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.
We're going to be talking about skepticism in medieval philosophy. And this seems like a kind of surprising topic. It would be easy to assume that medieval thinkers don't really get into the issue of skepticism because they have very confident beliefs, religious beliefs and other beliefs about metaphysics, the nature of the soul and all sorts of things. And we might suppose that it's really with early modern philosophy, with figures like Descartes, that skepticism comes back into the history of philosophy after playing a role in antiquity with the ancient skeptics. So what reason is there even to think that medieval philosophers have something to say about the topic of skepticism?
First of all, we should take into account that medieval thinkers were familiar with ancient sources. In particular, they knew quite well the academic tradition. They read Cicero's Academica and Augustine's Contra Academicos. That's why they were familiar with many skeptical scenarios, say sensory illusions, the dream hypothesis, or even the hypothesis that a divine being could intervene and manipulate our cognitive states. So, that gave rise to many skeptical debates. They started in the 13th century; Henry of Ghent is a famous person who read Augustine and commented on all the skeptical cases presented in Contra Academicos. Then we shouldn't forget that it was in fact the Christian tradition that gave rise to new types of skeptical scenarios. Let's take the example of demons, the so-called fallen angels, that could intervene and in a way tamper with our brain so that they create sensory images of things that do not exist. So, how can we know that we do have reliable cognitive states? Why couldn't it be the case that right now a demon is intervening and in a way changing something in my brain? In addition, Christian thinkers appealed to divine omnipotence, and they understood omnipotence in two ways. First of all, God can use his power according to the laws of nature, but second, he can also in a way ignore these laws and do whatever he likes. That was called the potentia absoluta. And by using this unlimited power, God can do everything, including everything that concerns our cognition. So, why couldn't it be that God creates in me a thought about a thing that doesn't exist while I mistakenly believe that this thing exists?
Actually, that sounds a lot like Descartes' evil demon hypothesis. A demon could, for example, be inserting a belief into the listener's mind that they're listening to a podcast, but actually they're not listening to a podcast and God could do the same thing, presumably.
In fact, and it was exactly this hypothesis that was thoroughly discussed in the 14th century. But that doesn't mean that skeptical thinkers drew the same conclusions as Descartes. Why not? First of all, they made the same remark as Descartes later made in the first meditation. God is a benevolent creator. That's why we shouldn't think that he wants to deceive us. So, all medieval thinkers said that in principle, God could use his unlimited power to deceive us, but since he's a good God, he does not want to deceive us. That's why they thought, well, if it's not God, then it's perhaps one of these demons that can intervene. And then they asked the question ‘but how far can it go?’ So, is it really true that a demon can do everything that concerns my intellect, or can it intervene on a lower level only, say on the level of inner senses, so that the demon can perhaps manipulate my sensory images, but not the way I abstract something from these sensory images? That's why they made a limited use of this skeptical scenario.
And maybe it's worth saying here that what we're talking about, I guess, is going to be uses of skeptical arguments or hypotheses. We're not actually going to find medieval philosophers who are full blown skeptics the way that ancient skeptics were skeptics or even the way that, say, David Hume is a skeptic.
Indeed, I think we should clearly distinguish between the methodological use of skeptical arguments and a skeptical position. That is, medieval thinkers were convinced that we do have and can have reliable knowledge, but they tried in a way to test our knowledge claims and said ‘but are you really certain that you have knowledge of material things or of supernatural things?’ And that's why they introduced skeptical hypotheses. But that was a methodological use, and we should not in a way think that they denied the possibility of knowledge completely. In addition, we should also distinguish between unlimited skepticism and what would be called local skepticism. That is, skepticism that concerns some types of knowledge claims, say about causal relations or about a certain range of entities, say material substances or whatever.
Or you might be skeptical about your senses, but not about your mind or something like that.
Exactly, exactly. And since the medievals worked with a hierarchy of faculties of the soul, they thought that you could be skeptical about the senses because they might not be reliable. Since we are not perfect beings, we don't have perfect senses. But nevertheless, we have an intellect that is a higher faculty, and, on that level, we can perhaps reach perfect knowledge.
Actually, one of the things I wanted to ask you was exactly about this, because they have these epistemological theories inherited mostly from the ancients, if you count Augustine as ancient, which I would. So, for example, you have the basically empiricist epistemology or theory of knowledge from Aristotle. You have the more illuminationist model of how we get knowledge from Augustine. And I'm wondering whether these positive theories of how we acquire knowledge actually themselves gave rise to discussions about skeptical hypotheses.
