Exploring the Definition of Knowledge

So we're starting this philosophy series with the question, what is knowledge? Now, on the face of it, that might sound like kind of a boring topic to start with. You know, I don't think many people are kept up at night worrying about the correct definition of knowledge, like they might question such as whether God exists or what the meaning of life is. But the definition of knowledge is a good topic to start with because it illustrates a lot of how philosophy works. When you boil a lot of these philosophical questions down, they ultimately just become questions about definitions of things. So if you think about moral philosophy for example, a theory such as the use of the terrorism is essentially just a definition of what good is. And so by looking at this story of the various definitions of knowledge that the people pose over the years and the issues with these definitions, it's a good way to illustrate how this philosophical game is played. It illustrates the difference between a good definition and a bad definition. It shows us how we can critique philosophical definitions and theories. And it's just a good all-round introduction to philosophical reasoning in general. So let's do some philosophy. So before we get into the definitions of knowledge themselves, let me first explain how this game is played. The classic way definitions work in philosophy is with this idea of necessary and sufficient conditions. Now, of course, there's many philosophers who reject this entire conception of language, such as Wittgenstein. And this is probably more in line with my own personal view. But at least for the purposes of today, we're going to play the game according to these rules. We're going to try and... provide a definition of knowledge in terms of necessary and sufficient conditions. So according to these rules, a good definition will contain conditions where each individual condition is necessary for knowledge and taken together all these conditions will be jointly sufficient for knowledge. Now what that means in simple terms is that everything that's an instance of knowledge will fit our definition of knowledge and everything that fits our definition of knowledge will be an instance of knowledge. If you think about it in terms of Venn diagrams, what that looks like is two circles that overlap perfectly. If we include conditions that aren't necessary, then our definition will be too narrow because it will rule out things that are knowledge. If our definition includes conditions that aren't necessary for knowledge, then our definition is going to be too narrow. In other words, there's going to be things that are knowledge that our definition excludes. On the other hand, if our conditions aren't sufficient for knowledge, then it's going to have the opposite problem. Our definition is going to be too broad. Or in other words, our definition is going to say things are knowledge when they are not knowledge. So one final thing to mention before we get on to the definitions of knowledge themselves, and that's to specify the kind of knowledge we're defining, because there's many different types of knowledge. So you've got ability knowledge. that's knowledge of skill such as knowing how to ride a bike or knowing how to juggle. When we're talking about the definition of knowledge we're not talking about the definition of ability knowledge. You've also got acquaintance knowledge which is knowledge of a person or a place so in the way you might know your home city well or you might know your best friend well. When we're talking about the definition of knowledge today we're not talking about acquaintance knowledge either. Instead when we're talking about the definition of knowledge today we're talking about the definition of propositional knowledge specifically. Now propositional knowledge is knowledge of facts and it's sometimes called descriptive knowledge. So my knowledge that I know that Paris is the capital of France or I know that two plus two equals four. These are examples of propositional knowledge and so when we're talking about the definition of knowledge going forward we're talking about the definition of propositional knowledge specifically. So the classic definition of knowledge then. is justified true belief and this is sometimes called the tripartite definition because there's three parts justified true and belief and this this definition originated all the way back in Plato's dialogue Theotetus written I think in the fourth century BC and this definition of knowledge this tripartite definition was generally accepted for over 2,000 years to provide both the necessary and sufficient conditions for knowledge. On the face of it, the tripartite definition works quite well. All of these conditions do appear to be necessary. Obviously, knowledge requires it to be true. If you say I know the moon is made of green cheese, then you wouldn't consider that knowledge because it's not true. So truth is necessary for knowledge. Likewise, belief. It just wouldn't make sense if you said, I know that Paris is the capital of France, but I don't believe that Paris is the capital of France. That just doesn't make sense. So belief is also considered to be a necessary condition of knowledge. Finally, justification. It's this justification condition that that differentiates knowledge from just a lucky guess. So if I were to ask you how many moons does Pluto have? Now unless you're an astronomy nerd or whatever you probably don't know the answer but if you just had a strong gut feeling that the answer was five you would be right but your belief that Pluto has five moons wouldn't be knowledge because you don't have a justification for it. So justification also appears to be necessary for knowledge. So justified true belief appears to work quite well as a definition of knowledge. Each of these conditions appears to be necessary for knowledge. Well, actually, I mean, there's some philosophers who argue that some of these conditions aren't necessary. Like you can get into wacky definitions about truth or describe weird scenarios where somebody knows something, but they don't believe it. Maybe it's subconscious or whatever. But generally speaking, justified true belief is considered to be. necessary for knowledge. The juicy criticism, and I'm of course talking about Getty Act cases, is arguing that the tripartite definition is not sufficient for knowledge. So like I mentioned earlier, the tripartite justified true belief definition of knowledge dated all the way back to Plato in the 4th century BC, and this definition basically went uncontested for over 2000 years. That is until 