There is a severe form of skepticism that has floated around philosophical issues for hundreds of years. These skeptics want to show that we have very little (if any) knowledge of the external world. Here's how the argument goes:
1) If I know some ordinary proposition O (e.g., I have hands) then I know that propositions inconsistent with O are false ~P (I know that I'm not just in the Matrix and being deceived that I have hands). 2) I don't know that I'm not in the Matrix. 3) So, I don't know O (e.g., that I have hands).
This argument is valid (3 follows from 1 and 2 by modus tollens) but it may need more explaining. For anything that I say that I know (we're calling this O) I have to know that ~P (not P), where P is some proposition that, if true, would make O false. So, if I know that I'm looking at a Ford Tempo, I know that I'm not looking at a Mercury Sable. If I know that I'm looking at a zebra, then I know I'm not looking at a cleverly disguised mule.
To put the argument in more understandable terms, let's say I ask you if you know where your car is (or pick anything that you own that you can't see right now. This statement is like O, from above). You say "Yes, of course. My car is parked right outside." I respond by asking if you know that your car hasn't been stolen in the past 5 minutes (here, we have P. You must know P is false because, if it were true, O would be false. If your car has been stolen, then you don't know where it is). Here, most people will concede they don't really know where their car is, but they're pretty sure. Of course, you could walk outside and go check to make sure your car is there. Then, presumably, you would know where your car is. Taking this back to the Matrix case, it seems like any knowledge of the external world would be false if we're plugged into the Matrix, or dreaming, or being deceived by an evil demon. Of course, we can't just walk outside and check to make sure we're not in the Matrix. The key point here is: we can NEVER know that we're in a skeptical scenario, like the Matrix But since we don't know P is false, we don't know O is true. And thus, the skeptic argues, we don't know very much at all.
There are some fairly standard philosophical moves to try to handle the argument, and they get far too complicated to go into here. But I believe the answer to the skeptic is more intuitive than philosophers have thought. So I'd like to get the AG community's thoughts on the problem. Which premise or premises from the above argument would you reject? Or do you think we really don't know ordinary propositions, like having hands?
I would like to see those, I think Freakenstein would also want too B)
If you're interested in some of the answers to skepticism, there are some excellent papers. I can mention them here, but to get a full understanding (which you really need in order to discuss these things) you should read them yourself.
One move is to deny the first premise, that if I know P then I know Q (or ~Q, in our case). Denying this premise is to deny that knowledge is closed under known entailment. I'm sure that clears everything up :P Basically this idea (championed by Fred Dretske) argues that if I know P and I know P entails Q, I don't necessarily know that Q. Again, it's just far too complicated to get into here, but it seems to be a genuine move. Although many philosophers hate the idea of denying what amounts to deductive knowledge. That's not a good place for us to be, epistemically.
One of the most famous responses to the skeptical argument is the Moorean shift - developed by G.E. Moore. He turns to argument around by denying the conclusion and using it as a premise. So the original argument: 1, 2, /3 turns into 1, ~3 /2 thus getting the conclusion that we can know that we're in a skeptical scenario. His argument is that we have way more proof that the external world exists than we have proof that we're in a skeptical scenario. On the face of it, this looks like question begging, but Moore seems to wiggle out of it (although I have my doubts).
Another famous (or maybe infamous) response is known as contextualism. One of its main proponents is Keith DeRose (super fun guy). Basically the word 'know' can change meanings in different contexts, according to him. What we mean by 'know' in easy cases is less stringent that what we mean by 'know' in hard cases.
Again, this is an extremely cursory glance at some very complicated arguments and theories. But do any of these seem plausible? Or at least more plausible than the others?
I always believed that there's pretty much an equal chance either way. Just because we don't know we don't live in a Matrix-type world doesn't mean we do, it means we don't know. Likelihood is unaffected by our lack of knowledge.
I used to think that through the use of Occam's Razor, it was safe to assume that reality was truly real.
Then I read this and had a change of heart. if you don't care to delve into it, suffice to say that there are 4 possilities which exist, the first two are eliminated outright, leaving 2. Either we are part of a simulation, or we will be the first civilisation to run a simulation, the latter being more likely. So we are either in a simulation. Or we are the products of another simulation, which sparked another simulation, and so on. Weird stuff.
So what's the difference?
