Could God have made a Picasso?

This somewhat silly but intriguing question comes from an essay in the most recent Journal of the History of Philosophy by Rondo Keele. Take any painting by Picasso — let’s say Guernica. Could God have created that painting, without using Picasso himself as an intermediate cause?

Some would say yes, since Guernica is just a creature, and an omnipotent being can create any creature, it seems.

But some would say no, since part of Guernica‘s identity is tied to the fact that Picasso painted it. Look at it this way: suppose a stroke-for-stroke duplicate of Guernica is created ex nihilo. Is it Guernica? No, at most it can only be a wonderful forgery of Guernica, because it didn’t come from Picasso.

So really the question is: can God make Guernica, or at most only a forgery of Guernica?

(This debate, by the way, originates from a dispute between the medieval philosophers William Ockham and Walter Chatton.)


31 thoughts on “Could God have made a Picasso?

  1. Huenemann Post author

    Didn’t quite get that, Vince. Are you saying God creates Picasso, and actually is the cause behind the movement of Picasso’s hand, and presumably the cause of the canvas and paints, etc., so in fact God did make Guernica? Sounds a bit like occasionalism — which is ok, though many people find it counter-intuitive.


  2. Huenemann Post author

    Oh boy. This is one of those topics that is truly inexhaustible. Let’s see if how much I can hold myself back!

    First, just about all traditional theists maintain that God sustains creation at each and every moment (but this should in no way make God or God’s actions temporal). So believing that is not sufficient for making one an occasionalist. On top of believing in God’s sustaining creation, most theists then think that some creatures do have causal powers themselves — so-called “secondary” causal power, which works in additio to God’s primary power — and so can bring about effects in other creatures (with God’s support). These causal powers, along with the creatures, are re-created each moment by God. This is where the occasionalists part company: they deny that any creature has any causal power (well, there were a few who made a few exceptions, but never mind). So every effect is caused by God, and there are no secondary causes.

    Second, the occasionalists nevertheless believed in free will, even though people aren’t genuine causes. People’s volitions (which are self-caused? I don’t know how this part of the story goes) serve as “occasions” or “requests” for God to exert causal powers and make an arm go up, a mouth open, etc.

    Third, on the historical stuff: I know Hume’s skepticism toward genuine causal forces in nature was preceded by Malebranche, but I don’t know whether Malebranche was familiar with al-Ash’ari. My guess is that the scholastics were familiar with him, because there are quite a few scholastic arguments against occasionalists, though exceedingly few scholastic occasionalists.

    Here’s a weblink for a discussion of Malebranche’s occasionalism:


  3. JT

    Thanks to an email from Mike, I checked out this interesting little post.

    (BTW Vince and Charlie, I’m working on Ockham, but of course Ockham spend all his time quoting at length and criticizing Scotus and Henry of Ghent, so I spend just as much time with Scotus and Henry as I do with Ockham.)

    Ockham is certainly the name most associated with this debate, but it was a common topic in scholastic philosophy going well back.

    Anyways, the problem for the scholastics runs like this. Suppose that A causes B, and B causes C. The question is: can A cause C without B?

    For example, a hand holding a stick moves the stick so as to push a ball along the ground. Here the hand’s movement is A, the stick’s movement is B, and the ball’s movement is C. Can A cause C without B? Of course. The hand can directly move the ball, without using a stick as an intermediary cause. Ockham argues that this is because a hand (or rather, the human controlling the hand) is causally powerful enough to cause C without B.

    (Notice that the stick cannot move the ball without the stick itself being caused to make the ball move. Thus, B is not causally powerful enough to cause C without A.)

    A case where this wouldn’t be true is a hand using a hammer to drive a nail into some wood. Here the hand’s movement is A, the hammer’s movement is B, and the nail’s movement is C. In this case, could A cause C without B? No. A hand is not causally powerful enough to cause C without B.

    The same principle applies for cases where God is A. For example, suppose God causes a stick to move so as to push a ball along the ground. Here God is A, the stick’s movement is B, and the ball’s movement is C. Can A cause C without B? Of course. God is powerful enough to cause a ball to move, so God could certainly do that without a stick as an intermediary cause.

    Ockham’s position is that God (A) can always cause C without B, except where it would be a contradiction for C to be caused by something other than B.

    The example I often give of such a contradiction is creating a son without a father. Obviously this is impossible, because a son must, by definition, be caused by a father, so God could not create a son without the son’s father.

    But there is a technical point here that needs to be cleared up. If Harold is the father of Peter, God could create Peter. After all, God can create any human he likes. What God cannot create is Peter’s relation being the son of Harold. So God can create Peter without Harold, he just can’t create Peter as the son of Harold without Harold.

    (This is where the ‘contradiction’ bit comes in. If God, without Harold, created Peter with the relation being the son of Harold, Peter would bear the relation being the son of Harold (because God created Peter with that relation), but Peter would also bear the relation not being the son of Harold (because Peter was not begotten by Harold), and that’s the the contradiction. But there’s nothing contradictory about God creating Peter without the relation being the son of Harold.)

