But I’m actually more interested in discussing “what this ‘I’ is” as you put it.
Ah, indeed. It’s kind of an elusive creature though. I was thinking it might be better to start with some of the things it is supposedly doing.
@Downbylaw
No, there isn’t such an assumption. A lot of things are assumed to be random, so if the starting conditions are duplicated very accurately, you will just get consistent probabilities—say 10% of the time it will do this, 90% of the time it will do something else and there won’t be any way to improve on those results.
Many things that involve quantum mechanics are believed to be intrinsically probabilistic. So a certain radioactive isotope might have a half-life of 4.98383 years. You can determine a half-life quite accurately like that, but it might be intrinsically impossible to predict exactly when any particular nucleus decays.
OK, I figured it was the case that these probabilities would be a part of it, but I just wanted to ask kind of openly just to see how you understood it exactly.
So I have some more questions then.
In that case is there an assumption that if all properties of a certain system were repeated exactly the same a huge number of times, you would get the exact same you would be able to predict very accurately for each component of the system the percentage of times different specific results would occur every time they were repeated. And thus, should be able to predict all the possible outcomes and the percentage each possible outcome would occur?
Other things seem to be “functionally random”. This happens in complex feedback systems where even extraordinary tiny differences in starting conditions soon get amplified into large scale effects. Weather is a classic example. You can do pretty good forecasting of most weather for 12 to 24 hours, but most of the time predictions 5 days out are not very accurate. Even if you could gather vastly more data about the atmosphere, and crank it through really gargantuan computers, you would only extend good predictions modestly.
Also, another question. Is there an assumption that to search for the causes of things, we need to keep breaking things down into smaller and smaller components? And then when we understand what is happening at the smallest levels we can put all of those things together and see how they all work together to add up to understand how the larger system works?
So in the case of what you call “functionally random” is only seemingly random because there are so many tiny differences that are all adding up that we could never have the resources and capabilities of actually being able to measure them all to make a prediction. However these things are not actually truely random because we are aware that the results still are just based on a lot of smaller processes that if mesured individually could be accurately predicted?
@RISS
These are good questions.
In that case is there an assumption that if all properties of a certain system were repeated exactly the same a huge number of times, you would get the exact same you would be able to predict very accurately for each component of the system the percentage of times different specific results would occur every time they were repeated. And thus, should be able to predict all the possible outcomes and the percentage each possible outcome would occur?
Yes, that’s right. The idea is maybe easier to grasp if you flip it around the other way: If you observe a change in the probability distribution, then the “properties of a certain system were repeated exactly” wasn’t true after all. Something was different.
Also, another question. Is there an assumption that to search for the causes of things, we need to keep breaking things down into smaller and smaller components? And then when we understand what is happening at the smallest levels we can put all of those things together and see how they all work together to add up to understand how the larger system works?
No, you won’t find many scientists who think that because detailed knowledge at fine scales often doesn’t illuminate what is interesting about larger scales. The weather is a good example again. The properties of a volume of air are well described by simple gas laws that use just volume, pressure, and temperature. But pressure and temperature are really just statistical averages from all the gas molecules colliding and bouncing off each other. The difference between these two scales, however, is immense. Even a small volume of air has a vast number of molecules, and if you then try to think about all the different ways two molecules can ricochet off each other…. So if you imagined somehow having a weather model that calculated all these collisions, you would have vast amounts of data about endless boring blobs of still dry air. Really interesting stuff like cold fronts and thunderstorms wouldn’t stand out.
Oh, and your weather model would have another serious problem. It certainly won’t run on your laptop. And it will be slow. No matter how big of a computer you built, it wouldn’t be able to calculate the exact weather 30 days out in less than thirty days. Actually Earth itself is the only “computer” that can do this calculation, and to calculate the exact weather 30 days from now takes, yes, exactly 30 days.
