Okay.
But if I’m giving a recommendation or advice, then I don’t need to add a bridging assumption, right? For example, putting your hand on the fire is dangerous, therefore I strongly recommend that you not put your hands there—there’s no naturalistic fallacy in that, right?
And in general, where do recommendation or advice fit in, legally speaking? Seemingly not in the normative sphere and not in the factual one either?

“Dangerous” is a vague concept.
Putting your hand on the fire causes 2 effects:
1. Pain.
2. A burn, disfigurement of the hand.
These are effects that under normal conditions normal human beings keep away from.

With fire it’s easy to prove the effects even to someone who has no experience with fire and doesn’t believe your claims. You simply start bringing the hand closer to the fire.

But for example with drugs it’ll be harder for you to prove to the other person why it’s not good, because many say it’s okay, and for that you need to build credibility regarding the rest of the prohibitions and advice that you give.

I didn’t ask about the claims themselves—they were only an illustration. I asked about drawing a certain conclusion from certain premises.

It depends on how you understand what a recommendation is. If a recommendation means this: if you want to avoid pain, don’t put your hand in the fire—then that is a fact independent of anything else. He will decide whether he wants to avoid pain or not. If you recommend that he not put his hand in the fire, then you are in effect assuming that pain is something to be avoided, and that has to be assumed.

In French there’s a famous saying: “à l’impossible nul n’est tenu,” which in Hebrew would come out something like “one is not obligated to do the impossible,” or in another formulation, “ought implies can,” meaning that it’s absurd to demand that someone do impossible things, like be in two different places at exactly the same time.

That rule that “ought implies can” can be formulated symbolically like this: “person X is obligated to do p if and only if he can do p,” and that last sentence is equivalent (logically, among other things) to “if person X cannot do p, then he is not obligated to do p.”

Here we derived a norm (“X is not obligated”) from a fact (“X cannot”). Didn’t we?

Ability is a necessary condition, not a sufficient one, for obligation. So “if and only if” is not correct here. The symbolic form is: obligated -> can. Cannot -> not obligated.
But implication is not an inference but a proposition. There is an assumption here that inability implies lack of obligation, but the lack of obligation is not a result of the inability.
If you formulated a parallel argument here:
Premise: Reuven cannot do x.
Conclusion: Reuven is not obligated to do x.
That argument is indeed invalid. An additional bridging assumption is required: whoever cannot do something is not obligated to do it. That assumption is the implication you’re talking about.

A very fundamental insight, thank you very much. It seems to me that this is the same distinction Russell made in order to resolve Carroll’s paradox in the dialogue “What the Tortoise Said to Achilles”: that there is a difference between the inference rule of Modus Ponens and material implication.
Am I right or am I mistaken?
In short, if p then q is different from p therefore q.
Right?

Exactly. The first is a proposition and the second is an argument.

After thinking it over again, there’s something I don’t understand. MP is an argument, fine—but what is an argument? It is itself a proposition of implication. That is, an argument is not valid absolutely; rather, an argument basically means: “if you accept the premises, then you must accept the conclusion.”
So after all, “P therefore Q” means “if you accept P, you must accept Q”; in other words, at the foundation of an inference there lies a proposition of implication.

Implication is a proposition, not an argument. It is measured in terms of truth or falsehood, not in terms of validity or invalidity. When you say that inability implies no obligation, you are not saying that this is necessarily true; rather, you are claiming that it is true. In the parallel argument—premise: cannot, conclusion: not obligated—you are claiming that the conclusion follows necessarily on its own.

Rabbi, what is the relation between Kant and Hume regarding the naturalistic fallacy? That is, who discovered this fallacy? Because both of them say somewhere that you can’t derive a conclusion of type X from premises of type Y. What is the difference between them?

Hume is considered the one who conceptualized the ought-is fallacy. Moore, much later, defined the naturalistic fallacy, and it is not exactly the same thing. But it is common to call Hume’s fallacy the naturalistic fallacy as well. Kant adopted it.

How, or perhaps whether, does it logically follow from the fact that a certain object is mine (that I am the owner of the object) that you are forbidden to steal it? Isn’t that a naturalistic fallacy? If some law (natural, moral, or religious) did not forbid this to you, would it be permitted for you to “steal” it?

Or rather, doesn’t every realist approach—moral realism, natural law, halakhic realism—suffer from the naturalistic fallacy?? Since this object is impure, it is forbidden to touch it. Since I am the owner, it is forbidden for you to take it. Since murder is evil, it is forbidden to murder.

Regarding the first question, for example Aryeh Edrei (whose commander you were 🙂) writes in his article on liability without fault: “The divine law constitutes an essential and inseparable part of the laws of nature… the validity of the law does not stem from the command, but rather, as stated, from its being part of the order embedded in the world.” I never understood it that way; I always thought the validity of the law stems from the command. And you emphasize that in your trilogy. So what is the focus, the basis, of the dispute? Or, can one be a halakhic realist and still hold that the validity of the laws lies in the command?

The “fact” that the object is mine is an ethical fact. With respect to that, there is no naturalistic fallacy. Just as from the ethical fact that there is a prohibition against murder it follows that murder is forbidden. Beyond that, even if you do not accept this, the prohibition is not derived from the fact; rather, there is a normative principle that forbids stealing other people’s property.
I do not know whether Edrei is a moral realist. The command is itself the creation of the ethical law/idea/fact.