Yes, in fact, they did give rise to skeptical hypotheses. Let's start with Aristotle, who claims, to summarize it briefly, that we first get sensory properties from material things, and then we form sensory images. Let's take an example, I see the table in front of me. So, I get the color or the size and the shape of the table, and then I form some kind of image of the table. Then Aristotelians claim that we are able to abstract something like the essential features of the table. And it's exactly at this point that some thinkers ask skeptical questions. Why, they ask, can we abstract essential features? What enables us to do that? And how can we be sure that we abstract exactly the essential features that are in the table? Couldn't it be that these essential features are not accessible to us, or that we don't have the cognitive means to reach them? So that's why they asked ‘can we really go beyond the level of sensory images?’ And perhaps do we need external help in order to do that? The illuminationists, for instance, said that, well, a human being cannot just get everything out from sensory images, so we need illumination. God, in a way, needs to assist us so that we really get the essence of the thing. And others, with more skeptical motivations, thought that perhaps that's something we can never reach. Perhaps our knowledge is limited.
So, maybe there's three possible views there. The Aristotelian view, which is sense perception will get you all the way to universal, necessary understanding of essences. The illuminationist view taken from Augustine, which is that sense perception won't do that. It's a more Platonist view, I guess. So, you need direct illumination from God. And then the third view would be, well, maybe we don't actually get knowledge of essences. But, actually, it does seem like a genuine skeptical position. And this is actually a position held by some medieval philosophers.
In fact, that is a skeptical position, although I wouldn't say that they would really claim that we can never go beyond the sensory level, since they also had the Christian belief that, in a way, we are more or less equipped with reliable faculties. They would say that, well, perhaps we cannot get exactly the essential features as they are, but we can get something out of the thing. And if we repeatedly go through a process of abstraction, then in the end, we can have at least an approximation. So, they didn't take the full-blown skeptical position.
You can get close enough, basically.
Exactly. And let's add another argument they often use. They said: if you start, as Aristotle says, with the grasp of sensible properties, then you have only access to properties. I mentioned the color of the table or the size. But how do I get the substance of the table? And this, again, was a source of skepticism, because some people said: well, if the properties are not identical with the substance, if they are—as many, especially in 14th century, said—accidents that are really distinct entities, then we get only these entities through this empirical process, but never the substance in itself. So, perhaps even if we have reliable faculties, our knowledge is very limited. We can know about color, size, and shape, but not about the thing itself that has color, size, and shape.
It strikes me that part of the problem there is that the ancient views of knowledge had set the bar very high for what would count as scientific knowledge or understanding. Aristotle thinks that understanding is necessary, universal, essential. So, Aristotle would reject as a kind of really good example of knowledge, he would reject something like ‘the microphone is sitting on the table,’ which nowadays seems like a very good example of knowledge. So, maybe part of what's going on here is that the medievals are thinking, well, the resources that we've been offered are pretty much satisfactory for understanding that there's a microphone sitting on the table, but they're not going to get us all the way to this kind of full-blown Aristotelian knowledge.
Definitely. And, terminologically, they distinguish between scientia, which would be science or knowledge in this full-blown sense, and mere cognitio, cognition. And most medieval said, well, perhaps we can never have full-blown knowledge, or if it's possible, only in some disciplines like mathematics. But that's even an exception. In most cases, we have just cognition, cognitio. And that's why they said, well, perhaps we shouldn't have this kind of very high standards. Perhaps we should be satisfied with cognition that is reliable in most cases, and limited to some types of things in the world, like properties, but not necessarily substances.
And that seems to, in a way, to prefigure what happens in early modern philosophy, where there are other thinkers who lower the bar a bit for what counts as knowledge. By the way, that contrast between scientia and cognitio, can we line that up with the contrast between knowledge and belief? Or is cognitio more like knowledge, but a lesser form of knowledge rather than merely true belief?
It depends on what you understand exactly by knowledge. If you say that knowledge is justified true belief, then cognitio is certainly not knowledge, because we don't have a justification just by having or acquiring a cognition. In most cases, it's just a belief, not even true belief, because most medievals would say that we can have wrong cognition. Why is that? Because we acquire cognition through a causal process, and many things can go wrong while we acquire a cognition. That's why a lot depends on how you acquire it. So, you first have a cognition, which is a mere belief, and then you can ask ‘how did I get it?’ And then you can perhaps say, well, I get it through this kind of process, so it must be adequate or correct. And then you can even ask ‘and how can I give a justification for what I have?’ So, I would say cognition is the first building block, and then you can go on and eventually reach knowledge in the sense of justified true belief, which is not yet scientia in the Aristotelian sense.
And to get scientia, you have to add something like justification based on first principles or...
Exactly, and the syllogistic order and whatever. So, demonstrative knowledge, as the medievals said.