So in the case of what you call “functionally random” is only seemingly random because there are so many tiny differences that are all adding up that we could never have the resources and capabilities of actually being able to measure them all to make a prediction. However these things are not actually truely random because we are aware that the results still are just based on a lot of smaller processes that if mesured individually could be accurately predicted?
That would be true if the smallest possible scale actually was perfectly deterministic. But quantum physics once again leads us to believe that is not so. Time and space become very different at such minuscule scales, and that puts a limit on how accurately something can be measured even in principle. So there really are intrinsic limits to how accurate predictions could ever be.
Yes, that’s right. The idea is maybe easier to grasp if you flip it around the other way: If you observe a change in the probability distribution, then the “properties of a certain system were repeated exactly” wasn’t true after all. Something was different.
So just to clarify for sure, a result is always entirely contingent on these probabilities?
No, you won’t find many scientists who think that because detailed knowledge at fine scales often doesn’t illuminate what is interesting about larger scales. The weather is a good example again. The properties of a volume of air are well described by simple gas laws that use just volume, pressure, and temperature. But pressure and temperature are really just statistical averages from all the gas molecules colliding and bouncing off each other. The difference between these two scales, however, is immense. Even a small volume of air has a vast number of molecules, and if you then try to think about all the different ways two molecules can ricochet off each other…. So if you imagined somehow having a weather model that calculated all these collisions, you would have vast amounts of data about endless boring blobs of still dry air. Really interesting stuff like cold fronts and thunderstorms wouldn’t stand out.
Oh, and your weather model would have another serious problem. It certainly won’t run on your laptop. And it will be slow. No matter how big of a computer you built, it wouldn’t be able to calculate the exact weather 30 days out in less than thirty days. Actually Earth itself is the only “computer” that can do this calculation, and to calculate the exact weather 30 days from now takes, yes, exactly 30 days.
What it seems to me like you are saying is that it is literally impossible for us as humans to get anywhere near to having the capability of creating a model that would predict almost any kind of complex system with any kind of accuracy. Not only because we are ourselves part of the system we would need to model, but also because we would need to have a system that could store more information than the universe itself. But if there was somehow some realm outside of the universe that had that capability, technically the way that it would be modeled would be to understand how things behave at the very smallest level first and then put them altogether. Does that go along with your position?
Or if it’s not true that how things behave at the smallest level are what ultimately entirely govern collectively the behavior of the entire system, could it then be true that there is some kind of larger entity that is in a way “more than the sum of it’s parts” or somehow “other than the sum of it’s parts” that actually has governing power onto or over the typical behaviors of the smallest parts?
That would be true if the smallest possible scale actually was perfectly deterministic. But quantum physics once again leads us to believe that is not so. Time and space become very different at such minuscule scales, and that puts a limit on how accurately something can be measured even in principle. So there really are intrinsic limits to how accurate predictions could ever be.
So in a previous post you said that many things in the universe are intrinsically probabilistic, and you also said that these things give consistent probabilities. So is this limit on how accurate predictions could be based on that fact that humans or a person within the system itself could never accurately measure it for the reason given above? Or is this limit on how accurate predictions could ever be based on the fact that the probabilities of things are not actually consistent. Or is it because not everything is intrinsically probablistic. Earlier you said “many” things are intrinsically probabilistic, but not all things. So what are the other things?
@RISS
So just to clarify for sure, a result is always entirely contingent on these probabilities?
Well to really clarify we need to remember the armchair philosopher’s assumptions that we are making such as knowing the complete probability distribution. As an empirical matter, you wouldn’t know about those probabilities that are much too rare to have shown up in the finite time you have been observing. An example of this is that some “unification theories” predict that protons eventually decay. Experiments have been looking for this for a few decades now, but have found no evidence that it happens. So as things stand now, the half-life of the proton must be at least 10^33 years. Or to write it out:
1,000,000,000,000,000,000,000,000,000,000,000 years
For comparison, the universe is only 10^10 years old:
14,000,000,000 years
The experiments have ruled out theories that expect a half-life of 10^31 years, but other theories predict a somewhat longer half-life that is still within the sensitivity of the device (the Super-Kamiokande in Japan) so the experiment continues. Still, they will pull the plug on it someday. All the experiment can do is set a lower limit on a possible half-life even if it is a stunning number of years.