    Likewise with the Picasso painting. Ockham would say God can create Guernica without Picasso, but God could not create Guernica’s relation being painted by Picasso without Picasso. God can create Guernica, but he can’t create Guernica as being painted by Picasso.

    One other point about Ockham’s position on this question. Ockham himself was not an occasionalist, nor is he trying to disprove it, so occasionalism is not the reason he asks this question. Ockham frequently asked questions of the form ‘what if God could, by his absolute power, do x?’ The purpose of these questions was to sort out, through a thought experiment, what exactly is at issue.

    In the case of A causing B causing C, Ockham asks this question to get clear about exactly what is essential to causes and their effects. Consider what I said about Peter and Harold. By asking whether God could create Peter without Harold, Ockham comes to the conclusion that the feature of Peter which essentially pertains to his causal origin is simply the relation being the son of Harold. This greatly clarifies the issue, and this is exactly what Ockham wants to do. Likewise, by asking this question about hands-sticks-balls and hands-hammers-nails, Ockham sees that some As require Bs to cause Cs, and some As do not. Again, this is a huge clarification to the problem, and this is exactly what Ockham wants to do. Hey, it’s Ockham’s razor baby.


  4. JT

    Some of my own thoughts on the question, from my (weak) Ockhamist’s perspective:

    The original post suggests, in the argument for the contrary, that the relation being produced by Picasso is tied to the identity of Guernica. But to what extent is this relation tied to Guernica’s identity? Necessarily (or essentially), or contingently?

    I think it makes the most sense to say that the relation being produced by Picasso is only contingently part of Guernica’s identity. It’s hard to see how this relation could be tied to Guernica’s identity necessarily, because one can easily imagine a possible world (or two) where Picasso did not paint the Guernica.

    Besides, why should we treat relations of causal origin, like being produced by Picasso, as essentially part of an object’s identity?

    For some objects like trees and stones, a relation of causal origin (call it R) does not seem to matter very much. For example, it doesn’t make any difference to me whether the damson tree in my back yard grew from a seed coming from a damson tree in the lot on my left or from a damson tree in the lot on my right. Nor would it make much difference to me whether the tree came from a tree-parent or whether God created it in my back yard directly. It’s just a tree, no matter where it came from, and I’d leave it at that.

    But for other objects like paintings and human persons, R does seem to matter. For example, we treat paintings which come from Picasso as more important than objects which come from an anonymous copiest. But why should this be so? Why should we privilege one causal origin over another? Why should Guernica be any better because it has the relation being produced by Picasso than if it had the relation being produced by an anonymous copiest?

    One might say that Guernica, if produced by Picasso, has some intrinsic perfection which could only be accomplished by a master such as Picasso. But imagine if God created a perfect replica of Guernica. In this case, Picasso’s Guernica and God’s Guernica would be exactly the same apart from their R (God’s Guernica would have the relation being produced by God, while Picasso’s Guernica would have the relation being produced by Picasso, but the two paintings would otherwise be the same). In this case, the intrinsic perfection of Guernica 1 and Guernica 2 would be identical, so why should we privilege one R over the other on the basis of the intrinsic perfection of the paintings?

    One might then say that although we can’t appeal to the intrinsic perfection of the painting itself, we could appeal to the perfection of the causal origin. That is, Picasso is just a more skillful causal painter than an anonymous copiest. But again, how would this argument fare against the example of God creating a perfect replica of Guernica? In that case, the causal origin (God) would obviously be much more skillful than Picasso, so we can’t really appeal to the causal origin’s perfection either. So again, why should we privilege one R over another?

    Why should we not just say that our culture has, for one reason or another, decided to privilege a certain R (such as being produced by Picasso) over others (such as being produced by an anonymous copiest)?

    My question, then, is this: why shouldn’t we just say that Guernica’s R is only contingently part of Guernica’s identity, and thus God (or evolution, or natural forces such as strings, or whatever) could have caused things to turn out differently?


  5. JT

    One more point (I’m bored!).

    The original post also makes the suggestion that God’s Guernica would be a copy or a forgery, just because it’s not created by Picasso. The assumption behind this might be about R, something like this:

    Def. of Forgery: some x is genuine (not a forgery) if it is produced by a legitimate producer, but it is a forgery if it is produced by an illegitimate producer.

    For example, a painting is genuine if it comes from Picasso, but only a forgery or a copy if it comes from someone else (e.g., an anonymous copiest).

    But what if Picasso painted the Guernica, and then painted a copy of his original? On the Def. of Forgery, Guernica 2 wouldn’t be a copy. But clearly Guernica 2 is a copy of Guernica 1, and it’s also a genuine Picasso. Now imagine that an anonymous copiest makes another Guernica. Guernica 3 would be a copy (of Guernica 1 or of 2?), but it would not be a genuine Picasso.