Let me ask you something else about this causal chain from the object to my beliefs or knowledge about the things that are in the outside world. It seems like a lot of these medieval thinkers, again, in the Aristotelian tradition, are imagining that what happens is that through my sense perception, I get a representation of the thing in the outside world. So, look at the microphone, and the microphone is not in my mind. What's in my mind is a representation, like an image of the microphone. And again, in early modern philosophy, there's this very widespread problem discussed by figures like Kant, Berkeley, Hume. So, the worry is, do my representations accurately represent the world? And if so, how do I know that? Can I have confidence that my representations are actually reliable? Is that another kind of skeptical worry that appears in the medieval tradition?
Yes, it did appear in the early 14th century already. Why is that? Most Aristotelians before that would say that we acquire a representation because we get the forms that are in the external things. So, let's go back to my example. I get not just an assemblage of properties, I get the so-called sensible forms that are really out there. And on the basis of these forms, I also get the substantial form of the table, if everything works the way it should. So, that's why up to the late 13th century, many authors said that we assimilate the forms as they really are. And that's why they said, well, if everything works the way it should, if there is a reliable causal chain, then it's perfect. But Ockham, for instance, said, well, why do we claim that we really assimilate forms? Perhaps things just make impressions on our senses. And on that basis, we form some kind of concepts. Ockham thought that these are mental terms and then mental propositions. But we do not, literally speaking, get the forms as they are in the things out there. We get some kind of equivalent to what is out there in the world. And that's why, of course, many authors after Ockham then had the worry that what we have in our mind does not perfectly match what is out there in the world. So, the so-called problem of representation, as it was later discussed by Descartes and other authors, is a problem that was already detected in the 14th century, precisely because these authors gave up the assimilation theory.
So, are you saying that Ockham thinks that what happens in my mind is that I form something that has propositional content? So, what happens in my mind is linguistically structured in something like sort of a sentence, ‘the table is brown’, something like that?
Definitely. Ockham defends the thesis that we have a mental language in the strict sense. That is, that our mental concepts have a linguistic structure. They have not just a semantic content, but also a syntactic structure, so that we come up with mental sentences. And when we make affirmations, we form mental sentences. But of course, these sentences have a linguistic structure, and that's not what we find in the world itself. That's why you can then ask, well, ‘do these sentences really match what we find in the world out there?’ To be sure, Ockham was not a skeptic. He thought that we acquire all these concepts and, in the end, sentences in a reliable way, so that we have some kind of correspondence between what is in the world and what is in our mind. But nevertheless, of course, one could ask, and many authors after Ockham did ask this question, how we can be sure that our sentences match the things and properties in the world.
Right. Actually, it seems like there's two ways of thinking about this representationalist view, both of which are open to different kinds of skeptical worry. I mean, there's a sort of picture or image view, where I have a sense impression, and I guess maybe Aquinas might think something more like this. So, there's a sense, it's almost like a picture in the soul. And then there's the view you just described that we find in Ockham, where it's more like a sentence. And with the picture view, the worry would be, well, my representation is like an inaccurate portrait. So, it's like a painting of someone and it doesn't look like the person. And how do I know that my pictures look like the things? And then in the case of the Ockham view, the worry would be something more like, well, ‘how do I know the propositions are true?’ Because what I'm getting is the impression that the table is brown, but maybe it's not brown.
Definitely. We do have these two views. And I would say that Aquinas, of course, has not only the picture view, because he would say we have pictures on the sensory level, but in addition to that, we have something on the intellectual level. Aquinas thought that we can, as I already mentioned, abstract essential features. And there we get the additional worry: if we have something more than an image, what exactly is it? Is it really true that we can have, in addition to the picture, the substantial form of the table? What should that be? How could that form exist in my mind? So that was very puzzling. And if we then look at Ockham, who in fact defends the view that we have something linguistic in our mind, then we can first ask ‘but how do we get these sentences, given that we have just sensory inputs?’, ‘what enables us to form mental sentences?’ Is there some guarantee that there's a reliable process for acquiring these sentences? And once we have them, in what sense do they represent things out there in the world?
Is that one reason why he went for this more propositionalist idea about how sensation works that at least you don't have to somehow cross this bridge from something like a picture in vision to something like a sentence in the mind, the way that Aquinas would have said? Is that part of the motivation for the view?
I think that's one part of the motivation. Another part is also that Ockham had a different theory of the soul. Aquinas still thought that there's one single soul in a human being, so that we have different levels, that these levels, in a way, are part of one and the same soul. There is no gap between senses and intellect. Ockham, on the other hand, thought that there's a sensory soul and a really distinct rational soul. That's why he couldn't say that, in a way, the intellect can use what the senses present as an image. So, Ockham would say we get, thanks to our sensory soul, just a sensory input. And then the rational soul has to do something with that, but it's a distinct cause and power that does, to speak loosely, its own job. So that's why the rational soul has to produce sentences.
Right. Okay. So, when I grasp something, then I have to grasp it propositionally using my rational soul.
Exactly. Exactly. Exactly.