1963 when Edmund Gettier described two scenarios now known as Gettier cases where a person has a justified true belief that isn't knowledge. In other words Gettier described two scenarios that show justified true belief is not sufficient for knowledge. The first Gettier case goes like this. Two guys Smith and Jones are interviewing for the same job. Smith hears the interviewer say, I'm going to give Jones the job. And Smith also sees Jones count 10 coins from his pocket. So based on this, Smith forms the belief that the man who will get the job has 10 coins in his pocket. So they go through the interview, but Smith gets the job and not Jones. And then Smith looks in his pocket and by pure coincidence, Smith sees he also has 10 coins in his pocket. So Smith's belief here was justified. It was justified by the fact that he heard the interviewer say he was going to give Jones the job. And it was justified by the fact that he saw Jones count 10 coins from his pocket. What's more, Smith's belief was true. The man who got the job... did have 10 coins in his pocket. The thing is though we don't want to say Smith's belief he accounted as knowledge because it was just luck that he was correct. Smith's true belief was formed on the basis that Jones would get the job and the fact that Jones had 10 coins in his pocket. It was just luck that Smith himself had 10 coins in his pocket. This shows that the justified true belief definition is too broad because Smith had a justified true belief but he didn't have knowledge. This isn't one of Getty Air's own examples, but it illustrates the same principle. So imagine you hear your phone call. You pull your phone out and you see you've got a message, but you also see that your phone was on silent and actually actually the noise you heard was somebody else's phone who has the same ringtone. So here, your belief that you had a message was justified by the noise you heard, and it was true, you did indeed have a message. But again, the justification was irrelevant to the truth of the belief. It was just lucky. So again, this is another example of a justified true belief that isn't knowledge. Gettyer's 1963 paper is about as groundbreaking as it gets in philosophy, which maybe isn't saying very much, but anyway. And since 1963, various philosophers have tried to propose new definitions of knowledge that avoid Gettyer cases and patch up the definition of knowledge such that it doesn't include Gettyer cases. So the first post-Gettier definition we're going to look at is called the no false lemmas definition. And what this does is it just adds an extra condition onto the tripartite definition. So the new definition becomes knowledge is justified true belief that is not inferred from a false lemma. A false lemma is essentially just a false belief or a false assumption in an argument. So going back to the original Gettier case, we can correctly say that Smith's belief does not count as knowledge because it was derived from the false lemma that Jones would get the job. Even though Smith's belief was true and justified, the no false lemmas definition would say it's not knowledge because the justification was based on the false lemma that Jones would get the job. So by adding this extra no false lemmas condition, we narrow the definition of knowledge and we rule out Gettyer cases. So it's an improvement. However, this no false lemmas definition didn't last very long before people started coming up with similar counter examples to the Gettyer case where somebody has a justified true belief that is not inferred from a false lemma but that nevertheless isn't knowledge. So the classic example used to illustrate this is fake Barn County. Imagine a place called Fake Barn County where the people create fake barns that look identical to real barns. A guy called Henry is driving through Fake Barn County but he doesn't know the locals do this. Henry often thinks that is a barn when he looks at the fake barns. And these beliefs obviously aren't knowledge because they're not true. The barns are fake barns. However, there is one real barn in Fake Barn County and when Henry looks at it he forms the exact same belief that that is a barn. In this case his belief is true, it's also justified by his visual perception and most importantly it's not inferred from a false lemma. So here Henry's belief is a justified true belief that isn't inferred from a false lemma. But like in the original Getty case we don't want to say Henry's belief here counts as knowledge because it's just luck that he was correct in this one instance. After all every time he was looking at the fake barns he was forming the exact same belief and so he doesn't have the ability to tell real Barnes from fake Barnes. Yet the no false lemmas definition says Henry's belief here counts as knowledge. So what this shows is that the no false lemmas definition is still too wide, a bit like the original tripartite definition. In other words, the no false lemmas definition is not sufficient either. So we saw that the justified true belief definition wasn't sufficient for knowledge. And while adding this no false lemmas condition improves this definition, the no false lemmas definition isn't sufficient for knowledge either. So perhaps Descartes can give us a better definition of knowledge. Well, Descartes doesn't really intend this as a definition of knowledge per se, but his approach to knowledge in meditations rules out both Gettyer cases and fake Barn County scenarios. For Descartes, something only counts as knowledge if it's absolutely certain, if it's rationally impossible to doubt. And so this approach is sometimes referred to as infallibilism. Applied to the fake Barn County case, there's all sorts of reasons why Henry could potentially doubt his belief that that's a barn. For example, Henry's eyesight might not be 100% accurate, or perhaps Henry's mistaken somebody's house for a barn, or maybe Henry's actually in bed and dreaming that he's looking at a barn and so his belief is false. Or maybe Henry's taking a bunch of drugs, I don't know, and he's just hallucinating barns. In any case, these are all potential reasons for Henry to doubt his belief that that's a barn. And while these might not seem particularly likely, they're at least possible. And if they're possible, they're potential reasons for Henry to doubt his belief. So according to infallibilism, Henry's belief doesn't count as knowledge because there are potential reasons to doubt it. Henry's belief there is a barn is not certain. It's a similar thing with the Getty case. There's all sorts of reasons why Smith could doubt his justification for his belief. For example, Smith