    This goes to show that the relations being produced by and being a copy of are different (call them RP and RC, respectively). Guernica 1 is related to Picasso by RP, but is not related to anything by RC. Guernica 2 is related to Picasso by RP, and is related to Guernica 1 by RC. Guernica 3 is not related to Picasso by RP, but is related to Guernica 1 (or 2?) by RC.

    So I wouldn’t call God’s Guernica a copy or a forgery. It would just have a different producer.


  6. Huenemann Post author

    Welcome aboard, JT! It’s great to have your insights.

    In the JHP article which raised this question, the author says the answer hinges on how real we take relations to be. Chatton thought that the relation “son of” consisted in some very real properties or thingies or doohickies attached to the relata. So the father has a kind of metaphysical tag attached to him which reads “father of Isaac” and the son has a similar tag reading “son of Abraham.” Ockham saw this as multiplying entities without necessity, so argued against the reality of relations, and I guess tried to reduce them to the bare relata + ???? (their arrangement?).

    But I also gather from the article that the Condemnation of 1277 clearly implies that God can create any creature without having to use anything else as an intermediary. Ockham apparently used this against Chatton’s view, since if Chatton were right, then theoretically God could create Abraham, with Abraham’s real property of being father of Isaac, and Isaac, with Isaac’s real property of being son of Abraham, without Abraham really being the father of Isaac. This, Ockham thought, was absurd. (I love it when philosophers have such strong intuitions about matters I can barely wrap my mind around!)

    Re. relations and identity: I don’t think that “painting Guernica” is essential to Picasso (so P would still have been P, had he not painted G), but I think — or at least it’s arguable — that “being painted by P” is essential to G. To see that it is, we have to try to conceive a possible world in which G is painted by someone else. Of course, we can imagine a painting *just like* G being painted by someone else, but would that “just like” painting actually be G? The negative answer relies on Kripke’s “Naming and Necessity” for its support, which argues that when a thing gets baptized, we are singling out that very object, with its own particular causal origins, and rigidly assigning a name to it. So, this perspective would say, if we were to discover to our great surprise that the painting we identify as “Guernica” was in fact painted by Matisse, we would have discovered that Guernica was in fact not Guernica, but some other painting of the same name!


  7. Mike

    In previous threads I’ve bashed a lot of thought that isn’t what I would consider engaging the world. You might think that would include discussions like this with the scholastics but the flip side to that is their whole perceptions of the world were so different that understanding one of them is one of the nearest points we get to seeing what a wholly other viewpoint would be. Mostly because of the depth of thought and how it’s very outside how we organize our thoughts in this age. We hinge things to science (usually subconsciously) that they never would and yet their world is in order to a similar level of depth (if not greater).

    In some ways they create an ethical impetus that says something like “Paint yourself a scholastic picture of the world and earn your pluralistic point of view!”. That’s the root of what I think it is to understand another and these hardcore scholastic thinkers are one avenue towards that. When you can fill in the gaps to their thinking in a way that makes it entirely reasonable for you then you’ve accomplished something. More than that though, you’re changed. You have earned a degree of enlightenment that most ‘liberals’ pretend to have but lack the resources to hold as a ‘true’ belief.


  8. JT

    Yep, Ockham denies that relations are extramental entities, ‘little things’ as he calls them in jest of Chatton’s view, that inhere in objects. (Note: he doesn’t deny this status of relations in the case of the trinity.) In this case then, when I talk about relations like Peter’s being the Son of Harold, I simply mean, from Ockham’s point of view, whatever it is that is the truth-maker of the statement. It needn’t be a ‘little thing’. It can just be the fact of causal origin instead.

    That’s interesting about Kripke and causal origin. I’m about to read Naming and Necessity again with a friend, so this will be a good conversation to keep in the back of my mind!


  9. Mike

    Vince said, “It is interesting that theistic fatalism comes to the same conclusion as the atheist Hume on cause and effect.”

    Not sure how I overlooked that earlier. “interesting” or pointing towards the futility of certain forms of thought?


  10. Mike

    Hume’s thoughts on cause and effect are pretty interesting and gave me a lot to think about when I encountered them. My question in more explicit form is “what does Hume’s point of view on causal relationships imply about how we should live?” Or does any strictly deterministic point of view necessarily degrade how we view existence or is there no necessary relationship? Another question is whether the choice between a deterministic point of view and non implies a greater discrepancy in human behavior than other points of view. If so, how so?

    I think for instance that not having the ability to do what you desire directly impacts existence while having a deterministic point of view might feel oppressive periodically if I think about it in a certain way but it generally doesn’t bother me. Some days I might feel free, other days not so much.


  11. Huenemann Post author

    Well, let’s hope that being a determinist doesn’t commit me to believing that we are forced to do what we don’t want to do more frequently than would be the case if determinism were false. (Did that make sense? I just mean that being a determinist shouldn’t make you feel ‘more forced.’) We’re determined by the conjunction of many forces, and some of these are our desires, and when we get to do what we desire to do, then that behavior is free.

