Free Will and Choice – Lesson 4

Table of Contents

• [0:01] Introduction and presentation of the debate between libertarianism and determinism
• [2:35] Three planes of discussion: philosophical, scientific, diagnostic
• [4:49] Divine knowledge and its connection to free choice
• [7:02] Newcomb's paradox and the box game
• [11:06] The paradox against the existence of a prophet and God's power
• [18:58] Logical determinism and a claim about the future
• [25:51] Defining the truth value of a proposition and its connection to the Talmud

Summary

General Overview

The speaker sets libertarianism against determinism and argues that the libertarian view is not a naive caricature, but recognizes environmental, genetic, and social pressures as influences that do not determine. He presents three possible planes for deciding the question of free choice: a priori philosophical arguments, scientific arguments, and a diagnostic plane of self-examination. He indicates that by the end of the series, in his view, there will not be a decisive resolution, but there is a fairly good probability in favor of libertarianism. After returning to the problem of divine knowledge, he uses Newcomb's paradox and a critique of logical determinism to undermine the combination of foreknowledge with free choice, and concludes by preparing the ground for the scientific plane through the causality argument and deterministic physics, while discussing judgment, intelligence, and trust in conclusions in a deterministic world.

The Framework of the Discussion: Libertarianism vs. Determinism

The speaker says that libertarianism recognizes that a person is subject to many pressures, but argues that these are influences and not determinations, and therefore they can tilt but do not necessarily decide. He presents a topographical parable in which a person operates within a contour that attracts, repels, and presses, but unlike a little ball, the person can choose to climb or not to roll downhill, whereas the determinist sees the contour as determining. He argues that the difference between the views is smaller than it seems, because all the statistics, psychology, and familiar phenomena can also exist within the libertarian picture, and the question of decision remains open.

The Three Planes of Decision

The speaker divides the discussion into three planes: an a priori philosophical one that tries to show contradictions or justify a position without relying on scientific facts; a scientific one, where it is claimed that findings from recent decades may decide the matter; and a diagnostic one, where even without a logical or scientific resolution, one can make use of indications through self-examination. He says he began with the philosophical plane, will move to the scientific plane, and then to the diagnostic one, and marks as a general conclusion that in his opinion there is no clear resolution, but there is a probabilistic tendency in favor of libertarianism.

The Knowledge of the Holy One, Blessed Be He, and Free Choice

The speaker says that the theological argument of the knowledge of the Holy One, Blessed Be He, reverses the usual picture of the debate, because it actually leads to a deterministic conclusion on theological grounds. He proposes various ways of dealing with it and presents the solution that he thinks is the most plausible: that there is free choice and the Holy One, Blessed Be He, does not know. He argues that this does not detract from divine omnipotence, because omnipotence means the ability to do anything that is defined and conceivable, not to perform nonsense like a “round triangle.” So too, “the Almighty cannot know information that does not exist.”

Newcomb's Paradox and Foreknowledge

The speaker presents a game with a “prophet” who knows in advance what will be chosen and prepares a closed box with a million dollars if only that box is taken, or leaves it empty if both boxes are taken together with an open box containing a thousand dollars. He describes a loop in which, on the one hand, it is better to take only the closed box in order to win the million, while on the other hand it seems rational to take the extra thousand as well, because whatever is in the closed box was already determined yesterday. He concludes that the assumption of the existence of a prophet who knows the choice in advance leads to a paradox, and therefore serves as a proof by contradiction that there is no such being “of any kind whatsoever” that knows in advance what will be chosen, and he includes the Holy One, Blessed Be He, in that as well.

What the Paradox Rules Out, and What It Does Not Prove

The speaker emphasizes that Newcomb's paradox does not prove that there is free choice, and by itself it is not evidence against determinism. Rather, it neutralizes the possibility that foreknowledge and free choice can both exist together. He presents two consistent possibilities: either there is no foreknowledge and there is free choice, or there is foreknowledge and there is no free choice. He adds that in the book On the Science of Freedom he argued that the paradox also works against determinism in a more fundamental sense, because the question is whether a strategy exists at all, not whether he himself in particular can choose. Therefore, rejecting the paradox requires arguing that the concept of “free choice” is internally contradictory, a claim he presents as far-reaching and, in his view, implausible.

Logical Determinism and the Truth Value of Future Propositions

The speaker presents the Aristotelian question about the proposition “Tomorrow there will be a sea battle” and asks what its truth value is today. He mentions Łukasiewicz, who proposed a three-valued logic in which future propositions are “unknown” or “not yet determined.” He rejects this and argues that a true proposition is a correspondence between the content of the proposition and the state of affairs it describes. Therefore, if there really is a sea battle tomorrow, then the proposition was true yesterday as well, and even “at the time of the Big Bang”; it simply was not known. He describes the claim of logical determinism, according to which the atemporal truth value is already fixed and therefore the future must unfold accordingly, but responds that the argument confuses information or fact with truth value as a logical definition.

The Distinction Between Information and Definition: The Talmudic Topic of Retroactive Clarification

The speaker argues that correct information in the present about the future dictates the future and therefore clashes with free choice, but the truth value of a proposition is not information and not a fact, but a definition, and there is no problem with a definition depending on a future event. He brings an example from the Talmud, tractate Gittin, in the topic of retroactive clarification, about a bill of divorce written “for whichever one of his two wives will go out first through the doorway tomorrow morning.” He explains that according to the view that there is retroactive clarification, this is considered a valid definition of the woman already today, even though the defining event will happen tomorrow. He concludes that if tomorrow there will be a sea battle, it will become clear retroactively that the proposition was always true; and if there will not be one, it will become clear that it was always false. But that does not dictate what will happen tomorrow; rather, what happens tomorrow retroactively clarifies the status of the proposition.

Judgment, Intelligence, and Determinism

The speaker asks what judgment means within a deterministic picture, and argues that in such a picture there is no judgment, only mechanical calculation. He attacks the tendency to attribute intelligence to animals and computers by means of an analogy to flowing water, which one could seemingly crown as intelligent because it “solves” the Navier-Stokes equations in real time, equations that people are unable to solve. He concludes that this is absurd, because the water is not calculating; it is simply flowing. He argues that a computer's intelligence is an indirect expression of the intelligence of the programmer, and that intelligence is defined only in relation to creatures that exercise judgment, just as one does not say that a point has “zero length,” but rather that it has no length at all.

