Transcript: 210. John Marenbon on Peter Abelard

Note: this transcription was produced by automatic voice recognition software. It has been corrected by hand, but may still contain errors. We are very grateful to Tim Wittenborg for his production of the automated transcripts and for the efforts of a team of volunteer listeners who corrected the texts.

Peter Adamson: Maybe first you can just remind us who Peter Abelard was. 

John Marenbon: Right. Abelard is probably the most important and certainly the most controversial philosopher of the 12th century. He was born in 1079 and his work really spans the first four decades of the 12th century. Up until about 1120, he's working mainly on logic. After that, he interested himself increasingly in theology, and he writes a variety of theological works and also one work was very striking in its form because it consists of a dialogue involving a philosopher. So, something like a pagan philosopher who talks first to a Jew and then to a Christian. 

Peter Adamson: But we're not going to be talking about that. 

John Marenbon: That's true, yes. 

Peter Adamson: Actually, I've already mentioned that anyway. What we are going to be talking about is freedom and ideas like contingency and necessity in Peter Abelard. This is something I've already tackled several times on the podcast and pretty recently, in fact, in the context of Latin medieval philosophy, because I talked about Eriugena’s views on predestination and when I was introducing Anselm, I talked about his views on freedom and what freedom of the will would consist in. Now, Abelard's not responding directly to Eriugena and Anselm, that's right? 

John Marenbon: That's right, yes. Abelard's discussion comes out of looking at Aristotle's On Interpretation, chapter nine of that, Boethius's commentary on the Aristotle, and then also Boethius's Consolation of Philosophy, book five, where Boethius himself is looking back at least in part to that Aristotelian logical tradition. 

Peter Adamson: And presumably he knows the Aristotle in Boethius's Latin translation as well. 

John Marenbon: Exactly, yes. 

Peter Adamson: And the reason why this comes up in this context is that in the famous ninth chapter of On Interpretation, Aristotle presents an argument for determinism, which he then refutes. And down to today, there's not a lot of agreement about what exactly the deterministic argument is, nor is there agreement about what the solution is. Can you sketch what the basic problem is, just to remind people, even though I have looked at this before? 

John Marenbon: Right, so the background problem, Aristotle's problem, is just that if we say there's going to be a sea battle, there'll be a sea battle tomorrow, and if we consider that every proposition is either false or true, so it seems that we've got to say, well, that's either false or true. We don't know which it is, but supposing it's true, then there's going to be a sea battle tomorrow, supposing it's false, there isn't. So it seems that just by thinking of logic, we've established that there's no contingency with regard to whether or not that sea battle takes place. 

Peter Adamson: Because if it's already true now, then it's too late to do anything about it as it were. 

John Marenbon: Exactly, yes. Of course, one way out of that, and some people think that that was the way Aristotle took, and some people deny it, is just to deny the principle of bivalence, that is to say deny that the proposition does have to be false or true, and say that with regard to future contingent propositions, they're neither. 

Peter Adamson: But I guess that solution is off the table if you have divine omniscience, right? 

John Marenbon: Exactly. So the problem which really concerns Abelard, and which concerned Boethius, especially in the Consolation of Philosophy, exactly is what happens when you bring in a God who knows all things. 

Peter Adamson: Why not just say that God doesn't know the future, though? I mean, why couldn't he know all the things that there are to know, which means all the past things and all the present things, and maybe the future things that are necessary, like one plus one will still equal two tomorrow? 

John Marenbon: It's difficult to answer that in that I don't think anybody took the view of saying that simply future contingents aren't among the things which are there to be known. It was always considered that it would detract from God's omniscience. And I think there's a certain sort of common sense in that. It does seem, if some being knows all that's going to happen in the future, this being seems obviously to know more than a being who doesn't. 

Peter Adamson: Right. So just the fact that there could be truths to know is already going to give them a big push in the direction of saying that God knows them. 

John Marenbon: Yeah. 

Peter Adamson: Okay. And how does Abelard then set up the problem? Is that basically his way into it? 

John Marenbon: Yeah. So that's basically his way into it. If we put his way of stating it so that we can then see how he gets out of it. I suppose it's quite important that in his way of stating it, he rather forgets the temporal elements. So we're in fact talking about God's knowledge of the future. But his way of stating the problem is to say that if God, or indeed if anybody knows a proposition, then that proposition is true because you can only know what is true in virtue of the meaning of know. This is surely something which is necessarily the case because we're not just talking about an accident. We're talking about what something must be in order to be known. So we can say that necessarily if God knows some proposition P, then P is true. And then the problem is put in terms of, well, doesn't that mean therefore that P is necessary? 

