Artificial Intelligence – Lesson 13 – Rabbi Michael Abraham

[Rabbi Michael Abraham] Okay, so I’ll begin — first of all I’ll begin with a short summary of what we did last time. I got a bit of feedback that it would be good to make a little more order there, because maybe it was a bit fast, or unfamiliar to some people. And again, for some of you this is certainly trivial, and then I apologize, but I do want to get us all on the same page so this can be understood better. So look, I’ll do this really briefly, I obviously won’t go over everything, but I will try to summarize it. Look. You can’t see this? One second. Look at these two things because they represent the issue in the best possible way. Look, we have here an electrical circuit. Here there’s a battery, here there’s a lamp, a bulb, and these are two switches. Okay? Now imagine I have two wires going into these switches. The switches are currently open. Now there’s an electrical wire going into this switch and another electrical wire going into this switch. Okay? Now I’m saying: if this electrical wire carries voltage, say five volts, then it closes the switch. If there are zero volts there, the switch stays open. Okay? There are two states: zero volts or five volts. Right? And that’s what will determine whether the switch is closed or open. Now for the sake of discussion we’ll assume that zero volts is zero and five volts is one in our binary language. Okay? So what is this circuit actually doing? This circuit is basically saying the following: when will the bulb light up? When one of the switches is closed, right? If one of them is closed, or both of them are closed, the bulb will light up. The bulb will not light up when both are open, in the state currently shown. Okay? What does that mean? If we now look at the two wires going into the switches, then say both these wires carry no voltage. Zero volts in both wires. Both switches remain open, and the result is a bulb that is off. A bulb that is off, according to the interpretation we gave it, is zero. A bulb that is on is one. So if zero and zero are fed into the two switches, the result is zero. If one goes in here and zero here, then this switch closes and this one stays open, and the bulb lights up. Yes. Yes. Because there’s a circuit here through the lower switch. Therefore if this is one and this is zero, the bulb is one. What happens if this is zero and this is one? Also one, right? The bulb lights up because the circuit is closed from above. What happens when both switches are closed? Meaning, when one goes into both of them? Then the bulb also lights up through both these paths, right? So what does that actually mean? That I have a relation between the two inputs and the output. Say this is the output, and the inputs are the input here and the input here. So if both inputs are zero, the result is zero; if one of the inputs is one, the result is one; if both are one, the result is one. Which is exactly the operation of “or,” right? Or this or this or this — this operation is exactly what’s described by the circuit here. Is that clear? Does anybody not understand? Whoever doesn’t, say so, because if you do understand then fine, but I don’t want to lose you, because this matters to me. These things are extremely basic, but it’s important to me that we understand these basic things before we start talking a little about the implications. Okay, so that’s one circuit. The second circuit — let’s look at this. This is called a parallel connection; you see, this switch is in parallel to this switch, so this is a parallel connection. Here we have a series connection. Why? Because these two are in series, one after the other. Now again imagine two wires going into the two switches; one closes the switch, zero leaves it open. Okay? Let’s now do the calculation. If both switches are closed, what happens to the bulb? It lights up.

[Speaker C] It lights up, right?

[Rabbi Michael Abraham] If one of them is open and one is closed? It goes off. The bulb is off. If both are open? Also off. What operation is that? “And.” The operation “and,” right? “And” — I need both of these to be on in order for the bulb to be on. Meaning only when the input is one and one is the result one. In every other input, the result is zero. Right? Now understand that this circuit is in fact a computer. It’s a very, very simple computer that performs one logical operation. In this case it’s the logical operation of “and,” and in this case it’s the logical operation of “or.” Okay? So basically you can already see. The operation of “or.” How does it happen? If I give a meaning to the voltage entering the circuit, a logical meaning — I say the voltage represents zero or one, yes, zero or five volts represents zero or one. A lit bulb represents an output of one, an unlit bulb represents an output of zero. Okay? And once I’ve given that meaning to the electrical phenomena at work here, I can see this thing as a computer. A computer that performs the calculation of the “or” operation. Okay? So if I now want — I have a compound sentence A or B. Now I want to know when that compound sentence is true. Okay? The answer is: it’s true when either A is true and B is not, or when B is true and A is not, or when both are true. You understand that this circuit represents exactly the logical operation of or. Meaning, we basically have here a computer that performs for us the logical operation of or. Or — that is “or” in English, yes. So this circuit represents in exactly the same way the logical operation of “and.” Because both inputs have to be true for the result to be true. That is exactly the property of the compound sentence “and.” So if I say, for example, “it is both night now and the door here is closed.” Okay? Now I ask: when is that compound sentence true? It is true only when both it is true that the door is closed and it is true that it is now nighttime. Okay? Only when both sentences that make up the compound sentence are true — only then is the compound “and” sentence true. So you understand that this circuit performs that calculation. Meaning, this switch will express whether the door is closed or not. If the door is closed, then the switch is closed; if the door is open, it is open. If it is now night, the switch is closed; if it is not now night, the switch is open. Now you understand that this is basically performing the same calculation we do in our heads when we try to think whether the compound sentence is true or false. You can represent it by the circuit described here. So far is that okay, clear? Now in the next stage we moved to a kind of calculation that isn’t — these are logical calculations. You can do this for all logical operations; these are examples, but you understand the principle. I think it’s enough to understand the principle. Every logical operation can be done this way. Now I want to move to arithmetic operations, meaning calculations, not logical operations. In calculations I want, say, to add numbers, subtract numbers, multiply numbers, and so on. So what we did last time — I tried to show how, by means of circuits, say two circuits of the kind I just showed you, one can build a computer that adds numbers. How do you do that? Let’s remind ourselves for a second. Wait. Here. Okay, how do you do it? So look. We know how we want to add numbers. If this is zero — these two are the input and this is the output. Zero plus zero gives zero and no carry. Zero plus one gives one and no carry. One plus zero gives one and no carry. One plus one is actually two, but in base two that is ten — we talked about this in binary — so that means the result is zero and the carry is one. One moves over to the tens digit. Okay? So that is the table I want to implement. Now how do I build an electrical circuit that will do this addition for me? This is called a half-adder, yes — a partial adder or half-adder. Okay? So look, here it’s basically built like this. I now want to do — these two are the input; you see here input one and input two come in. Okay? These are basically the two digits you see here: zero-zero, zero-one, one-zero, and one-one. They come in here; it’s either zero volts or five volts, which represents zero or one. Okay? Now what does it do? This is what this circuit does, this box. Yes, this is called a logic gate. Okay, so this is an XOR gate — that’s the operation being done. Now I also want the carry. The carry we already know; we don’t even need to call it by name. Right? The carry is this column. Remember what operation that is? For zero-zero it gives zero, for zero-one it gives zero, for one-zero it gives zero, for one-one it gives one. What operation is that? AND. AND, right, “and.” Okay, so this is an AND gate. You see? It’s an “and” gate and it gives me the carry. You understand that what we’ve done here is take the two electrical circuits I showed you before and build out of them a computer that knows how to add one digit. Meaning, one digit plus one digit. Okay? What do we do now in order to add several digits? Basically, now we have to take the two numbers — say I have 101 plus 100. I need to add them digit by digit, right? So 101 plus 100 — we start with the units digit. So one plus zero. One plus zero — we use this to add it. Okay? Then the tens digit: zero plus zero. Again, we feed them into another such unit and add them. But now the carry will also enter, because maybe a carry came to us from before. Then we move to the hundreds digit, and again another such unit that adds the hundreds digit. And this is actually one of the units that does that. Yes? You see here input one, input two, and the carry from the previous addition already come in. Okay? And this basically performs the operation, producing a sum and a carry. Okay? Now take such a unit for every pair of digits, and you can add any two numbers. This is already really a calculator. This is no longer just a primitive computer like the one I described to you here — it’s a calculator. You now know how to build a calculator. That calculator knows how to add any number you want. And more than that: if you know how to add, then you also know how to multiply, right? Because what is six times three? It is adding three to itself six times, right? So in fact, if we have a computer that knows how to do addition, then we also have a computer that knows how to multiply. Now if we manage to do something in reverse, then of course we also have a computer that divides. We take the output and run it backward, turn it into the input. Okay? So basically we know how to divide. I’m obviously doing this in a very schematic and simplistic way, but I want to show you that basically from what you’ve seen up to this point, you now know how to build a computer at the conceptual level. No, you don’t need anything except what we’ve seen up to now. That is the whole genius of this thing. Meaning, we take several very, very simple operations, and by different combinations of them we build the most complex operations there can be. And there are even theorems in mathematics about what kinds of operations can be done with these basic operations, and it turns out to be basically almost every operation. Meaning, Turing’s theorems and the Turing machine — I won’t present that to you now — but there there are actual theorems showing that with these few simple operations you can perform any calculation you want, of any kind whatsoever. Okay? That really already sounds very far-reaching, and I want to use this understanding as an infrastructure, because from here on I won’t describe the details. But it was important to me that you see these basic building blocks — how we get from them to computers. Okay? Wait one second. Now I want to show you a continuation of this, and this time really not in detail, but so that you get the idea. So look — basically now I’ll go on, and again some of you may not know what a derivative and an integral are, and that doesn’t matter to me. Meaning — there is a mathematical operation that is already more complex than addition or subtraction and multiplication and the like, much more complex, and it’s called derivative and integral. It doesn’t matter to me right now. Here there’s a lamp — sorry, a battery — and here these are basically lamps; it’s actually a resistor, but think of it as lamps, okay? Now I say the following: here I put — all this is code language, I don’t care right now if you don’t know it — but I put here a capacitor. A capacitor has the property that its current is the derivative of its voltage. Fine? Proportional to the derivative of the voltage. That’s all. I use that now to create an electrical circuit. I feed in an input voltage here, Vin, input voltage, and I measure the external voltage here — or in other words, I look whether this bulb — this is a bulb — whether this bulb is on or off. Fine? For the sake of discussion. And then the claim is that what I will measure here is the derivative of the voltage from here. Meaning, if for example what enters here — Vin is the voltage that enters — what is the derivative? Whoever knows: here the derivative is constant because this is a straight line, positive. Here the derivative is constant negative. The output will be this. Meaning, if this is the input, here you get its derivative. Now again, it really doesn’t matter even if you don’t know what a derivative is — that’s not important. I’m trying to show you that we take advantage of a physical phenomenon, such a capacitor that in fact has an interesting relation between voltage and current, namely that the current is the derivative of the voltage, okay? And we basically say, okay, if we measure the current, then we know what the derivative is — say, for the voltage here we put in a certain function, say this function, and we want to know its derivative. No problem — look at the output coming out here; that is the derivative of that function. So basically we have here a computer that knows how to differentiate. And if instead of a resistor you put an inductor here, then this computer will do integration, not differentiation — it will do an integral. Because in an inductor the current is the integral of the voltage, not the derivative of the voltage. Okay? But again, as far as I’m concerned this can all be Chinese — it doesn’t matter if you don’t know the concepts. I only want to show you that the basic idea we saw in the things we know — adding digits or doing or and and — everybody knows that. The exact same idea happens in the more complex functions. Meaning, I can simply make a computer that differentiates, that performs integrals — things that many people can’t do on their own, things that are very complicated, can be very complicated. Okay? And there is a computer that will do it easily. I can build that computer and it will do it easily. So in addition, the computer only saves me time, because I too know how to add on my own. But with integrals and derivatives the computer can already do things I don’t know how to do at all. Not only do it faster — it can do things I don’t know how to do at all. Okay? And this…