Let me go back to yet another avenue of skeptical argument, something you mentioned towards the beginning of the discussion, which is this worry that a demon or God could somehow induce a false belief in me. And I mean, obviously, you don't have to believe in demons to worry about this. The modern equivalent might be a sort of virtual reality kind of scenario where a mad scientist is putting false beliefs in me. So, what did the medievals have to say about this kind of hypothesis?
They took it very seriously, but tried to, I would say, not eliminate it, but neutralize it by appealing to some metaphysical principles. Let me mention two of these principles. The first could be called the principle of hierarchy: they would say, yes, demons can intervene, but they are not omnipotent. So, their power is limited. They may manipulate what is going on in the sensory soul, but they can never touch the intellectual soul. Let me give you an example. Perhaps a demon could right now create in me the sensory image of an elephant that is flying through this room. Why not?
Crazy image, but could be the case.
If that actually happens, tell me and I'll call the doctor.
Sure. But then my intellect would come into play and say ‘look, is it really possible that there are elephants that are flying?’, ‘can it be that an animal that is that heavy can fly through a room?’ So, my intellect would analyze what is presented to me in this image and eventually reach the conclusion that this is impossible. I ascribe properties to a thing that can never be in this thing. That's why I can eventually reject the idea that there is really an elephant flying through this room. So, the medievals would have said that it's possible to be deceived on the sensory level, but thanks to our intellect that can analyze what is presented on this level, we can in the end neutralize that. And that's why we are not deceived. Another principle they used was the principle of what they called natural evidence. Buridan, for instance, a 14th century author, used this principle when he discussed skeptical scenarios. And he said that, yes, it's possible in principle to be deceived by a demon, but given the natural evidence we have acquired in many cases, we can compare this case with ordinary cases, and test if it really fits into what we have experienced so far. And if it turns out that this is just an exception, then we can eliminate it. So, it's not a case that in a way one single case of possible deception would rule out the possibility of knowledge at all.
It sounds like they only really worried about a demon or God giving you a certain false belief on a certain occasion. They're not actually worried about this kind of virtual reality hypothesis where all your beliefs are systematically false. Is that right?
In fact, in fact, and that's where I see a crucial difference between medieval debates and early modern debates, because all the medieval authors, who were, of course, strongly inspired by Aristotle, thought that, in principle, we have reliable cognitive capacities. So, we might be deceived in exceptional cases, but not in a normal case. And we can trust our senses. And of course, we can also trust our intellect. Only later when people started to question the reliability of our cognitive capacities, did they come up with the radical skeptical hypothesis that all our beliefs might be wrong, because we could be deceived in each and every situation. So, it seems to me that's also another reason why the medievals were not full-blown skeptics, since they worked with the scenario that, in principle, we have reliable cognitive capacities, they didn't take seriously the hypothesis that general deception is possible.
Yeah, that seems like kind of a philosophical failing on that part, because once you've had the idea, and they talked about it very explicitly, that a demon could give me a false sensory experience, it seems very natural to worry ‘oh, gosh, maybe all my sensory experiences are being manipulated by a demon, right?’ So, is this just a lack of imagination on their part? Or is there something more principled behind it?
I don't think that it's just a lack of sophistication or something like that. They also had an additional reason, namely that if you think that you are wrong in each and every situation, you cannot give any explanation of what is going on in the world. So, the world becomes something that lacks coherence and your own attitude toward the world is a world of utter ignorance. But they thought, inspired again by the Christian background, that we have a well-ordered universe. So, there is order in the material world, but there is also order in our mental world. And that's why in a way, they thought it's unimaginable to think that we have just many beliefs that pop up in our intellect, but then that might be wrong or completely lead us astray. So, it seems to me that it's the idea of rational order that inspired them to reject a general skeptical hypothesis.
Yeah, actually, if you look ahead to Descartes, the way he eventually defeats the general skeptical hypothesis is to prove that God exists and then say, oh, God wouldn't let that happen. And so in a sense, all he's doing is making explicit a more tacit assumption that was guiding the medievals and leading them to only worry about local skeptical hypotheses and not global skepticism.
Exactly. I mean, Descartes is so interesting because he is still part of the scholastic world and partly also he belongs to the modern world. He's part of the scholastic world because he thinks exactly as he said that, well, yes, in principle, God could make us have all kinds of wrong beliefs, but since God doesn't want to deceive us, he doesn't do it. So that's the medieval part. But where he's no longer medieval is in his anti-aristotelian attitude, namely that we cannot take it for granted that we have reliable capacities. We need to give a proof for that. And that's why he does all he can in the Third Meditation to prove God's existence so that we have someone who then guarantees the reliability of our cognitive capacities. So, it seems to me that it's by giving up some kind of idea of cosmological order in general, but also order in a human being as far as the cognitive capacities are concerned, that enables Descartes to go one step further.