Free Will and Choice – Lesson 11

This transcription was produced automatically using artificial intelligence. There may be inaccuracies in the transcribed content and in speaker identification.

🔗 Link to the original lecture

🔗 Link to the transcription on Sofer.AI

[Rabbi Michael Abraham] They didn’t check anything yet. Just a moment—I’ll come back to these things. Don’t jump ahead. Right now I just want to explain the experiment. So basically Libet’s claim was: I need to find a way to extract from the person the moment on the timeline when he made the decision, and place it here on the axis. How do you do that? One of the big problems in researching mental phenomena is that we have no way to access the mental phenomenon except through the person’s own report. The person himself has to report what he feels, what he thinks, what he wants, when he decided—even things like that—because I have no other way of knowing. I can know what is happening electrically in the brain; at best I have various tools to examine electrical phenomena in the brain. But what is happening in the spiritual, psychological, mental sense—feelings, desires, thoughts, decisions, memories, imaginings, things of that sort—they are all the product of the person’s own report. We have no direct way, aside from his report, to understand whether he wants, whether he decided, whether he this or that. We can see electrical expressions of all these phenomena, but we cannot access the mental processes directly. And so Libet says: we have to find a way to extract from the person the moment when he made a decision, so that he gives me as precise a report as possible about the moment he made the decision. What did he do? He did this: he sat a person in front of a table, and on the table there was a button, and he said to him: whenever you decide, press this button—whenever you want. Fine. Now on the wall in front of him—again, I’ll mute this for a moment—on the wall in front of him there was a clock running. The person sees a clock in front of him. And the moment he decides—he’s constantly looking at the clock—the moment he decides to press, he has to remember where the hand of the clock was. Of course this isn’t an ordinary clock, because it doesn’t have high enough resolution; it has to be a clock with one hand moving pretty fast, and you have to tell me where the hand was when you decided, not when you pressed. When you decided to press. Okay? Because that’s what the experimenter needs. Then after you give me that report, I’ll place it on the timeline. I’ll compare it—meanwhile you’re sitting there with an EEG helmet on. I know when the readiness potential signal appeared, the RP. I know, of course, when you pressed; I can simply see it. And then I can place these three events on the timeline. The moment of the press I just measure—I see when you pressed. The appearance of the readiness potential I get from the EEG, and so I know where on the timeline it appeared, or when a significant magnitude of it appeared. And now when you give me a report of when you decided—when the decision was made—that’s the third point on the timeline that I need. Now of course this is very, very far from simple. Because think for a moment: if you had to report to someone when you decided to press the button, how would you define the moment of decision? A decision is something that builds up. It’s not at all clear that there is a well-defined moment that I understand as: now I decided. Libet did two things. First, he trained the people many times in dry runs before approaching the experiment, so that they would learn to identify for themselves when they had decided. And he had various indications that this was progressing—that is, that he was succeeding. They were giving reasonably good reports. Okay, I won’t go into all the details here. Second, he made that clock in front of them run relatively fast. Because understand: if the clock runs slowly—say, think of a clock ticking second, second, second, second—we are talking here about tens of milliseconds. Tens of milliseconds—you can’t point out to me where the clock was when you made the decision, because it will fall on the same point as the signal and the same point as the choice. It doesn’t have enough resolution. So what they do is make the clock run relatively fast. If it runs relatively fast, then with every millisecond you already see it ten degrees farther along. And therefore, when you make a decision at some point on the timeline, you can tell me what the angle of the clock was at that moment. When the clock runs fast, that increases the resolution of the report. Okay, those are the technical details. It matters, because afterward all these things became subject to criticism of the results of the experiment. I won’t get into all the details, but I’m just showing you roughly what this experiment involved. It’s not such a simple experiment as it may seem. Fine. After he trained the people, and set the proper speed of the clock, he did the experiment. I’m going back to the shared file. Notice: now I’m drawing the three data points here on the timeline. I put this arrow at the point in time when the subject made the decision. Okay, here I place the origin. You see—and everything is an average, of course. You see that the signal is already at a very high level. In other words, the experimenter can already know here that the person is going to perform the action. The person himself decides only a few more tens of milliseconds later, and after that the action is carried out. The difference between, say, the significant rise of the signal and the moment of making the decision—that is, this interval on the timeline—was roughly half a second or thirty—sorry, okay, I think something like half a second or a tenth of a second, on the order of a hundred milliseconds or something like that, plus or minus. Okay, in any case, that was the gap, and it was a clear gap. In other words, it was clear that the signal appears significantly—it already grows significantly—before the person made the decision. And what determines whether the action is deterministic or not is not this moment at all. The dotted line is not interesting. What matters is the relation between this line, the moment of making the decision, and the development of this signal. If this line appears after the signal is already significantly developed, that means that what we are observing here is a deterministic mechanism. If this line had been located here, or even here if you want, then no—because then the making of the decision would have caused the creation of the signal and not the other way around. But here we see that the creation of the signal produced in the person the feeling