could have misheard the interviewer, or again, Smith could be dreaming or hallucinating. And so again, infallibilism would say this isn't knowledge. And so at first glance, infallibilism seems to be an improvement on both the justified true belief and no false definition. Because where these... These definitions aren't sufficient. They say that Henry's and Smith's beliefs are knowledge when they're not. Infallibilism correctly says that Henry's and Smith's beliefs aren't knowledge because they're not knowledge. However, whereas the justified true belief and no false lemmas definitions were too broad in their approach, they're not sufficient in philosophical terms. Infallibilism faces the opposite problem. It basically rules out anything as knowledge, and so it's too narrow in its definition of knowledge. It seems obvious that I do know all sorts of things. For example, I know that 2 plus 3 equals 4, that Paris is the capital of France, and that Plato was a Greek philosopher. But There are all sorts of reasons why I could potentially doubt these beliefs. For example, I could have been the victim of some wacky conspiracy where everybody was lying to me about Plato being Greek and actually he was British or something. Now, even if that doesn't seem particularly likely, it's at least possible. And so it's a potential reason to doubt my belief that Plato was a Greek philosopher. And so according to infallibilism, I don't know this. We can come up with weird scenarios to doubt pretty much every belief. Even something as fundamental as 1 plus 1 equals 2. You know, it's logically possible that an evil demon is just manipulating your mind and making you think that 1 plus 1 equals 2, when actually 1 plus 1 equals 3. Now, again, I'm not saying that's particularly likely, but it's least possible. It's a potential reason to doubt your belief. And so according to infallibilism, you don't know that 1 plus 1 equals 2. But this just seems ridiculous because clearly we do know all sorts of things. And so if we're to... Maintaining any reasonable understanding of knowledge, we're going to have to reject the infallibus definition. It seems that certainty or the impossibility of doubt is not a necessary condition of knowledge. The next contender for a definition of knowledge is virtue epistemology. Now, this is a broad category like infallibilism, but In general, virtue epistemology definitions draw a parallel between moral virtue and intellectual or epistemic virtue. So let's start with the case of moral virtue first. An act of moral virtue achieves a virtuous outcome. So let's say you help an old lady across the road. Now that is a virtuous thing to do, but is it an act of moral virtue if the only reason you helped the old lady across the road was because you thought she'd give you some money? Presumably not. And so for something to be a proper act of moral virtue, it has to not only achieve a virtuous outcome, but it has to achieve the virtuous outcome because of the moral virtues of the person committing the action. Virtual epistemology says a similar thing about knowledge. It's not enough for a belief to achieve a virtuous outcome, i.e. be true. It has to achieve this virtuous outcome because of the intellectual virtues of the believer. So intellectual virtues are things like caring about the truth, thinking rationally and so on. And so applied to the Gettier case we could say that Smith's belief while true and while achieving a virtuous outcome is not an act of intellectual virtue because the truth of the belief is not a result of Smith's intellectual virtues. The truth of the belief was a result of luck. As such virtue epistemology would correctly say that Smith's belief doesn't count as knowledge. In the same way that helping the old lady across the road to get money wasn't a proper act of moral virtue, Smith's belief that the man who gets the job has 10 coins in his pocket, while true, wasn't a proper act of intellectual virtue because it wasn't true for the right reasons. Nailed it. Ernest Sosa illustrates this idea of intellectual virtue with a nice analogy to do with archery. So what is a good shot in archery? Well, firstly, a good shot in archery is accurate. It hits the target. But an accurate shot in archery doesn't automatically mean a good shot, because the arrow might have just hit the target through luck. And similarly, a true belief doesn't automatically count as knowledge, because the truth of the belief might have been lucky as well. The Getty case is an example of this. Instead, what makes something knowledge is that the belief is true because of the person's intellectual virtues, in the same way that a good shot in archery hits the target because of the archer's skill and not because of luck. By requiring this link between the truth of the belief and the way it's formed or intellectual virtues. Virtue epistemology rules out Gettier cases without going full infallibilism and basically ruling everything out as knowledge. So according to virtue epistemology I can know that Paris is the capital of France or that one plus one equals two or that Plato was a Greek philosopher. Even though I can't be 100% certain of these things, as long as they're true and I arrive at these beliefs by my intellectual virtues, because of my intellectual virtues, then they do count as knowledge. But as always with these definitions of knowledge, there's always a counter example that you can use to critique the theory. And in the case of virtue epistemology, the one I've got in mind is the example of children and animals. So typically we do think that children and animals can possess knowledge. For example, a dog might know its name or a dog might know that there is food in the bowl. Likewise, a toddler might know that a dog has four legs, for example. But it's hard to see how a toddler, much less a dog, could possess intellectual virtues such as caring about the truth or even thinking rationally. So according to virtue epistemology, children and animals can't have knowledge. And yet we typically think children and animals can have knowledge. So virtue epistemology is perhaps still too narrow a definition. Perhaps epistemic virtue or intellectual virtue is not necessary for knowledge. And actually, this children and animals objection can be extended to all the other definitions of knowledge we've looked at so far. For example, even justified true belief. It's hard to see how an animal could justify its belief or how a toddler could justify its belief. And yet again, we think they have knowledge. It's certain that there's wrinkles in the shell, given that maybe some shells don't have wrinkles. And I'm getting