But if there was somehow some realm outside of the universe that had that capability, technically the way that it would be modeled would be to understand how things behave at the very smallest level first and then put them altogether. Does that go along with your position?
So now we are sitting in a really BIG armchair, huh? I’m not inclined to claim that such an outside perspective even makes sense. But if the question is something like “Is the maximal possible amount of information about the state of the universe enough to predict its exact future?”, I’d suggest probably not.
could it then be true that there is some kind of larger entity that is in a way “more than the sum of it’s parts” or somehow “other than the sum of it’s parts” that actually has governing power onto or over the typical behaviors of the smallest parts?
No, at least not in the sense I think you are suggesting. There is, of course, the issue of environment. So an example here might be electronics where we have electrons zipping around through billions of transistors over and over. A huge problem in such systems is not letting your signal get swamped by the accumulating noise from all these interactions. That’s one of the virtues of digital—instead of letting your signal smear out into useless thermal noise, you keep repackaging it into nice crisp bits. Electronics are carefully designed to exclude the intermediate states that would be digitally ambiguous. You might have a wire that is at, say, 2 volts or 4 volts, but never at 3 volts. You build the environment so that the voltage cannot get to 3 volts.
Earlier you said “many” things are intrinsically probabilistic, but not all things. So what are the other things?
What I said (or at least meant to say) was that there are all these things that we have to treat as being probabilistic whether or not the universe at its core is probabilistic. But I suspect that it is.
Ok, this is only tangentially related, but I just watched this video that I liked. It is on entropy and the “arrow of time”, and makes these ideas much easier to understand:
Great contribution to the thread so far Downbylaw. Here’s a couple of easy to understand videos I found touching on the same subject as your video.
@Downbylaw
Well to really clarify we need to remember the armchair philosopher’s assumptions that we are making such as knowing the complete probability distribution. As an empirical matter, you wouldn’t know about those probabilities that are much too rare to have shown up in the finite time you have been observing. An example of this is that some “unification theories” predict that protons eventually decay. Experiments have been looking for this for a few decades now, but have found no evidence that it happens. So as things stand now, the half-life of the proton must be at least 10^33 years. Or to write it out:
1,000,000,000,000,000,000,000,000,000,000,000 years
For comparison, the universe is only 10^10 years old:
14,000,000,000 years
The experiments have ruled out theories that expect a half-life of 10^31 years, but other theories predict a somewhat longer half-life that is still within the sensitivity of the device (the Super-Kamiokande in Japan) so the experiment continues. Still, they will pull the plug on it someday. All the experiment can do is set a lower limit on a possible half-life even if it is a stunning number of years.
So is the answer “yes” or “no”?
So now we are sitting in a really BIG armchair, huh? I’m not inclined to claim that such an outside perspective even makes sense. But if the question is something like “Is the maximal possible amount of information about the state of the universe enough to predict its exact future?”, I’d suggest probably not.:
No, at least not in the sense I think you are suggesting. There is, of course, the issue of environment. So an example here might be electronics where we have electrons zipping around through billions of transistors over and over. A huge problem in such systems is not letting your signal get swamped by the accumulating noise from all these interactions. That’s one of the virtues of digital—instead of letting your signal smear out into useless thermal noise, you keep repackaging it into nice crisp bits. Electronics are carefully designed to exclude the intermediate states that would be digitally ambiguous. You might have a wire that is at, say, 2 volts or 4 volts, but never at 3 volts. You build the environment so that the voltage cannot get to 3 volts.
So it’s neither way? Then which way is it? Is there some kind of a third way? Which way did you think I was suggesting? Since we are talking about conciousness did you think I was talking about the human mind? Because I was.
