TheFaithfulStone Posted October 6, 2006 Report Posted October 6, 2006 K we were talking about causation being part of logic with the whole decapitation thing not whether or not decapitation causes death. I take it by trying to pretend the discussion was about something else you have given up there. What? The whole idea was that I said that statements of cause imply the phrase "all things being equal" and I used the decapitation example. You had previously argued the person could be a robot, or have their head sewn back on, or whatever - but that violates the fundamental assumption which is all things being equal. And I ceded the point to Q, that logic does not of necessity need causation so much as implication, but that in order to be actually useful it assumes it's existence. TFS
Qfwfq Posted October 6, 2006 Report Posted October 6, 2006 Anyone that says that induction failing is unlikely doesn't understand the issue.I did not say that induction failing is unlikely. You just keep missing points.
Kriminal99 Posted October 6, 2006 Author Report Posted October 6, 2006 Then explain to me how to relate the movement of the moon with a bottle rolling off my desk or an apply falling from a tree, without using causal notions. I don't claim that the bottle causes the moon to move, or visa-versa, just that they are fundamentally related. -Will You can relate them by correlation. If the moon is sometimes "the straw that broke the camel's back" and causes the bottle to roll off, but not often, then the correlation between the moon moving in a situation where the moon makes a difference and the bottle rolling off would be noticable.
Erasmus00 Posted October 6, 2006 Report Posted October 6, 2006 You can relate them by correlation. If the moon is sometimes "the straw that broke the camel's back" and causes the bottle to roll off, but not often, then the correlation between the moon moving in a situation where the moon makes a difference and the bottle rolling off would be noticable. You are completely missing my point. The bottle falling off the desk and the moon oribitting the Earth are fundamentally related: they have the same causal agent (Earth's gravity). However, if we can only talk about correlations, there are no causal agents, and no such connection can be made. Since these (causal) unifying theories are widely considered to be among the highwater marks of human thought, it seems silly to do away with the idea of causality. -Will
Kriminal99 Posted October 7, 2006 Author Report Posted October 7, 2006 -Yawn-.... I've been over that. That situation is sipmly represented by lesser correlations.
Erasmus00 Posted October 7, 2006 Report Posted October 7, 2006 -Yawn-.... I've been over that. That situation is sipmly represented by lesser correlations. No, you haven't, in your previous response you missed my point. An apple falling from a tree, a ball rolling down a ramp, the moon orbitting the Earth, etc. These are all UNCORELLATED phenomena that can be described using a very simple causal structure. (an inverse square law "force"). However, because these events are uncorellated, a theory that fails to look for causes can never achieve what Newton's gravitational theory already has. Any theory of correlations, by its very nature, has to treat these seperately. However, under causal theories we recognize these as different sides of the same coin. -Will
Kriminal99 Posted October 8, 2006 Author Report Posted October 8, 2006 No, you haven't, in your previous response you missed my point. An apple falling from a tree, a ball rolling down a ramp, the moon orbitting the Earth, etc. These are all UNCORELLATED phenomena that can be described using a very simple causal structure. (an inverse square law "force"). However, because these events are uncorellated, a theory that fails to look for causes can never achieve what Newton's gravitational theory already has. Any theory of correlations, by its very nature, has to treat these seperately. However, under causal theories we recognize these as different sides of the same coin. -Will No they are not uncorrelated. You just have to know how to organize the events to see the correlation. In general if X is highly correlated with Y and Z subsequently occuring, then Y and Z are going to be correlated with each other. And yes I had already been over that.