bogged down in that example. The point is, I don't know what the point is, actually. Anyway, the next definition is is reliabilism. So one way we can get around this issue is by doing away with the condition of justification or intellectual virtue and replacing it with something like formed via a reliable method. What we mean by a reliable method here is one that reliably produces true beliefs. So if your eyesight is 99% accurate, then that would count as a reliable method. Or, I don't want to talk about newspapers because they're not very reliable. If hypothetically there was a newspaper that was reliable, and it printed the truth, then that would also count as a reliable method if you formed your belief on the basis of this hypothetical newspaper. What's important here is that a reliable method doesn't require a justification. As long as something produces true beliefs in a reliable fashion, then it counts as a reliable method, even if you can't justify it or explain how it works. So our new definition of knowledge could be something like knowledge is true belief formed via a reliable method. And the advantage of this definition would be that it allows that children and animals can possess knowledge. For example, let's say a baby has parents who are knowledgeable and tell it the truth. And these parents tell the baby that Paris is the capital of France. And the baby believes the parents and believes that Paris is the capital of France. Well, in this case, the baby's belief does count as knowledge, even if the baby forgets that the parents told it it. So it can't justify its belief. It still has this belief, and the belief was formed via a reliable method, i.e. the parents. And so it does count as knowledge. And you can run a similar story with animals. So a dog, for example, let's say has good eyesight and good memory. Well, these could count as reliable methods. And so a dog can also possess knowledge, even if the dog can't explain how it could get to that knowledge, or it can't. justify its knowledge as long as those beliefs were formed via a reliable method then this definition allows that these beliefs do count as knowledge but it's never that simple with these definitions of knowledge and in this case you can use the getty air case to argue against reliablism because we could say that smith's belief that the man who get the job has 10 coins in his pocket was formed via a reliable method Let's say that Smith has 100% perfect eyesight when he saw Jones count 10 coins from his pocket. And let's say that Smith has 100% perfect hearing when he heard the interviewer say, I'm going to give Jones the job. Well, in this case, Smith's belief that the man who will get the job has 10 coins in his pocket was formed via a reliable method. And if you remember, it turned out to be true, but for the wrong reasons. It turned out to be true because Smith himself had 10 coins in his pocket. And so we've gone full circle here. We're back to square one, back to Gettier cases, and we don't have a definition of knowledge. So we've come full circle. At the start of this video, we set out to provide a definition of knowledge in terms of necessary and sufficient conditions. We started with the justified true belief definition, but we saw how Gettier cases showed that this definition wasn't sufficient. Then we looked at four post-Gettier definitions that seek to patch up this definition. We looked at the no false lemmas definition. We looked at infallibilism. We looked at virtue epistemology. And finally, we looked at reliabilism. However, with each of these alternative definitions, we found counterexamples that disproved these too, either showing these conditions weren't sufficient for knowledge or that they weren't necessary. There's way more to this definitions of knowledge debate than I've covered here. But the thing is with academic philosophy is it can get very pedantic very quickly. But suffice to say that The defenders of Reliabilism, for example, or Virtue Epistemology wouldn't leave the debate there. For example, we could have gone into how Goldman defines a reliable method in such a way that rules out Gettyer cases. Or we could have done a bit more of a deep dive into Sosa's Virtue Epistemology and look at how Sosa's definition of intellectual virtues allows that children and animals can possess knowledge. The point though wasn't to argue that any particular definition of knowledge was correct. Really, the point was more to illustrate how this philosophical game works, to illustrate how conceptual analysis in philosophy works. So we looked at this idea of necessary and sufficient conditions and how this is what constitutes a good definition. We looked at how we can critique these definitions by coming up with counter examples or thought experiments that show either these definitions are too broad, i.e. they are not sufficient, or they're too narrow in that they include conditions that aren't necessary. So that about does it for definitions of knowledge. Thank you so much for watching. I hope you learned something. If you have any questions, let me know in the comments. Or if you have your own definition of knowledge that you think will settle this debate once and for all, let me know in the comments. If you wanted to research this topic a bit further, you can go. I'll link the page on my website that talks about definitions of knowledge in more detail. And I can recommend a few books as well. Obviously, my own book. That provides a good, slightly more in-depth overview, perhaps, of these different definitions of knowledge. But if you're looking for something a bit more academic, perhaps, I can recommend an introduction to contemporary epistemology by Jonathan Dancy. I don't know who this guy is on the front, but he looks like Blackbeard. So you can get that book. I also I'm going to link Gettier's 1963 Is Justified True Belief Knowledge paper. in the description. It's not a particularly long one so if you want to engage first hand with some philosophy then be my guest. Then the other primary sources I mentioned, not really great for definitions and knowledge because they cover a lot of other things but I can recommend them. Plato's Theaetetus, I can't say I've read the whole thing but Plato's dialogues in general are quite readable so that might be worth picking up. And the other one I mentioned the classic Descartes meditations now this is really a page turner as far as philosophy books go so I can I can thoroughly recommend Descartes meditations it's it's a real classic um so yeah that about does it for this episode thank you so much again for watching and uh yeah I'll see you in the next one