    I think that with both the Chatton/Ockham debate and debates over Hume what happens is this. You start with an obviously important question, like the extent of God’s power, what’s necessary in the universe, whether causal relations are necessary, etc. (Or, if these questions don’t seem gripping, just trace out some of the consequences of the various positions one might take on them.) Then you start hammering away on the questions, and pretty soon you’re arguing over something like whether God could have created Guernica without creating Picasso. Anyone walking in on the discussion at that point would say, “What a load of crap! And what a flock of useless intellectuals!” But that’s because they weren’t in on the first part of the conversation, and so they have missed the connection to important questions.

    Mike is right that the trick is to keep in mind how/why the question is significant, and in exploring historical texts, making sure that (in Gadamer’s metaphor) their horizon begins to merge with our own.


  12. Mike

    I like Vince’s box analogy in regard to the limitations of science.

    Is there some sort of dialog that can be had between Aristotle and Hume in regard to causation? I’m not familiar enough with Aristotle to know where he would address that sort of thing.


  13. JT

    Excellent question. All of this turns on Ockham’s theory of properties. Let me try to explain it with some brevity.

    In Ockham’s day, everybody agreed that there are properties such as being human (humanity). And everybody agreed that humanity is a property of each individual human. So you have the property ‘humanity’ and I have the property ‘humanity’.

    The disagreement was over how many humanities there are. Take your humanity and my humanity. How many humanities are there? Realists such as Walter Chatton said there is one. Nominalists such as Ockham said there are two.

    If we say there is only one humanity in you and me, then we are saying this: there is one humanity, which is shared by you and me, and it’s not divided between you and me. In other words, properties have the characteristics of being singular (not many), being shared by many things, and being undivided by the things that share them.

    If we say there are two humanities in you and me (one in you, and one in me), then we are saying this: each of us has our own humanity, which is not shared between us. In other words, Ockham would say that properties have the characterists of being many (not singular), and being individual for (not shared by) each thing.

    Today, we tend to call properties of the first type ‘properties’, and we call properties of the second type ‘tropes’. For clarity, I will stick to the modern terminology in what follows.

    Now, Ockham believes in tropes (or at least he believes in some tropes), but he rejects properties. He thinks it is absurd to think that you and I each share the numerically same humanity property. Instead, we each have our own humanity trope.

    However, Ockham does think that our concept of humanity is singular and shared in the way that Chatton believes properties are. Ockham believes that although in the world outside my mind you have your humanity trope and I have my humanity trope, I do not have two humanity concepts, each corresponding to each of our humanity tropes. Instead, he thinks we only have one concept of humanity which applies equally to you and to me.

    Because the concept of humanity is singular and shared in the way that Chatton believes properties are, Ockham says that properties exist only in the mind. That is, Ockham’s claim is that the charecteristics of singularity and sharedness that Chatton ascribes to properties in the real world outside my mind — those characteristics only exist of concepts.

    In a nutshell, Ockham’s position is this: humanity tropes exist in reality, but not as concepts. Humanity properties exist as concepts, but not in reality.


  14. JT

    A little more on Ockham’s position.

    The property ‘humanity’ is the type of property that we would today call a ‘natural kind’. Natural-kinds are properties that we utilize to answer the question ‘what kind of thing is that?’ But the scholastics recognized other types of properties. For example, they also recognized properties of space and time, of mass, of performing an action and receiving an action, of relation, and so forth.

    The nature of each of these types of properties were hotly debated by the scholastics, just as the nature of natural-kind properties were debated. The issues here are complex, and I don’t really understand it all, so I’ll only talk about one aspect of the relations debate here.

    Suppose that you are taller than me, and let’s agree that we can say that you have the (relational) property being taller than me. When I speak of this property, what exactly am I referring to? What is the truth-maker of the phrase ‘you are taller than me’?

    Realists such as John Duns Scotus and Walter Chatton hold that your property being taller than me is really a property that inheres in you. If you can imagine that it is a little, invisible thread stretching between you and me (and if you can then imagine such a thing without spatial and temporal characteristics), then you have a rough idea of what Scotus and Chatton are saying.

    Relational properties for Chatton and Scotus are, as Ockham puts it, ‘little things’ that inhere in objects. So when I say ‘you are taller than me’, that phrase refers to the little thread that stretches between you and me. You actually having that little thread stuck to you is the truth-maker of the phrase.

    Ockham thinks this is absurd. The relational property being taller than me is not a ‘little thing’ that inheres in you. There is no little invisible thread stretching between you and me. Relations are not properties like that.

    Instead, Ockham would say that you are of such and such a height, I’m of such and such a height, and when we compare these two heights in our brains, we see that one is less than the other. The comparison (i.e., the actual relating of you to me) occurs in the mind.

    That’s all there is to it for you to be taller than me, there’s nothing further. There is no little thread that stretches between you and me which is the truth-maker of the phrase ‘you are taller than me’. Instead, your having a height that is greater than mine (when the two heights are compared by someone’s mind) is the truth-maker of the phrase ‘you are taller than me’.