[Speaker E] But here this is a completely different technology, it’s not a switch opening and closing, it’s…

[Rabbi Michael Abraham] Right, that’s why I jumped there and I’m only showing it on the side. It’s not on the axis I’m going to continue with. But I’m only trying to show… What?

[Speaker C] Rabbi, I have to tell you something — these last two lectures are taking me back forty-five years. Your father taught me this. Your father?

[Rabbi Michael Abraham] Of blessed memory. Okay, yes.

[Speaker C] He taught me, and the first time too…

[Rabbi Michael Abraham] I learned it from him.

[Speaker C] And from him to this day I remember Karnaugh maps and applications and everything, and the derivative, integral, capacitor and inductor — that’s what I remember. At university I went over it again, but it wasn’t like the way he taught it.

[Rabbi Michael Abraham] Okay, good.

[Speaker B] It’s the first electronics course, yes.

[Speaker C] But I’m telling you, already in ninth grade…

[Rabbi Michael Abraham] I studied from the booklets, the booklets for technicians and so on that my father wrote before he got to your school. He was in charge of electricity matters in the Ministry of Energy. So he examined technicians who came from abroad in order to give them licenses and so on, and he also wrote teaching materials for technician courses. So those early materials for digital electronics were booklets he wrote in Hebrew.

[Speaker C] Where did he teach?

[Speaker B] And I…

[Rabbi Michael Abraham] He taught at Basmat for a certain period, but he wrote it without teaching.

[Speaker C] No, I studied with him at the technological institute at Bar-Ilan.

[Rabbi Michael Abraham] Yes, yes, but that was after he had written the booklets. But I read the booklets at home. That’s where I first learned this.

[Speaker C] Blessed be His name.

[Speaker B] Fortunate are you. And everyone learned from that — everyone who studied electronics at the university went through it, they sat over it.

[Speaker C] But this wasn’t forty-five years ahead of its time — he mastered it at levels where we barely understood what he was talking about.

[Rabbi Michael Abraham] In any case,

[Speaker E] So what unites these two kinds of computers into one category? Because it still seems…

[Rabbi Michael Abraham] No, no, I’ll explain — now I’m making the unification. Look, what we’ve seen up to this point is basically not the computer you know. Can’t hear, can’t hear. What? Can’t hear me?

[Speaker E] Before, when the Rabbi spoke, you could hear.

[Rabbi Michael Abraham] Okay, wait, I’ll also put on the spotlight. Okay. What we’ve actually seen up to now is what’s called an analog computer. It seems to me that today they may not even teach this anymore, I think. But once they did. When I was still in university they taught it. An analog computer is basically a computer based on analogy. Meaning, just as in nature the current of a capacitor is the derivative of its voltage, so this forms an analogy to the operation of differentiation. So I perform a physical operation that is an analogy to some mathematical operation. And when I want to carry out the mathematical operation, I simply activate a mechanism that does the same thing and just measure the physical result, and that’s how I obtain the result of the mathematical operation. So if I want to do an addition, for example, then I connect all the circuits as I described earlier, send electrical currents and voltages through them, measure the result — whether the bulb is on or off, say — and in effect that is an analog computer. Because there is an analogy between a lit bulb and one, or between an unlit bulb and zero. The moment I established that analogy, physics is already doing the calculations for me. But understand: physics is not doing any calculations. The calculations exist only after I have established an analogy between what happens on the physical plane and some mathematical operation that interests me. Okay? Physics knows nothing. Physics sends currents and voltages; it knows nothing at all. It knows neither mathematics nor anything else. Remember the water — that’s why I gave the introduction about the water. Remember that water “solves” the Navier-Stokes equations? Right. Now, it doesn’t solve any equations — it just flows. Right? But the water is an analog computer. If someone wants to solve the Navier-Stokes equation under certain boundary conditions, let him construct those boundary conditions, let water flow there, and measure the water flow. The result he measured is the solution to the Navier-Stokes equation. That is an analog computer. An analog computer is a computer that simulates the mathematical operation in the physical world. And if there is something in physics to which I can make an analogy with the mathematical operation, then I can activate the physical mechanism, let the water flow, measure the result, and that result is the result of the mathematical operation. It’s as if I solved the Navier-Stokes equation. If I have a way to measure the speed of the water, then I have in fact performed a calculation that solved the Navier-Stokes equation. That is the idea of an analog computer. Everything we have seen up to this point is basically an analog computer. Because both the electrical circuits of the sum-up and the logical operations, and also the electrical circuits of the derivative, are all just built on analogy. I create an analogy between what happens in the capacitor — the voltage on the capacitor is the function I am differentiating, and the current of the capacitor is the result of the derivative. So if I input a voltage — say I want to differentiate the function y equals x squared — then I feed in a voltage that behaves like x squared, I measure the current that comes out, and the current function is the result of the derivative. So in fact I performed differentiation without doing a mathematical operation. Nature did it for me. Same thing as in the electrical circuits. This is a very important point, which is why it’s important to me that you understand what I’m saying now. This is almost the focus of the whole move. The first electrical circuits I described are also built on analogy.

[Speaker C] But it’s like you found a mechanism in nature that behaves according to your theory.

[Rabbi Michael Abraham] Exactly. And the moment I’m convinced that it really is analogous, then I no longer need to do the mathematical operation; we let nature do the mathematical operation for me. I only measure what comes out there, and there I have the result of the operation. Whether it’s addition or differentiation or multiplication — it doesn’t matter. Whatever you like, so long as you succeed in building a circuit that does that operation. Now, many times we have an equation, say, and I want to solve it using an analog computer. There are mathematical and engineering methods by which I can build circuits that will solve an equation for me. Say I want to solve what’s called a differential equation. A certain function equals eight times its derivative plus its derivative squared minus twice the function. I can build an electrical circuit that represents that, and it will solve the equation for me. Everywhere there is a derivative I’ll put a capacitor; for a second derivative I’ll differentiate the result of the first derivative, meaning I’ll put another capacitor; and thus I basically build a computer that can solve any equation I want. There are methods for building these circuits for any equation you like. Nobody uses this today because it’s an analog computer, so it’s no longer of interest. We’ll get to what people do use today, but it is important to me that you understand the principle, because that principle is much easier to understand in an analog computer than in a digital computer. Okay? So here the understanding is very important to me. Up to this point, is this idea of an analog computer clear? I simply find a natural process whose mathematical description is exactly the problem I want to solve, I run that natural thing, and I measure what comes out. What comes out is basically the result of the mathematical operation. That is the idea of an analog computer. Now of course, in order to build analog computers you need an entire theory — how to build a computer for a given problem. I have a given problem I want to solve; let’s see how to build a physical system that behaves exactly like that given problem. And there are whole theories about how to build that. In control theory — whoever studies introduction to control or something like that in engineering — that is what they learn there. Okay? More or less what they studied when I studied, I don’t know, maybe today they no longer study analog matters. In any case, that is an analog computer. Now I want to show you an amazing analog computer — but one you can see with your eyes, no calculations, nothing will frighten you, don’t worry, everything will be fine. It’s just — once I saw this thing, it was unforgettable. Wait.

[Speaker C] Something with water probably, liquids or something.

[Rabbi Michael Abraham] No, no, not with water. Wait. Now look, I won’t show the whole video, only some images from it, because it’s twenty minutes. But I’ll send you the link if you want; it’s also in one of my columns. Wait. This phenomenon is simply unbelievable. By the way, yes.

[Speaker B] But actually one-zero and binary thinking are characteristics of a digital computer; an analog computer is more about differential equations. Wait, wait, it’s coming. Okay.

[Rabbi Michael Abraham] Can you see? Yes. This monster that you see is basically a kind of analog computer. It’s a computer that simulates the walking of animals. But understand — this is basically an analog computer. I build animal walking using, notice, purely mechanical means. There is no electricity here, nothing at all.

[Speaker I] Here he teaches a bit of the mechanics of the matter — how he builds the joints and how the whole thing works. This thing is unbelievable. Wait.