Trust in Conclusions in a Deterministic World and Evolution

The speaker proposes that if there is no freedom and no judgment, but only blind mechanical calculation, there is no way to justify trusting cognitive conclusions, because there is no vantage point “outside the mechanism” from which one could check whether it is reliable. He argues that the chance that a blind random mechanism would produce correct answers is negligible, and therefore the results of thought in a deterministic world are at most survival-oriented, not epistemic. He rejects the attempt to rely on evolution as a mechanism that guarantees correct thinking, and argues that evolution guarantees thinking that is useful for survival, not necessarily true. He adds that trust in the existence of evolution itself relies on that same mechanism of thought, and if it is blind, there is no basis for trusting it.

Transition to the Scientific Plane: Causality and the Laws of Physics

The speaker concludes that on the philosophical plane the considerations remain without a decisive resolution, though in his view they tend toward libertarianism. But every argument rests on assumptions that can be rejected. He opens the scientific plane with the causality argument, according to which the ordinary laws of physics are deterministic, and therefore given a particular state they determine what will happen. This contradicts the definition of free choice as the possibility of several futures emerging from the same state. He illustrates this with a physical chain of “throwing a punch,” through muscles, nerves, electrical currents, and fields in the brain, and argues that introducing choice means that at some stage an electron moves contrary to the electric field, that is, a breaking of the laws of physics. He declares that this discussion stands on the seam between philosophy and science and will be dealt with in the next lecture.

Questions from the Audience and Closing

The speaker answers a question about “deciding which electricity to send through” and says that apparently this contradicts physics as he presented it, and he will discuss that next time. He confirms that the determinist can say that there is no intelligence even in a human being, or define it differently so that it exists in everyone. He clarifies that the claim about Navier-Stokes was empirical, not a mathematical proof of unsolvability, and answers a question about Newcomb's paradox by saying that the problem is not merely “there is no strategy,” but that there are decisive considerations for two contradictory strategies. In response to a question about “stupid” athletes who perform complex calculations, he says it is an excellent question and he does not know how to decide whether this is judgment or automatic instinct in the style of “like water,” and he concludes with the blessing: Sabbath peace.

[Rabbi Michael Abraham] Okay, so for now I’m muting everyone, with your permission, as usual, and at the end of the lecture I’ll open things up and anyone who wants is welcome to ask questions or comment. Good. We were at the point of where we stand in the overall course of things. I presented the two theses, or the thesis and antithesis, opposite one another: libertarianism and determinism. I showed that there are overly childish, or caricatured, conceptions of the libertarian view. The libertarian view does recognize various influences on a person; it just argues that these are influences, not determinations. Meaning, these influences can tilt us or pressure us to act or not to act in certain directions, but they do not determine what we will do. I spoke about the "either way" argument, which really could have come up at this stage in the series as well. The argument that says: either you have a reason or you don’t have a reason, and if you have a reason then that’s determinism. Wait, what’s going on here? Now let’s check. If you have a reason, that’s determinism; if you don’t have a reason, that’s indeterminism. I said that’s not true—there’s a third route. That’s Van Inwagen’s "either way" argument. And then I presented free choice as, let’s say metaphorically, an action of a person within some topographical layout that obviously exerts various pressures on him and pulls him here and repels him there and applies all kinds of pressures. And of course that exists in the libertarian picture too: environmental, genetic, social pressures, whatever you want. But unlike a little ball that’s in such a situation, a person in such a situation can choose to climb the mountain or not to slide down into the valley. And in that sense the libertarian differs from the determinist. The determinist sees the topographical layout as determining what I will do, and the libertarian sees it as influencing what I will do. In other words, these are two different conceptions, and therefore the gap between them is narrower. All the statistics, the psychology, all the phenomena we know, obviously exist in the libertarian picture too. And now we have to see what we do with that—meaning, who is right. So I also talked about the implications: fatalism, morality, judgment, but I’m not going back to that now. Last time I began discussing the question of who is right, and I said that this question can be discussed on three planes. I introduced this at the start: it can be discussed on three planes. The first plane is a priori philosophical arguments. That is, arguments that find a contradiction in one of the sides, or philosophical arguments—not scientific ones, not based on certain facts, although the boundary isn’t sharp. The second kind of plane on which we need to examine the issue, the second kind of arguments, is scientific arguments. There have been claims in the last few decades that there are scientific findings that can decide in favor of one side or the other. And the third plane of discussion is what I’ll later call—I’ll explain this more—the diagnostic plane. Meaning, suppose I have no decisive conclusion, neither philosophical-logical nor scientific; how do I nevertheless make decisions? And there are some indications that can help me make a decision on a third plane, a plane beyond those first two, through various kinds of self-examination. So that’s the diagnostic plane. I began discussing the philosophical plane, and really that’s going to take us to the end of the series. I’ve now started with the philosophical plane, the next stage will be the scientific plane, and the third stage will be the diagnostic plane. And in the end I want to show that first of all, I think there is no clear decision, but it seems to me there’s a fairly good probability in favor of libertarianism. Okay. So last time, when I began with arguments on the philosophical plane, I spoke about the argument from the knowledge of the Holy One, blessed be He. In the meantime all kinds of adventures on the website have developed around this issue; I’m just about to publish the third column there on this topic. And the claim basically was that contrary to what is usually accepted—as if believers are libertarians and neo-Darwinians, materialists, are determinists, they think there is no free will and no spirit and so on—in the first argument I discussed, namely divine foreknowledge that seemingly dictates what we will do, that argument actually turns the tables. It basically says that from theological considerations I arrive at a deterministic conclusion, not from anti-theological or atheistic considerations, which is a bit unusual. In other words, it’s even surprising for anyone familiar with the picture of the debate, at least today. So the claim was: if God knows, then how do we have free will? I presented several possible ways of dealing with this. In the end, at least I think—and that’s the last column that will go up on my site, though I’m not completely certain about this, I say it somewhat cautiously—but I think the most plausible way to handle the issue is to say that I have free will and the Holy One, blessed be He, does not know. And if someone is troubled by the question of God’s omnipotence, I think I spoke a bit about that and I’ll also write about it on the site, so I won’t go into it here. It doesn’t impair His omnipotence, because omnipotence means being able to do everything that is defined, everything that is conceivable. But omnipotence doesn’t mean doing nonsense, doing things that are undefined, making a round triangle. In the same way, even the omnipotent cannot know information that does not exist. To know information that does not exist is like a stone the omnipotent cannot lift, like a round triangle, and like all those scholastic tricks that try to catch the concept of infinity by the tail and don’t quite succeed. They circle around it instead of turning it around themselves. So that conclusion shouldn’t worry us too much. That’s regarding the first argument, and I’ll leave it here.