Erasmus00 Posted October 8, 2006 Report Posted October 8, 2006 No they are not uncorrelated. You just have to know how to organize the events to see the correlation. In general if X is highly correlated with Y and Z subsequently occuring, then Y and Z are going to be correlated with each other. And yes I had already been over that. Please point me toward where you discussed uncorrelated events with essentially the same causal structure. (i.e. all of the events for which the root cause is gravity). -Will
Kriminal99 Posted October 8, 2006 Author Report Posted October 8, 2006 Any 2 events that have the same cause are going to be correlated with each other, unless the cause causes either event A OR event B but not both. Pretty simple reasoning... if the cause occurs both A and B are likely to occur and therefore A and B will be correlated. pointing it out in old posts is pointless since it is your lack of agreement with how what was said shows this that is the issue. What? The whole idea was that I said that statements of cause imply the phrase "all things being equal" and I used the decapitation example. You had previously argued the person could be a robot, or have their head sewn back on, or whatever - but that violates the fundamental assumption which is all things being equal. And I ceded the point to Q, that logic does not of necessity need causation so much as implication, but that in order to be actually useful it assumes it's existence. TFS All things being equal to what? You already know all things are not equal If you wanted to go down that road the first step would be to say (assuming you were a determinist) if everything in the universe was exactly the same as last time, it occured at exactly the same point in space in time etc it should cause the same result. Of course it will never happen that way, for starters because if it happened at the same point in time then you wouldn't have experience of it because that would have had to happen in the past. And everything in the universe is never going to be exactly the same, but more than that you wouldn't even know if it was. Generality problem So if everything can't be equal, then the next thing to try would be to say everything relevant to determining the result of the potential cause. Of course, you can never know for sure what that would entail. There may be things you know could be different that could effect the outcome, but there also may be things you don't know that would effect the outcome. You only know about the latter when the result is different even though there is nothing you are aware of that would cause it to be so. Sayying "all things being equal" (which is certainly NOT a fundamental assumption of saying something like decapitation causes death) is equivalent to trying to define an event as the cause of another event. It's fine to have a concept like electrocution, where if the person doesn't die then they didn't get electrocuted - they were just shocked (an alternate result). But it is a violation of the arguments mentioned in the above paragraph to try and define ALL electric shock to cause death. Not only are you not allowing for some way in which a person could survive that you are not aware of (which is why in general definitions are limited to one event object or concept rather than to groups of such that are percieved as connected) but you are also would be ignoring reasoning which would suggest it might be possible to survive being shocked. In the case of someone's head being removed, you know death is correlated with a lack of oxygen to the brain which you know need not (which means if it is, your measurement is invalid) be 100% correlated with a head being removed from the body.
TheFaithfulStone Posted October 8, 2006 Report Posted October 8, 2006 Sayying "all things being equal" (which is certainly NOT a fundamental assumption of saying something like decapitation causes death) is equivalent to trying to define an event as the cause of another event. Although I don't consider wiki to be an authoritative voice on the subject - au contraire! Serves as a good introduction. TFS
Kriminal99 Posted October 9, 2006 Author Report Posted October 9, 2006 Although I don't consider wiki to be an authoritative voice on the subject - au contraire! Serves as a good introduction. TFS Why did you link to that? 1) I agree wiki is not an authoritative source, in fact IMO there is no such thing as an authoritative source that can be used in an argument especially not on a philosophy subject. Authoritative source refers to something that you personally choose to trust in lieu of analyzing it yourself. To try to force a third person to accept something that you accept as an authoritative source is a fallacy. 2) The wiki entry actually outlines the same issues as I did in my previous post with misusing the concept. It says quote: "For example, in the philosophy of science it is common to say that there is a natural law that events of kind A cause events of kind B if and only if an event of type A, ceteris paribus, is always followed by an event of kind B — in order to rule out the possibility of other causal phenomena overriding the ordinary effect of the event of type A. But in order to eliminate the ceteris paribus clause in this analysis, a philosopher would need to know every sort of causal event that could possibly override any other sort of causal event — and even if there is in principle some finite list that exhausts all of these possibilities (a philosophically controversial claim), that list is certainly not known to the person who is claiming to be giving a definition of causality" It talks about some "all else equal" claims being eliminable, the example used being that there are 28 days in february "ceteris paribus", which can be reduced to "if it is not a leap year" This attempted example demonstrates the utter uselessness of the concept. If you know that leap year is what could cause februrary to not have 28 days you should have just said that, if you don't know that then it is rediculous to say "all else equal" because you don't even know what would have to be equal to achieve the same result (a 28 day february) In either case february is simply defined to have 28 days except on leap year. The only situation the concept is at all useful for is when there are many factors you KNOW are capable of altering the outcome and you just want to say hold the other known factors the same. You cannot say this to hold unknown factors the same.