[Speaker G] You see here this is the basic blueprint, yes, how to produce…

[Speaker C] Walking — how to produce a joint that walks.

[Speaker I] It’s a mathematical method; there aren’t many structures that succeed in doing this.

[Speaker G] It’s a structure that has to satisfy…

[Speaker C] Very specific properties.

[Rabbi Michael Abraham] It’s real mechanical genius. Wait.

[Speaker C] Wow.

[Speaker H] But Rabbi, how does it know when to, as it were — how does it know to push? I mean, how does it start moving? From the wind.

[Rabbi Michael Abraham] From the wind, it has sails. With the wind — it has sails.

[Speaker B] Yes, that’s precisely why he put it there.

[Speaker I] Then it’s much easier for it to stay on top of the sand. But they also fulfill a less obvious function. You’ve got these wires here and what this does together with the larger surface area is it provides a lot more contact in the long run. But the beast also faces more violent threats. Here, look, come see. Like the storm may be better cut by the other stump bases. Look how it looks here. Walking a Strandbeest up a Dutch beach. What an epic job. Here he built the — it’s simply insane.

[Speaker K] Strandbeests are in a never-ending fight with the elements.

[Rabbi Michael Abraham] Here, here. We visit in currents of the wind or water. Here, minute fifteen. So now that you’ve seen the Strandbeest crawl slowly…

[Speaker E] But now there’s one more thing. In the wrong direction because they are being carried by the wind, it actually senses…

[Rabbi Michael Abraham] When it’s touching the water and then hopefully… Now can you hear me? Yes, yes. Now I’ll show you just the beginning. He builds a mechanical neuron. Meaning, there is no electricity here at all — it’s all mechanical, with bamboo, nothing. Now he builds a neuron. What does that mean?

[Speaker C] When this thing encounters the wind it…

[Rabbi Michael Abraham] Turns right, or when it touches water it moves away from it — it turns right. Which is basically a kind of neuron, or a computer that receives input — there is water here — and gives a mechanical instruction to this whole monster: turn right. Now it has many such neurons, and it really simulates a brain — all mechanical, there’s nothing electrical here. With a connection of brain cells and muscles, it can then course-correct to go away from the water back onto the beach. And of course it’s important to know that you’re close to the sea, right? This is why I have a water feeler here. Here is the neuron. It simply fills with water and then moves that thing. About that great. Can you hold? It goes over the ground about this high and as soon as it comes into the sea, it swallows the water and feels the resistance of the water. It still doesn’t work. Well, I must regret. Oh now it’s working. So happy. It really simulates a human brain encountering water. When you touch water, here you see — it’s like a brain, here is the neuron. After you’ve seen the water you should be able to use your muscles. So what you need is a nervous system. But if I push in this piston here, it’s blocked. So in fact what you see here is only a crancher, only a valve. Open, close, right? Now I’m going to operate that crancher with air. So if I blow air in here, this closes and no air comes out of here. If you see this as the output and this the input, the output is opposite from the input. Here he simply closes the passage and then the wind flows left. In other words, you can switch zeros and ones just like in a computer. Really just like in a computer. Truly brilliant, simply brilliant. When it touches the water the neuron jumps.

[Speaker D] This is a part of the Strandbeest. Do you have an idea how many brain cells you’d need to be able to make that work?

[Rabbi Michael Abraham] Able to make that work?

[Speaker D] Over time, the beasts also slowly degrade.

[Rabbi Michael Abraham] They lose their color and parts…

[Speaker D] Might break off.

[Rabbi Michael Abraham] On the scale of the global microplastic pollution and also the Netherlands from the rising… and the life. So you might ask, is it still really about protecting the Netherlands? And the answer is no, not really. It’s about a much more human need that most of us have — the desire to be remembered after we’re gone. I mean, it’s ominous to imagine that we’ll die and be forgotten someday, so a lot of us go to great lengths to avoid this by having kids, donating large sums of money…

[Speaker B] To get a building named after you, or writing a book.

[Rabbi Michael Abraham] These sorts of ideas. For Jansen it’s making Strandbeests. And when they fulfill their final goal of living independently on the beach, then he wrote, I can die with peace of mind. Okay, that’s enough. I’ll send you the link to watch the video — it’s simply unbelievable, brilliant, just brilliant. Now, this is supposedly what God did with us — we are basically the analog computers of the divine formulas. Exactly. But that’s the question: are we analog computers? Put in an electrical neuron — meaning something electrical that when it touches water, something jumps, receives a signal, or something like that, and it changes direction. All these things, of course, can be…

[Speaker B] Done electrically…

[Rabbi Michael Abraham] Or…

[Speaker C] Mechanically, or whatever. What do they have in common?

[Rabbi Michael Abraham] That the mechanical mechanism and the electrical mechanism are both physical mechanisms, and they simply simulate the mathematical phenomenon we want to represent. Therefore it is simply an analog computer. Whether you implement it mechanically or electrically really doesn’t matter, so long as the analogy is preserved. For example, in our introductory control course in engineering, he showed us that every electrical circuit can also be made into a parallel mechanical circuit. Instead of an inductor there is a spring; instead of a capacitor there is — I no longer even remember what — and you set up such a circuit where here there’s a spring and here there are weights and springs and things like that, and it solves the same differential equations solved by the electrical circuit. It’s the same thing, because the properties are analogous properties. You can make a full analogy between the mechanical circuit and the electrical circuit. And the whole idea is the analogy. Meaning, we are using here a physical mechanism — it can be mechanical or electrical but physical — that is analogous to a mathematical exercise or some sort of computation we want to perform. That’s all. And we use that analogy to solve the exercise we want to solve. The analogy still isn’t very clear to me — I mean, I don’t see the difference between this and, say, a ball rolling. What did I understand? I’m having trouble understanding the analogy between this and computation and the fact that it’s… A rolling ball is an analog computer. Every physical phenomenon is an analog computer. But when does it become an interesting analog computer? When you succeed in building a physical phenomenon that is analogous to the mathematical problem that interests you. But every physical phenomenon is analogous to some mathematical problem or other — it’s just not always the problem that interests me. When I use a computer, understand — physics is the means, not the goal. Usually as physicists we use mathematics to solve problems in physics. An analog computer does the opposite: it uses physics to solve mathematical problems. Because it’s just the same thing in reverse. Right, exactly. The question is what is the given and what is the result, what is the output and what is the input. So physicists take physics and try to describe it using mathematics. Once they found the description, one can use their description to do the opposite operation: to use the physical process to solve mathematical problems. Once physicists showed me that a capacitor differentiates, then now every derivative I have, I’ll use a capacitor to perform the calculation that interests me. You know, it’s like — do you know the difference, the connection between a motor and a generator? They both create energy. What? No, in one the magnetic field becomes… In a generator you put in mechanical force and what comes out is electricity. The generator produces electricity — how? By moving things, right? Mechanical motion produces electricity. In a motor, electricity produces the mechanical motion. Right? It’s simply the opposite direction. A generator and a motor are the same thing — only the output is the input and the input is the output, that’s all. And that’s the whole idea of an analog computer. The idea of an analog computer is that we basically know the forward direction — we know how to describe physics with mathematics, that’s the physicists’ job. The engineers do the reverse path: they build physical systems according to the data the physicists give them. They build physical systems that solve mathematical problems for them — or computers, basically. That’s what a computer does — an analog computer for now, by the way. Okay? That’s the basic idea. Now, up to this point we’ve seen an analog computer. Now let’s try to see how we move to the computers familiar to us. So what happens next is the following phenomenon. When we want to produce a more complex operation — say, adding numbers of any size — you understand that we need to make many, many circuits of the type we saw before, right? With lots and lots of lamps and switches and batteries and so on. For every digit you have to make several such circuits; to do, I don’t know, thousands of circuits, tens of thousands, millions of circuits, in order to add large numbers as we wish. Okay? Now what will happen if we also want multiplication, not only addition? Fine, then we need to add many more circuits, right? Because multiplication is repeated addition. And if we want other operations too, many more circuits. This already becomes impossible. And if we want a computer that won’t perform one simple operation but a computer that will be multitasking, where you can do many operations on it, we can no longer do it in this way. So what do we do? We build the computer, we build a processing unit or computational unit that contains the basic logical operations. And now we operate it via software. An analog computer is a computer that has only hardware. There is no such thing as software for an analog computer. In an analog computer, the physical system performs the calculation. There is no software here. You built the computer for one specific purpose; this computer solves one specific equation and that’s it. It doesn’t know how to do anything else. An analog computer performs one operation, that’s all, no more. Now if you want that same golem, yes? that same unit to perform many different types of operations — not in parallel, simply to be able to solve many kinds of problems — then you need to move to a different way of thinking, and that is basically called a digital computer. What does that mean? That we take the basic units, the circuits I described before — of course we implement them using electronics; they’re not electrical circuits exactly, but they’ll be small and efficient — but the principle is this. We build, say, gates that create AND, OR, XOR, these basic gates; out of them we build units that know how to do, I don’t know, addition of two digits, multiplication of something, basic things, and from there on what runs the show is the software. Basically now I need to add an input-output unit, which an analog computer doesn’t have. I need to add an input unit, which basically says the following: it takes the — say I, the programmer, take — say I want to solve the problem two plus three equals what, okay? using a computer. So I tell it: I need to write a program that tells the computer which parts to activate each time according to the problem I want to solve. So say I want to solve two plus three. I tell the computer, look, first represent two in binary form. Then represent three in binary form. Now take two and three, feed them into the XOR gate to perform the addition, feed them into the AND gate to create the carry, then feed the result into — but I’m not building all this with wires; I’m giving it as instructions. Here I move from hardware to software. Meaning, the gates are inside the computer, but the order in which they operate, and who operates and who goes where and who passes information to whom and when and what one does with that information — all that is software instructions. That is the whole leap of the digital computer. The leap of the digital computer basically says: take these basic logical units, many units so there’s something to work with, but now you can solve many problems with them. Every piece of software, every program, is meant to solve a particular problem. A different program will solve a different problem, but on the same computer.