So if you think about moral philosophy for example, a theory such as the use of the terrorism is essentially just a definition of what good is. And so by looking at this story of the various definitions of knowledge that the people pose over the years and the issues with these definitions, it's a good way to illustrate how this philosophical game is played. It illustrates the difference between a good definition and a bad definition. It shows us how we can critique philosophical definitions and theories.

And it's just a good all-round introduction to philosophical reasoning in general. So let's do some philosophy. So before we get into the definitions of knowledge themselves, let me first explain how this game is played.

The classic way definitions work in philosophy is with this idea of necessary and sufficient conditions. Now, of course, there's many philosophers who reject this entire conception of language, such as Wittgenstein. And this is probably more in line with my own personal view. But at least for the purposes of today, we're going to play the game according to these rules. We're going to try and...

provide a definition of knowledge in terms of necessary and sufficient conditions. So according to these rules, a good definition will contain conditions where each individual condition is necessary for knowledge and taken together all these conditions will be jointly sufficient for knowledge. Now what that means in simple terms is that everything that's an instance of knowledge will fit our definition of knowledge and everything that fits our definition of knowledge will be an instance of knowledge.

If you think about it in terms of Venn diagrams, what that looks like is two circles that overlap perfectly. If we include conditions that aren't necessary, then our definition will be too narrow because it will rule out things that are knowledge. If our definition includes conditions that aren't necessary for knowledge, then our definition is going to be too narrow.

In other words, there's going to be things that are knowledge that our definition excludes. On the other hand, if our conditions aren't sufficient for knowledge, then it's going to have the opposite problem. Our definition is going to be too broad.

Or in other words, our definition is going to say things are knowledge when they are not knowledge. So one final thing to mention before we get on to the definitions of knowledge themselves, and that's to specify the kind of knowledge we're defining, because there's many different types of knowledge. So you've got ability knowledge. that's knowledge of skill such as knowing how to ride a bike or knowing how to juggle.

When we're talking about the definition of knowledge we're not talking about the definition of ability knowledge. You've also got acquaintance knowledge which is knowledge of a person or a place so in the way you might know your home city well or you might know your best friend well. When we're talking about the definition of knowledge today we're not talking about acquaintance knowledge either. Instead when we're talking about the definition of knowledge today we're talking about the definition of propositional knowledge specifically.

Now propositional knowledge is knowledge of facts and it's sometimes called descriptive knowledge. So my knowledge that I know that Paris is the capital of France or I know that two plus two equals four. These are examples of propositional knowledge and so when we're talking about the definition of knowledge going forward we're talking about the definition of propositional knowledge specifically. So the classic definition of knowledge then. is justified true belief and this is sometimes called the tripartite definition because there's three parts justified true and belief and this this definition originated all the way back in Plato's dialogue Theotetus written I think in the fourth century BC and this definition of knowledge this tripartite definition was generally accepted for over 2,000 years to provide both the necessary and sufficient conditions for knowledge.

On the face of it, the tripartite definition works quite well. All of these conditions do appear to be necessary. Obviously, knowledge requires it to be true.

If you say I know the moon is made of green cheese, then you wouldn't consider that knowledge because it's not true. So truth is necessary for knowledge. Likewise, belief.

It just wouldn't make sense if you said, I know that Paris is the capital of France, but I don't believe that Paris is the capital of France. That just doesn't make sense. So belief is also considered to be a necessary condition of knowledge. Finally, justification.

It's this justification condition that that differentiates knowledge from just a lucky guess. So if I were to ask you how many moons does Pluto have? Now unless you're an astronomy nerd or whatever you probably don't know the answer but if you just had a strong gut feeling that the answer was five you would be right but your belief that Pluto has five moons wouldn't be knowledge because you don't have a justification for it. So justification also appears to be necessary for knowledge.

So justified true belief appears to work quite well as a definition of knowledge. Each of these conditions appears to be necessary for knowledge. Well, actually, I mean, there's some philosophers who argue that some of these conditions aren't necessary. Like you can get into wacky definitions about truth or describe weird scenarios where somebody knows something, but they don't believe it. Maybe it's subconscious or whatever.

But generally speaking, justified true belief is considered to be. necessary for knowledge. The juicy criticism, and I'm of course talking about Getty Act cases, is arguing that the tripartite definition is not sufficient for knowledge. So like I mentioned earlier, the tripartite justified true belief definition of knowledge dated all the way back to Plato in the 4th century BC, and this definition basically went uncontested for over 2000 years.

That is until 1963 when Edmund Gettier described two scenarios now known as Gettier cases where a person has a justified true belief that isn't knowledge. In other words Gettier described two scenarios that show justified true belief is not sufficient for knowledge. The first Gettier case goes like this. Two guys Smith and Jones are interviewing for the same job.

Smith hears the interviewer say, I'm going to give Jones the job. And Smith also sees Jones count 10 coins from his pocket. So based on this, Smith forms the belief that the man who will get the job has 10 coins in his pocket. So they go through the interview, but Smith gets the job and not Jones.