of decision, not the other way around. In other words, he wants what he does, and does not do what he wants, the way we delude ourselves into thinking. That, basically, was Libet’s result. Fine. Since then, this result stirred up a tremendous uproar. In the world of brain research, and among philosophers and scientists and so on, this was a revolution. The feeling was: here we have an empirical scientific result showing us that our sense of choice is an illusion. The world is a deterministic world. And therefore, in every discussion of neuroscience and free choice, Libet’s experiments are always center stage. Now, “Libet experiments” today has already become the name of a type of experiment. Why? Because Libet himself performed several different experiments. As I said, at the end of the 1970s and beginning of the 1980s of the previous century. Since then they have never stopped doing experiments of this kind, right up to our own day. Each one refines the others in this nuance or that nuance—hundreds and hundreds of experiments. All over the world, each time a lot of criticisms come up—scientific, philosophical, and others—and the next experiments try to correct those criticisms. And they tested this from every angle and every direction and every sort and variety they could manage. And so many people today live with the feeling that we’re already past the point of debate. The later experiments confirmed Libet’s results—not only confirmed them, they even strengthened them. In 2008, for example, there was an experiment by Soon and his colleagues, and there they discovered something amazing. They basically told a person to decide whether to raise his right hand or his left hand. Instead of pressing a button, to raise a hand—it doesn’t matter—raise the right hand or the left hand. They discovered that the EEG can predict the person’s decision six or eight seconds—not milliseconds—six or eight seconds before he decides. That’s already not a matter of some scientific dispute—maybe you missed it, maybe he didn’t know exactly when he decided. Six or eight seconds is forever. It’s forever. That’s a huge amount of time. If I know it that long before the person decides, that basically means the whole thing is completely deterministic. There are two types of prediction there: whether he will raise a hand, and which hand he will raise. As for whether he will raise a hand, that was six seconds in advance; as for which hand he will raise, that was—or the reverse. As for whether he will raise it, that was eight seconds in advance; as for which hand he will raise, that was six seconds in advance. In other words, the decision to raise the right hand is a decision that is made gradually. One center of the brain, eight seconds earlier, produces a readiness potential; two seconds later another center of the brain says: it will be the right hand. Six seconds later I decide to raise the hand, and another second later I raise it. So these are already really unequivocal results. A very, very long time interval. Six or eight seconds. There are very many experiments of all different kinds, and the conclusion of many people was that our free will is really an illusion. Neuroscience settled this ancient question by knockout. Now there were many criticisms of Libet’s experiment, and also of the later experiments and its results. There were criticisms of it—you have to understand, when I describe this here, I’m describing something in a very, very, very simplified way. In the end there are many decisions here, and we take averages, and this signal is not nearly as clean as it seems. There is a lot of noise around it. So there are very, very complicated mathematical and scientific techniques here just to extract the results, even before analyzing the results. What are the results of such an experiment, anyway? Very far from simple. Very far from simple, and you have to understand that this is critical. It is critical for defining the concept of free choice and for determining whether there is free choice here or not. I’ll explain in a moment why this is so critical. There were those who argued about the statistical reliability, about the question whether a person can really determine when he decides. Some wanted to argue that in fact the person decides unconsciously, but the decision is still a free one. His awareness arrives at a certain stage, but he made the decision before he became aware that he had decided. And so it could be that the decision was made before the readiness potential began to grow. My awareness comes afterward. That is a claim that recurs among quite a few critics. I think this claim is irrelevant. It’s irrelevant to the essential discussion. It may perhaps be true, but it’s irrelevant to the essential discussion. Why? Because the question whether I decide freely or not—why is it interesting? What makes it important? It’s important, for example, if I want to ask myself whether I assign moral responsibility to a person who performs an act. Did he choose it, or was it forced on him? If it was forced on him, then there is no moral responsibility. Now it is clear that even if I make a decision freely—not through my physics or my physiology—but I do it unconsciously, you can’t impose responsibility on me. Moral responsibility can be imposed on a person who made a conscious decision to perform some action. Therefore, even if the critics are right when they say that there is some free decision here but it is unconscious, and it preceded the development of the readiness potential, that still doesn’t really protect the thesis of free will; it throws the baby out with the bathwater. You defended the thesis of free will, but emptied it of content. Because in the end it turns out that maybe there is some free will in the sense that we are not subject to physics, but it has no implications. Because as far as assigning responsibility to me is concerned, you still can’t assign responsibility to me. That’s not what we call free will. What we call free will is a situation in which a person makes a conscious decision for this purpose. Therefore, as long as the decision is unconscious, no—it contributes nothing to the discussion. Okay. Now I’m moving on to—I’m going to present a series of three, maybe even four, levels of criticism which, in my view, are the substantive criticisms. Yes, they came up in various forms in the literature too. I don’t know all the literature either—it’s an enormous literature. After I wrote this chapter in the book, Judith Ronen from Beit Avi Chai sent me—she has a doctorate in philosophy—her son did a master’s in psychology on Libet’s experiments under the supervision of Benny Shanon from the Hebrew University, and he