[Speaker E] It simply takes the units that are in this computer and activates them in different ways.

[Rabbi Michael Abraham] First activate this unit, feed the result into that unit, then move it to this unit, multiply these two by means of that unit, and give me the result. That’s one piece of software. Another piece of software takes those same units and says no — first take this unit, not that one; take its result and feed it into this unit; then add these two; then AND this with that result, XOR with the other result, and give me the output. And you understand that there are infinitely many possible problems such that, if I know how to build those problems on the basis of the basic operations the units in the computer know how to do — addition, multiplication, or the basic things. Okay? But what is software? What does it mean that I tell the computer something? The software goes through the input unit. I enter — I type some instruction into it, say. Okay? Once it used to be with cards, but now we type. You type some instruction, and that typing is basically electrical currents. These electrical currents are translated into instructions to the processing unit — what to do. And now the unit receives instructions: now activate this unit, take this result, feed it into that unit, multiply that result by the result of this unit, sum these two, OR with what’s happening here, and output the result. And that is the output. In the output unit, by the way, there is an opposite translation. After a result comes out — say the bulb is on, okay? Say that is the output. The output is one, the result is one. We see the digit one written on the screen, right? We don’t see a lit bulb. We see the digit one written. How does that happen? There is simply an output unit that performs the reverse operation of the input unit. It takes a lit bulb and turns it into the number one lighting up on the screen. Simply in an electrical, electronic way, okay? But that is only to represent for me that the bulb inside is on. But for me, I want to see that the result came out one. So they draw for me on the screen the shape of a one. That’s all — when the bulb is on. And if there is lit-lit-off-lit, then they draw 1101 for me. That’s all — they translate it into shapes on the screen. That is basically the screen. The screen is built — at least it used to be, I don’t know how it is today — but once it was built from electron beams, beams of electrons shot onto a fluorescent screen that lights up when the electrons strike it, and basically when you want it to write a one for you, you fire the electrons in such a way that it lights exactly those dots in the shape of a one. That’s all. And all this is caused by the fact that inside there is a lit bulb. There is a device that says: the moment the bulb is on, fire the electron beams such that on the screen there will be formed the result one, the image one. That’s all. But what happens inside the device is what I described to you. But the software basically passes through the input unit, which translates the programmer’s instructions, yes? It goes through a high-level language, then an internal language, it doesn’t matter, I’m no longer up to date with what happens today either, but once it was like that and it doesn’t matter — it goes through all kinds of very complex translation processes. But the schema is this schema. You give instructions in your language; those instructions are translated into electrical signals; those electrical signals determine a sequence of operations inside the computer — what it does. The result is lit-off-off-lit bulbs; on the screen you get 1001 lighting up. That is the result of the addition operation. Rabbi? Yes?

[Speaker B] When you speak about the input unit, is that what you call bits? Why?

[Rabbi Michael Abraham] Bits. Bits are not a unit of anything; bits are a notion, a concept.

[Speaker B] No, because today every bit is a sort of memory cell built inside and storing our commands. So is that what you’re saying now?

[Rabbi Michael Abraham] The memory represents the bit. Right. We give these things meaning. What happens in the computer is electrical currents and voltages, that’s it. Nothing besides that. We give it the meaning. In the end, even with the software and everything, in the very end the principle is what I showed you at the beginning. It is no different in principle. It’s much more complex and you can do many tasks with it on the same computer. Yes? Think about the computer that that fellow built there — it only knows how to do the one operation for which it was built. Now if I suddenly want it to start flying, I need to equip it with another system, right? But if it had a software mechanism, if it were a digital computer and not an analog computer, then I would tell it: activate the joints in such a way that it flies, and another instruction would tell it: activate the joints in such a way that it runs, or turns right, or turns left. And then with the same device itself — mechanical in that case — with different instructions, it would perform different actions. But not yet… maybe only there…

[Speaker E] But at the base there is still some initial analog unit that is nonetheless physical?

[Rabbi Michael Abraham] The basic units, what are called the gates — yes, you can call them analog computers. They are basically electrical circuits like I drew for you here in… like we saw at the beginning: series connection, parallel connection. It’s implemented electronically, not important, but that’s what is there inside. Those are the basic units. Everything else is just combinations of them — many, many such units and translation processes, that’s all. That’s why I say it was very important that you understand the beginning, because all the rest is just combinations, but the idea remains the same idea. And therefore, for example, if I now ask you whether this thing thinks — now we already have a digital computer with software and everything, not artificial intelligence, we’re still in the previous generation of computers. We haven’t yet reached artificial intelligence, okay? Now I ask you: can one say that this thing thinks? “No.”

[Speaker E] “In my opinion it doesn’t think.”

[Rabbi Michael Abraham] It seems to me pretty clear that it doesn’t, right? And the fact that it is much more complex than an electrical circuit with a parallel connection doesn’t matter. It has many such circuits — so what? But it’s still electrical circuits that light bulbs by means of voltage, that’s all. I built them in such a way that they are analogous to the mathematical problem I want to solve. That’s all. No one would dream of saying that this thing thinks. Right? Now even if this thing reaches a level of sophistication where it passes the Turing test — and that’s the important point. Suppose this thing passes the Turing test. Remember the Turing test? We talked about it. I have two screens and I am talking with two entities that communicate with me through these screens, each through a different screen. One of them is a computer, one of them is a human being. And I converse with them and they answer me. And I have to know which is the computer and which is the human. So Turing says: the moment I am unable to distinguish which is the computer and which is the person, that is the stage at which the computer is already a person. That is the Turing test. Now today we already know computers — not today, twenty years ago, ten years ago — we already know computers that pass the Turing test. Computers like the ones I described to you here. Still not anything with artificial intelligence. Ordinary computers, completely deterministic computers, okay? They pass the Turing test. It seems to me that today, once we’ve built these computers, it is completely clear that there is no basis for attributing thought to them. Just as I didn’t attribute thought to the electrical circuit or to the water. There is no difference between a computer and water. No difference, except that the computer…

[Speaker L] “Wait, does a quantum computer change anything in this respect?” “I’m asking whether a quantum computer changes anything in this respect?”

[Rabbi Michael Abraham] Not really — a quantum computer, all the physics you’re using is quantum physics, so what? The idea is exactly the same. It’s still physics.

[Speaker B] “Except that today there’s a computer that talks with the computer.” So what? “No, I’m just saying, we’ve reached even that.”

[Rabbi Michael Abraham] Okay. But again, I’m not talking right now about artificial intelligence. I’m deliberately building this step by step. We’ll get to artificial intelligence and we’ll see that it poses an additional challenge. Right now I’m talking about an ordinary deterministic computer, like what I’ve described to you until now. Basically, from what I’ve described to you until now, you know what a computer is. Not the details, but you completely know what a computer is. And now when I ask you — say, like Turing’s question — you are sitting here in front of two screens, you are talking with such a creature and with a human being, and you cannot tell the difference. It has reached a sophisticated enough level that you cannot tell. Would you be willing to say that this thing is already a human being? You understand that once we understand well what this thing is, such an initial assumption doesn’t even arise. Clearly not. It is simply a collection of bulbs and switches. What difference does it make if it’s very, very many bulbs and switches? So what? Why should that matter? Of course it matters in one sense — it’s very interesting — but not on this philosophical level of whether it is a person.

[Speaker B] Okay? “Wait, doesn’t this come with your prior assumptions about human nature?” “Can’t hear.” “What you’re saying — I agree with it — but it seems like you’re adding…

[Speaker D] Some further assumptions about human nature. That is, if you are a complete materialist and you see the human being as a creature of…”

[Speaker J] “That’s why I gave the introduction.

[Rabbi Michael Abraham] We already went through about twelve lectures on the way. Exactly for that reason I gave the introduction. After we did the introduction, you can agree or disagree, but there I explained exactly the assumptions I’m making. Now take everything we discussed in the introduction, apply it to what you know as the digital computer, and tell me whether there is any initial assumption at all to say that this thing really thinks like a human being.”

[Speaker C] “Yes, but Rabbi, I can also think that this human being I’m speaking with, who really is a person just as it seems to me he is a person — maybe he’s not either… I think only I think; maybe this other person isn’t a person either?”

[Rabbi Michael Abraham] “We’re going back… we’re going back…” So we’re going back again to all the introductions I gave — I talked about all this. Exactly for that reason I gave all those introductions. Because if this were the issue right now, then in the framework of this argument I’d have to go back and give ten lectures. And that’s why I say no — first I gave those ten lectures, and now let’s use everything we saw there for this problem of what this computer is. Okay? I don’t want to go back there now because that would just reopen all those matters. We already covered them.

[Speaker D] Amiti, a question about what you explained now regarding a mechanical computer — can it have software? Because it sounded like you were saying that the only software possible for a mechanical computer is the laws of physics.

[Rabbi Michael Abraham] No. In theory, in theory it could have software. You would have to find a translation system, an interface, that takes a program you write and turns it into mechanical instructions, and of course you would have to build basic mechanical units whose different combinations perform many operations. At the conceptual level it’s not absurd.

[Speaker D] Okay, that confused me earlier because it sounded to me like you…

[Rabbi Michael Abraham] No, no, I wasn’t making a principled claim. Practically it’s impossible. But that’s not a principled claim. In principle, yes — in principle, everything you can do electrically you can do mechanically. It’s just that the computer — I don’t know what — a mechanical calculator would probably take up the size of the Milky Way galaxy. Meaning, if you want to make a mechanical calculator. But in principle it’s possible.

