On Halakhic Autonomy, Sovereignty, and Infinity


Let’s begin our discussion from afar. As a reader of British philosopher Derek Parfit, I’ve come upon a list of his positions on various ethical dilemmas. In particular, there is the famous murderous trolley dilemma, in which the staunch utilitarian quite strongly (unusual for him, btw) asserts that the driver should choose to save the lives of 5 at the expense of 1.

What about Halakhah (Jewish Rabbinic law)? There is mathematical mnemonic device that is useful in understanding of what’s going on here: in Halakhah, certainly post-Mishnaic era, a life of one Jew is being notionally equated with infinity. Hence, a life of a sinful or stupid Jew ought not be less valuable than a tzadik’s or a Rebbe’s, if one considers trading the former’s life to save the latter. From that perspective, killing or causing death of a Jew to save another Jew is prohibited, and value judgments on life are non-trivial (whether it works out in practice is a different story).

However, we need to understand if there is anything equivalent in Talmud that discusses the proper course of action in a dilemma similar to the murderous trolley problem.

Question: should Jews give up a fellow criminal Jew if non-Jews demand it, under threat of extermination of the entire Jewish community hosting him?

Before I begin actually answering, let me just say that by “criminal Jew” I refer to a Jew who would be considered to be a transgressor of Halakhah, not some other nation’s law or foreign ruler’s opinion. This is actually a strong requirement, because I am considering cases for which even the Jews know that their fellow is guilty, not just the outside party.

Our first and main stop is Tosefta, Terumot 7:23 (R. Shimon’s opinion is the most honored one):

A group of [Jews] to whom gentiles say, “Give us one of you and we shall kill him, and if not, behold, we will kill all of them”; they should let themselves be killed and not deliver them one soul from Israel. But if they designated [the person] to them – for example, Sheva ben Bichri – they should give him to them and not let themselves be killed. Rabbi Yehuda said, “When do these words apply? In a case when he is [inside and they are] outside [a fortified city]; but in a case when he is inside and they are inside, since he will be killed and the [other Jews] will be killed, they should give him to them and not let themselves all be killed. And so did it state (II Samuel 20:22), ‘And the woman come to all of the people in her wisdom, etc.’ – she said to them, ‘Since he will be killed and you will be killed, give him to them and do not kill all of you.’” Rabbi Shimon [b Yohai] says, “So did she say [to them], ‘Anyone who rebels against the monarchy [of the House of David] is liable to [receive] the death penalty.’”

In this case R. Shimon implies that only the [legitimate] Jewish King has the right to demand the handing of the Halakhically guilty/evil Jew [Sheva ben Bichri] for execution. (Btw: the latter was handed to King’s men already executed.)

Now let’s move on, as this isn’t the end of the story. Talmud Yerushalmi, Terumot 8:4 describes a kind of sequel to this, as found in a debate between the next generation of sages: R Jochanan [b Nappaha] and his pupil-cum-famous-rabbi Resh Lakish (R Shim’on b Lakish). In this case, the issue is somewhat reformulated: gentiles encounter traveling party of Jews and demand handing out of a Jew or else they massacre the entire group. R Jochanan says, that request to hand out should be satisfied as long as it is for a named/particular individual (i.e. his opinion superficially seems to follow R Yehuda’s). However, Resh Lakish says: request should be satisfied *only* if the individual is someone like Sheva b Bichri, i.e. someone already deserving death penalty.

By the way, Resh Lakish was a mensch, but this a separate story altogether.

Here is the relevant passage I extracted from Talmud Yerushalmi, version translated by Neusner:

[IV:2 A] It is taught as a Tannaite statement: [As to] a group of men who were journeying on the road, who were met by gentiles, who said, “Give us one of your number that we may kill him, and if not, lo, we will kill all of you” — [B] let them kill all of them, but let them not give over to them a single Israelite [see M. 8:12]. [C] But if they singled one out, [D] such as they singled out Sheba the son of Bichri [2 Sam. 20] — [E] let them give him to them, that they not all be killed [T. Ter. 7:20]. [F] Said R. Simeon b. Laqish, “And that is the case if he is already subject to the death penalty as was Sheba the son of Bichri.” [G] R. Yohanan said, “That is the case even if he is not already subject to the death penalty as was Sheba the son of Bichri.”

But, likely added by later redactors, Talmud Yerushalmi Terumot 8:6 (46b) continues with an anecdote which, in essence, declares that if you give up even a named, Halakhically guilty fellow Jew, you are a despicable man, *even if* you acted in accordance with Terumot 8:4, Resh Lakish’s ruling.