wrote a master’s thesis on it. She sent me his thesis, so in the end I even found some references and discussions, but it didn’t substantially change what I wrote. In any case, if anyone wants, there’s a pretty decent overview there—again, correct as of 2012 or something like that. Since then there have already been developments, at least one of which I’ll still mention. In any case, I want to propose three arguments against the conclusion from Libet’s experiments that point to determinism. The first argument was raised by Libet himself. Libet himself, as I already mentioned, was a libertarian. He believed in free choice. So his experiment, which on the face of it reflected a deterministic mechanism, was a blow for him. In other words, it required him either to give up the libertarian thesis or to offer some explanation or reconciliation for these results. It’s very interesting, by the way, because ostensibly he should have thought about this when he did the experiment, since he did the experiment in order to test whether we have free choice or not. If you design an experiment that tests whether we have free choice, it can’t be that after it shows that we don’t have free choice, you explain that it doesn’t show that. If it doesn’t show that, then why did you do it? In other words, if in any case the experiment won’t give me an indication, why did you do it? Which shows us that human beings are often driven by, let’s call them, intellectual interests. That is, I believe in free will, I wanted to prove it, I brought an experiment, the experiment slapped me in the face, so now I start thinking—I have excuses. I already explain why it isn’t right. Okay, but fine, that’s how it works; probably none of us is free of that problem. And what he argued was the following. He said: in fact the experiment showed that every time before the person pressed the button, a readiness potential arrived that preceded his decision. But the experiment did not show the reverse—that every time there is a readiness potential, a decision appears afterward. A very subtle point. In other words, there may be situations—he didn’t test this, though in later experiments they in fact found such situations—Libet only argued that there could be, and therefore there is no proof from the experiment’s results—that there could be situations in which there is a readiness potential, and afterward no action is carried out. True, if an action was carried out, before it there was always a readiness potential, then a decision, then an action—determinism. But that’s only whenever an action was carried out. Libet says: I didn’t examine situations in which a readiness potential developed but no decision was formed and the person did not press the button. What does that mean? Basically what it means is that a person has veto power. In other words, the readiness potential will always come first—if there was a press, then before it there was always a readiness potential—but if there is a readiness potential, I am not compelled to press the button. I can decide to impose a veto and not press the button. And Libet says: if that is so, then this result does not show determinism. It shows half-determinism. That is, it says: every one of our decisions is indeed driven by physiological mechanisms, by the electrical system in the brain. But it is not true that the electrical system in the brain forces the decision on us. That is not true. Because it may be that the electrical system wants us to do something, and we will impose a veto and not do it. We do not have the ability to decide positively; we have the ability to veto decisions that the brain is trying to lead us toward. That is basically the claim. Right, someone comments here that this is remorse, so two days before Yom Kippur this is really a lesson for the season. A nice comment, and a correct one. But a person can—even though this is not remorse over a decision; it is remorse over a desire that arises spontaneously, so it’s a bit different from what we speak about when we talk about repentance—but yes, it’s an interesting idea. In any event, Libet’s claim was that the very fact that we did not rule out the possibility of a veto means that this experiment did not show determinism. It does not necessarily express determinism. A person has the possibility of veto. He has no possibility of making a decision independently, but he does have the possibility of imposing a veto. That’s the claim. In other words, it’s a kind of incomplete determinism. You can still assign blame—say, to a person who murdered. You can assign blame to him because, true, his brain wanted to make him murder, but he could have imposed a veto and not murdered. And therefore responsibility still rests on him. In that sense he saved the moral essence of free choice. Half of free choice is enough for the moral essence. Now, imposing a veto is also indeed a kind of choice, but with regard to that there is no experiment about readiness potential. So of course you are just pushing the question back one step. Notice: even if after Libet they really found that there are situations where a person imposes a veto—that is, a readiness potential is created and the action is not performed—that of course still does not prove there is free will, because it may be that the imposition of the veto itself is driven by some other electrical signal. That too may be the result of a deterministic brain mechanism. It only means that Libet’s experiment did not prove determinism. It does not mean that there is proof here of libertarianism. There is no proof. The question remains an open philosophical question. That is basically the claim. Now I just want to clarify: the chapter I’m dealing with today actually summarizes everything we’ve done so far in the series. Because understand what veto power basically means. I remind you again of a whole lesson that I devoted to explaining what free choice is, because many people don’t understand what it is. People think free choice means action in a vacuum—no influences on it: not genetics, not environment, not society, not education, nothing. Rather, I decide completely freely, arbitrarily, randomly, whatever you want, in a vacuum. And then they always say: wait a second, but there is genetics that determines whether you’ll be this way or that way, there are brain influences, environmental influences, educational influences—so that means determinism. Not true. Libertarianism does not deny the existence of environmental influences, genetics, education, and so on. What it denies is that the environment dictates the result. There is a difference between influencing and dictating. What is the difference? Influence means that you are trying to move