The second argument is called logical determinism. But actually, you know what, maybe I’ll start with the third argument—I’ll start with a different argument. I’ll talk about Newcomb’s paradox. Newcomb’s paradox basically says this: let’s assume there is a prophet who knows everything that is going to happen, including what people will choose. And he offers you a game. In front of you are two boxes: one open, with a thousand dollars, and one closed, and you don’t know what’s inside. You have two options: either take the closed one alone, or take the closed one and the open one. Now the prophet tells you what his policy is: he prepared the closed box the day before, and of course he already knows whether you’ll take both or whether you’ll take only the closed one. And then he tells you, look, if you take only the closed box, you’re content with less, so I fill it in advance with a million dollars. All right? I put a million dollars inside when I prepared it yesterday. If I see that you’re greedy and you take the open box too, then I leave the closed box empty. That’s my policy. Now he tells me this policy. Meaning, he offers me this game and says, look, that’s my policy. And now he prepares the closed box, puts a thousand dollars in the open box, places them before me, and goes home. Now when he prepared the boxes, he knew what I would choose, right? He’s a prophet, so he knows what I’ll choose. He has direct information from the Holy One, blessed be He, and he knows. Now the question is: what should I do? On the face of it, if the assumption is that he’s an all-knowing prophet, then I should take only the closed box. Because if I take only the closed box, and he knew that in advance, then he filled it with a million dollars, so I gained a million dollars. If I’m greedy and I take both boxes, also the open one with the extra thousand dollars, then when I open the closed one I’ll discover that it’s empty. Because he knew in advance that’s what I would do, and he left the closed box empty, and so my profit is a thousand dollars. So clearly it’s preferable to take the closed box and earn a million dollars rather than take both and earn a thousand dollars. Okay, that’s apparently the case.

On the other hand, right now there are two boxes in front of me: one closed, whose contents I don’t know, and one open. Now suppose I take only the closed one because I want to earn a million dollars—that was the conclusion of the initial line of reasoning. But now there’s another box in front of me, an open box with a thousand dollars. Why shouldn’t I take it? After all, whatever is inside the closed box is already there. I can’t change anything, right? He already prepared it yesterday. He knew everything, fine, but he already prepared it yesterday. What’s in the closed box is already there; it won’t change. He’s all-knowing, not omnipotent. Meaning, he can’t empty the box now in light of something I do. Okay? What’s inside the box is already there. So now I’ll take the open one too, earn another thousand dollars—what’s bad about that? Therefore even if he prepared the contents of the box—well, then he prepared the closed box empty. And in the end I’ll earn a thousand dollars, meaning I’ll earn less, not more. Basically, we’re entering a kind of loop here. And this loop—again, one can discuss it a lot and quibble, because it has all kinds of folds and nuances, and I won’t get into all those folds here because it’s a bit tedious—but in principle this argument says that we don’t have, you can’t define a strategy here. Meaning, I can’t say to a person what he should do. Seemingly, if this prophet is all-knowing, you should take only the closed box. On the other hand, there’s no logic to that—after all, you have another box with a thousand dollars, and whatever is in the closed box is already there. So take the open box and earn another thousand dollars. So what should you do? Take only the closed box and earn a million, or also take the open one and earn, I don’t know, either a thousand or a million and one thousand? Right—it’s impossible, and that’s exactly the paradox.

The claim that basically emerges from this is, first of all, on two levels. On the first level, it says that apparently there is no such prophet. Meaning, because the assumption that there exists such a prophet led us to a paradox. That means this is a proof by contradiction that there is no such prophet. Right? If a certain assumption leads us to a paradox, that’s what’s called a proof by contradiction. Meaning, the assumption led us to a paradox, so apparently it’s not true. That’s how all kinds of things are proven by contradiction. Okay, even in mathematics things are proved by contradiction: you assume something, show that it leads to a contradiction or a paradox, and in that way show that the assumption was false. So here too, same thing. It means there is no such prophet. Now first I’ll comment on this—what does it mean that there’s no such prophet? After all, I never said who or what this prophet is: whether he’s a human being, the Olympic octopus that supposedly knew in advance which football team would win, or whether it’s the Holy One, blessed be He. I didn’t define who this prophet is. My claim basically was: there cannot be a being of any kind whatsoever, with legs, without legs, with eight legs or zero legs, there cannot be anyone who knows in advance what I will choose, because if there is someone like that, I can force him into a contradiction, into a paradox. And therefore there cannot be anyone like that. Now that “someone” includes the Holy One, blessed be He—you have to understand that. And so this only strengthens the conclusion I stated in the previous lecture: there cannot be anyone, whoever it may be, who knows in advance what I will do, because otherwise that someone could play the box game with me. Even the Holy One, blessed be He, can put boxes in front of me and ask me what I’ll do. And He promises me in advance, I don’t touch the boxes—because He is omnipotent, not just all-knowing. So He could also remove from the closed box the million He prepared for me. Don’t worry, I prepared it yesterday and I’m not touching it anymore. But I did it according to what I knew you would do tomorrow. Now the question is what I do with that. I enter a loop. And once I’ve entered a loop, that means the assumption that the Holy One, blessed be He, knows is an assumption that leads me to a contradiction, and that means He cannot know—which strengthens the conclusion from the previous lecture.