What separates it is the human ability to analyse a scenario which best fits the data. For example, I could be in a house and have my car parked outside, be talking to a friend, who asks where my car is. I would be valid in saying my car's parked outside, and here's why:
The skepticism argument seems to rely on the presupposition that validity of any claim can only be gained empirically. I dispute this. Although empirical evidence is ideal, rationalism is a good enough substitute. To apply this to the car example, when my friend asks where's your car parked, I assume it is still outside, because the possibility of it being stolen would be unlikely without my knowledge. Assuming I'd locked the door (central locking systems rule!) I can also assume that breaking in would require lots of tools or lots of noise. Now the chances of a car jacker with said tools happening to be walking down the street and taking advantage of my car is very unlikely. The chances of someone smashing a window and me not hearing it is very unlikely. Therefore, without any of these stimuli to indicate something has happened to the car, why should I assume it is gone, since it is much more logical to assume it is there.
Kind of like, other life in the universe theories. We have no empirical evidence aliens exist. Most people think they do, simply because the universe is so big. Absense of evidence is not necessarily evidence of absense.
The skepticism argument seems to rely on the presupposition that validity of any claim can only be gained empirically. I dispute this. Although empirical evidence is ideal, rationalism is a good enough substitute. To apply this to the car example, when my friend asks where's your car parked, I assume it is still outside, because the possibility of it being stolen would be unlikely without my knowledge.
I see what you're saying. You have a belief (maybe knowledge?) here that your car is still parked outside. But your justification for this belief is still empirically counterfactual. Because you didn't hear your car alarm or someone breaking in, you are justified in believing that your car is outside. But this justification still seems empirical. What's happened in this case is that you're able to rule out a relevant alternative through some degree of evidence. What interesting here is that your argument seems to implicitly suggest that you (like most of us) think that your car being stolen is, in fact, a relevant alternative. It's just that in most cases we can justifiably believe that this alternative has not obtained. But in a skeptical scenario, we have no justification (the skeptic argues) for believing this alternative to be false. So the most promising move here is to argue that skeptical scenarios aren't relevant - at least not in the same way that your car getting stolen is.
The chances of someone smashing a window and me not hearing it is very unlikely. Therefore, without any of these stimuli to indicate something has happened to the car, why should I assume it is gone, since it is much more logical to assume it is there.
Another good point. But there are oftentimes knowledge claims that, however likely they are to be true, we still don't consider knowledge. Remember it's knowledge that we're ultimately addressing. Let's say I buy a lottery ticket and I form the belief that I'm going to lose the lottery. The odds are certainly in favor of my belief being true and, as it turns out, I do in fact lose the lottery. But most of us wouldn't say that I know I lost the lottery (here, the issue of why gets complicated). My point being that just because some event is highly unlikely (like a skeptical scenario) doesn't mean that we can know it doesn't obtain.
So the most promising move here is to argue that skeptical scenarios aren't relevant - at least not in the same way that your car getting stolen is.
Exactly. We get noweher by making presumptions which don't best suit the available data. I mean I suppose it would be nice to be empiricaly sure of everything all the time, but since that's not going to happen, what's wrong with some good old fashioned pragmatism?
My point being that just because some event is highly unlikely (like a skeptical scenario) doesn't mean that we can know it doesn't obtain.
That is a good point. Just to go back to pragmatism though, does it really matter that much if unlikely events can occur. We may not be able to claim knowledge that they won't occur, but this doesn't change our behaviour. I guess what I'm asking is what important implications thislack of knowledge can cause. So far, outside of an awesome topic for discussion, I can't think of anythingwhere the gamble is not worth the risk.
The skepticism argument seems to rely on the presupposition that validity of any claim can only be gained empirically. I dispute this. Although empirical evidence is ideal, rationalism is a good enough substitute. To apply this to the car example, when my friend asks where's your car parked, I assume it is still outside, because the possibility of it being stolen would be unlikely without my knowledge. Assuming I'd locked the door (central locking systems rule!) I can also assume that breaking in would require lots of tools or lots of noise. Now the chances of a car jacker with said tools happening to be walking down the street and taking advantage of my car is very unlikely. The chances of someone smashing a window and me not hearing it is very unlikely. Therefore, without any of these stimuli to indicate something has happened to the car, why should I assume it is gone, since it is much more logical to assume it is there.
But you are using empirical evidence to come to these conclusions. You have past observation and experience that the car is parked outside. So using this evidence you can come to the conclusion that unless the situation changes (loud noises for one) that you will arrive at the same outcome (the car being parked outside).