me to do something. There is a greater chance that I will indeed do it, but I still have the ability to impose a veto and not do it. If you remember, I gave the analogy of the topographic map: a person moving around like a little ball, or water—a stream of water flowing on some topographic surface, mountains, hills, valleys, and so on. And the claim is that the stream of water will arrive at the place that the topography leads it to. That is determinism. The environment determines what you will do. Also a little ball rolling there will arrive at the place of minimum potential energy—at least local minimum. But what happens with a human being? The libertarian claims that a person who is in that same situation can of course slide down to the valley, to the minimum potential energy, but he can also decide that he is unwilling to do that, or he can decide to climb the mountain even though that costs him potential energy. And in the analogy, if we take all our environment—psychology, education, the brain, genetics, the home, society, everything—and draw all of that as a map of pressures or forces acting on the person, then basically it is as though I act within some topographic terrain. There is a certain direction where there is a mountain—it is hard for me to climb, so it will be hard for me to go in that direction. There is pressure on me not to go in that direction. In another direction there is a valley—it is very easy for me to slide down there. So the topographic map is an analogy for the collection of influences and pressures acting on me. This is accepted even by the libertarian. The libertarian also agrees that there are influences on me, and these influences can even be examined statistically—I spoke there about how statistics is possible for something that is not random—but there are statistics for these things; psychology works statistically. But where is the difference between the libertarian and the determinist? The libertarian says that this is influence, not determination. Circumstances do not dictate what I will do; they influence, they exert force on me, it is harder for me to resist, but I still can. And to resist, translated into Libet’s language, is to impose a veto. That is what it means to impose a veto. Or in other words, if I translate the model I described there into Libet’s result, into Libet’s proposal that was later also confirmed in experiments, what we are basically saying here is that the sum total of all the environmental pressures and influences on me creates a readiness potential to press a button, to raise the right hand, the left hand. It creates the readiness potential. That is the summary of all the deterministic forces trying to make me perform an action. And therefore, if I do not impose a veto, I will indeed perform that action. I will decide and I will perform that action. But I, as a human being, unlike a little ball or a stream of water, can also decide to impose a veto. I am not going to fall into the valley that the environment wants to take me to, to make me fall into. Or I do want to climb the mountain that the environment is trying to keep me from climbing because it is hard—it is hard for me to climb it. I can still decide to do that. That is called imposing a veto. And the readiness potential basically marks—it is really a one-to-one expression of the model I described at the beginning. Because basically the readiness potential describes what I would do if I were a little ball or a stream of water. There, the circumstances would dictate what I do. If there is a valley here, I would go down into the valley; I would not climb the mountain. That is the expected result. But all is foreseen, and freedom is given. And you can impose a veto and say: I am not a little ball or a stream of water. I decide to climb the mountain even though it is hard, and a little ball or a stream of water would not do that. But I decide to climb the mountain, and even though I lose energy by doing so—or gain potential energy, or lose—meaning, I accumulate energy, I work in order to accumulate potential energy, okay?—still I decide to do it. That is exactly the meaning of veto. To impose a veto is to go against my inclinations—we return to Yom Kippur. Fine? To go against my inclinations; they want to make me do something, and I impose a veto and refuse to do it. If I committed a transgression, it is always the result of inclinations. But the fact that there are inclinations does not automatically mean that I will necessarily commit the transgression. I can impose a veto and not do it, overcome my inclination. I think this is really a wonderful reflection—the results and the veto interpretation—a wonderful reflection of libertarian discourse exactly. Because that is exactly what the libertarian says—the sober libertarian, not the naïve libertarian. The naïve libertarian says there are no influences, no actions, the person is an angel. Nothing in the environment affects him. There is no topographic terrain, nothing. That’s nonsense. But the sober libertarian says: of course there are influences, and of course most people will also go with them in most cases. But I have the possibility of imposing a veto and not going along with them. That is the meaning of choice. So if we take a hundred people who have an inclination to sin, and a hundred people who do not have an inclination to sin, obviously in the first hundred there will be more sinners. But not all hundred will sin. There will be some who don’t. In other words, statistically it is true, but deterministically it is not true. That is exactly the point. Therefore there is no proof here of determinism. I’ll say more than that: I mentioned earlier Libet’s averaging. Basically, in order to present these beautiful results, he averaged over lots of noise and lots of decisions—you average over lots of cases. And averaging is a very dangerous thing. Why? Because if I took, say, a hundred cases and I check what the average result is, it is enough that in one of them a person imposed a veto in order to show that he has free choice. And in the other ninety-nine he went along with his inclination. So the average will show that the person clearly goes along with his inclination. But the average does not reflect reality, because in order to show free choice it is enough to show that one time out of a hundred he did not do it. And therefore such an experiment, by definition, cannot test determinism simply by virtue of the fact that it averages, even before the question of veto. Simply by virtue of averaging, it cannot test that, because the point is not what usually happens. The question is whether in principle I can avoid going along with the circumstances, avoid