Peter Adamson: Right. So this wouldn't only apply to future facts. So it would also mean if God knows that there was a sea battle yesterday, then necessarily there was a sea battle yesterday as well. 

John Marenbon: That's right. But of course that doesn't seem to be such a problem because although there's a certain sense in which we want to say that yesterday's sea battle wasn't necessary, there's also a certain sense from now in which it is necessary and there's nothing we can do about it. And one could say the same about the present too. But when you consider this with regard to the future, it seems to be a problem because we want to say now it's open whether or not the sea battle is going to take place.

Peter Adamson: When you frame the argument this way, how much work is being done by the fact that we're talking about knowledge? Imagine that I say I now have a true belief, but not knowledge, that there will be a sea battle tomorrow. There's evidence concerning it, and I've checked with the generals and the admirals and they all have assured me that there will be a sea battle. So I believe now truly that there will be a sea battle tomorrow. Would that not raise the same problems? Because it's true now.

John Marenbon: Yeah. It would raise the same problems if we can take it that it's true. So a lot of accounts of knowledge are that it's true belief plus something or other. And the important thing I think is setting up this problem is the truth of it. However, the way, when you're saying, well, the admirals have assured me, it seems you're not talking about a true belief, but simply about a belief for which we have a very great deal of evidence. So it's highly probable that it will turn out to be true. And of course there's no problem about that. 

Peter Adamson: Well, what I was imagining was a case where it's a justified belief. And in addition, it happens to be true. And although I can't be sure that it's true, in fact, it is true. What I was wondering, in other words, is whether the necessity is supposed to flow from the thought that anything I know must be incapable of being otherwise because that's what knowledge is like. And that wouldn't be true of true belief. 

John Marenbon: Well, yeah, yes. Perhaps casting that in terms of true belief in that way is a first step on the way to the, in a way rather simple, but also as it turns out, an adequate solution that Abelard proposes. Because what Abelard has, which Boethius doesn't have, is a notion of operating on propositions. So if we take a complex proposition such as, if it's day, it's light, and then we think about negating it, Boethius, it seemed, could only think of negating each or both of its parts. So, if it's not day, it's light, or if it's day, it's not light, or if it's not day, it's not light. But what Abelard realizes is that you can do something which is different, which you can say that the whole of that isn't true. That's to say this doesn't follow from that. Once you start to think in that way, you can see that in this problem, there's a distinction between the true proposition that necessarily, the whole of what follows is necessary, if someone knows P, then P, and we accept that. And the distinction between that and putting the necessity operator before P. So it doesn't follow from that, that necessarily P. 

Peter Adamson: So it's like the difference between saying necessarily colon, if it's now true that there will be a sea battle tomorrow, then there will be a sea battle tomorrow, and on the other hand, saying if it's now true that there will be a sea battle tomorrow, then necessarily there will be a sea battle tomorrow. 

John Marenbon: Exactly, yes. 

Peter Adamson: It’s where you put the necessity, whether it's applying to the whole inference or the actual event of the sea battle. And then Abelard applies this to get out of the deterministic argument, so how exactly does that work? 

John Marenbon: Well, so, I mean, his point would be that just because God knows everything, including whether there'll be a sea battle tomorrow or not, and there is a connection of necessity between God's knowing things and their happening, this doesn't mean that they happen necessarily. 

Peter Adamson: It's just necessary that if he knows them, then they happen. 

John Marenbon: Yeah, just as it's necessary that if I know something, it will happen, or it happens. The thing is, though, I can't know future contingents. 

Peter Adamson: Whereas God can. 

John Marenbon: Whereas God can. But Abelard considers that he's solved the problem, but he's only really solved, or thinks he's solved the problem, because he's, as I said at the beginning, removed the temporal element from it. So I think one can see it commonsensically, that it's not nearly so worrying that somebody should know what's happening now. And we don't feel there's a problem about that moving contingency. Whereas there does seem really to be a problem about somebody knowing what's happening in the future, supposing that's contingent, that that might or might not happen. Because it seems, if we're insisting that God now knows what's going to happen in the future and that that's contingent in the future, then it seems that something in the future must have the power to make what God supposedly knows now into not knowledge, but false belief. Otherwise how could it be contingent?

Peter Adamson: So what you're saying is, if, let's change the example, because I can't really stage sea battles, I don't know about you, but I do have the power to have eggs for breakfast tomorrow morning. Taking that example, I guess what you're saying is something like this. Given that tomorrow morning it will be up to me whether or not I have eggs, it would be in my power tomorrow morning retroactively to make God have had a false belief today about whether I would have eggs or not. 