So that’s regarding the question whether God’s knowledge is compatible with our freedom of will. This is another argument saying that it is not. But here one has to pay close attention: I haven’t proven that we have freedom of will. I haven’t proven that you have freedom; this is not a proof against determinism. It only says that if we have free choice, then the Holy One, blessed be He, does not know. And there’s another alternative as well: that the Holy One, blessed be He, does know, and in fact I have no freedom of choice. It’s not a question of strategy, what I’ll choose—I have no free choice, that’s all. What I’ll have to do is what He knew in advance, and that’s what I’ll do. And then too, things are consistent. In other words, Newcomb’s paradox merely eliminates the possibility of assuming both foreknowledge and my having free choice. That can’t be. What can be? One of two things: either there is no foreknowledge and I do have free choice—that’s the option I chose last time—or the deterministic option, which says that the Holy One, blessed be He, knows in advance and I have no free choice. That too is possible. The only thing that cannot be is both: that He knows in advance and that I have free choice. That’s what Newcomb’s paradox eliminates. So on the face of it this is not evidence against determinism. It only rules out the possibility of having both together. Either determinism or free choice, but not divine foreknowledge and free choice together.

Now in my book The Science of Freedom, I argued that there is also an argument here against determinism, not just an argument that rules out the possibility of adopting both divine foreknowledge and free choice—or foreknowledge in general, not specifically divine—the idea that information in advance exists and that I have free choice. Free choice. Because my claim was: suppose I too have no free choice, but in principle there could be some person who does have choice, some being that has free choice. Human beings don’t, but the Holy One, blessed be He, could have created a world in which there is free choice; ours He didn’t make that way, or couldn’t. Now I ask: what would that being’s strategy be in the box game? After all, my question is not what a person will do; the question is whether there exists a strategy. Even if I am a determinist and can’t choose one strategy out of two—I’m a determinist, I live in a deterministic world, everything is forced on me and all is fine, everything is compelled and permission is not granted—but the principal question is whether there is a strategy, not whether I can do it. And that’s separate from the question whether I have free choice or not. The only thing one could say here is that the concept of free choice is internally contradictory, and that seems implausible. In other words, I think very few determinists make that claim. They say it’s implausible, it contradicts the laws of nature, it doesn’t make sense, I don’t know, all sorts of reasons. But to say that it’s a logical contradiction, that the concept itself is contradictory—that’s a very far-reaching claim. It doesn’t look like a contradictory concept. And again, I’m saying this is an argument from plausibility. Someone can come and say yes, this proves that the concept is contradictory. But I say: if I do not accept that the concept is contradictory, then even if I assume determinism, I still haven’t solved the problem. I still have to assume that the Holy One, blessed be He, does not know in advance what I will do, because if He did know in advance, I wouldn’t have a strategy. Not I, sorry—there wouldn’t be a strategy. Right, even though I might be a deterministic creature, so the question may not pertain to me. But still I ask about another creature to whom it does pertain—did the Holy One, blessed be He Himself, have a strategy? He’s playing a game with the Other Side; both are omnipotent, they’re playing the box game, both have free choice. Okay—sorry for the mythological language. So yes, the claim is basically that there cannot be a strategy in such a situation. And here there are two possibilities: either to assume that the concept of free choice is inherently contradictory—which, as I say, is of course possible, I can’t say it isn’t, but it doesn’t seem that way. It may be that it does not exist, it may be that it contradicts the laws of physics, but conceptually I do not see a contradiction within this picture or this description of free choice. But if it is not a contradictory concept, then I can imagine a world in which there is a being with free choice, and now I’ll ask what the box game would yield against that being. And then this basically means that Newcomb’s argument still stands, meaning the Newcomb problem still exists, and apparently if the concept of free choice is not contradictory, then you can’t dismiss it by means of Newcomb’s paradox and you have to assume there is no such prophet, no such foreknowledge. And in that sense, it’s an argument against determinism. Of course, again, there are some assumptions here that one need not accept, just as one can claim that free choice is a contradictory concept, as I said, and so on. That’s always the nature of philosophical arguments: there are always assumptions, and one can always refuse to accept the assumptions. Then one has to decide what sounds more plausible and what sounds less plausible. So that’s the first argument—second argument, sorry.

So we spoke about the theological argument—that the Holy One, blessed be He, knows everything in advance, and therefore we have no free choice—that I discussed in the previous lecture. Now I mentioned Newcomb’s paradox, which first, strengthens the conclusion I discussed in the previous lecture, that if I have free choice then the Holy One, blessed be He, does not know; and second, I think it also serves as an argument against determinism. The third argument I want to discuss is what is called logical determinism. Logical determinism basically makes the following argument. It starts with Aristotle. Aristotle asked a question about tomorrow. He asked: take the statement “Tomorrow there will be a sea battle.” What can I say about it today? I can say about it today that it is true, and I can say about it today that it is false, right? I don’t know—let’s wait for tomorrow and see. I waited for tomorrow and it turned out that there was a sea battle. Fine. Now tomorrow I already know that there was a sea battle. What is the status of the statement “Tomorrow there will be a sea battle” as of yesterday? What was its logical status? Was it true or was it not true? Yesterday I didn’t know, but today I already know. So what is the logical status of the statement as of yesterday? On the face of it, that statement is a true statement; it’s just that yesterday I didn’t know it.

There is a proposal by a Polish logician named Łukasiewicz—you know, Poland is a logical superpower. People don’t really know this, but there’s Polish notation in logic. Poland is a superpower, a superpower in the field of logic. At least it was; I don’t know what its status is today. So Łukasiewicz, one of the well-known Polish logicians, argued that for statements about the future you can assign neither a truth value nor a false value, and therefore he built a three-valued logic. For example, in this case he says the value is “unknown.” Regarding a sentence that speaks about tomorrow, I can say it is true, I can say it is false, but in fact what should be said is that it is unknown. Because you can’t—it has not yet been determined, if you like. Either unknown or not yet determined. But to my mind this is implausible. It’s implausible because when we speak, how do we define a true statement? A true statement means a statement whose content matches the state of affairs it describes, right? If I say now, “It’s light outside,” the statement is true if I look outside and indeed see that it is light. That matches the content of the statement, so the statement is true. In other words, the determination whether a statement is true or not is the result of a comparison. So let’s make the comparison. After all, the assumption is that tomorrow—meaning, I’m already at tomorrow—so the sea battle occurred. Now I ask: what was the truth value of the statement today, Friday, “A sea battle occurred”? Apparently you didn’t hear the news, but a sea battle occurred today, on Friday. Now I ask, what is the status of the statement uttered on Thursday, “On Friday a sea battle will occur”? I claim that clearly this statement is a true statement. The fact that I didn’t know it when I was yesterday—true, I didn’t know. There are many true facts that I don’t know. That doesn’t make them untrue. They are true facts that I don’t know. How do I know that this statement was true? Because I compare the content of the statement with the state of affairs it describes. What does the statement describe? That on Friday there will be a sea battle. I look on Friday and see that indeed there was a sea battle. True, I can make that comparison only today; yesterday I couldn’t. So yesterday I didn’t know the truth value of the statement. But today, when I can make the comparison, then I know: the truth value of the statement is that it is true. And accordingly it was true yesterday too, and the day before, and at the time of the Big Bang. It was always true; I just couldn’t know until it happened, because I have no information about the future, because I’m not the Holy One, blessed be He. I have no information about the future. So therefore I didn’t know the statement was true, but the Holy One, blessed be He, already knows—if we go back to the earlier formulations. Right, He can know the future, and so He already knows that this statement is a true statement.