going along with the readiness potential that is trying to push me in a certain direction. It is enough that I do that once out of a thousand, or five times out of a thousand, to say that I have free choice. The fact that in 995 cases out of a thousand I did go with the inclination, as it were, with my natural tendency, does not mean I have no free choice. The person who believes in free choice also agrees that very often a person goes with his impulses. He only claims that a person can choose not to go with them. And the determinist does not accept that. Therefore the dispute is only about what happens in every single case to the last one; the average simply cannot be a measure for that question. This is a very, very important point that people unfamiliar with this experiment do not understand. Now I want to make a second argument. The second argument is also connected to things I discussed in previous lessons. Everything is connected; this chapter summarizes everything I said. Because everything I just said will find expression in where the failures in the interpretation of the experiments lie. The failures come from imprecision or lack of philosophical skill. That is exactly the point. The second problem is the problem of picking versus choosing. Picking is just taking something randomly, and choosing is choice. Okay? What does that mean? When a person stands at this table—or sits at this table—and has to decide whether to press the button, or whether to raise the right hand or the left hand, or decisions of this sort, these are banal decisions. They involve no dilemma. There is no conflict. I am not torn between two opposing values or something of that sort. Okay? Now what happens in a case like that? That is why this is called a state of picking, as opposed to a state like a moral dilemma, which is a state of choosing. What is the difference between the two situations? A critical difference. Why? Because in a state of picking, the person—say a readiness potential is formed and tells me now press the button—why should I impose a veto? There is no dilemma. Pressing the button has no particular consequences that I am weighing for or against. So what’s the problem? If I ask myself when to press a button, whenever my readiness potential pops up, I press. Why not? There is no preference for one moment over another. It is a neutral action. It is a banal action and completely devoid of significance. So in deciding when to do it, of course I let my readiness potential decide when to do it. Whenever my readiness potential pops up, I decide and perform the action. So what is the surprise that you won’t find veto in banal actions of this kind? Why should I impose a veto? Why should I go against my inclination if the inclination is not an evil inclination? Rather, I just feel like eating something tasty, and there is no prohibition against eating the tasty thing, okay? And it’s not fattening either. And there is no problem at all. It’s just tasty. So I have an inclination to eat that thing. Is there any logic in imposing a veto on that? Why would I impose a veto on that? It’s tasty, it’s pleasant, it’s not forbidden, everything is fine—so I eat it. So obviously one can predict in advance that if I have an inclination to eat it, I will indeed eat it. Is that astonishing? It’s obvious. Let’s see how clever you are, Libet, with a choosing experiment. Remember the trolley dilemma? Whether to divert the train here, divert the train there. Fine? The train is traveling on a track and is about to run over five people. Now I am standing next to a switch, yes, at a fork in the tracks, and I can divert the train to the side. Then the train will run over only one person who is standing on the other track, but I killed him with my own hands because I diverted the train and sent it to kill him. So that is not a simple dilemma. On the one hand I kill one and save five; on the other hand I kill him with my own hands, whereas otherwise I am only failing to save. And the person is in a dilemma. This is what is called the trolley dilemma, which Philippa Foot and other philosophers and psychologists and neuroscientists all deal with. If I attached to the person standing by that switch handle an EEG helmet and Libet’s clock, and I asked him, when did you decide, and I checked the relation between the time of decision and the readiness potential—that could be an indication. Why? Because here, if the readiness potential really tells me in advance what he will do before he has decided, that means that even decisions in value-laden dilemmas are not determined by free choice but by a brain signal. That is already a state of choosing, not a state of picking. Therefore an experiment of that type might perhaps bring us closer to the question whether we live in a deterministic world or not. Libet’s experiment examined a state of picking, and all the experiments until the most recent one examine a state of picking. Until 2013, no experiment at all had been done on a state of choosing. Hundreds of experiments, and not one was on a state of choosing. At the time, when I wrote the book in 2012–13, a friend from the Hebrew University told me—he had been the teaching assistant in the neuroscience course I took there—that they were now working at the Hebrew University on a choosing experiment. Okay? Now later I’ll try to explain that even if a choosing experiment is done and shows that the readiness potential precedes the decision, even that will not settle the deterministic question. That is the third layer of my challenge. But before that I won’t deprive you of the following gem, what I mentioned earlier. In 2013 my book came out. I wrote it in 2012–13; it came out in 2013. In 2015, two years later, an experiment was conducted that included dozens of leading brain researchers from all over the world, among them Liad Mudrik—yes, she had been a journalist on Army Radio; today she is at Tel Aviv University in neuroscience—and she took part in this experiment. And this experiment tried to create a state of choosing. And look what a beautiful trick they came up with, because it is not simple at all to do such an experiment. Understand that I need to do an experiment that places the person in a moral dilemma. That is, if he does something there will be a bad result; if he does not do something there will be a good result. Now this has to be done according to the Helsinki Declaration. You can’t do an experiment like that where you give a person the option of harming someone, killing him, I don’t know, injuring him, I don’t know exactly, doing something bad. How do you do an experiment that will be choosing, but that can also be done technically, morally, legally, judicially?