Erasmus00 Posted October 9, 2006 Report Posted October 9, 2006 Any 2 events that have the same cause are going to be correlated with each other, unless the cause causes either event A OR event B but not both. Pretty simple reasoning... if the cause occurs both A and B are likely to occur and therefore A and B will be correlated. pointing it out in old posts is pointless since it is your lack of agreement with how what was said shows this that is the issue. This sort of logic only allows one to investigate proximal causes, and nothing deeper. This is the point I was trying to get at. Somethings may have the same ROOT cause, but different proximal causes. As such, they are uncorrelated despite being fundamentally related. -Will
TheFaithfulStone Posted October 10, 2006 Report Posted October 10, 2006 This attempted example demonstrates the utter uselessness of the concept. Sigh. Krim, I gotta hand it to you - your ability to seize on a relatively minor point and change the subject to it is nonpariel. Without ceteris paribus clauses you are of course, correct - for any explanation there exists a coherent universe where the explanation is not true. But it's a sneaky semantic trick - because it ISN'T coherent in the original universe. We are for instance, not concerned with what the gravitational constant is in the next universe over. Nor are we concerned with what happens to a replicant or a person wearing a force field if we try to cut off his head. Furthermore, as has been repeatedly explained - just because The Cause of The Effect may not be clear (ie, the proximal cause) does not mean that causality is not a meaningful concept. What does it prove in any case if you are right that cause and effect is meaningless? That black people have more babies out of wedlock? And that being black means you are likely to have been born out of wedlock? Let us assume you want to solve the problem. In concrete terms what would you do? TFS InfiniteNow 1
Kriminal99 Posted October 10, 2006 Author Report Posted October 10, 2006 This sort of logic only allows one to investigate proximal causes, and nothing deeper. This is the point I was trying to get at. Somethings may have the same ROOT cause, but different proximal causes. As such, they are uncorrelated despite being fundamentally related. -Will Who told you that? The result is the same regardless. Maybe if you are confused by the lack of proximity to the alleged cause and just organize the correlation measurement wrong. Sigh. Krim, I gotta hand it to you - your ability to seize on a relatively minor point and change the subject to it is nonpariel. Without ceteris paribus clauses you are of course, correct - for any explanation there exists a coherent universe where the explanation is not true. But it's a sneaky semantic trick - because it ISN'T coherent in the original universe. We are for instance, not concerned with what the gravitational constant is in the next universe over. Nor are we concerned with what happens to a replicant or a person wearing a force field if we try to cut off his head. Furthermore, as has been repeatedly explained - just because The Cause of The Effect may not be clear (ie, the proximal cause) does not mean that causality is not a meaningful concept. What does it prove in any case if you are right that cause and effect is meaningless? That black people have more babies out of wedlock? And that being black means you are likely to have been born out of wedlock? Let us assume you want to solve the problem. In concrete terms what would you do? TFS It isn't a semantic trick, you misunderstood my argument. The coherent model of the universe statement is meant to include everything that we know about our current universe. It means that there is a way given all the information we have recieved that any belief can be false. Often such universes which disprove what we believe and still make sense given everything else we already know contain some kind of explanation why we would be decieved into believing what we did before. They range from completely arbitrary (there is an evil deceiver making us think everything we see is real, even though the concept of an evil deceiver is based on our experience in the world we are now claiming is false) to things that we now know are true about our world, are now widely understood, and were only disproofs of beliefs sets people had prior to some important discovery. The evil deciever example just goes to show you no matter how much faith you have in a given belief it can still be disrpoved by additional information (since everything known is disproven in the evil deciever case). If the evil deciever example is to radical for you, then just imagine that you were dreaming. It happens all the time, you thought everything around you was real and the next minute you wake up and realize it was a dream. So yeah, there could be replicants you don't know about, or force fields or different gravitational constants and that still be completely coherent with the current universe. But more than that I don't think it would even be beyond current science to decapitate someone and have them live in a controlled enviornment. proximal cause... lol ok mr bandwagoner. Cause and effect is not a meaningless concept its just a vague concept that requires us to rely heavily on our unattended subconsious to use. Which probably just defines causality as some function of correlation (and maybe order of events) anyways. So why not take control of the process and use correlation directly so we can have more control and precision. I already told you in concrete terms what I would do. I would find the factor with the highest correlation to OOW births that I could alter (taking into account other issues such as cost) and alter it. What more do you want? You want me to speculate on what that factor is? It's not a problem that I am concerned with or have done research on. My issue is with poor reason by egalitarianism biased scientists not with black people.
TheFaithfulStone Posted October 11, 2006 Report Posted October 11, 2006 I would find the factor with the highest correlation to OOW births that I could alter (taking into account other issues such as cost) and alter it. The two highest -correlating- factors are income and race, which are independently correlated with each other. I don't know to what degree. For the sake of argument, assume r=.7 for race, r=.9 for income, and r=.6 for race v income. That is 70% of OOW births are to black mothers, 90% are to poor mothers, and 60% of black people are poor. There are other factors. Which ones do you adjust? Admittedly, it's a tricky problem with no single correct answer, but I want to see how you attack it if you can't say "A is caused by B." I want a concrete answer. I want to know what the highest correlating factor you can control is, and I want to know how you alter it. TFS PS - PLEASE do not take those statistics as fact - they are for demonstration purposes only.
Kriminal99 Posted October 11, 2006 Author Report Posted October 11, 2006 The two highest -correlating- factors are income and race, which are independently correlated with each other. I don't know to what degree. For the sake of argument, assume r=.7 for race, r=.9 for income, and r=.6 for race v income. That is 70% of OOW births are to black mothers, 90% are to poor mothers, and 60% of black people are poor. There are other factors. Which ones do you adjust? Admittedly, it's a tricky problem with no single correct answer, but I want to see how you attack it if you can't say "A is caused by B." I want a concrete answer. I want to know what the highest correlating factor you can control is, and I want to know how you alter it. TFS PS - PLEASE do not take those statistics as fact - they are for demonstration purposes only. Ok. In this scenario I would privatize secondary education.
Recommended Posts