So using this evidence you can come to the conclusion that unless the situation changes (loud noises for one) that you will arrive at the same outcome (the car being parked outside).
I cannot be empirically certain my abswer unless I check where my car is, as soon as my friend asks me where it is. Empirically it is still uncertain. It is however rationalism that underpins my belief that my car is still there, since I assume I would know if it wasn't. This assumption is what distinguishes the rational from the empirical. Empirically, assumptions do not need to be made, since you observing an objective truth. Rationally you are making possibly assumptions, based on logic and mathematical probability. Assumptions which could be incorrect.
I cannot be empirically certain my abswer unless I check where my car is, as soon as my friend asks me where it is. Empirically it is still uncertain. It is however rationalism that underpins my belief that my car is still there, since I assume I would know if it wasn't. This assumption is what distinguishes the rational from the empirical. Empirically, assumptions do not need to be made, since you observing an objective truth. Rationally you are making possibly assumptions, based on logic and mathematical probability. Assumptions which could be incorrect.
Empirical evidence is based on observation and/or experience that can be verified. Given you would have past experience that your car would be outside and you could look outside and verify this your rational assumption would be based on empirical evidence.
I think Mage is right here. You can rationally decide what empirical evidence will count for or against a proposition, but it will ultimately be empirical evidence (or lack thereof) that justifies the belief. Thus, it is an empirical belief. To put this in context, consider how you would justify your belief that your car hasn't been stolen. Presumably, you would say something like you didn't hear it being broken into. This lack of empirical evidence justifies your belief. But you're right in that it is a rational process that determines what evidence counts. Your point though, Firefly, on pragmatism is well made. But where do we draw the line? At things we can't possibly know? That seems reasonable, but there are still a near infinite number of alternatives that could make a particular belief end up being false (even if it is justified). And I, for one, would certainly act differently if I discovered this reality were merely a simulation. I don't know how many other people feel this way.
Given you would have past experience that your car would be outside and you could look outside and verify this
But that's the point. I'm arguing that I could hold knowledge that my car was still outside without empirically verifying it and looking. Why? Because it is rational to assume so. To extend the car analogy further, if I got a sudden urge to go to Brixton and left it there with the door unlocked for a week, I would also be justified in assuming it would have been stolen. Not because I could verify it, but because it is rationalism to assume this is the case, based on mathematical probability, something that isn't affected by empiricism.
Well, now that's a very good point. I think that you and I think about the car example in different ways, so I wasn't understanding your point of view until I just read this. Of course, the belief that your car would be stolen in Brixton is certainly empirically founded, but I don't think that will affect your claim about the car's location right now. It does seem to be a rationally, rather than empirically, justified belief.
With this new understanding, could you relate this argument back to the skeptical scenario? The reason I ask is because there still seems to be a large disconnect between the two cases. I can accept a rational justification in the car example, but not in response to the skeptic. I think this may have to do with the fact that the skeptic's argument is one that is rational, rather than empirical. In fact, most of the empirical evidence we have would seem to contradict the skeptic outright. I will say, though, that I find the pragmatic response to the skeptic nothing more than ad-hoc. I think very few people would accept pragmatism as a whole, so it shouldn't be allowed to be applied in this one instance. At least, not without an argument as to why pragmatism is a valid response to the skeptic. Although, I imagine the argument for pragmatism would have to be some sort of pragmatic argument, which would beg the question against the skeptic. Maybe not. What do you think?
And I, for one, would certainly act differently if I discovered this reality were merely a simulation. I don't know how many other people feel this way.
Oh I forgot to address this part of your last post. I'm not so sure about this. If there was some option to actually find out about the 'real' world a la blue pill red pill scenario, I'd jump at the chance. Other than that though, we are trapped in the illusion. I don't think my decision making would change in any big way, since I still feel pain and joy, I'd still try and maximise the latter and minimise the former.
I will say, though, that I find the pragmatic response to the skeptic nothing more than ad-hoc. I think very few people would accept pragmatism as a whole, so it shouldn't be allowed to be applied in this one instance.
I think pragmatists would argue that the assertion of pragmatism was created for an instance such as this. Although not completely cogent as a theory, I think that as an antithesis to skepticism it is effective. I do acknowledge however that it doesn't fully deal with the root of the problem of the dangers of assuming knowledge without truly empirical evidence, rather trying to limit the damage they can cause.
Although, I imagine the argument for pragmatism would have to be some sort of pragmatic argument, which would beg the question against the skeptic. Maybe not. What do you think?