Fine. So if one really accepts that, then it means that assuming today a sea battle occurred, then the statement “On Friday, on such-and-such a date, a sea battle will occur” was true even 500 years ago, and also 2,000 years ago, and yesterday, and today, and tomorrow. Or in other words, the truth value of a statement is a timeless concept. If a statement is true, it is always true; if it is false, it is always false. The truth value of a statement does not depend on time. Now if that is indeed so, then the logical determinist argues that basically, assuming today a sea battle occurred, then already 500 years ago the statement “On Friday, on such-and-such a date, a sea battle will occur” was true. But if it was already true then, then it can’t be that now a sea battle will not occur. Because after all, 500 years ago it was already true. Granted, I didn’t know it was true, okay—but now I have reached the time where I can know, and what should become clear to me? It must become clear to me that the statement is true, because it was already that way 500 years ago. Only then I didn’t know, and now it has come to my knowledge. So the logical determinist argues that because the truth value of statements is timeless, it doesn’t depend on time, then basically a statement about the future is already true or not true—I don’t know—but its truth value already exists today, I just don’t know it. Before Heaven it is revealed; the Holy One, blessed be He, does know. Okay? But if that’s so, then basically when that future arrives, there is no escape: it will have to behave according to the truth value of the statement that is already fixed today. And that is an argument in favor of determinism.

Why is this argument not correct? Because I think it confuses two levels of discussion. If I know today that there will be a sea battle tomorrow—say, the Holy One, blessed be He, in line with the previous arguments—and if that information is correct, and assuming that when the Holy One, blessed be He, knows, He knows correct information, then necessarily tomorrow a sea battle must occur. And if tomorrow we have a choice whether or not to wage a sea battle, that means the Holy One, blessed be He, does not know today. That’s the conclusion from the previous argument. But what the logical determinist says is not that the information about tomorrow exists today. What he claims is that the truth value of the statement also exists today. That is not information. The truth value of a statement is just a definition of logicians. It is not a claim about the world. When I say that the truth value of this statement is true or false, that is not a claim about the world. It is a logical definition of my relation to the statement. Why is that important? Because with respect to information, information is basically the result of events. The events are the cause and the information is the effect. Therefore, if the events are created in the future, the information can’t be in the past, because the effect always comes after the cause in time. But the truth value of a statement is not information and not a fact; it is simply a definition. There is no problem with a future event determining the truth value of a statement in the past. Because the truth value of a statement is not a fact. Facts cannot be causally determined by future events. That’s true. Events that determine facts have to occur before those facts. But the truth value of a statement is not a fact; it is just a definition. What’s the problem? Can’t I define today a definition that depends on what will happen tomorrow? This is in the Talmud, tractate Gittin. The Talmud there discusses the topic of retroactive clarification. The Talmud says there: someone who writes a bill of divorce for whichever one of his two wives goes out first through the door tomorrow morning. So the Talmud brings a dispute there—there is discussion—but on the fundamental level, according to the opinion that accepts retroactive clarification, then this bill of divorce is perfectly valid. It is considered a bill of divorce written for the sake of that woman. Now the event that the woman will go out tomorrow morning will only happen tomorrow. Does such a woman exist today? Did I write the bill of divorce today for the sake of a defined woman? Because that is a requirement. A bill of divorce has to be written specifically for her, for a defined woman. Now I defined the woman for whom I wrote the bill of divorce on the basis of a future event, an event that will happen tomorrow. Is that considered a definition of the woman? The Talmud says yes. Why? Because as a definition, there is no problem at all with a future event determining a definition.

Suppose there are three people before me today. I say: I want to grant my property from now, all the million dollars I have, to the one who tomorrow morning is the first to leave the house. Fine? Now is that transfer a legitimate transfer? Of course it is. What’s the problem? We don’t know today who it will be—fine, we’ll wait until tomorrow, see who leaves first through the doorway, and then it will become clear that I already defined him today as the beneficiary of the gift. Right? There is no problem at all defining someone today by means of a future event. Why? Because a definition is not information, not a fact. A fact cannot come from a cause that happens after it, from a future cause. But a definition can depend on a future event. I can define something on the basis of a future event; there is no obstacle to doing that. It’s a definition. Definitions—I can define whatever I want. The truth value of a statement is a definition. It is not a factual claim. And there is no problem at all grounding a definition on the basis of a future event—a definition of something in the present. Therefore the logical determinist is wrong. He mixes up two levels. He assumes that a future event cannot generate a present truth value. But that’s not true—it can. And therefore what is the answer, basically, to the logical determinist? If tomorrow there is a sea battle, then it will become clear retroactively that from time immemorial the statement “On Friday there will be a sea battle” was true. If on Friday there is no sea battle, then it will become clear that from time immemorial the statement “On Friday there will be a sea battle” was not true. That’s all. But it was always the same; it’s just that this will be brought to our knowledge tomorrow. Does that dictate what will happen tomorrow? Absolutely not. What happens tomorrow will clarify the definitional status of the statement throughout history. That’s all. It says nothing about whether what happens tomorrow is fixed or not fixed.