    The same thing goes for relations of causal origin. According to Ockham, Peter’s relation being the son of Harold would not be a little thread stretching between Peter and Harold (whereas for Scotus and Chatton it would be). Instead, Harold exists at a certain place and time, and he produces Peter at a certain place and time, and so when we consider this, we can compare Peter to Harold as Harold’s offspring. And that’s all there is to it for Peter to be the son of Harold, there’s nothing further. The mere fact that Harold produced Peter is the truth-maker of the statement ‘Peter is the son of Harold’, not some little thread stretching between Peter and Harold.

    Relations are not, for Ockham, ‘little threads’ that exist in reality, outside the mind. They are nothing but comparisons we perform in our minds. Recall my description of natural-kind properties above. Ockham believes that natural-kind properties only exist in the mind, they do not exist outside the mind. The same thing goes here. Relations (little threads stretching between related things) only exist in the mind, they do not exist outside the mind.

    Notice that the statements ‘you are taller than me’ or ‘Peter is the son of Harold’ are still, on Ockham’s view, true. This is really just a debate over the truth-maker of those statements. For the realists such as Scotus and Chatton, these statements are true because there are ‘little things’ outside the mind which these statements refer to. Ockham rejects this. Instead, the mere facts of you being taller than me and Peter being the son of Harold are what the statements refer to.

    So when Ockham denies that properties and relations exist (in reality outside the mind), he’s not saying that you and I are not humans, nor is he saying that you aren’t taller than me. He’s simply saying that what Scotus and Chatton posit as the truth-makers for those statements — those kinds of things do not exist (outside the mind), they only exist in the mind.


  15. JT

    Whew! I got Ockham’s position out. So finally, to answer your excellent questions, Vince.

    1. Both the realists like Chatton and nominalists like Ockham would agree that Guernica 1 and Guernica 2 are distinct objects. For both realists and nominalists, objects are always individual (and therefore distinct from each other).

    The debate is about whether their properties are distinct. Recall the earlier question: if we consider the humanity of you and me, is there one humanity or two? Similarly, if we consider the being Guernica ( Guernicality) of G1 and G2, is there one Guernicality or two?

    If you’re a realist like Chatton or Scotus, you’d say that Guernicality is a property, so it’s singular, shared by G1 and G2, without being divided between them (just as humanity is singular, shared by you and me, without being divided between us).

    If you’re a nominalist like Ockham, you’d say Guernicality is just a trope, and so G1 has its own Guernicality trope, and G2 has its own Guernicality trope. (This makes it easy to see why Ockham has no problem thinking God could create Guernica without Picasso. Guernica is just an object with the trope being Guernica.)

    2. As for the relations, you’re right that relational properties are for Ockham just non-physical tags. But they are still true tags. For Ockham, this relation not being identical with would not be an invisible thread stretching between G1 and G2. The statement ‘G1 is not identical with G2’ is true simply because G1 and G2 are not identical, not because the statement refers to some invisible thread (and the same goes for G1’s and G2’s spatial and temporal relations).

    3. As for the Buber thing, Ockham would undoubtedly find Buber’s position strange. Since Ockham doesn’t think relations exist in the world outside my mind, it would be hard for Ockham to believe that, say, Guernica is actually just a collection of relations.

    But again, the state of affairs which we indicate by relational statements are still, on Ockham’s view, true. It’s just that they’re truth-makers are not invisible threads. They’re facts about the way that individuals exist together.

    So could we read Buber as an Ockhomian (I just coined that word)? Might there be a way to find a Buberian-Ockhomian utopia? What do you think?


  16. Huenemann Post author

    There’s a place just outside Nibley where we can form a Buber-Ockham utopia, if anyone’s interested.

    JT, your explanation of Ockam’s nominalism vs. Chatton’s realism are really clear and helpful. Thanks!


  17. Huenemann Post author

    Vince – I’m surprised you, as a physicist, think that scientists have followed Ockham’s lead on the reality of relations. My poor understanding of relativity physics is that the only thing remaining constant among all frames of reference is the space time interval, which is just a relation among events (i.e., whether the events are connectible through time, space, or light). And I would think that the Copenhagen interpretation treats the reality of a particle’s property in relation to its measurement.

    Don’t envy the Wittgensteinians; they just blame language for the problems they can’t think through!


  18. Mike

    I haven’t learned my Wittgensteinian lessons well enough. I just keep talking and can’t seem to pass over much of anything in silence.

    On the language point I think Charlie used to say Wittgensteinian method can get you to resolve some philosophical difficulties but it requires that you actually do the work linguistically to show where the problem exactly lies. Once that process is through Charlie thinks genuine philosophical problems may still exist that cannot be resolved in that manner. I love speaking for Charlie. :) That point of view on that particular philosophical claim seems fair enough to me.