[Speaker E] What exactly was Turing’s initial assumption then? He didn’t…

[Rabbi Michael Abraham] No, so now this is an important point. Look, I want to go back for a moment to the Turing test because it’s a very important lesson. And by the way it matters in many contexts, not only here. Turing did not imagine what a computer that could pass his test would look like. He was dealing with an entirely hypothetical situation. When you deal with a situation that is entirely hypothetical, you can raise all sorts of speculations one way or another. Presumably you assume that if we build something so sophisticated that it can pass the Turing test, then it probably will also have a soul and will be a human being — I don’t know what. Or without a soul — it will be a human being. But today we already know. We have built this thing, we already talk with it. It is before us. We know it. This is no longer a hypothetical situation. In such a situation, what seemed obvious to Turing seems obviously false to us. And that is a very interesting point because it has implications, by the way, for many matters of halakhic ruling as well. I also wrote about this a lot, and in general in various discussions, because the question is how much a discussion of a situation that for you is totally hypothetical — you’ve never experienced it directly — how much in such a discussion can you really hit the truth? And you can be a genius like Turing — and he was a first-rate genius, no one disputes that — but he lived, he did not live the situation; he never encountered such a device. He didn’t know how a device capable of doing such madness would be built. Therefore he says: fine, if it meets the tests, then it is probably a person. But today we see that by the very technique Turing himself helped establish, one can continue in a more sophisticated way and with greater miniaturization and all the new technologies — we can, with his own technology, build the very hypothetical device he was talking about. And then suddenly it turns out that his criterion is worth nothing. That his test is simply nonsense. Now this is a very important lesson. I’m deliberately spending a few minutes on it because it’s an important lesson not only for this series but in general. This is a wonderful example of how a genius reaches a conclusion, and when you encounter the situation, you understand that he was talking nonsense. Yes, but we’re talking about artificial intelligence now. Notice — I’m talking only about the ordinary digital computer. Not artificial intelligence.

[Speaker E] So isn’t this really a classic proof of the importance of particularity, the question of particularism of cases? Because really, to say the law of the stubborn and rebellious son, as we talked about last week, or when you don’t live the…

[Rabbi Michael Abraham] I’m going to get into that now for a few minutes and see to what extent yes,

[Speaker C] But Rabbi, this test of his wasn’t some law of physics like Einstein writing formulas; he only proposed a testing parameter. I don’t think he attached that much importance to it.

[Rabbi Michael Abraham] Correct, and that testing parameter is nonsense.

[Speaker C] Fine, so he was wrong about that, who cares.

[Rabbi Michael Abraham] It matters,

[Speaker E] Why doesn’t it matter?

[Speaker C] No, he didn’t determine that this is how… no, but I don’t think…

[Rabbi Michael Abraham] He didn’t write that E equals MC squared and that was nonsense.

[Speaker C] No, I’m not…

[Rabbi Michael Abraham] I didn’t come to criticize Turing and consign his name to disgrace forever. I’m saying that if a genius made a mistake, that ought to teach us something. If an idiot made a mistake, that teaches us nothing. But if a genius made a mistake, that does teach us something. Something very important.

[Speaker C] That he’s human.

[Rabbi Michael Abraham] No, not that he’s human. We’re human too. The question is whether we are doing the same things that led him — what led him — into error. And that’s the important lesson here. And that is what I want to learn here. Think, for example, about Maimonides when he writes the Laws of the Foundations of the Torah. He uses some strange Aristotelian physics there, with separate intellects and all sorts of nonsense like spheres and I don’t know what, all kinds of things like that. That’s all nonsense of course, but a genius like Maimonides uses it, so one has to understand what message I can still derive from that even though the content itself can be thrown in the trash. But there is some message here. For example, in that case, the message is how to relate to the scientific knowledge of your time. Is it relevant to use it in order to formulate a worldview? Maimonides taught us that yes. The scientific knowledge in itself, and also the worldview built on it, may not be worth anything. But the very fact that Maimonides did something here teaches us something, even if he was wrong. Now here I’ll say — I’m returning to Jewish law. Look. When I discuss a halakhic question where I have not the faintest idea how it is perceived by a person who is inside that situation, in my opinion I am forbidden to issue a halakhic ruling about it. I wrote about this in articles discussing halakhic ruling in the Holocaust. I gave there an example about monetary law — Rabbi Gibrelter, a Jew who was in the Kovno ghetto and issued rulings in monetary law, all sorts of very strange rulings, and some rabbi, some judge in monetary law, some Torah scholar, criticized him and wrote a critique: obviously he’s wrong, and so on. And it outraged me. Why? Because when you live the situation Rabbi Gibrelter lived inside the ghetto, you understand that he was right. And when you sit here with your theories and know all the books and have a lot of knowledge, you are talking nonsense because you do not understand what that situation means.

[Speaker C] Let me…

[Rabbi Michael Abraham] I’ll give you less extreme examples. If you go ask a conservative, Haredi rabbi whether it is permissible to go to a performance by a woman singer — yes? a female singer. Okay? Now presumably he’s never been there, because he thinks it is forbidden. So he hasn’t been there. He doesn’t understand the situation at all. He doesn’t understand why people go there, he doesn’t understand what people experience when they are there, and therefore he cannot issue a ruling. He’ll tell you it’s forbidden, but he’s talking nonsense. Not because it’s permitted — maybe it really is forbidden — but he cannot say that it’s forbidden because he doesn’t know the situation. He is completely convinced that whoever goes there goes only to arouse all kinds of sexual fantasies in himself. He does not know the phenomenon that you go to hear a female singer because she simply sings beautifully, and it is very beautiful art, and I enjoy the art in the place.

[Speaker D] Doesn’t that make him more objective?

[Rabbi Michael Abraham] If he doesn’t have that desire? I’ll say — I’ll get to that other side in a second. So that’s one side. Therefore the same thing also came up once when I talked about the binding of Isaac, yes? There are all kinds of critics — both our commentators and various philosophers — who criticize Abraham: how could he accept the divine command there in the binding? Against morality, against reason — what do you mean? Maybe it was just some misleading demon and not a divine command. How could he allow himself to go and murder his son? Now what I keep thinking about in that matter is: I was never a prophet. And I think that if I were a prophet, it might be that when I hear the voice of God speaking to me, it is completely obvious to me that it is Him. Not some misleading demon, nothing — I simply know. Now whoever never experienced that is right: from his perspective, who knows, maybe it’s just a delusion, maybe it’s just some hallucination. But for the one to whom it happens, it is like explaining to a blind person that you see something. Now he has never seen in his life; he doesn’t know what it means to see. And he says, ah, but who told you that sight is reliable? Maybe it isn’t. But look — if you had experienced it, you would understand that sight is something reliable. I have no way to convey that to you; you need to experience the thing directly. And so too regarding halakhic ruling. That is one side. And on the other side, someone here correctly remarked earlier that there is value in keeping a distance from the situation. There is value in keeping a distance from the situation because your direct involvement in the situation can lead you to mistakes, biases, errors. Therefore, for example, I am very much in favor of the Haredi model of the Haredi parties, in which there is a Council of Torah Sages and there are Knesset members. The Council of Torah Sages makes the principled decisions — they don’t specifically have to be rabbis, but they should be wise people with values we trust — who are not mired in the mud, in interests and intrigues, but rather look at things from some broader value-oriented perspective, and they should give the principled directives. Let the Knesset members handle the implementation. I am very much in favor of that model. I only think that the Council of Torah Sages should include people who understand the world they are talking about — which is not what happens. Because they understand nothing. So you need familiarity. The situation should not be far from your world. But in the specific situation you are judging — there, do not be involved; be distant. But understand well what such a situation means. And that is the point. So there are two sides here that seem a bit contradictory, but they are not contradictory. But these are lessons I think are true for philosophical questions, true for halakhic rulings, true for many things. There is value in being involved and there is value in being detached. And I think the most correct combination is to be involved in the intellectual sense — that is, to know such situations, to experience such situations yourself — and then to step far from the specific situation you are dealing with, to be far from it, not involved there, and determine a position regarding it. I think that is the ultimate combination, as far as one can formulate any sweeping rule. And I think the example of Turing is an excellent example. Because Turing was dealing with a situation that from his point of view was completely hypothetical. Therefore he took a position, and this test accompanied humanity for many years. To this day many people are captive to that test. But today many people already understand that the test is simply not relevant. It is not relevant because it is completely clear that this collection of switches and bulbs, even if it is very, very complex and reaches achievements Turing never dreamed of, is still just a collection of electrical wires and bulbs. That’s all.

[Speaker B] But that can also happen with us. Seventy years have passed since Turing died — seventy years — and another seventy years from now somebody could say, guys, you were talking nonsense, look what… we don’t know what the development will be.

[Rabbi Michael Abraham] I don’t know either. Therefore obviously one should take what we’re doing now in that same perspective. Right, I completely agree. But still, you know, a judge has only what his eyes can see. We can discuss what we can discuss and learn lessons from history. We cannot learn lessons from the future because it hasn’t yet arrived. But we can learn lessons from history. It’s like someone who discusses the question of how a four-dimensional world behaves. What do you do in order to discuss that? You look at how a two-dimensional world behaves and ask yourself how a two-dimensional creature perceives your world, which is one dimension higher. Then you try to use that to say how I, as a three-dimensional creature, can understand what goes on in a four-dimensional world. It’s like learning from the past in order to try not to err about the future. There is no guarantee we’ll succeed, but it’s the best we can do. Okay?

[Speaker M] Rabbi, I have a question. Yes. Is there any practical implication in this series? Is there some practical implication from the fact that in the end, I assume the Rabbi’s view is that a person is essentially different in his thinking from AI, no matter how smart it becomes? So is there any practical implication besides saying, okay, we’re different?