I think if you locked a pragmatist and a skepticist in a cupboard and asked them to discuss this, they would end up going round in circles on this point, and would probably have to agree to disagree in the end. Neither side can really win, because both judge knowledge by different criteria. Unless one could convince the other that their conception of knowledge is correct, I don't think their argument is solveable, because it isn't really compatible.
From our objective stance though, I think we can have a much better stab at solving the issue by properly defining when it is justifiable to believe in something. Where the skeptics and the pragmatists differ is that for skeptics it never seems to be justifiable until it has been observed. For pragmatists, there is more leeway.
Actually, now that I think more about it, I'm starting to go more for the pragmatist notion of knowledge. If you use Schroedinger's Cat to show that observation affects truth, empiricism doesn't stand up to muster in the same way that a pragmatic approach does.
I really like the idea of pragmatism and I really hope someone one day can convince me that it's correct. But there are a number of serious problems with this view (of course, that's certainly not unique to pragmatism). What I find most problematic is that pragmatism is able to count as true or meaningful certain statements that we know to be false or meaningless. If we take pragmatism to be the thesis that what is true is what is expedient to believe, then some problems arise. For example, what if I really really want to murder my neighbor because he's a loud annoying a-word. It would be expedient for me to believe in this case that murder is fine (at least in this instance). But we all know this belief is false. In fact, it would be extremely difficult for me to adopt this belief - even if it is the most expedient way to believe.
I realize this isn't the only way to define pragmatism. It seems you take more of a meaning approach. Perhaps something along the lines of: "meaningful statements are only those which, if known, would have an effect on our lives or choices" (there's a more sophisticated definition, but I can't think of it right now). But here we have an arbitrary assessment of meaningfulness. Case in point: our disagreement about how learning we're in a skeptical scenario would affect us. Your contention is that your behavior wouldn't really change that much - pain still hurts and joy still feels good. But for me, the consequences are dire. I don't really know what I'd do if I found out I was in the Matrix or whatever.
Ah, the definition I was trying to remember is something like: a statement is meaningful only if it has some practical evaluative status for you. But again, this seems like it would be arbitrary based on who you're talking to. And doesn't our being in a skeptical scenario at least seem meaningful - if only on a commonsense view of the word 'meaningful'? As intuitive as pragmatism seems, it just has weirdly unintuitive consequences like this.
It would be expedient for me to believe in this case that murder is fine (at least in this instance). But we all know this belief is false. In fact, it would be extremely difficult for me to adopt this belief - even if it is the most expedient way to believe.
Surely a pragmatist argues that the most expedient course of action would be to knock on the door and ask his neighbour to turn it down, install double glazing? I see the point you're trying to make, but surely murder as a solution is pragmatically a lousy one? I mean, apart from asll the hassle of carrying out the act, the risk of punishment would be anough to dissuade any rational pragmatist.
But here we have an arbitrary assessment of meaningfulness. Case in point: our disagreement about how learning we're in a skeptical scenario would affect us. Your contention is that your behavior wouldn't really change that much - pain still hurts and joy still feels good. But for me, the consequences are dire. I don't really know what I'd do if I found out I was in the Matrix or whatever.
So could it be argued that nearly any belief system operates on essentially pragmatic principles, but differ only with regards to their definition of 'meaningfull'? But I digress. I think you yourself put forward a universal conception of meaningfull when you said: ''"meaningful statements are only those which, if known, would have an effect on our lives or choices"'', or indeed your later suggestion ''a statement is meaningful only if it has some practical evaluative status for you.''. To me that seems universal, since although the degree to which people would care about a certain proposition (like the Matrix example) it would surely still effect everyone's life to a certain degree.
Well, the pragmatists still have to have a principled account of what a meaningful statement is. Those two statements you quoted from me are some standard ways for the pragmatist to put forth their account of meaning. As attractive as pragmatism seems on the surface, it has some very deep, very troubling philosophical problems associated with it. I would almost prefer denying that knowledge is closed under known entailment before I would accept pragmatism with all its inherent flaws. Almost...
Well, the pragmatists still have to have a principled account of what a meaningful statement is.
If that's the case, I don't think it's possible, because of the inherent sunjectivity of the term 'meaningful'. In my opinion though, I still am not convinced that pragmatic solutions are not appropriate for some problems. In essence play it by ear. Of course this inconsistency is a weakness in the theory, but not to the point where it should be completely disregarded, at least not in my opinion.