[Speaker C] Look at their idea. I’m sharing with you a clip from a video, okay? On YouTube. One moment. Yes. This is shared with audio, so you’ll be able to listen. Listen, I’m opening it for something like 10 minutes—listen to it in her own words.

[Speaker E] All these Libet experiments and the ones that came after them—the subjects were not making decisions like “to be or not to be.” They were just deciding whether they wanted to move or not move. And in our view, that’s not the question. Meaning, that’s not what we should be asking when we deal with free will. Why? Because these findings were used by people to make claims about moral responsibility. So if you want to make claims connected to moral responsibility, then let’s examine decisions that have some connection to morality—decisions with meaning, decisions that consciousness might actually have a reason to be occupied with—not whether I want to move my right hand or my left hand. After all, what happens in these experiments? The subjects sit in a lab. Okay, they move—so what? It has no significance for them. Maybe it’s not surprising that their consciousness trails somewhere behind the decision process. So what we noticed was that there are actually two branches of knowledge here. There is the research field of voluntary action or free will, and they deal only with picking, only with meaningless decisions. And there is the field called neuroeconomics, and they are very interested in meaningful decisions, but they are not interested in consciousness. They are not interested in the question of the connection between brain activity and the experience of decision. So we essentially tried to put those together and conduct an experiment that would allow us to isolate the question of the connection between brain activity and the conscious experience in the context of meaningless decisions and meaningful decisions. So what we did was take subjects—this was still in the United States, at Caltech—we took subjects who described themselves as socially involved people, active in nonprofit organizations, volunteered at least three times a year, voted in the last election. At the beginning we presented them with various organizations, various nonprofits. They could go into the pages of those organizations, learn about them, and as you can see there were organizations that were more consensual, like the fight against cancer or The Hunger Project, which wants to fight hunger in the world, but there were also organizations that were more controversial. Because if you want to donate money to an organization that supports abortion, you certainly won’t want to donate money to an organization that opposes abortion. So we basically wanted to get a broad scale of preferences from our subjects. And the subjects basically rated how much they supported each and every one of those organizations. They got time to learn about them, think about them, make the decision. And then, for each subject, we tailored an experiment that fit his own preferences, and there were trials in which he received a hard decision—that is, he had to choose between two organizations whose ratings differed from one another by one point. And there were trials in which we gave them easy decisions, where the two organizations differed from one another in his preferences by at least four points. And the subjects had to decide which of the two organizations appearing on the screen—they had the two goals there—they wanted to donate money to. And each subject knew that at the end of the experiment we would hold a lottery and donate forty real dollars to whichever side was selected by the lottery. And more than that—we told them, and that’s also what we did, and I was terribly anxious they’d choose a bad organization—at the end of the experiment we held a lottery among all the subjects, and for the subject who was selected, to his chosen side we donated a thousand dollars to that organization. So the subjects knew that every decision they made could potentially lead to a donation of a thousand dollars to one of the organizations. That was choosing, but there was also picking. What was the picking? We presented exactly the same two goals on the screen, and the subjects still had to press right or left, but this time it didn’t matter which organization they chose, or which side—whether they pressed right or left—both organizations would receive five hundred dollars if that side was selected, or twenty dollars for each of the subjects. So here we have the same visual input—that is, the same names on the screen—we have the same motor output, always pressing right or left, but the meaning of the decision is completely different, because in choosing they are choosing and there are reasons and consequences, while in picking they are just pressing right or left and it affects nothing and they have no reason or basis to prefer one alternative over the other. And it’s a kind of two-by-two setup, meaning that each decision—or each easy decision—appeared once as choosing and once as picking, and of course that alternated. So just so you understand what this experiment looked like, the subjects are sitting there, they see two names on the screen and have to choose which one they want—here in this case it’s picking, sorry. In order to make sure they were still looking at the screen, from time to time we gave them a little memory test. In a moment you’ll see that we ask them: what was shown there before, just to make sure they are reading the names and actually paying attention, because they could simply have closed their eyes and done right and left. So that was in order more or less to equate the conditions. Of course it’s not the same thing, because here there is a meaningful decision and here there is not. So we basically had these blocks of decisions that were either meaningful or meaningless, either easy or difficult, and from time to time a memory test. The first thing we wanted to see was how the subjects behaved—that is, whether their behavior indicated that we had really caused them to make meaningless decisions and meaningful decisions. So what you see here are the subjects’ reaction times. Every dot here is one subject. In red, that’s in the condition of meaningful decisions; in blue, that’s in the condition of meaningless decisions. And the location of the dot in this coordinate system is determined by the reaction times in the easy decisions. Let’s say this subject took an average of two seconds to make an easy decision, and for a hard decision this subject took about two and a half seconds. So what do we see here from this picture? First of all, we see that they made the meaningless decisions much more quickly. You see, the average here is about one second, a little less than a second. The meaningful decisions they made more slowly. The average here is 2.2 and here 2.5. The second thing we see is that the meaningless decisions lie on the diagonal. What does it mean that they’re on the diagonal? That there is no difference between the reaction times for easy decisions and hard decisions. Why is there no difference? Because it’s meaningless. By contrast, in meaningful decisions the subjects are slower when the decision is hard than when the decision is easy, which makes perfect sense. We also checked how consistent they were, how consistent they were with the ranking they had given us before the experiment. And we see that in meaningful decisions, if it’s an easy decision, they are very, very consistent. They almost one hundred percent always choose what they had initially preferred; there is a four-point difference compared to the second alternative. By contrast, in hard decisions they are around eighty-seven percent, if I remember correctly, which makes sense. Because if earlier I said that I prefer the fight against cancer and I give it seven points, and to education I give six points, and now when you show me one against the other I suddenly say: wait a second, maybe I actually prefer education over the fight against cancer. What matters is that in the meaningless decisions they are around fifty percent. In other words, they do not show consistency with their prior preferences. So behaviorally we were thrilled: we succeeded in getting them to make genuinely meaningless decisions and genuinely meaningful decisions.