In other words, while Halakhah normatively follows Resh Lakish and R Jochanan’s rulings, it still upholds R Shimon’s extreme principle as the righteous one: even if it’s a bad Jew who deserves the death penalty, you *don’t hand him out* to gentiles for killing, when they demand him, even under threat of complete massacre of your fellow community. If you, under such tremendous duress, do hand out the Halakhically guilty Jew, you are not trespassing Halakhah directly, but you are still committing a grave act.

Talmud Yerushalmi, was likely first assembled by R. Jochanan based on collected rulings of other rabbis, including of Resh Lakish. Let us now closely re-read R. Johanan’s opinion from Terumot 8:4 above: Jews, under deathly duress from gentiles, should not only hand a Halakhically guilty man like Sheva b Bichri, but even an innocent (Halakhically or otherwise) fellow Jew! Here R Johanan not only opines against R Shimon b Yohai and Resh Lakish’s rulings, but — upon attentive reading — he also neglected both R Judah’s clause on potential defense and presumption of the Halakhically guilty status of the surrendered Jew! From now on, any named Jew can be surrendered for survival of the rest…

Which way does post-Talmudic Halakhah lean? Does it support R Shimon’s or R Yehuda’s or Resh Lakish’s or R Jochanan’s opinion? Maimonides (Rambam) unequivocally leans on R Yehuda’s side, in his Mishne Torah, a compendium summarizing the Talmud. However, R Meiri, a rabbi from Catalan and a student of Rambam disagrees: his opinion follows R. Johanan’s.

So, in today’s world: is it acceptable to hand a guilty (whether by Halakhah or outside law) Jew to the gentile authorities, in case where dina d’malkhuta dina (the law of the land is the law) applies, in galut communities? If Talmud is considered Torah, it is not righteous to act this way. However, the unfortunate reality was that while the historical setting of Mishnah took place in close-to-autonomous Jewish communities, albeit under occupation of Judeah by Romans, such near-autonomy had started to become a rarity with the beginning of galut, especially after the Bar-Kokhba revolt . Hence, the newer reality imposed its own requirements. More concretely, this has to be decided on a case-by-case basis from within the galut communities themselves, and R Jochanan’s is the preferred ruling to follow.

For example: for “benevolent” America, contemporary R Fink considers dina d’malkhuta dina to extend all the way to capital cases, based on a few other preceding sources. That is, R. Jochanan’s and R. Meiri rulings are being upheld as the gold standard.

From my understanding, given the Halakhic sources I’ve outlined, and that Rambam, who lived in galut and, if not directly suffered, was much inconvenienced because of it, concurs with Resh Lakish and R Shimon b Yochai’s spirit — every effort should be made by good Jews to not give away one of their own, even in reality of galut. From Rambam’s perspective, blindly following dina d’malkhuta dina would be evil, if opportunity allows to save a fellow Jew without repercussions. More so, if he is Halakhically innocent.

Let’s now consider Halakhah as it is applicable to contemporary Israel: should Israeli Jews both on the level of community and level of government hand out a Halakhically guilty Jew to gentiles upon demand? Answer: certainly not — in either sovereign or even partly-occupied Israel: eg international extradition is Halakhically illegal. Furthermore, by Halakhah, killing an innocent gentile does not warrant death, which might lead to an interesting conundrum: if a Jew does kill one, in Israel, should his community hand him to the secular authorities, if he is likely to be either punished excessively or executed by secular court, which does not directly obey Halakhah and might have gentiles on the adjudicating panel? Here, the answer is also no: Jews living in Israeli communities should *not* hand their fellow Jew to higher authorities for execution/excessive punishment, even if the higher authorities are nominally Jewish.

Is there a pattern here? Can we completely resolve R. Yochanan ben Nappaha’s utilitarian stance with R. Simon b Yochai’s heroic one? Can we have one Halakhah with two seemingly conflicting decisions. Yes! As I’ve noted before, the debate between R. Judah and R. Simon b Yochai, which took place before or around the Bar Kokhba revolt, was presupposing a defensible Jewish community; in this case, a fortress inside a (semi-)independent Judeah at war. However, the debate between Resh Lakish and Jochanan b Nappaha took place later, after Jews were very much under control of Romans and no longer had own organized defense and a much reduced political autonomy. Moreover, the terms of debate are centered around a traveling Jewish party, a group of Jews “out in the wild.” Hence, the following is what resolves it:

When Jews live in foreign countries or at mercy of foreign rulers and authorities, the right decision to abide by is R. Jochanan b Nappaha’s.

When Jews live as part of autonomous community where they are politically and militarily independent, the right decision to abide by is R. Simon b Yochai’s.