[Rabbi Michael Abraham] We’ll get there, we’ll get to the practical implications later. Okay. Okay, up to here. Any comments or questions?

[Speaker D] I wanted to ask a question that isn’t so related to the topic, about your moral conception.

[Rabbi Michael Abraham] Okay.

[Speaker D] Regarding how you understand Jewish law in relation to morality. And at the same time, if I understood correctly, your conception is Kantian when it comes to morality. The opposite.

[Rabbi Michael Abraham] Kantian both in relation to morality and in relation to Jewish law.

[Speaker C] What does it mean, Kantian in relation to Jewish law?

[Rabbi Michael Abraham] That in Jewish law too, the halakhic value of an act exists only if you do it in order to fulfill your obligation and carry out the command.

[Speaker D] Okay, I understand. Regarding the halakhic claims themselves, I didn’t manage to understand what gives them their value, their objective value. So what gives them that value? I didn’t understand. Say when it comes to moral matters, Kant would say it derives from the universality of reason. That when I act as a rational creature, I cannot will the — I cannot, when I act in a non-universal way, I act against my own rational nature.

[Rabbi Michael Abraham] Kant does not offer an explanation for the value of a moral act. He assumes there is value to a moral act and tries to uncover what that can mean, assuming there is value.

[Speaker D] No, I’m saying…

[Rabbi Michael Abraham] It’s the opposite move.

[Speaker D] I probably spoke incorrectly; I meant what gives validity to the command itself. How do you get the categorical imperative? Kant has no explanation for that. Why? He brings it in the Groundwork of the Metaphysics… Right, there is no explanation there that…

[Rabbi Michael Abraham] I think specifically he brings… Kant goes here in the opposite direction. Kant assumes there is some validity and asks himself: how can that be? What does that mean? There must be some command, and that command is categorical, and its content is: act only according to that maxim you could will to become universal law, otherwise it cannot have validity. But why does it have validity? There is no explanation.

[Speaker D] That’s only in the second chapter, but in the third chapter he speaks… No… it doesn’t really matter for my question. The point is that Kant does give a justification for the categorical imperative. Do you at least agree with me on that? That he speaks specifically about it as deriving from the universality of reason.

[Rabbi Michael Abraham] What is “a justification”? He assumes that anything with validity must be universal, and therefore he says: if it has validity, it must be universal. That is exactly the point. He goes in the reverse direction from the one you’re describing. He starts from there being some validity, and he assumes that only something universal can have validity; therefore it…

[Speaker D] Must be universal.

[Rabbi Michael Abraham] He always goes backward from the validity.

[Speaker D] An anthropological move and not a philosophical one. Maybe I misunderstood him, but then how do you understand the third chapter where he speaks of man as a free being, and when you act according to your freedom and according to your universality, the only way to act is according to the categorical imperative? So then what do you see there? He says that when a person acts, he must act according to the idea of freedom. And when he acts according to that idea, he acts… and where does that come from? He brings there the argument according to which action itself is an act of choice.

[Rabbi Michael Abraham] Fine, but again I’m saying: clearly he assumes that this thing is valid, and then uncovers all the assumptions it must satisfy in order to be valid. His whole move is like in the Critique of Pure Reason. His move in morality is similar to his move in epistemology. He goes backward. His considerations are: I assume there are synthetic a priori judgments, and now I ask myself how can that be? I assume there is valid morality, and now I ask myself what it is based on, how can that be? That is never an explanation of why it is valid.

[Speaker D] I strongly disagree because I understood him completely differently from you. I actually understood the aesthetic as an example that, look, there are synthetic a priori judgments like space. That he proves it as an example, yes? But he proves it — isn’t that right?

[Rabbi Michael Abraham] No, he doesn’t prove it, he assumes it. Why?

[Speaker D] Actually here I remember the details very precisely, that at the beginning of the aesthetic…

[Rabbi Michael Abraham] I don’t remember the details precisely, but if you want we can discuss some passage you have in mind. But the move, in my opinion, is that kind of move. I call it a theological move and not a philosophical move. Meaning, a move that goes from the conclusions backward to the premises, and not from the premises to the conclusions.

[Speaker D] I can give the example he brings about space. He says, suppose space cannot be something we learn a posteriori, since the very perception — yes? the very experience of perceiving one object in one place and another object in another place and myself in another place already requires the concept of space, and therefore it is given a priori, and of course it is synthetic because it structures experience. But who told you this datum exists? Experience. Maybe we’re just confused?

[Rabbi Michael Abraham] Kant here is already subjective… No, but that’s exactly the point. He’s not…

[Speaker D] Asking that — this is the ABC of the Critique of Pure Reason.

[Rabbi Michael Abraham] Kant does not ask whether synthetic a priori judgments are possible; he asks how they are possible. Of course they are possible — the only question is how. Of course they are possible because there are such things. He proves they are possible because he gives an example of one. No, but the fact that there are such judgments is not what’s interesting. Of course there are such judgments, no one argues that there are such judgments. Hume didn’t argue that there were such judgments either. Synthetic a priori — he did argue. No, he didn’t. He only claimed that such judgments are delusions.

[Speaker D] But one can formulate synthetic a priori judgments in language. Okay, so I’ll ask my question about obligation without the issue of… because it seems this will take us into a whirlpool. My question specifically about Jewish law is this: if, from what I heard in the podcast, in your conversation with Shalom Tzadik, you understand it as coming from God — meaning He created you and therefore has the authority to give you Jewish law. Okay. Did I understand correctly up to here? Yes. I’d be interested to know how you answer this, the Euthyphro dilemma — does He say it because it is right, or is it right because…

[Rabbi Michael Abraham] Look at columns 456 and 457. In 457 I talk about the Euthyphro dilemma exactly in this context, and I make a distinction there. I argue there that morality — and maybe Jewish law too, although I don’t deal there with Jewish law but with morality, but it is the same structure; there is a later column where I expand this also to Jewish law — that morality is imposed even on the Holy One, blessed be He, but what gives it binding force for us is the command of the Holy One, blessed be He. Fine, you need to look there.

[Speaker N] Rabbi, could you perhaps give an example of something that is different between a human being and the machine of the computer — briefly, some kind of example?

[Rabbi Michael Abraham] What do you mean, an example? I didn’t understand. Who said there is something different?

[Speaker N] For example, abduction or something like that — maybe the human being is also some kind of machine. Maybe. It could be.

[Speaker B] By the way, Miki, in one of Mario Livio’s lectures…

[Rabbi Michael Abraham] Wait, wait a second, let’s finish the previous discussion.

[Speaker N] It could be — why not? Maybe the human being also operates with…

[Rabbi Michael Abraham] Electrical currents…

[Speaker N] And things like that.

[Rabbi Michael Abraham] Could be. Who says not?

[Speaker N] No, because then there’s no difference at all between a human being and a computer.

[Rabbi Michael Abraham] What difference? Who said there is one? The Rabbi said…

[Speaker N] There is a difference earlier in the lecture.

[Rabbi Michael Abraham] Where? When did I say that? I said that I think machines do not think. So if you accept that, then there you have it: thought is the difference. What example do you want? And if you don’t accept that, then fine, free of charge: maybe a human being is also a machine. What example were you expecting to get?

[Speaker C] Only you think; everyone else are machines. You can always think that only you think.

[Rabbi Michael Abraham] I can think that you think while in fact you are a machine.

[Speaker C] Yes, okay.

[Speaker B] No, but that’s what I wanted to say about that, also about the big change. I assume everyone knows who Mario Livio is, so in one of his lectures they asked him about life. After all, he deals with space and life and so on. So he said that many times we think about life as we are. And based on the artificial intelligence that exists today, maybe one day we’ll get to some place where there is life, but not like us — rather artificial life. I talked about that, I talked about it in previous lectures. Right, but he added there that beyond life and electrons and neurons and all that, consciousness is very, very important. Consciousness. He says that aside from the fact that we don’t know how life was created — I mean beyond RNA and the like, yes? — consciousness, consciousness is basically what today divides between us.

[Rabbi Michael Abraham] I don’t agree with that definition. And as I said in the introductory lecture, one has to distinguish between vitalism, meaning the question of life on the biological plane, and the question of soul, which is dualism. Vitalism is not dualism; they are two different things. And consciousness belongs to the question of dualism, not to the question of vitalism.

[Speaker B] The question is whether the computer can understand that it is the computer and that it is alive.

[Rabbi Michael Abraham] That is a question of consciousness; it is not a question of vitalism.

[Speaker B] That’s what I said — consciousness, consciousness in the machine.

[Rabbi Michael Abraham] A different question. It’s not related to the question of life.

[Speaker B] No, but it’s the question that he…

[Rabbi Michael Abraham] Someone asked before, what is the difference. So I think that’s a very big difference. A substantive difference. I said it could be. So I’m saying: if you accept that human beings have consciousness and a computer does not have consciousness, or that human beings think and a computer does not think, then that is the difference. That’s what I was talking about now. But if you want examples — examples of what? I’m saying the human being thinks and the computer does not think. That’s the example. What example is needed? And if you think that the human being also doesn’t think or that the computer does think, fine, then perhaps it is the same thing. But we’ll get to these things further on as well, because right now we’re only dealing with understanding what a computer even is. I’m just constantly accompanying the understanding of the computer with these questions so that you keep noticing what the things we’re learning mean for these questions. I’ll return to them later and deal with them. But I want to keep reminding you of what we discussed in the introductory lectures so that it accompanies you here in the lecture as well. I don’t want you to lose the forest for the trees, yes? Pay attention. My goal is basically to return to the fundamental questions: is there something in us beyond what there is in a computer? I still haven’t stated in a principled or categorical way what my opinion is. I said here that I don’t think — it isn’t likely that anyone would say that a computer thinks in the way a human being does. Fine. But we’ll return to that matter later in a more orderly way.