John Marenbon: Exactly. Yes. 

Peter Adamson: That's really bad. 

John Marenbon: Yes. And that's really bad. And if you want to see why Abelard’s analysis won't get round that, then what you need to consider is that, let's consider God's belief about your eggs. And it's not just a matter that God now has a belief about how many eggs you're going to have tomorrow morning. God had that belief yesterday, in fact, he had it for all eternity. There's a sense in which that belief is necessary, not necessary in a straightforward sense, but as people said, accidentally necessary because it's in the past. God has come to that belief. And so it's not just a belief, it's knowledge. So God has that knowledge, and that's because the past is necessary. So what we have is both that necessarily if God knows P, then P, but also we've got that necessarily God knows P with the necessity of the past. And most people would admit that you can, in this case, as it were, transfer necessity, so that you can then validly deduce necessarily P. 

Peter Adamson: Okay, so thinking about that in a slightly less technical way, we could say that if God already knew yesterday and indeed from eternity that I'll have eggs tomorrow for breakfast, then although it's not too late for me to do anything about having eggs tomorrow for breakfast, it is too late for me to stop God from knowing that I'll have eggs tomorrow for breakfast. And if I can't stop him from knowing that, then I can't do otherwise with respect to having the eggs. Is that basically the problem? 

John Marenbon: Yeah, exactly. So we need its knowledge and we need that it’s in the past so you can't change it.

Peter Adamson: Because he can't have been wrong and it's past. 

John Marenbon: So in fact, after all, and despite Abelard's helpful bit of logic, you've got to have those six eggs or whatever.

Peter Adamson: Oh no, I'm glad it's just you and not God telling me that. And of course, that's good in a sense because future medieval philosophers are going to keep discussing this, so this leaves room for them to say something interesting about it. 

John Marenbon: Exactly. Yes. And I mean, they do see that Abelard didn't succeed in dealing with the problem. And they come up with all sorts of clever ways of trying to deal with it, though whether any of them are successful or not, it's hard to say. But I'd quite like to say something about another aspect of the problem of freedom where Abelard might be more successful. And he's very successful in formulating an interesting problem. And I think that he proposes an interesting solution to it. This is a problem about freedom which comes from a somewhat different direction, not just from thinking about the logic of propositions that are true or false and then putting in God's knowledge, but thinking about the nature of God and his ways of acting. Because it's agreed among all medieval Christians that God is omnipotent and also that he's completely benevolent. So he always wants to do the best thing. But then Abelard says, and this is something which he becomes very interested in in the middle of his career, when he's moved away from just doing logic, and so if you like, moved on from the discussion which we were looking at before. So let's consider God and how he acts. Well, at every juncture, he has to do whatever's best. But if he has to do whatever's best, then he has no alternative choices. 

Peter Adamson: Because he knows what's best, too. 

John Marenbon: Exactly, he knows what's best. He can't get it wrong. And there's also no chance of him saying, well, I'd like to do this, but I can't because God's omnipotent. 

Peter Adamson: Oh, right. Okay. He's stuck. 

John Marenbon: And God cannot do other than he does. 

Peter Adamson: And so actually, we possess more freedom than he does. Actually, he doesn't possess any freedom at all. Because he has to do exactly the best thing at every moment throughout eternity.

John Marenbon: Yes, I mean, that's right. And what you just said, I think, is indeed what Abelard will come down to saying in the end. But you might think that that's not the case. And you might think that because God is just acting in necessity, and we're used to thinking when we have a Christian thinker that God in some way is arranging all things. So if God acts of necessity, and God is omnipotent, then surely there's going to be no room for us to act. 

Peter Adamson: Oh, so it's even worse. So we're not free either. 