What is the mnemonic for this? In case of possibility of real defense, in the context of sovereign State, Jews are at luxury to perform “calculations” with regards to infinity, where action that leads to killing of innocent Jew to save a few is meaningless, because the infinities add to the same quantity. Even when killing a Halakhically guilty Jew, the action of handing out the Jew (even his body!) to outsiders is threatening to Jewish sovereignty, and it is Jewish sovereignty that is being protected here, not Jewish life per se. The most salient argument here is this: since Jewish sovereignty is higher in value than a single Jewish life — here we are talking about the infinity of the power set of natural kind, which is a higher order infinity than infinity of natural kind (I’ll  delve into why/how this is so later) — killing and/or handing out even a Halakhically guilty Jew is forbidden. In case of no possibility of defense, outside of sovereign State, sacrificing many Jews to save one, even an innocent Jew, is akin to mass suicide. Calculations with regards to “infinity” suggest that the action to protect one Jew, i.e. going against “forces above,” is akin to active murder of all Jews, a graver sin than protection of that Jew.

To summarize:

According to Rabbis from Mishna, Talmud and Medieval times, when Jews live in the State of Israel, they should not give up one of their own, even under threat of death. Jews who live in galut must bear the evil of giving their own, in accordance with the doctrine of dina d’malkhuta dina.

And last but not least, the final verdict on the murderous trolley dilemma: according to Talmud’s sovereignty doctrine, one may not, under deliberation, actively divert the trolley to save five Jews at a cost of murdering another Jew. This doesn’t, however, say anything about an action done on an impulsive basis… After all, how many people can perform such calculations in their head if they see five people ahead and impulsively turn the wheel/flip a switch to avoid hitting them?! Hence, if the action is done impulsively, under duress of urgent circumstances, without timely awareness of the one (ultimately killed) Jew, then the doctrine of no-sovereignty/no-autonomy/limited-agency/”forces above”/”forces beyond own control” applies. In latter situation, such action might be forgiven, as decided on case-by-case basis in Jewish court.

Robin Hanson’s brain emulations


I’ve been reading Robin Hanson’s blog, “Overcoming Bias,” on and off for a few years now. Among other topics, he likes to write about the coming Age of Brain Emulations. Basically, the idea is that he (and I tend to agree here) thinks that general AI will not become a reality from first principles, that it will be easier, at least at some point, to simply emulate human brains in extremely powerful parallel hardware.

I think that the argument for and against possibility of general AI is a weaker version of the argument between two people on the opposite fences of the Newcomb paradox. General-AI-by-emulation is a form of perfect-cloning: but the more you think about how to perfect-clone someone, the more you realize how much you need to clone from-outside of that individual. Depending on game-theoretic criteria, i.e. is the predictor/general-AI mechanism adversarial or friendly, its behavior will be one or another. While some kind of an approximation to “intelligence” is viable, at least in certain context, the answer is very much dependent on both the designer’s intent and the limitations and trade offs imposed by fine-tuning to be a practical device. Just like a perfect oracle is both imaginable but not necessarily implementable, I cannot disprove that general AI is impossible, but neither can you prove that it is possible. My short take: it will always remain an error-prone approximation, limited in what it can be and do.

(Along with that assertion, by the way, my answer to the Newcomb paradox is to try to understand whether the oracle is adversarial or friendly. The default answer is “adversarial.” In this case, I’d always choose two boxes, because I am convinced that an adversarial oracle — unless it’s specifically tuned into my brain right before I’m to utter my answer — does not have any special insight into my game, other than estimating likelihoods of my choices from purely game-theoretic principles that minimize its loss… However, if I determine that the oracle is a friendly one, I’d almost-always choose one box before the game commences.)

Here is my longer take: although it’s impossible to prove within my lifetime, Hanson’s brain emulation (“em”) paradigm is not likely to be a near-term likelihood, certainly not even in this century, if at all. Even if something like this does happen in a few hundred years or millennia, these emulations will *not* completely displace humans and their cultures. The reasons are twofold.

The first reason I rely on in my thesis was best expressed by T.B. Lee. Basically, Lee’s argument is along the lines that computing em is of same or, possibly, even worse complexity than that of weather prediction: we might be able to predict weather, in some cases ahead to a few days. However, weather prognosis happens with sharply decreasing probabilities of correctness, as one measures the accuracy on ever finer space and bigger timescale basis — anything past a week or so is close to  being completely meaningless. Albeit Hanson tried to address it, I am not convinced: I think this is an extremely hard problem to solve.