[Speaker M] Rabbi, I have a small question. You said that because Turing was a genius, therefore it does teach us something, as if it proves some point. What does it help me that Turing was a genius? In the end all we learned from it is that he was wrong.

[Rabbi Michael Abraham] If Turing was such a genius — if he were not a genius, then you would say he talked nonsense because he’s an idiot. What would there be to learn from that? Don’t be an idiot? Fine. But here I’m saying: if a smart person talks nonsense, then there was a reason there that apparently caused it, that misled him. Now if such a reason can mislead a smart person, then it is very important that each of us beware of that kind of deception. That is the lesson.

[Speaker M] I understand, I understand.

[Speaker E] I also wanted to take the opportunity to ask the Rabbi something a little political. I’m referring to a remark the Rabbi made at the end, saying that he actually very much likes the Haredi mode of conduct in terms of the separation between the Knesset members who deal with the concrete matters and the Council of Torah Sages who deal with the more principled level. Right? I wanted to ask the Rabbi something fundamental I’ve wanted to ask. This week I was at a conference — I went, may it atone for my sins, I went to a Ben Gvir conference in Netanya. Two hours listening to him and…

[Speaker C] If you’re considering voting for him?

[Speaker E] No — yes, right. And you sit there and see the diverse public that was there.

[Speaker C] Ben Gvir? You went to his lecture?

[Speaker E] Yes, yes. I registered, they checked me, put this bracelet on me, and I went in.

[Speaker L] Did you at least get the children’s book?

[Speaker E] He tried to push it on me and somehow — I have enough books — but he projected it on the screens. But what I wanted to ask is this: you see a public there, teenagers and adults of all kinds and types, and in my eyes they are captives raised among non-Jews. What do I mean? The rabbis of the Council of Torah Sages, as well as the rabbis of the hardline Religious Zionist world, know how to speak — in the political realm they are fully inside the craziness, and they state their opinions and tie them to their Torah and they push it, and they have students. The more moderate rabbis, whatever we want to call them, we don’t need to get into semantics, they won’t say political things. Rabbi Sherlo, Rabbi Stav, rabbis I very much love and respect, won’t say political things because it’s uncomfortable. Why should I get myself into trouble with parents and students? And besides, it’s not accepted. I want to say only weekly Torah portions, sermons, and theoretical analysis of some Talmudic passage. But life itself, which is politics — no, I won’t bring my Torah into that. The result is captives raised among non-Jews, and it’s destructive, really destructive. The Rabbi is much more in that direction, but there is no voice and no answer, no one who will come out.

[Rabbi Michael Abraham] I’ll tell you a few things. First, I agree with the criticism. Meaning, my own way is indeed to speak out. And unfortunately I really do not see other rabbis who belong more to the more liberal, modern directions — whatever you want to call it — speaking out in a sufficiently unequivocal way. But on the other hand, there are two things here to take into account. They do try to speak out. They just don’t come out with unequivocal cries and preaching and calls of anguish, but all their students know exactly what they think. Meaning, there are no questions left vague here. It’s just that their students know how to interpret and understand what they think. There’s Rabbi Blumentzweig in Yeruham. There was always this tension, say, with Yeshivat Mitzpe Ramon, which is one of those hardline “Kav” yeshivot, and we had connections with them because it was very close. And they would say to them: they have a line, we don’t have a line. In Yeruham there is no line. So Rabbi Blumentzweig tried to explain to them that that is our line. Meaning, our line is to see things broadly, and still I think one can understand from that what we think, but we don’t think we need to dictate to anyone what we think. Therefore there is here a kind of policy of broader, more inclusive, less direct expression. Okay, that’s one point. A second point — or third, actually — is that we’re dealing with a public such that even if you come out and preach and urge them on, they don’t react like a conservative hardline Religious Zionist or Haredi public. Meaning, a hardline Religious Zionist or Haredi public, when you urge it on, it goes. Meaning, a lot of it goes. There it works. In a more open, more liberal public, even if their rabbis came out and became “Haredi” in a liberal direction, it still wouldn’t spur their public in the same way. Meaning, they don’t function the same way. Neither the public nor the rabbis. And that’s a package deal, you know, for better and for worse. Meaning, I do have some criticisms of it, but on the other hand that’s part of the charm of that world.

[Speaker E] Rabbi, the Rabbi didn’t completely understand me. What? I agree with everything the Rabbi said, but the Rabbi didn’t completely understand me. I’m not worried about the Rabbi’s students or Rabbi Sherlo’s students or Rabbi Stav’s students. I’m not worried about them. They more or less know how to think for themselves, relatively speaking, and they’ll be okay. But those people who are not their direct students, who only hear…

[Rabbi Michael Abraham] All day…

[Speaker E] Rabbi Yitzhak Yosef and Rabbi Maya and Rabbi this one…

[Rabbi Michael Abraham] And that rabbi over there — what will happen to those poor captives? That is a big question. I don’t know how much effect there would be if liberal rabbis came out with very forceful and unequivocal messages — how much of that other public would be influenced by it. I’m not sure.

[Speaker E] At least there would be a polemic. There would be a polemic, there would be some tension in the air, where people hear opinions…

[Rabbi Michael Abraham] Both of which come in the name of…

[Speaker E] Torah and Judaism.

[Rabbi Michael Abraham] That criticism I accept. Meaning, say — the fact that nobody explains that Yitzhak Yosef is corrupt and an idiot, for example, and anyone who calls him a rabbi is also corrupt and an idiot — no liberal rabbi allows himself to say that. And in home meetings too I met people, and we talked about the “third path” and all that, and we spoke about these matters, and Yitzhak Yosef came up again in one of the meetings. I told them my opinion of him in rather unequivocal terms, and they were shocked. They were really alarmed: how can you say such a thing? And the fact that he says the same thing about other rabbis doesn’t trouble them. And the fact that he is the one who is wrong and they are right — and that when he says it about them, he himself is the idiot — that if they were to say of him that he is an idiot, they would be right. Only they don’t say it about him. Meaning, this really is one of the prices. And I think it’s one of the reasons I often speak bluntly, even though people get offended and say, wait, one doesn’t speak like that — precisely to break this one-way barrier, this half-permeable barrier, which says that one side is allowed to attack the other without limits, while the other side is supposed to go on behaving politely and showing respect. Now there is in that, by the way, a certain kind of power — meaning, for some people it may perhaps affect them even more. But you are right that there are not a few people for whom, if those idiots were put in their place, perhaps it would have some effect. I don’t know.

[Speaker E] And unfortunately, in my view the motivations of the more liberal rabbis for not coming out are not because of the very correct considerations the Rabbi mentions, but rather because of, unfortunately, lower considerations. Because obviously, to come and say sharp things and then maybe lose an Israel Prize or not be invited to… Right, Yosef, you’re right, Leibowitz was willing to do that.

[Rabbi Michael Abraham] Leibowitz was willing to pay the price. I’m not sure you’re right. Maybe. But you know, people are complicated creatures, unlike computers. Human beings have various considerations. And it could be that the consideration of policy from the outset is also mixed with this issue of fear that you raised. But there is also an approach here that is principled from the outset. Some approach that really does show respect even to bitter enemies. There is also an approach here. Meaning, it’s not only concern for interests. In that sense, despite the criticism I give them more credit than you do.

[Speaker E] I agree with the Rabbi; I just think the real test is always whether you sacrifice for something. If I see someone sacrificing something, paying personal prices, then I say there’s something here, even if he’s wrong, but there’s something genuine in his path. But if he only accumulates benefits, then it already seems to me almost meaningless, even if he…

[Rabbi Michael Abraham] Is right, even if he’s right. Look, the amount of rage that Rabbi Sherlo and Rabbi Stav absorb shows that they are also sacrificing. They speak more politely than I do; they also have more to lose than I do, but they too are sacrificing. They have paid and are paying prices for it. It’s not that they’re unwilling to sacrifice. They have their own way. Again, I also have some criticism — part of your criticism I accept — but I also understand their direction very well. It’s not only fears; there is also a certain approach here.

[Speaker E] Which of them would give up an Israel Prize the way Tomer Persico did? What? Which of them would give up an Israel Prize? I’m not… I’m not speaking specifically about these names, Rabbi, just generally. I haven’t yet seen any of them give one up, nor professors of philosophy.

[Rabbi Michael Abraham] I don’t know. Leibowitz gave up the Israel Prize because he criticized the body that gave the Israel Prize. Say if Yitzhak Yosef — Yitzhak Yosef, not “Rabbi” — were to give me a prize, I would refuse it, I’m telling you, even if it were the Nobel Prize.

[Speaker B] You are on totally different levels. Rabbi, Shmuel’s question is in place and he’s right, but you’re forgetting that the Haredi public does not serve the Holy One, blessed be He; it serves the rabbis, and they keep them there precisely for that.

[Rabbi Michael Abraham] Fine, okay — again, that criticism too is too extreme. I have sharp criticism of the Haredi public, but that is too extreme.

[Speaker C] But I think it stems from the essence of religion in general. A person who goes to hear religious people wants to hear absolute, one-dimensional truth, and that is what he hears from those rabbis.

[Rabbi Michael Abraham] That is one of the problems. I try to fight against that because I don’t think that identification is correct. I don’t think religiosity has to be fundamentalist and fanatical. But that is what is out there, and there’s a model that is prevalent on all sides of the divide, yes? The Haredim, the Religious Zionists, the secular — everyone identifies religiosity with fanaticism.

[Speaker C] Yes, because they say: he says the truth, he holds it firmly, he knows what he wants, they want something charismatic. Right. Then along comes some rabbi saying, okay, this is allowed, that is allowed — it doesn’t sound good.

[Speaker B] I think for that there needs to be a separate series, not artificial intelligence.