And then Smith looks in his pocket and by pure coincidence, Smith sees he also has 10 coins in his pocket. So Smith's belief here was justified. It was justified by the fact that he heard the interviewer say he was going to give Jones the job. And it was justified by the fact that he saw Jones count 10 coins from his pocket.

What's more, Smith's belief was true. The man who got the job... did have 10 coins in his pocket. The thing is though we don't want to say Smith's belief he accounted as knowledge because it was just luck that he was correct. Smith's true belief was formed on the basis that Jones would get the job and the fact that Jones had 10 coins in his pocket.

It was just luck that Smith himself had 10 coins in his pocket. This shows that the justified true belief definition is too broad because Smith had a justified true belief but he didn't have knowledge. This isn't one of Getty Air's own examples, but it illustrates the same principle.

So imagine you hear your phone call. You pull your phone out and you see you've got a message, but you also see that your phone was on silent and actually actually the noise you heard was somebody else's phone who has the same ringtone. So here, your belief that you had a message was justified by the noise you heard, and it was true, you did indeed have a message. But again, the justification was irrelevant to the truth of the belief.

It was just lucky. So again, this is another example of a justified true belief that isn't knowledge. Gettyer's 1963 paper is about as groundbreaking as it gets in philosophy, which maybe isn't saying very much, but anyway. And since 1963, various philosophers have tried to propose new definitions of knowledge that avoid Gettyer cases and patch up the definition of knowledge such that it doesn't include Gettyer cases.

So the first post-Gettier definition we're going to look at is called the no false lemmas definition. And what this does is it just adds an extra condition onto the tripartite definition. So the new definition becomes knowledge is justified true belief that is not inferred from a false lemma.

A false lemma is essentially just a false belief or a false assumption in an argument. So going back to the original Gettier case, we can correctly say that Smith's belief does not count as knowledge because it was derived from the false lemma that Jones would get the job. Even though Smith's belief was true and justified, the no false lemmas definition would say it's not knowledge because the justification was based on the false lemma that Jones would get the job. So by adding this extra no false lemmas condition, we narrow the definition of knowledge and we rule out Gettyer cases.

So it's an improvement. However, this no false lemmas definition didn't last very long before people started coming up with similar counter examples to the Gettyer case where somebody has a justified true belief that is not inferred from a false lemma but that nevertheless isn't knowledge. So the classic example used to illustrate this is fake Barn County. Imagine a place called Fake Barn County where the people create fake barns that look identical to real barns.

A guy called Henry is driving through Fake Barn County but he doesn't know the locals do this. Henry often thinks that is a barn when he looks at the fake barns. And these beliefs obviously aren't knowledge because they're not true. The barns are fake barns.

However, there is one real barn in Fake Barn County and when Henry looks at it he forms the exact same belief that that is a barn. In this case his belief is true, it's also justified by his visual perception and most importantly it's not inferred from a false lemma. So here Henry's belief is a justified true belief that isn't inferred from a false lemma. But like in the original Getty case we don't want to say Henry's belief here counts as knowledge because it's just luck that he was correct in this one instance.

After all every time he was looking at the fake barns he was forming the exact same belief and so he doesn't have the ability to tell real Barnes from fake Barnes. Yet the no false lemmas definition says Henry's belief here counts as knowledge. So what this shows is that the no false lemmas definition is still too wide, a bit like the original tripartite definition. In other words, the no false lemmas definition is not sufficient either.

So we saw that the justified true belief definition wasn't sufficient for knowledge. And while adding this no false lemmas condition improves this definition, the no false lemmas definition isn't sufficient for knowledge either. So perhaps Descartes can give us a better definition of knowledge. Well, Descartes doesn't really intend this as a definition of knowledge per se, but his approach to knowledge in meditations rules out both Gettyer cases and fake Barn County scenarios.

For Descartes, something only counts as knowledge if it's absolutely certain, if it's rationally impossible to doubt. And so this approach is sometimes referred to as infallibilism. Applied to the fake Barn County case, there's all sorts of reasons why Henry could potentially doubt his belief that that's a barn. For example, Henry's eyesight might not be 100% accurate, or perhaps Henry's mistaken somebody's house for a barn, or maybe Henry's actually in bed and dreaming that he's looking at a barn and so his belief is false. Or maybe Henry's taking a bunch of drugs, I don't know, and he's just hallucinating barns.

In any case, these are all potential reasons for Henry to doubt his belief that that's a barn. And while these might not seem particularly likely, they're at least possible. And if they're possible, they're potential reasons for Henry to doubt his belief.

So according to infallibilism, Henry's belief doesn't count as knowledge because there are potential reasons to doubt it. Henry's belief there is a barn is not certain. It's a similar thing with the Getty case. There's all sorts of reasons why Smith could doubt his justification for his belief. For example, Smith could have misheard the interviewer, or again, Smith could be dreaming or hallucinating.

And so again, infallibilism would say this isn't knowledge. And so at first glance, infallibilism seems to be an improvement on both the justified true belief and no false definition. Because where these... These definitions aren't sufficient.