So here I want to distinguish this argument from the previous two. Newcomb, and also the argument from God’s knowledge—there is a connection among all these arguments, but a particular kind of connection. The previous arguments, the first two, spoke about information that exists in the present. Information that exists in the present—assuming it is correct, and information in the possession of the Holy One, blessed be He, is presumably correct—dictates what will happen in the future. That seems inescapable. Therefore, if the information exists now, then tomorrow I don’t have free choice. If tomorrow I do have free choice, then the information does not exist now. The logical determinist goes one step further. He wants to argue that even without the information existing today, the mere fact that there is a logical definition of the status of the statement—meaning, it is already true or false today—that too dictates what will happen tomorrow. That’s simply not true. Here it can be that both things exist in parallel. What I was unwilling to accept in Newcomb and with respect to the Holy One, blessed be He, here it is certainly true. There is no problem at all. It can be that the truth value of the statement is already true today, that it already exists today, and tomorrow there is free choice whether to wage a sea battle or not. What I do tomorrow will retroactively determine the truth value of the statement throughout history. There is no obstacle to doing such a thing.

Now I want perhaps to give one last a priori argument, and then I’ll begin touching the scientific plane. Another argument asks: in the deterministic picture, what is the meaning of judgment or deliberation? When we exercise judgment regarding some question, I weigh considerations this way and that way, whether it’s a value question and perhaps also a factual question—say I have a dilemma about which scientific theory is correct. So I weigh facts on this side, facts on that side, what the logic is, I apply considerations this way and that way, I exercise some kind of judgment, and then I arrive at a conclusion. How am I supposed to relate to this judgment? In principle, in a deterministic picture there is no such thing as judgment; what there is is calculation, not judgment. What is the difference between calculation and judgment? Or at least that’s how we usually think there is a difference. The determinist in fact says there is no difference.

I’ll give you an example or examples. Very often in articles published in the popular press—it’s very popular to report such findings—people talk about the intelligence of animals. What monkeys know, what sheep know, what birds that can navigate know, or all sorts of things like that, which show that animals have considerable intelligence. There are even different rankings of which animal is more intelligent and which less so. More than that: there is also computer intelligence. There is a computer with some software, and one can try to measure its intelligence, meaning what it can perform on the human IQ scale. You can test a computer with an IQ test and check how much IQ it has. And then in artificial intelligence they talk about different measures of intelligence and apply them to all sorts of creatures, not specifically to human beings. Human beings are considered intelligent creatures, but let’s say computers, certainly in certain areas, are already reaching higher levels. But something here is still troubling, at least to me and I think to others too. My feeling is that no creature has intelligence except human beings. Not that its intelligence is zero—the concept simply doesn’t exist with respect to other creatures. If we accept computer intelligence or the intelligence of a navigating bird, then by the same token I can also talk about the intelligence of water. Water is very intelligent. You know why? Because when water flows, and we talked earlier about a topographical layout, when water flows through a given topographical layout, the equations that govern the flow of water are called the Navier-Stokes equations. Very complicated equations. No physicist or mathematician, however brilliant, knows how to solve these equations except in some toy-model case, some super-simple isolated special case with no complications—maybe there perhaps one can solve them. In any case beyond the completely trivial, nobody knows how to solve these equations. But water solves them at every moment in real time. It flows through the terrain and look—it succeeds in calculating how it’s supposed to flow there, and it does it. In other words, water is more intelligent than Einstein or, I don’t know, Gödel, or all the great mathematicians and physicists. Water knows how to perform a calculation that no genius mathematician or physicist can do, and it does it with utter contempt. It doesn’t even sweat. I mean, it just—well, water doesn’t need to sweat. It just flows on the basis of the given terrain and carries out these crazy calculations that no one could ever do, easily. And so water is a thousand times more intelligent than all the physicists and mathematicians you know. That sounds implausible. Why? Because water does no calculations at all. Water does no calculations; water simply flows. At most, if you insist, the one who did those calculations is the Holy One, blessed be He, who created the water—or programmed the water, let’s put it that way, if we treat it as some kind of computer. The intelligence of the computer is ultimately only an indirect expression of the intelligence of its programmer. The computer simply moves electrons by virtue of different voltage gradients—that’s all the computer does. The computer does not perform calculations. The computer does not do the calculation “one plus three equals four.” The computer moves electrons. I give an interpretation: I built the computer so that when it moves electrons in certain ways, from my point of view this will be perceived as calculating one plus three, and I’ll see that the result is four.

That means that the computer basically has no intelligence, just as water has no intelligence, just as birds have no intelligence. Intelligence belongs to beings that can and need to weigh, to exercise judgment, and decide whether this is right or that is right, to weigh considerations and determine what outweighs what—to exercise judgment. You understand that behind the concept of judgment, if I am right in my definition here, there basically stands some notion of freedom. In other words, deterministic systems, systems whose behavior is simply the automatic mechanical product of how they are built, are not intelligent systems in this definition. Okay? Now whoever accepts this definition of judgment must admit that there is in human beings something that does not exist in birds or water: the ability to exercise judgment, or the ability to choose a tool from a toolbox and try to solve the Navier-Stokes equations. And whoever succeeds in doing that, in a nontrivial case, really is a genius. Water does not choose a tool from a toolbox and succeed in solving the Navier-Stokes equations and then flow. It simply—the one who solved those equations is the Holy One, blessed be He, who created the water, if at all. That’s a metaphorical description. It’s not even clear that one must say of Him that He solved those equations; He created the equations. But one cannot attribute intelligence to water in this definition. Intelligence involves some exercise of judgment. I decide that this consideration is correct, I think of some new argument, and then I arrive at some conclusion. That means that there is within me some kind of freedom that allows me to decide this way or that way. And that is a necessary condition, though not a sufficient one, for the existence of intelligence. Of course, you also have to apply it correctly, and then you really are an intelligent person. But if you’re not a human being, you can’t be intelligent. It’s not that you are unintelligent; you cannot be. Intelligence simply isn’t defined with respect to you. To say that water is intelligent or to say that a computer is intelligent is like saying that virtue is triangular. The concept “triangular” does not characterize virtues. It’s not a false sentence. It’s not that the degree of triangularity of virtue is zero. Triangularity is simply irrelevant to virtues. You are attaching an attribute to a subject to which that attribute is not relevant; it is not in the relevant semantic field. I claim that intelligence is not in the semantic field of creatures that do not have judgment. Not that they have zero intelligence. They do not have intelligence. Just as a point does not have zero length. A point has no length because it is in zero dimensions; it is not of zero length. Zero length would be when you have a one-dimensional line and its length is zero. But a point has no length—not that its length is zero, but that length is not defined for it, because it lies in a dimension in which length is not defined; that is zero dimensions. So likewise, creatures that do not have judgment are not creatures with zero intelligence; rather they are not creatures about whose intelligence one can speak at all. The concept of intelligence involves judgment. Therefore, I say again, whoever accepts this claim basically accepts that there is in the human being some kind of freedom—freedom of will, or in this case freedom of judgment—that does not exist in inanimate, vegetative, or animate systems—probably, I don’t know, maybe in animals too, it doesn’t matter, I’m not entering that debate now—but regarding creatures that do not have this freedom, one cannot speak of their intelligence.