    My point of view is a weird conglomeration of Wittgenstein, Pragmatism and some sort of (mostly Nietzsche and Kierkegaard inspired) existentialism where I want to “ground” the world or trace the beginning of the world back to the day I was born (more likely the day I “woke up”) rather than come up with some sort of theory of everything that would demand a God’s eye view. But even that explanation of how I see things isn’t nearly as accurate as just saying I value method and approach over particular philosophical claims. I’m much more likely to pick friends who seem to approach problems in similar ways to me than pick friends who share my particular beliefs at any given time.

    In philosophical conversation I think I’m a bit different too in that I want to try to represent my own view so that I can be wrong. I can try to discuss a lot of other points of view but I just love being wrong so much that I try to keep pushing my own.


  19. Mike

    What would be nice is to have a topical thought map where we could place thinkers and learn which are most relevant to our particular interests. At least for me I get a need to figure out a particular thought and get as much info as I can on that subject. A necessary component of philosophical therapy.

    I wouldn’t mind going through a few of free-will/determinism essays and see how those correlate to scientific approaches. As if I have a lot of time at the moment…


  20. JT

    Good stuff Vince.

    Ockham would agree that things are defined by their relationships, or at least by certain relationships. I have certain characteristics (being funny, wishing to be a ninja all the time, being a former drum player, etc.). Since I ‘have’ or ‘exemplify’ these characteristics, we can state my relationship to these characteristics with a relation: I am related to characteristic x by the relation having characteristic x (call it RC for ‘Relation to a Characteristic’).

    (If you’re a realist like Scotus, you think that RC is a little thread stretching between me and my characteristic, and that little thread is the truth-maker of the phrase ‘I have characteristic x‘. If you’re Ockham, there are no such little threads, so the truth-maker of the phrase is simply the fact that I am x.)

    To a large degree, my RC to these characteristics defines my personal identity. And if you think about it, most of these characteristics are relational characteristics. So, for a scholastic like Scotus or Ockham, much of my personal identity turns out to be determined by relational characteristics.

    I think Buber would probably agree with this much. It seems we’re working towards our Buberian-Ockhomian utopia after all!


    There are two places Ockham would disagree with Buber.

    (1) If Buber wants to say that relations are little things, little invisible threads, then Ockham would disagree. But is it necessary to read Buber as a realist in this way? What do you think?

    (2) If Buber were to say that objects depend ontologically on relations, then Ockham would disagree. Ockham argues that relations depend ontologically on their relata. If there is a relation, then there must be two things to be related in the first place. So the two things are ontologically primary, and the relation depends ontologically on them. It’s not the other way around. But again, is it necessary to read Buber as saying that relata are ontologically dependent on their relations? I don’t know. It’s been too long since I read Buber. What do you think?


    Also, you mentioned that electrons are only defined by their action in relation to other things. This is just a modern version of a standard scholastic (and Aristotelian) principle: we know things by their actions. And since for Aristotle actions are always in relation to the recipients of those actions (Aristotle defines the action and the recipient of the action as ‘action’ and ‘passion’), it follows that we only know things by their relation to other things. That’s standard Aristotelian (and scholastic) epistemology. It’s amusing to me that this still applies in science.


    Miguel, I of course agree with much of what you say about method. I far prefer, even academically/professionally, acquaintances who are willing to think carefully through an issue rather than acquaintances who agree with me. Hell, I don’t really know what I think anyways, so it’s hard to find people who agree with me.


  21. Kleiner

    I’ve been out of town, so have missed out on this great discussion!

    At least on my reading (one that is admittedly influenced by my affinity for Heidegger and Levinas), I do think that Buber would insist that “relata are ontologically dependent on their relations”. Buber is on the front edge of the deconstruction of “substance metaphysics” – a metaphysics that insists that substances are self-identical/independent entities whose relations are accidents.

    I think Buber would argue that relations are constituitive. His is a relational ontology. Indeed, without relation there would be nothing to speak of at all! But I don’t think he would want to say that relations are “little things”. Again, I think he would resist that tendency to think in terms of independence (the usual way in which we think of “things”).

    For those that don’t know Buber as well, I think Kierkegaard anticipated much of what is in Buber. Recall that for Kierkegaard the self is something that arises out of relation (see Sickness Unto Death).


  22. Charlie Huenemann

    So, JT, can you say a bit more on how Ockham regards relations? Above you said he denied they are extra-mental entities; so I would guess that the reality of a relation is in the mind of the observer (or contemplator). But you also have said that he’;d agree that objects are ‘defined’ by their relations (though not ontologically dependent on them).

    I probably need to think harder about relations on my own. I think Mercury is closer to the Sun than Earth is, and I think that will be true even after all the thinkers are extinct. Does this commit me to little invisible threads?


  23. jtpaasch

    Hmmm…..what more to say about Ockham’s take on relations. Maybe I can get a little more technical on this. I’m not sure how else to explain it (the medieval debate on relations is pretty complicated, as is the modern debate).