[Speaker C] It doesn’t matter that that audience is much more traditional and doesn’t obey everything they are told.

[Rabbi Michael Abraham] Great rabbis at least ought to have intelligence. Whether it is artificial or not artificial, that can be discussed afterward. Unfortunately many of them don’t even have that.

[Speaker B] There’s that joke where one person meets another who works in high-tech and asks him, are you already working with artificial intelligence? He says, with natural intelligence we still haven’t started, and you’re asking about artificial intelligence.

[Rabbi Michael Abraham] That’s Levi Eshkol, you know, who said about a five-day work week instead of six: first let them start working one day.

[Speaker B] Yes, right.

[Speaker D] Well Rabbi, regarding the issue with ethics, I found the idea in Kant that I was talking about. Can we go back to that topic for a moment? Earlier I didn’t hear the last thing you said; my Zoom crashed. Okay, can you share a screen, or what do you want to do? I’m holding the book. So what do you want us to do? I can read it — it’s basically one line. It’s exactly before he brings the first categorical imperative about acting in a way that doesn’t contradict yourself. He says — there are about twenty points before that — but such a principle must necessarily be found in every rational being and be able to arise from his will, and therefore the principle of this will… and so on. Meaning, before he gives the categorical imperative, before he gets into what I recall he called practical anthropology, the laws of morality themselves, he grounds it in the universality of reason, and then on the fact that when he acts as a rational creature he acts in a way that does not contradict his will. And when he acts in a way that contradicts his will, that is basically immoral action.

[Rabbi Michael Abraham] Again — you said this before too, and I answered it. I claim that after he knows it is valid, he uncovers the conditions for valid morality. And there he says the conditions are that it be universal and that there be free choice and so on. But you cannot justify why it is valid.

[Speaker B] That’s an assumption, he…

[Speaker D] Here he’s not now justifying why it’s valid. I’m not managing to understand you, I mean with your citation of his morality.

[Rabbi Michael Abraham] Obviously not, what do you mean?

[Speaker D] Why not? Because where…

[Rabbi Michael Abraham] Explain to me what the justification is for morality being valid. Give me an argument in your own words that explains why morality is valid. It can’t be a naturalistic argument, right? It can’t be a naturalistic argument, and it can’t be based on God. Okay, so explain to me, not by means of facts, what could ground the validity of morality. In my opinion, freedom — the freedom of the rational. Give me an argument in your own words.

[Speaker E] Why? Freedom is precisely what lets you be immoral most effectively.

[Speaker D] No, no — whether it’s possible or impossible is a factual question; that’s not the relevant issue. Of course several moral assumptions are needed, but given those…

[Rabbi Michael Abraham] I’m asking what gives them validity — that’s exactly the point. He says you can’t derive an ought. Meaning, facts are neutral as regards validity or invalidity. Therefore one cannot derive an ought from an is. Right? You need to derive ought from ought. Right? So I’m asking: where will the first ought come from? That will be your ought.

[Speaker D] I think that question is equivalent to the question of where the first is comes from. Right? So I would say: one ought not act in a way that contradicts one’s nature.

[Rabbi Michael Abraham] Why on earth? On the contrary.

[Speaker E] Why? On the contrary.

[Speaker D] Nature…

[Rabbi Michael Abraham] My nature, for example, is to speak gossip. And I think that one should not…

[Speaker D] So here we have several… I mean nature in Kant’s sense, as a rational being.

[Rabbi Michael Abraham] I didn’t understand. What does “my nature” mean here?

[Speaker D] I mean your nature as your essence as a rational being.

[Rabbi Michael Abraham] You’re bringing in things here that have no basis whatsoever.

[Speaker D] I’m not managing to understand why.

[Rabbi Michael Abraham] Because you haven’t shown me any basis. Explain to me where the first ought comes from.

[Speaker D] Look, I agree with you that an axiomatic system is required, but wouldn’t you agree with me that not…

[Rabbi Michael Abraham] Every axiomatic system is equal. No — what is needed is an explanation that there is a category of ought at all.

[Speaker D] Why does that require a source? I’m not managing to understand.

[Rabbi Michael Abraham] Because why should there be such a category at all? How do you know? Where does it come from? It cannot derive from any is.

[Speaker D] I think that… our views are so different. I think the opposite. I think that the existence of ought is self-evident. It seems to me that ought comes before is.

[Rabbi Michael Abraham] Ah! And is too is self-evident. That’s exactly the point. That’s where it begins and that’s where it ends. Meaning, you assume the obligation, and now you only uncover what lies at the basis of the matter. You assume the obligation just as you assume the existence…

[Speaker D] Of obligation. Okay, but once you make that step, you can do…

[Rabbi Michael Abraham] Once you assume it, you can do whatever you like — but you are assuming it.

[Speaker D] Yes, so in your case… so if with Kant it is grounded on reason and freedom, then with you it’s on religion and Torah. No, it isn’t grounded — no. You’re mistaken about Kant. What’s my mistake?

[Rabbi Michael Abraham] Your mistake is that it is not grounded on freedom. It is grounded first of all on…

[Speaker D] The perception…

[Rabbi Michael Abraham] The immediate intuition that there is such a thing as valid morality. And he says: let’s try to understand what could be the basis of such a thing. Then you go backward and say, fine, if there is valid morality, necessarily it applies only to a free human being, and necessarily it applies only to universal rules, and necessarily it is based on such-and-such things. But you begin from the end. That is the argument I call a theological argument and not a philosophical one. It’s a different kind of argument. And it is completely parallel to the first Critique, the Critique of Pure Reason.

[Speaker D] I’m looking here at what he says, and I see him saying the opposite. For example, right at the beginning of the third chapter he already talks about freedom and says: “Since morality serves us as a law only insofar as we are rational beings, it must necessarily apply to all rational beings, and since it is to be inferred only from the property of freedom, it must be proved that freedom is also a property of the will of all rational beings.” Meaning, he grounds morality on reason and freedom.

[Rabbi Michael Abraham] But even in your quotation it says: since obligation can apply only to rational beings and to freedom. What is that “since”? Meaning, he assumes there is something obligatory, and then says: but what is obligatory can apply only to rational beings and only universally. There — in the very sentence you read, it appears.

[Speaker D] No, I don’t think so, because what he’s doing is looking at the thing and asking what is necessary for it to exist. Exactly! But he begins from the fact that it exists. Exactly the point. And that…

[Rabbi Michael Abraham] I always illustrate this through the argument about morality. This basically says — and I think that’s how you began the question. So I argue that for morality to be valid there needs to be a divine command; there needs to be some authority standing behind it, and for me that is a proof of God’s existence. Now, how is that proof built? It isn’t pragmatism. Pragmatism says: since I want there to be valid morality, let’s invent God so that we’ll have valid morality, because it’s convenient for us that it should exist. No — this is a logical argument. A logical argument that says: I assume there is valid morality, because I intuitively understand that, and then I say, fine, but after all there cannot be valid morality unless there is a God who commands it. Conclusion: there is a God. Do you understand? That is a theological move, a move that goes first of all by assuming the result.

[Speaker D] Maybe it seems as though he takes one more step. To me it seems as though he doesn’t assume it, but it’s more like an if-then claim.

[Rabbi Michael Abraham] There’s an… I have a disagreement with you and with many interpreters of Kant — you’re in good company — and I think they’re all mistaken.

[Speaker D] It’s nice to be in good company, but to me it looks like an if-then, like a conditional proof: suppose P, then if we suppose P we get Q, so if P then Q. Then he shows some… some equivalence between them and then…

[Rabbi Michael Abraham] Wait, but if your implication is “if P then Q,” then I say as follows: I assume Q, and I say — but only if P then Q, so therefore P. That is his move.

[Speaker D] Okay, fine. So in that sense he does base his conclusion.

[Rabbi Michael Abraham] No, because Q is valid morality. But valid morality can be based only on your being free — that is P. Only on your being free and on universal rules. So apparently you are free and universal and from that it follows…

[Speaker D] No, I remember… he specifically… I remember him specifically proving that to behave morally, to be… the existence of morality is equivalent to free conduct according to reason.

[Rabbi Michael Abraham] What do you mean “equivalent”? It cannot exist without that.

[Speaker D] Equivalent. No, no — equivalent. I mean exactly the same thing, simply logically equivalent.

[Rabbi Michael Abraham] No, it cannot exist without it. What… what does “equivalent” mean? The equivalence is not… relevant here. It cannot exist without it. I mean equivalence, equivalence. I don’t mean what you’re saying now, that it can’t exist without it. I mean that when you behave morally, you behave freely according to reason, and when you behave freely according to…

[Speaker D] Reason, you behave morally. Why? Why on earth? And there I ask: why should one? Why should one behave freely? Why should one behave morally? Why should one behave according to reason? And why do freedom and reason entail morality? Why do freedom and reason entail… They don’t entail — they don’t entail, it’s an equivalence.

[Rabbi Michael Abraham] Okay. Now I ask why all these equivalences are true.

[Speaker D] I agree with you that there is an axiom here regarding…

[Rabbi Michael Abraham] But it’s a normative axiom. An axiom of ought, not of is. But if an axiom of ought is needed, then there you are: you assumed the existence of obligation, the existence of ought, you are assuming that sphere of ought.

[Speaker D] Ah, I see the… okay, right. Now it’s… I don’t know… yes, I drifted a bit there… right.

[Rabbi Michael Abraham] Fine, we’ve gone overtime by quite a lot already. Okay.

[Speaker C] Gemini explains that one.

[Speaker B] It was very interesting, thank you.

[Speaker C] Thank you.

[Speaker D] Thank you very much.

[Speaker B] Goodbye.

[Speaker C] Okay, Sabbath peace. Sabbath peace.

[Speaker B] Sabbath peace. Thank you very much. Sabbath peace.