Creatures that activate autonomous capacities—I manage to breathe and run all my physiological systems in my body in a really brilliant way. It’s just that these are autonomous muscles; I am not really choosing whether to do that. Nobody would think of saying that I have tremendous autonomous intelligence. That’s not intelligence; it just functions. The one who built me is intelligent, not me. Right? So even in a human being, when we talk about the parts regarding which he does not exercise judgment, it is not the result of his initiative and decision, we do not measure his intelligence through them. We can measure his complexity, but his complexity indicates the intelligence of whoever created him, not his own intelligence. My intelligence can be measured through the solving of problems that I solve. I activate a set of tools and I solve the problems; someone else did not solve them for me, or in me, or through me. So this is another consideration, I think, in favor of the existence of freedom in human beings. And again, as with all these considerations, someone can deny this and say, fine, so be it—the concept of intelligence is defined mathematically and formally, and really there is no difference between human beings, and water too has intelligence. Who says not? Fine, no problem, water too has intelligence. Whoever accepts that, then of course my argument does not address him. Each person has to decide what sounds more plausible to him. Is there really some different concept of intelligence here, or are we forced to choose this cold and empty mathematical concept, in my view, this definition of intelligence?

There are somewhat different formulations that carry this argument further and basically say: if I do not have intelligence in this sense, if I do not have freedom or judgment that I can exercise, then I also cannot accept my own deterministic conclusion. Because I also cannot accept that this conclusion is a conclusion I arrive at. Why? Because basically this conclusion is not the result of free judgment that I exercised, but only the result of a mechanical calculation that I performed. Now the question is whether my calculating machine is a reliable machine—does it yield good results or not? I have no idea, and no way of knowing, because after all I have no judgment by which to determine that in a deterministic world. I simply am as I am, because that’s how I am. I have no way of judging this, stepping outside myself and judging it, and seeing whether this deterministic computer, which is my thinking, yields good results or bad ones. Even my decision that these results seem good to me is itself the result of operating that same computer. It won’t help; I cannot step outside it. And then what comes out is that if I live in a deterministic world, I have no judgment, and if I have no judgment then I cannot accept any conclusion that I reach about the world. Because that conclusion is the result of the operation of a blind mechanism, and I have no way of knowing whether it leads to good results or not. And if we were to choose at random among different blind mechanisms, 99.999 percent of blind mechanisms lead to the wrong result. Right? Take a computer, put random software into it, and now ask the software what one plus two is. What is the chance that it outputs three? Zero. The chance that a mechanism chosen randomly and arbitrarily, operating without judgment, will yield the correct result is negligible. Therefore, if I continue this argument from the standpoint of judgment, then I’m basically saying: if I adopt the second view I mentioned, that the concept of intelligence really is only the cold concept, and water and birds and computers all have intelligence, then basically I’m claiming that the world truly is deterministic and everything is calculations. It’s not judgment—I spoke earlier about the relation between calculation and judgment. It’s all calculations, mechanical operations like what water does. Okay? Now I ask myself: when I arrive at some conclusion about something in the world, is there any point in trusting the conclusion I arrive at? I think there is no logic in trusting it, because the probability that a blind mechanism will yield the right result is negligible.

By the way, some people bring in evolution as a counterargument. Meaning, once you have evolution, that basically means there is a natural mechanism that causes me to think correctly. And therefore, although it’s a mechanical and blind mechanism, you can trust its outputs. Because it underwent natural selection, right? Evolution created it as a reliable mechanism. Okay? But one has to remember that evolution too is something I concluded exists in the world by exercising my judgment. And if I have no trust in my judgment, then I don’t know how I can trust that there is evolution in the world. And another point: evolution is not a mechanism that guarantees correct thinking. Evolution is a mechanism that guarantees useful thinking. And that is not the same thing. Let me put it this way: there is logic in being afraid of things that are not fearsome if I am thereby cautious of them, if it brings me better survival outcomes. Because those frightening things don’t need to frighten me at all, but they might cause me some other kind of harm. So evolution will cause me to fear them even though there is nothing to fear, but it will aid my survival because I—well, or I don’t know. Those who are not afraid may perhaps survive better, perhaps not, I don’t know, one can debate it, and then they won’t be afraid of something that really should be feared. Survival considerations are not always determined by considerations of truth. And if thinking is based on evolution, then I can trust its outputs from the standpoint of survival value. I cannot trust its outputs from the standpoint of their cognitive, epistemic value. In other words, who says it’s true? If I see a tiger in front of me, I arrive at the conclusion that there is probably a tiger here. Who says? Maybe there is no tiger here, but whoever reaches the conclusion that there is a tiger and runs away survives. Why? Because what is really here is fire. And if I hadn’t run away it would have burned me. So my conclusion that there is a tiger here is complete nonsense, but it greatly contributed to my survival. Whoever did not reach that conclusion would not survive. Does that mean that the survival mechanism guarantees that I think correctly? Absolutely not. So why trust my thoughts? I can trust them in a utilitarian sense at most, but not in an epistemic sense. If I reach some conclusion about the world, there is no reason to assume it is a correct conclusion.

And this is another extension of the same consideration that says: if you have no judgment, if in effect everything you do is mechanical calculation, then you can’t really take any of your conclusions seriously. In the end, the conclusions you arrive at are survival-oriented conclusions, not true, epistemic conclusions. And as I say, this includes the conclusion that there is evolution. That conclusion too is one I reached through various considerations and observations and so on. And who says it is true, if this whole thing is the result of a blind process?