    Some predicates can be true of only one subject. For example, I can use the predicate ‘pale’ of John: I can say ‘John is pale.’, without the need to consider another subject. But other predicates cannot be true of only one subject. They require two predicates. For example, I cannot use the predicate ‘taller than’ of John: I cannot say ‘John is taller than’, and then just stop the sentence there. I have to include another subject, e.g., George: ‘John is taller than George’. So ‘____ is pale’ is a one-place predicate, because we make it true by filling in the one blank, but ‘_____ is taller than ____’ is a two-place predicate because we can only make it true if we fill in both blanks. (We can have n-place predicates too, since some predicates can only be true of n predicates, but I’ll just talk about two-place predicates for simplicity.)

    But at this point, the question becomes: why exactly do two place predicates require two subjects, while one-place predicates require one subject?

    One fairly intuitive answer is to say that just as things have properties which make one-place predicates true of them, so too do things have properties which make two-place predicates true of them. Consider one-place predicates first. It seems entirely intuitive and natural to say that one of John’s features or properties is that he is pale, and because he has that property, the statement ‘John is pale’ is true. And since this is so intuitive and easy to understand, why should we not just say the same thing for two-place predicates? Why not just say that some properties make two-place predicates true in the same way?

    So let’s picture exactly what that would look like. First consider the properties that make one-place predicates true. The one-place predicate ‘being pale’ is true of John because the predicate refers to a property ‘being pale’, which sticks to John. This makes a one-place predicate true because it sticks to one thing (namely, John). Now transfer this over to properties that make two-place properties true. It would follow that a two-place predicate like ‘is taller than’ is true of John and George because the predicate refers to a property ‘being taller than’, which sticks to John and to George. This makes the two-place predicate true because it sticks to two things (namely, John and George).

    Of course, it’s hard to picture this. It’s easy to picture John’s paleness, because we can just imagine a sort of free floating white patch. But paleness is visual in the first place, so it’s easy to imagine. Non-visual things like ‘being taller than’ aren’t like that, so it’s hard to imagine ‘being taller than’ floating free from John and George like we can imagine a white patch floating free from John. The best we can really do is to imagine it as if it were visible, and so we imagine a sort of invisible thread stretching between John and George.

    But nevertheless, the point is that one-place predicates are so easily explained by saying there is a property that sticks in one thing, and so we should just use the same model to explain two-place predicates: they are explained because there is a property that sticks in two things. Today, we call those properties that stick in one thing (or better: require only one thing to be instantiated) ‘monadic properties’, and we call those properties that stick in two things (or better: are instantiated by two things jointly) ‘polyadic properties’. This latter term is just a technical name for ‘relations’.

    By admitting polyadic properties, we now have an easy way to explain two-place predicates (relational predicates). If you ask me what is the truth-maker of ‘John is taller than George’, I can just say that there is a property (‘being taller than’) which sticks in both of them, stretching between them, and that is why we have to predicate the phrase ‘is taller than’ of two subjects. After all, it sticks in two subjects, not one. And this is all just like how ‘John is pale’ works. The property ‘being pale’ sticks in one subject, so we have to predicate the phrase ‘is pale’ of only one subject. Nice and easy, and pretty intuitive.

    This is, essentially, the realist solution for people like Scotus. It is, I think, guided largely by the intuition that a feature or property of something is what makes statements true about it.

    The nominalist position for Ockham is to reject that polyadic properties exist outside the mind. We can accept monadic properties, but not polyadic properties. We should not say that there is some property (‘being taller than’) stretching between John and George. Instead, John has his own monadic property ‘being such and such a height’, and George has his own monadic property ‘being such and such a height’, and our minds compare these two properties to see that one is greater than the other. That’s what explains the phrase ‘John is taller than George’.

    To put this in a more technical way, the truth-maker of a two-place predicate is not one (polyadic) property which stretches between two things. It is two (monadic) properties, each instantiated by their own thing, and then a mental comparison between those two properties. The only thing that ‘stretches between two things’ is the mental act comparing the two things.

    I think the nominalist position is guided largely by a desire for ontological parsimony. Ockham wants to find a way that we can explain the truth of two-place predicates without positing unnecessary extramental entities. To explain two-place predicates, Scotus posits both the two things, their individual properties, and the property (the relation) that stretches between them. Ockham says we don’t need this extra step, because we can explain it simply by appealing to the two things, their individual properties, and a mental act of comparison.

    (Additionally, one of Ockham’s great arguments against Scotus is this. If there are polyadic properties, then how would I explain my spatial relation to the rest of the universe? I would have an infinite (or at least very large) number of polyadic properties instantiated jointly by me and every point in space to which I am spatially related. Additionally, when I move, all these polyadic properties would be destroyed, and I and every point in space to which I am spatially related would jointly instantiate an infinite (or at least very large) number of new polyadic properties. Talk about multiplying extramental entities beyond necessity!)