John Marenbon: That could be the problem. And what's interesting is Abelard puts that to himself in the form of a particular objection to the view that he's trying to propound. And he recognizes that this view that God cannot do other than he does is a very strange view. And other thinkers haven't held it in the past. And he knows he's going to be criticized for it, as indeed he was. But nonetheless, he wants to put it forward. But he does put forward this objection to it. He's talking about somebody who, as a matter of fact, is going to be damned, because he's led an evil life and he deserves to be damned. So take this person who's going to be damned. He calls it this damnandus. Christian doctrine requires nonetheless that we say that it's possible for him to be saved by God. Because if he's going to be saved, it's going to be by God. And if we were to deny that it was possible for him to be saved, then we'd be saying that whatever he did, he couldn't be saved. And that would certainly be against Christian doctrine. We know as a matter of fact he's not going to do the right things, and so he's not going to be saved, but he might do them. And so if it's a case that it's possible he'll be saved by God, surely it's the case that it's possible that God will save him. Because for him to be saved by God means the same as for God to save him. And if it's possible for God to save him, but in fact God isn't going to save him, then we've shown that God can do what he doesn't do. So Abelard found a very good argument against his own position. However, Abelard denies that consequence, and he denies the consequence by denying that in fact it's possible for him to be saved by God means the same as it's possible for God to save him. He agrees that if we just take the simple statement, God saves him, he is saved by God, they have exactly the same meaning, they're just two forms of words saying the same thing. But when you talk about possibility, that's not the case. Because he says that when we're talking about that it's possible for him to be saved by God, we're referring possibility to him and to his capacities. And there Abelard would say, as Christian doctrine demands, that of course it's possible for him to be saved by God because it's possible for any human being to be saved by God. That possibility remains until the moment of their death and damnation. But when we're talking about whether it's possible for God to save him, then it's different because we're talking about what's possible for God. And given that in fact he has done evil and it would be unjust to save him, it's not possible for God to save him. 

Peter Adamson: Because he'd have to do something worse. 

John Marenbon: He'd have to do something which was wrong, something unjust, which first of all would be against God's nature and secondly wouldn't be the best action that he could take. The best action in this case is to do what's just and damn the man. 

Peter Adamson: Let me ask you a more basic question about Abelard's whole position here. Why not say that when God's making a decision like this, for example whether to damn the sinner or not, he evaluates the situation, he sees what's best to do, and then he decides freely to do the best thing. In other words, why not say that his knowledge of what's best doesn't force him to do what's best. It just gives him a reason to do what's best, and then he acts on that reason. Even though he'd have the possibility of not acting on that reason. 

John Marenbon: Because Abelard doesn't believe that he does have the ability of not acting on that reason, because not acting on that reason would mean that he wasn't doing the best thing. 

Peter Adamson: But that seems kind of circular. So what I'm saying is that, I mean you can't prove that God has no ability not to do the best thing just by insisting that he must do the best thing. 

John Marenbon: If God is omniscient, omnipotent and entirely good…

Peter Adamson: So it actually flows from his nature. 

John Marenbon: It flows from his nature. Of course, an obvious objection is you might say yes, but supposing there are two things which are good for him to do. 

Peter Adamson: Oh right, okay, that’s even better.

John Marenbon: So then there must be some cases where, but no, Abelard says there couldn't be a case like that. Because supposing there were, then there'd be no reason why God should do one rather than the other. And so God would be acting without reason. And the universe couldn't be like that. It could never be. 

Peter Adamson: So just as God can't do something unjust or wrong, he can't do anything arbitrary either. 

John Marenbon: Exactly. 

Peter Adamson: So there must always be a best thing for him to do and he must do it. 

John Marenbon: One best thing. 

Peter Adamson: Just by his very nature as an all-powerful, all-good, all-knowing deity. I'm guessing that this position was not received with universal acclaim among later philosophers. 

John Marenbon: I mean, it was received with universal disdain. So what happened was there's actually an interesting discussion by Hugh of St. Victor. So a thinker of a very different cast, but quite an intelligent man. Quite near the time where he goes through, he knows the argument, probably not from any source that actually survives now, but he knows in some detail and he discusses it quite well. But then it gets taken up by Peter the Lombard in his sentences, and this is a book which comes to be enormously influential and is used in all the theology faculties of the universities. Peter the Lombard takes up a bit of the argument in a somewhat garbled version and rejects it out of hand. Everybody else knows the argument through Peter the Lombard. Peter the Lombard doesn't mention Abelard's name. He clearly has Abelard in mind. So people know this argument as a position which the Lombard rejects, and they also reject it. Sometimes they devise more elaborate arguments than the Lombard himself had. The one person who seems perhaps to have some inkling that it's by Abelard is Aquinas, who in I think it's his questions on power, refers to this position as having been taken by a certain Peter Amalario, which, it seems, there's some sort of text this has come through to him. But anyway, it's something which is universally rejected, until the one thinker who in some ways follows it, but thinks he's rejecting it, is the first significant thinker who restores the parentage of the argument to Abelard, and that's Leibniz. Leibniz in the end didn't actually read Abelard's Theologia, Theologia Scholarium or Theologia Christiana in which this argument is put. He read an abbreviation of it. He discusses it quite fully, and he rejects it, and he suggests that Abelard's view is completely different from his own, but actually Leibniz finds himself in much the same sort of position because he has the same sort of view about God as always having to choose the best. So although Leibniz might say that when Abelard puts forward this argument he's just playing with words, actually it's something which is rather central to Leibniz's own thought and a difficulty for Leibniz himself.