[Speaker F] Decisions that have significance are easy, because this secondary harm isn’t caused there. Either this one wins or that one wins.

[Speaker E] Listen, listen, there was there, for example,

[Speaker F] You’re saying there are harder dilemmas.

[Speaker E] You’re right, but first of all we chose subjects for that very reason who were socially involved, meaning people who care. Second, look, there was for example pro-Israel and pro-Hamas there. I was very worried that the pro-Hamas side would get it. Or those who oppose abortion or support abortion—for them it’s not a matter of this one wins and that one wins. That’s exactly why we included controversial organizations there; we wanted the subjects to care about their decisions. But you’re right that in an experiment there’s simply a limit to how many decisions you can make real. So what about the brain activity then? Will we find this readiness potential, this activity that precedes the decision, in both kinds of decisions? I’ll first show you the data from the individual subjects before taking an average. When you average, everything looks prettier. But precisely for that reason I want to show you what individual data looks like. So you see here—it’s a bit thin so maybe it’s hard to see—you see that this is in the meaningless condition. Each such line, each wave, is the electrical brain activity as recorded on the surface of the skull, on the scalp. Zero is the time of the decision. And we told the subjects: decide, and as soon as you feel the conscious urge that you’ve decided, respond immediately. And we see, as expected, about a second before that, this readiness potential, this potential of preparation that precedes the decision. What happens in decisions with significance? Well, to our surprise, nothing happens. It’s completely flat. I’ll show you now in the average. You see here we did a regression to check the slope; here it’s highly significant, here it’s not significant at all. When looking at the average, you see a very, very nice wave with the characteristic distribution across the scalp in meaningless decisions, and in meaningful decisions it’s no different from zero, it’s not significant, and the distribution doesn’t look like what we would expect. In other words, it seems that if Libet had done his experiments on decisions with significance thirty years ago, maybe at least some of the discussions about free will being an illusion and things like that could have been spared. Now I’m not saying we proved that there is free will—not at all. But what we did show is that the strongest scientific finding we have today against free will is limited to meaningless decisions. When you use meaningful decisions that have content, that finding is not obtained.

[Rabbi Michael Abraham] Now, this is really the main segment. The whole lecture is devoted to this, but this is really the main segment. I sent you the link to this lecture in the chat for anyone who wants to listen to it in full. I remind you, this lecture was given some three or four years after my book came out. I called her—I only discovered this lecture about two years ago, I think—I called her to ask whether she had read it and whether there was some connection, but there isn’t. She told me she didn’t know it. In any case, it reflects this issue beautifully. They found what I think is a wonderful idea for how to do an experiment involving a moral dilemma between good and evil in a way that is legally permitted and acceptable under the Helsinki convention and so on. And they found the results that indeed all this fuss about Libet applies only to acts of picking. And in acts of choosing there is nothing. It’s quite amazing, this thing. Meaning, I think that even if it doesn’t prove that there is free will, let’s say it proves very clearly that there is a difference between actions we do out of value-based judgment and actions we do just like that. And that much you can already see. Is that difference specifically free choice or not? That’s an interesting question. But it seems to me that after this, admittedly it’s not one hundred percent in favor of free choice, but it’s also no longer fifty-fifty. It’s already, say, seventy-thirty, or however you want to quantify it. Because there is something here that really shows that actions of judgment—if you take the trolley dilemma, yes?—if you do a Libet experiment on the trolley dilemma, you’ll discover that you can’t predict what the person will decide. And that says exactly that all the Libet experiments, at their substantive level, say nothing on the philosophical level. They say interesting things for brain researchers; absolutely, it may well be that there are unconscious decisions or conscious decisions or whatever. But in terms of moral responsibility, the implications of the libertarian view, it seems to me that this experiment—and apparently, as far as I understand, every experiment gets criticized, but this is a very well-founded experiment—she describes very well there everything they checked. Researchers of all kinds took part in it; they also brought in philosophers, and they understood that what’s needed here is a combination of various disciplines. And the finding is unequivocal. The finding is unequivocal, and she also addresses here all the points I talked about earlier, from the standpoint that averaging always ruins the issue here, because once you average, even if there are Libet-style results, it won’t say anything. The question is what happens in each individual case, not what happens in the average result, and everything I discussed earlier. So if I summarize, and I’ll stop here, then we’ve encountered the first two objections. The first objection: maybe there is a veto phenomenon. Libet says, I didn’t test that. Once there is a veto, we return to the free-choice model that I described with the topographic plane and everything I talked about earlier. A person has responsibility, a person has free choice, regardless of Libet’s results, even if the Libet experiment had been done on her experiment. Meaning, even if her experiment had shown that there is a readiness potential before the decision, it still would not have meant that there is no free will, because there still may be states of veto, and that the experiment did not test. That’s the first objection. The second objection says that even if there are no veto states—and each such objection stands on its own, meaning even if you got Libet-style results in a choosing experiment, that still wouldn’t prove anything, because maybe there is a veto. The second objection says that even if you tested all the cases and there is no veto, but you only did it on a picking experiment and not on a choosing experiment, again you proved nothing. Meaning, each of these claims stands on its own, and there are two very strong claims against Libet’s interpretations. And notice, as I said, both claims were raised regarding Libet as hypotheticals. Maybe there is a veto, or maybe in a choosing experiment it would look different. In later experiments, both claims were found to be correct. Not just hypothetical—it was measured, that in choosing experiments it indeed doesn’t work. And it was also measured that there are veto results. That too was measured in other experiments. Meaning, if in Libet’s time these were two hypothetical objections, later experiments show that indeed the objections were correct. And again, that doesn’t mean we have free will. It only means that Libet’s experiments cannot test the question of whether we have free will. As I said, next time I’ll begin with the third-level objection, and there I’ll argue that even if we accept, get past, overcome these first two obstacles—meaning we prove that there is no veto in any case, and we prove that even in choosing experiments the readiness potential precedes the decision—even then we still have not proved that there is free will. Fine, but I’ll leave you in suspense about that until next time. If anyone wants to comment or ask, you can do that now.

[Speaker B] More power to you. A good year, and may you be sealed in the Book of Life.

[Rabbi Michael Abraham] A good year, and may you be sealed in the Book of Life as well. Goodbye. May you be sealed in the Book of Life.