Now here, again, one can quibble, and in the book I do a bit of that, because one can… I can say: okay, so what am I saying, basically? That if the world is deterministic, then I cannot trust my judgment, including regarding determinism itself. But that’s not correct, because if the world is deterministic, then obviously I can trust my judgment when it says the world is deterministic. Because either way, if the world is not deterministic then my judgment is correct, and if the world is deterministic, then although one cannot trust my judgment, it is still true that the world is deterministic. So either way, the world is deterministic. Fine—that’s sophistry for people with a strong philosophical stomach. I’m not going into those loops here because it’s a bit exhausting. You can think about it afterward.

So I’m basically finishing here the first plane of discussion in this dilemma between libertarianism and determinism, and that is the philosophical plane. You see that in terms of a priori considerations, in the end these considerations somehow leave us with a tie. In my view, even leaning toward libertarianism, but certainly there is no deterministic conclusion on the basis of a priori considerations. If anything, in my opinion it leans strongly in favor of libertarianism. But as we said, a priori considerations are the result of foundational assumptions—what seems plausible to you and what does not, what you think and what you do not think. And if someone attacks my foundational assumptions, then obviously no argument can persuade him. In other words, I raise arguments this way and that way, but all these arguments rest on foundational assumptions. Therefore on the philosophical plane one can always somehow remain in the opposite position and say, okay, then I don’t accept your assumption, even though maybe it seems plausible to you, maybe it once seemed plausible to me too, but now I don’t accept it. You generally won’t settle the debate by means of a priori considerations.

Now if that is indeed so, we have to move to the scientific plane. And on the scientific plane—that’s the second plane of inquiry, and it will occupy us for the next few lectures—the scientific plane of inquiry actually begins with an argument that one could say is not really scientific, although it appears in scientific discussion, but in fact still belongs to the a priori plane. And this is the argument from causality. We know the laws of nature in the world, the laws of physics in the world, and these are deterministic laws. Leave quantum theory aside for the moment—we’ll still discuss what it says about the matter—but mechanics, the ordinary laws of physics, are deterministic laws. And since that is so, what this means is that given a certain set of conditions, the laws of physics determine what will happen in a completely deterministic way. Therefore, even if we do not know the laws of physics, that doesn’t matter, but the very assumption that the laws of physics are deterministic—that given the cause I can determine the effect—this conflicts with free choice, with the thesis of free will. Why? Because free will means that I am in a given state and I have two possibilities, or several possibilities, for how to proceed from here on. Now of course my proceeding from here on is a physical event, right? I am a material, bodily creature. What happens here—whether I walk to the right or to the left—is an action with mechanical significance. That is, a certain body of a certain mass goes here or goes there, at this speed or that speed; the laws of physics govern that. Okay? Now if I have free choice whether to go right or left, that means that given the circumstances in which I stand at the intersection, I have two possibilities. But physics says I do not. Physics says: give me all the data about the state at the moment I am standing at the intersection, and physics will tell me what I will do. Therefore, the principle of causality, the principle that says everything has a cause and that given the cause the effect will also occur, is, I think, the strongest argument in favor of determinism. It is an argument that apparently—not apparently, but directly—contradicts the libertarian view, the view that we have free choice.

One has to understand: free choice by definition is a deviation from physics. It cannot be something that coheres with physics. And whoever accepts the physical assumptions, the laws of nature, whoever accepts the validity of the laws of physics, seemingly cannot accept that we have free will. The moment a person does something, that is a physical action, so it came from some prior force, and if that prior force existed, then this is what he was compelled to do. He could not have chosen to do something else, because the laws of physics dictate that this is what he would do. Or if I formulate it a little more vividly, I’d say this. Suppose someone provoked me, and now I am debating whether to punch him or restrain myself and not respond. So that’s a deliberation, and according to the libertarian I exercise free choice, right? I exercise free choice and decide whether to do this or that. When I punch him—suppose I decide to punch him—that act of punching is a mechanical action. My hand moved. For my hand to move, the muscle has to move it. So the muscle too contracted and stretched. What contracted and stretched the muscle? Electrical currents from the neurons, from our nerves, right? Our nerves activated electrical currents and that contracted and stretched the muscle. Where did those electrical currents come from? From the brain. Right—the brain activated an electric field that sent an electric current through the nerves, which are like electrical wires that run from the brain to the muscle, and that contracted and stretched the muscle, and that’s how I threw a punch. Now I ask: what created the electric field in the brain that caused that current? Some neural activity of the brain, right? Meaning something—there are electrical currents, the brain is a mechanical thing. There are electrical currents in the brain that created some electric field there, which created an electrical current in the nerves, which stretched and contracted the muscle, which produced a punch—one kid after another, right? In the end, this whole business is a deterministic chain. The electrical currents in the brain too are the result of electric fields in the brain. What created those fields? The laws of physics. Electric fields are generated by the laws of physics. Therefore in the end there is a process here that is entirely physical, entirely mechanical—maybe there’s a little chemistry in the middle, it doesn’t matter—but it is entirely physical, and therefore I can go backward, backward all the time, and for everything that happened examine what physical cause produced it, and keep going all the way back until I reach the Big Bang. The human being doesn’t really play any role here at all. The human being is like water in the previous example—he is acted upon. He is part of this deterministic physical chain that began in the Big Bang and rolled forward until it reached the punch I gave the person standing in front of me. Where did my choice enter? Where could I have broken this? If I had broken this chain, that would mean that at some stage in the brain a certain electron moved right instead of left, even though the electric field is trying to make it move left. Then the laws of physics were broken. Because the laws of physics say that if the electric field points there, the electron is supposed to flow there, not somewhere else. That means I broke the laws of physics. It contradicts the principle of causality and contradicts the laws of physics. Contradicting the laws of physics belongs to the scientific plane. Contradicting the principle of causality, as we’ll see later, belongs to the philosophical plane. Therefore this discussion stands at the seam between the philosophical plane and the scientific plane, and we’ll have to handle it carefully on both of those planes in parallel. So that’s what we’ll do next time.

So I’m unmuting the microphones now. If there are questions, you’re welcome. There are all kinds of agitated chats here; I don’t look at the chats while I’m talking, only if something jumps out and I happen to see it, because it disrupts the flow of speaking. So anyone who wants to ask—now’s the chance. Rabbi?