They say that Henry's and Smith's beliefs are knowledge when they're not. Infallibilism correctly says that Henry's and Smith's beliefs aren't knowledge because they're not knowledge. However, whereas the justified true belief and no false lemmas definitions were too broad in their approach, they're not sufficient in philosophical terms. Infallibilism faces the opposite problem.

It basically rules out anything as knowledge, and so it's too narrow in its definition of knowledge. It seems obvious that I do know all sorts of things. For example, I know that 2 plus 3 equals 4, that Paris is the capital of France, and that Plato was a Greek philosopher.

But There are all sorts of reasons why I could potentially doubt these beliefs. For example, I could have been the victim of some wacky conspiracy where everybody was lying to me about Plato being Greek and actually he was British or something. Now, even if that doesn't seem particularly likely, it's at least possible.

And so it's a potential reason to doubt my belief that Plato was a Greek philosopher. And so according to infallibilism, I don't know this. We can come up with weird scenarios to doubt pretty much every belief.

Even something as fundamental as 1 plus 1 equals 2. You know, it's logically possible that an evil demon is just manipulating your mind and making you think that 1 plus 1 equals 2, when actually 1 plus 1 equals 3. Now, again, I'm not saying that's particularly likely, but it's least possible. It's a potential reason to doubt your belief. And so according to infallibilism, you don't know that 1 plus 1 equals 2. But this just seems ridiculous because clearly we do know all sorts of things. And so if we're to... Maintaining any reasonable understanding of knowledge, we're going to have to reject the infallibus definition.

It seems that certainty or the impossibility of doubt is not a necessary condition of knowledge. The next contender for a definition of knowledge is virtue epistemology. Now, this is a broad category like infallibilism, but In general, virtue epistemology definitions draw a parallel between moral virtue and intellectual or epistemic virtue.

So let's start with the case of moral virtue first. An act of moral virtue achieves a virtuous outcome. So let's say you help an old lady across the road.

Now that is a virtuous thing to do, but is it an act of moral virtue if the only reason you helped the old lady across the road was because you thought she'd give you some money? Presumably not. And so for something to be a proper act of moral virtue, it has to not only achieve a virtuous outcome, but it has to achieve the virtuous outcome because of the moral virtues of the person committing the action.

Virtual epistemology says a similar thing about knowledge. It's not enough for a belief to achieve a virtuous outcome, i.e. be true. It has to achieve this virtuous outcome because of the intellectual virtues of the believer.

So intellectual virtues are things like caring about the truth, thinking rationally and so on. And so applied to the Gettier case we could say that Smith's belief while true and while achieving a virtuous outcome is not an act of intellectual virtue because the truth of the belief is not a result of Smith's intellectual virtues. The truth of the belief was a result of luck. As such virtue epistemology would correctly say that Smith's belief doesn't count as knowledge.

In the same way that helping the old lady across the road to get money wasn't a proper act of moral virtue, Smith's belief that the man who gets the job has 10 coins in his pocket, while true, wasn't a proper act of intellectual virtue because it wasn't true for the right reasons. Nailed it. Ernest Sosa illustrates this idea of intellectual virtue with a nice analogy to do with archery.

So what is a good shot in archery? Well, firstly, a good shot in archery is accurate. It hits the target. But an accurate shot in archery doesn't automatically mean a good shot, because the arrow might have just hit the target through luck.

And similarly, a true belief doesn't automatically count as knowledge, because the truth of the belief might have been lucky as well. The Getty case is an example of this. Instead, what makes something knowledge is that the belief is true because of the person's intellectual virtues, in the same way that a good shot in archery hits the target because of the archer's skill and not because of luck.

By requiring this link between the truth of the belief and the way it's formed or intellectual virtues. Virtue epistemology rules out Gettier cases without going full infallibilism and basically ruling everything out as knowledge. So according to virtue epistemology I can know that Paris is the capital of France or that one plus one equals two or that Plato was a Greek philosopher. Even though I can't be 100% certain of these things, as long as they're true and I arrive at these beliefs by my intellectual virtues, because of my intellectual virtues, then they do count as knowledge. But as always with these definitions of knowledge, there's always a counter example that you can use to critique the theory.

And in the case of virtue epistemology, the one I've got in mind is the example of children and animals. So typically we do think that children and animals can possess knowledge. For example, a dog might know its name or a dog might know that there is food in the bowl.

Likewise, a toddler might know that a dog has four legs, for example. But it's hard to see how a toddler, much less a dog, could possess intellectual virtues such as caring about the truth or even thinking rationally. So according to virtue epistemology, children and animals can't have knowledge. And yet we typically think children and animals can have knowledge.

So virtue epistemology is perhaps still too narrow a definition. Perhaps epistemic virtue or intellectual virtue is not necessary for knowledge. And actually, this children and animals objection can be extended to all the other definitions of knowledge we've looked at so far. For example, even justified true belief. It's hard to see how an animal could justify its belief or how a toddler could justify its belief.

And yet again, we think they have knowledge. It's certain that there's wrinkles in the shell, given that maybe some shells don't have wrinkles. And I'm getting bogged down in that example.

The point is, I don't know what the point is, actually. Anyway, the next definition is is reliabilism. So one way we can get around this issue is by doing away with the condition of justification or intellectual virtue and replacing it with something like formed via a reliable method.

