depends upon destructive intent, but questioned whether the reshut ha-rabim exemption stems from the classification of a given incident as regel, or from the presumed right for an animal to conduct itself normally in a public domain.
In this shiur we discussed the precise definition of keren and encountered a contradiction between two sugyot, yielding two different positions among the Rishonim. One approach defines keren based on the motive behind the action: if the animal acted with malicious intent, the incident qualifies as keren. According to the second view, the animal's intent has no effect on the classification of the given damage. It is rather the factor of deviation from normal behavior that defines an incident as keren.
These two approaches reflect two fundamentally different perspectives on the owner's liability for keren damage. The factor of abnormality relates primarily to the extent of the owner's negligence and failure to protect against damage caused by his property. An animal's owner must anticipate standard conduct and is therefore obliged to take precautions to prevent damages that may result from such behavior. To a much lesser extent, however, would we expect an owner to foresee abnormal conduct on the part of his animal. It would seem, then, that if the definition of keren involves the factor of abnormality, then we deal here with the issue of negligence, which indirectly caused the damage. With regard to this issue of responsibility, the animal's intent bears no relevance. (On the other hand, see Tosefot, 3b s.v. lo re'i.) And if we do, indeed, define keren based on the animal's purposeful intent, presumably we must explain that the Torah holds the owner responsible only if the animal commits an act of destruction, which requires specific intention to cause harm. My property performed a destructive act and I as owner am responsible to cover damages. It is my responsibility as owner that obligates me, not my negligence.
With these two perspectives in mind, let us return to the Gemara's comments on 5b to which we referred towards the beginning of the shiur. The Gemara cites two views as to whether ke liability is more or less intuitive than the other three categories of damages. If we deal with the issue of negligence then we have less reason to hold an owner liable forkeren damages, since the animal initially does nohave the propensity to such conduct. If, howeve, the owner must pay because his property committed an act of damage, and he, as the owner, must cover the loss, then we are more likely to obligate him to pay in cases of keren, where the animal specifically intended to cause damage, than in shein, where the animal intended for its own personal benefit, or regel, where there is no intent at all.
Mekorot for next week's shiur:
Is it forbidden to damage property with intent to repay?
1) Mishna Bava Kama (51a), gemara ... aveira.
Kiddushin (42b) ve-ha di-tnan shalach ... aveira.
2) Rambam Hilkhot Nizkei Mammon 5:1.
Tur – beginning of siman 378 Choshen Mishpat.
Rashba Bava Kama (2b) s.v. Eima.
3) Yad Rama Bava Batra perek 2 siman 26
Teshuvat Chatam Sofer YD 241 – first paragraph.
1. Which gemara suggests that indeed an issur does exist? Do they specify whether a potential issur is de-oraita or de-rabanan ?
2. What is the source of such a prohibition?
3. How might the nature of the issur differ depending upon the source?