    While Ockham’s position might have ontological parsimony, Scotus’s position has explanatory parsimony. On Scotus’s view, a statement like ‘John is produced by George’ is straightforwardly true. The subjects directly refer to John and George, and the predicate directly refers to a property stretching between them. On Ockham’s view, the statement needs some re-phrasing to state it more accurately. Ockham will want to re-phrase the statement to read something like ‘At some point in time, George caused John to come to exist’. This includes a lot more information than the original statement (including temporal stuff), and one might get suspicious that this kind of drastic re-phrasing of a statement just to make it literally true smells of a desperate attempt to ‘explain away’ the obvious.


  24. jtpaasch

    Oops, forgot to answer your question Charlie. =)

    So my answer is: no. Saying that things are really related in the ways we say they are does not commit one to little threads. It just means we have to do a little work to say why the statement is true.

    For the realist like Scotus, as I said, a relational statement is true because there is a little thread that makes it true. To put this another way, the relational statement expresses true things about the little thread.

    For the nominalist like Ockham, a relational statement is true because the mind can compare certain properties or features of the relata. To put this another way, the relational statement expresses true facts or comparisons about the relata.


  25. jtpaasch

    As for the Buber and substance/relation metaphysics thing.

    The problem I find with trying to compare modern (continental) claims about relationality with classical philosophical claims about relationality is this: the terms we use today are often loaded with extra little meanings that aren’t necessarily present in the classical discussion, and this leads to misconstruing the classical position pretty badly.

    For example, it is common to say the self arises out of relation. But what do we mean by ‘self’? This seems to me a highly specialized term that entails, among other things, self-awareness, consciousnesses, intentionality, and so forth. The idea seems to be that if an object exhibits these features, then we can call it ‘a self’.

    But the scholastics agree that these self-features are relational. For example, most scholastics argue that one cannot have intentionality without an object (of intention), so intentionally ontologically requires an object (of intention). If self-features are constituted relationally, and if ‘a self’ is anything which has self-features, then everybody agrees that ‘a self’ is dependent on the relations which constitute ‘the self’.

    Another example is the use of the word ‘constitutes’ (as in ‘a self is constituted by relations). Here, ‘constitute’ seems to mean something like ‘is an essential or necessary condition/circumstance/requirement’. If this is right, then if we say that the self is constituted by its relations, we really mean that those relations are essential or necessary conditions or requirements of being ‘a self’. Intentionality, to stick to our example, cannot occur without an object, and so being related to an object intentionally is an essential or necessary condition for intentionality. There can be no intentionality without that relation.

    But again, the scholastics completely hold to this. They think there are plenty of polyadic properties (such as intentionality) that things have which are essential and/or necessary for a thing to be the thing it is. A human, for example, can’t be human without intentionality. And the scholastics all argue that these sorts of essential polyadic properties constitute the thing (precisely because that thing cannot be without those polyadic properties).

    Kleiner mentions that the relational-metaphysics people are reacting to ‘substance metaphysics’, which is basically a rejection of the claim that relations are accidents. I think this is dead on the money. From what I’ve read in people like Buber, Heidegger, Merleau-Ponty, and Levinas (and possibly Derrida), this is exactly what’s going on.

    The insight here seems to be that (some) relations are not accidents, so anybody who thinks relations are accidents must be wrong. We could bring up intentionality again to make the point: intentionality cannot be intentionality without being related to an object, so ‘being related to an object’ cannot be accidental. It is, rather, essential and necessary for intentionality to exist at all.

    But I don’t know any scholastic (or classical philosopher) who argues that relations are only accidents. The scholastics believe there are plenty of relations which are essential and constitutive of certain kinds of substances (for example, intentionality, consciousness, or even existence itself — since the scholastics thought we were all created such that existence itself is constituted by a relation to a creator). So I’m not really sure who it is that the relational-metaphysicians have in mind when they react to ‘substance metaphysics’.

    Additionally, there is a very basic sense in which all of Aristotle’s metaphysics is relational. For Arisotle, everything comes to be and passes away according to act and potency. A lump of clay, for example, is potentially a statue, and when someone shapes it into a statue, that potentiality is actualized. But for Aristotle, act and potency are relational (Notice that act and potency are two-place predicates). So for Aristotle, nothing at all can even exist without being related.

    Yeah, I suspect there’s just a lot of misunderstanding in the substance/relation metaphysics stuff.


  26. Mike

    Additionally, one of Ockham’s great arguments against Scotus is this. If there are polyadic properties, then how would I explain my spatial relation to the rest of the universe? I would have an infinite (or at least very large) number of polyadic properties instantiated jointly by me and every point in space to which I am spatially related. Additionally, when I move, all these polyadic properties would be destroyed, and I and every point in space to which I am spatially related would jointly instantiate an infinite (or at least very large) number of new polyadic properties. Talk about multiplying extramental entities beyond necessity!

    Brilliant. I’m tempted to adopt Scotus’s view so that I can think of the world in that way. I would also like to tell people “hold on a sec, my polyadic properties need to be jointly instantiated”.

    Funny that the use of “pale” as an example so readily brings up questions about scientific vs common sense realism and makes color skepticism a relevant topic.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s