What we mean by a reliable method here is one that reliably produces true beliefs. So if your eyesight is 99% accurate, then that would count as a reliable method. Or, I don't want to talk about newspapers because they're not very reliable.

If hypothetically there was a newspaper that was reliable, and it printed the truth, then that would also count as a reliable method if you formed your belief on the basis of this hypothetical newspaper. What's important here is that a reliable method doesn't require a justification. As long as something produces true beliefs in a reliable fashion, then it counts as a reliable method, even if you can't justify it or explain how it works.

So our new definition of knowledge could be something like knowledge is true belief formed via a reliable method. And the advantage of this definition would be that it allows that children and animals can possess knowledge. For example, let's say a baby has parents who are knowledgeable and tell it the truth.

And these parents tell the baby that Paris is the capital of France. And the baby believes the parents and believes that Paris is the capital of France. Well, in this case, the baby's belief does count as knowledge, even if the baby forgets that the parents told it it. So it can't justify its belief. It still has this belief, and the belief was formed via a reliable method, i.e. the parents.

And so it does count as knowledge. And you can run a similar story with animals. So a dog, for example, let's say has good eyesight and good memory.

Well, these could count as reliable methods. And so a dog can also possess knowledge, even if the dog can't explain how it could get to that knowledge, or it can't. justify its knowledge as long as those beliefs were formed via a reliable method then this definition allows that these beliefs do count as knowledge but it's never that simple with these definitions of knowledge and in this case you can use the getty air case to argue against reliablism because we could say that smith's belief that the man who get the job has 10 coins in his pocket was formed via a reliable method Let's say that Smith has 100% perfect eyesight when he saw Jones count 10 coins from his pocket.

And let's say that Smith has 100% perfect hearing when he heard the interviewer say, I'm going to give Jones the job. Well, in this case, Smith's belief that the man who will get the job has 10 coins in his pocket was formed via a reliable method. And if you remember, it turned out to be true, but for the wrong reasons. It turned out to be true because Smith himself had 10 coins in his pocket. And so we've gone full circle here.

We're back to square one, back to Gettier cases, and we don't have a definition of knowledge. So we've come full circle. At the start of this video, we set out to provide a definition of knowledge in terms of necessary and sufficient conditions.

We started with the justified true belief definition, but we saw how Gettier cases showed that this definition wasn't sufficient. Then we looked at four post-Gettier definitions that seek to patch up this definition. We looked at the no false lemmas definition.

We looked at infallibilism. We looked at virtue epistemology. And finally, we looked at reliabilism.

However, with each of these alternative definitions, we found counterexamples that disproved these too, either showing these conditions weren't sufficient for knowledge or that they weren't necessary. There's way more to this definitions of knowledge debate than I've covered here. But the thing is with academic philosophy is it can get very pedantic very quickly. But suffice to say that The defenders of Reliabilism, for example, or Virtue Epistemology wouldn't leave the debate there.

For example, we could have gone into how Goldman defines a reliable method in such a way that rules out Gettyer cases. Or we could have done a bit more of a deep dive into Sosa's Virtue Epistemology and look at how Sosa's definition of intellectual virtues allows that children and animals can possess knowledge. The point though wasn't to argue that any particular definition of knowledge was correct.

Really, the point was more to illustrate how this philosophical game works, to illustrate how conceptual analysis in philosophy works. So we looked at this idea of necessary and sufficient conditions and how this is what constitutes a good definition. We looked at how we can critique these definitions by coming up with counter examples or thought experiments that show either these definitions are too broad, i.e. they are not sufficient, or they're too narrow in that they include conditions that aren't necessary. So that about does it for definitions of knowledge.

Thank you so much for watching. I hope you learned something. If you have any questions, let me know in the comments.

Or if you have your own definition of knowledge that you think will settle this debate once and for all, let me know in the comments. If you wanted to research this topic a bit further, you can go. I'll link the page on my website that talks about definitions of knowledge in more detail.

And I can recommend a few books as well. Obviously, my own book. That provides a good, slightly more in-depth overview, perhaps, of these different definitions of knowledge. But if you're looking for something a bit more academic, perhaps, I can recommend an introduction to contemporary epistemology by Jonathan Dancy.

I don't know who this guy is on the front, but he looks like Blackbeard. So you can get that book. I also I'm going to link Gettier's 1963 Is Justified True Belief Knowledge paper.

in the description. It's not a particularly long one so if you want to engage first hand with some philosophy then be my guest. Then the other primary sources I mentioned, not really great for definitions and knowledge because they cover a lot of other things but I can recommend them.

Plato's Theaetetus, I can't say I've read the whole thing but Plato's dialogues in general are quite readable so that might be worth picking up. And the other one I mentioned the classic Descartes meditations now this is really a page turner as far as philosophy books go so I can I can thoroughly recommend Descartes meditations it's it's a real classic um so yeah that about does it for this episode thank you so much again for watching and uh yeah I'll see you in the next one

Transcript for:Exploring the Definition of Knowledge

Transcript for:
Exploring the Definition of Knowledge