[Speaker B] Rabbi, this is a wonderful explanation of “all is foreseen, yet permission is granted,” if we say there is a right of veto.

[Rabbi Michael Abraham] The same explanation I gave in the topographic model. Exactly the same thing. This is simply the electrical or scientific expression of the topographic explanation that I proposed there. That’s what I’m claiming. It’s simply a visual presentation of the libertarian model in this experiment.

[Speaker G] If there is choice, then how can it be that most of them act—that’s how you presented things? I didn’t understand. You presented it as though there is choice, and statistically most act, meaning most do not impose a veto. That’s how you presented things, and it’s not clear. If there is choice, then ostensibly it should be fifty-fifty.

[Rabbi Michael Abraham] Not at all. That’s exactly the mistake, exactly the philosophical miss of the scientists. That’s exactly where Libet’s interpreters went wrong. Because the libertarian, as I said earlier, also agrees that circumstances influence me. Meaning, if there are a hundred people who have an urge to murder, and a hundred people who don’t have an urge to murder, then if we have free choice it won’t be fifty-fifty in both groups.

[Speaker G] That certainly not. The way I understand it, it should be like this: those who don’t have an urge to murder, then zero percent murders; and those who do have an urge to murder, it should be fifty-fifty. That’s how it seems to me.

[Rabbi Michael Abraham] No. Those who have an urge to murder—it could also be seventy-thirty, eighty-twenty, depending on how strong the urge is and how important it is to you not to murder. But there will be some who do not murder—that’s what I can say.

[Speaker G] I would define it maybe a little differently. I’d say that in order to overcome the urge and not murder, I need to go through several stages. I need both to decide, and then at the next stage, after the urge overcomes the rational mind, to overcome it. It’s several stages, so statistically that’s why it becomes less than fifty.

[Rabbi Michael Abraham] You can define it that way; I don’t agree, but it doesn’t matter. Statistics are not fifty-fifty exactly. Whoever thinks free choice means statistics of fifty-fifty—that’s one of the mistakes I prepared in the previous lessons to explain why it’s wrong. That’s what I called the naive libertarian, who thinks we operate in a vacuum, and then it’s fifty-fifty. I claim that the sober libertarian understands that if I have an urge, the chance that I will sin is greater. The libertarian’s claim is not that the chance isn’t greater, but that it doesn’t determine that I will murder; rather, I also have the possibility not to murder. That is what the whole statistics of psychological phenomena are built on. There are statistics. After all, psychological phenomena show a correlation between causes and action. The libertarian doesn’t deny that either. He claims that correlation does not mean determination, only influence.

[Speaker F] One more question: is there a connection to Rabbi Dessler’s point of choice?

[Rabbi Michael Abraham] A certain connection. A certain connection, because I explained this issue then too. When I spoke about it then, I said there are several situations in which even the libertarian would agree that a person did not act out of choice. One is lack of attention. The second is where he had an irresistible urge. And the third is picking. Picking isn’t interesting because then there is no choice. Those are the three situations. Therefore I said: this lesson really sums up everything we’ve done. The three situations. Now, what you’re talking about—what is above the window of choice is an irresistible urge. I have no chance of coping with it, so I’m coerced. What is below the window of choice is a situation where I don’t even have an urge. It’s not a dilemma; I will certainly choose the good because I have no urge at all to do otherwise, I’m above that level. The window of choice says that there is a middle situation in which I am in a dilemma, and what I decide depends on me. So in that sense there is a connection.

[Speaker G] And according to what is that window determined?

[Rabbi Michael Abraham] Your own work, tendencies, genetics, lots of things.

[Speaker G] What I mean to ask is: how can there be an irresistible urge? If I have the reasons to do the thing, and we say that nevertheless I can impose a veto, then according to what is it determined on which thing I can impose a veto and on which thing I can’t impose a veto?

[Rabbi Michael Abraham] So I said: in a place where—let’s take the parable of the topographic outline, I explained this then. Suppose that in the topographic outline there is a very steep mountain in the direction of restraining myself and not hitting back at someone who hit me. Right? That’s the direction, that direction of response. What that basically means, in the parable, is that it depends on how much ability I have to climb the mountain and how steep the mountain is. If the mountain is so steep that even if I decide to climb it I won’t succeed, that is called an irresistible urge. In other words, there are situations in which it is so hard to impose a veto that the person cannot. But that does not mean the picture is deterministic. It means there are situations in which the outcome is forced on me. The determinist claims that all situations are like that. That is an important point, and on that the libertarian does not agree with him.

[Speaker G] The parable is wonderful, but what is the analogue?

[Rabbi Michael Abraham] My urges. When my urges are so strong that in order to overcome them I would need inner strength that I do not have, in such situations I won’t impose a veto. I can’t impose a veto. But that doesn’t mean that in all situations I don’t impose a veto. That is exactly the dispute. The dispute between the libertarian and the determinist is not over whether there are such situations, but over whether all situations are like that.

[Speaker G] I wanted to ask one more thing. Buridan’s donkey—if that same situation were with a person, would he be able to choose what to eat?

[Rabbi Michael Abraham] We’ll get there—that’s the last chapter. I call it Buridan’s man. Anyone else? Okay, then once again, may you be sealed in the Book of Life. I sent the link in the chat for anyone who wants that lecture. Goodbye.