Second reason, let me describe. Even if the above gets partially surmounted and that we figure out a way to make those ems to productively last for a few hours, the point of those devices is to make production as cheap as possible. At some point, however, consumption needs to happen. Short of burning the products, it is difficult to imagine who would be consuming, as the masses will be unemployed, and just a cost. This goes back into motivation: if no one is driving the devices (i.e. people die), these devices need to form a culture. This culture will need to be self-evolvable and constantly expanding. Since life is an open system, completely new challenges arise that require reaction and self-modification (see Gödel’s First Incompleteness Theorem and “Meta-Halakhah” by Moshe Koppel). Since (even superb) emulations are mere approximations/fuzzy algorithms of (human) biological systems, they will still need to interface with humans and human culture, with these both forming a “vital organ,” a blueprint for the necessary “update stage.” Whether the update stage happens every few decades or few millennia or even somehow continuously, in a pipelined fashion, it will still need to be happening…


So, no worries on my part: the apocalyptic scenario is unlikely to happen; people are not going to disappear, not even into the Matrix! However, what might happen is that certain people and certain cultures of people will disappear. The ones who do disappear will likely be, initially, from outside of the culture which first comes to possess such em technology and, later, from cultures where ems overtake humans completely, more likely due to latter’s nihilism rather than a Skynet-type scenario, thus stalling said cultures (as they again are mere mathematical models, golems).

Basically, because of the nature of em devices, they will clearly relegate human mental and physical activity towards more contemplative, abstract and longterm ones. It is quite likely that humans will, by that time, become both smarter and live much longer, due to our ability to repair our cell tissues and organella, like mitochondria, via genetic engineering, virotherapy, and proliferating stem cells directly. Quite possibly, humans in 10,000 years will be able to live longer than 200 years, maybe even thousands of years (will post on that separately), while staying more youthful in the process. Rigorously imagining the fulfilled human and, more importantly, societal potential that comes along with those types of outcomes is something that merits its own post.

For now, I conjure that humans in the next 10,000 years will mostly be mothers, fathers, mathematicians, scientists, engineers, artists, lawmakers, and higher level executives — as these processes require more concentration and long-term thinking and acting. For example, typical expensive cooperating em will only be needed for one day’s worth of labor, for household needs. There will probably be different gradations of these ems. Ones that run for approx. 6 seconds till decomposition/decoherence (for example, something that gets triggered to pick up vegetables from a counter) will be energetically the cheapest to emulate. There will be ones that will run for O(6 minutes)-till-decoherence, for mundane tasks like cleaning a desk. There would be ones that run for O(6 hours), for things that require composing and analyzing a tax return spreadsheet based on various other documents, for example.

If em is as hard as weather prediction, running for O(6 hours) in a coherent fashion might be orders of magnitude more expensive than coherently running for O(6 minutes). Hence, the human (and em, as bootstrapping-like device!) approach to em computing will probably be concentrated on ruthlessly weeding out opportunities for optimizing the time, space, and energy requirements thereof. There will likely be a lot of binning for said capabilities, based on variously competing requirements and necessitated compromises.

As ems become better engineered and cheaper over time, it’s possible that ems that affordably cohere for longer than a few minutes to a few hours might become a reality. When I say “affordably,” I’m comparing it with the price of human worker. It is possible that there would be ems that will cohere for weeks, if not months. The question here becomes: are they going to be cheaper to build/initialize and run than same human workers? Is human cognition based on human existence cheaper than the em one at a certain timespan bin? That is, are there any fundamental (practical) computational/physical limits to how cheap a cohering em can get? How needed will be an em that can cohere for a week but is 1000x more expensive than an em that can cohere for about 6 hours? Can most human menial and low-level work be compartmentalized into 6 hour bins? These are open questions, and they need to be addressed. For now, based on T.B. Lee’s writeup and the fact that biology has optimized the operation of cell machinery (including neurons) my bet is that it’s going to be very expensive to have something that emulates a capable human that’s runnable coherently for more than a few hours or days: as, increasingly, per-cell metabolism, mitosis, hormons etc modeling will all need to be factored in — in their finest details. Couple that with ability of humans to live coherently for many decades and, as noted above, with help of next-gen medicine, possibly even millennia — and we have our division of labor.

In the unlikely situation that ems become a danger, human societies that don’t take control of such beasts might suffer collapse. However, that is not a finality. By the time that this happens (if it happens) human cultures will splinter into many sub-cultures, with some being very spacially removed (being potentially even on different planets/worlds). This guarantees evolution choosing cultures in which physical-human-based cultures, being the inherently open, Universe-embedded systems that they are, will be selected for. In such viable cultures, ems will still be present, as they could be hugely beneficial for production, which is what would make them into sensible cultures also. However, the model of em-human interaction will be akin to a very cooperating peon-aristocracy relationship of yore, as I’ve described above. This way, golems should remain both a sensible and a viable proposition. By the way, in the very unlikely situation that destruction of humans is total, biological life will, still, always find a way, and this includes intelligent biological civilization — even if it originates in another planet/world. This is the nature of the Universe itself (see my penultimate two links there, if you’re impatient).

The other opportunity  for brain emulation might come from interfacing directly with living humans, as tissue augment, i.e the cyborg/Borg model. That, however, would require another post to discuss.