What I said (or at least meant to say) was that there are all these things that we have to treat as being probabilistic whether or not the universe at its core is probabilistic. But I suspect that it is.
Why do we have to treat them this way if it’s just a suspicion?
So is the answer “yes” or “no”?
Yes, on theoretical grounds. Empirically, you could say with a high certainty (which could be improved if you are willing to invest in the necessary research).
So it’s neither way? Then which way is it? Is there some kind of a third way? Which way did you think I was suggesting? Since we are talking about conciousness did you think I was talking about the human mind? Because I was.
Right, “higher level” structures don’t disrupt the behaviors of the underlying stuff. So there is not some “special chemistry” in living things that violates the rules of “regular chemistry”. Chemistry is chemistry. Living things are notable for having lots of very delicate and complex chemistry that we are far from fully understanding. But there is not the slightest theoretical or empirical reason to think any of it is “irregular”. The same goes for brains. There are no reasons to suspect that anything funny is going on in a brain when we think there are conscious states than when we think there are not. Consciousness (or the lack of it) has no relevance for understanding neurological, biological, or chemical processes.
Why do we have to treat them this way if it’s just a suspicion?
Because you can never know enough about the system to be able to treat in any other way. So in the case of weather models, most models themselves run deterministically. But a single run is never considered useful by itself. So you run your model over and over making slight changes in the starting data each time. If ¾ of the runs show rain in the mountains on Friday, then you predict a 75% chance of rain in the mountains on Friday (to simplify a bit).
This is very much the case with what goes on in brains. A lot of neural processes work on the basis of a threshold: a large number of inputs are going up and down, but nothing happens till they collectively cross some threshold level. Since there is also a lot of noise, the best you can do is say that the probability of crossing the threshold is going up or down. This is very different from how we design electronic circuits where we put together large numbers of small units that all behave in a very predictable manner.
Yeah, so back to free will. Sam Harris offers this: “The popular conception of free will seems to rest on two assumptions: (1) that each of us could have behaved differently than we did in the past, and (2) that we are the conscious source of most of our thoughts and actions in the present.”
Does this capture your idea of free will? Does it get at what House was talking about? And does it distinguish, say, between people with Tourette’s and people without—what would be “acts of free will” for most of us are not such for Tourette’s tics?
@Gustavo
Brian Cox is pretty good at it, isn’t he?
Yes, on theoretical grounds. Empirically, you could say with a high certainty (which could be improved if you are willing to invest in the necessary research).
Gotcha.
Right, “higher level” structures don’t disrupt the behaviors of the underlying stuff. So there is not some “special chemistry” in living things that violates the rules of “regular chemistry”. Chemistry is chemistry. Living things are notable for having lots of very delicate and complex chemistry that we are far from fully understanding. But there is not the slightest theoretical or empirical reason to think any of it is “irregular”. The same goes for brains. There are no reasons to suspect that anything funny is going on in a brain when we think there are conscious states than when we think there are not. Consciousness (or the lack of it) has no relevance for understanding neurological, biological, or chemical processes.
Right, “higher level” structures don’t disrupt the behaviors of the underlying stuff. So there is not some “special chemistry” in living things that violates the rules of “regular chemistry”. Chemistry is chemistry. Living things are notable for having lots of very delicate and complex chemistry that we are far from fully understanding. But there is not the slightest theoretical or empirical reason to think any of it is “irregular”. The same goes for brains. There are no reasons to suspect that anything funny is going on in a brain when we think there are conscious states than when we think there are not. Consciousness (or the lack of it) has no relevance for understanding neurological, biological, or chemical processes.
So the question is where we look to see the cause of behavior of a complex system. There are some different ideas that we can affirm, reject, or be not sure about.
One is that we can break down a system into smaller and smaller bits and if we could get an idea of what seems to cause these smaller bits to behave the way they do, then we could add them all up together and see that the behavior of the whole system is just the total of these smaller bits all put together. I’m still not clear on whether this is an idea that you affirm, reject, or are not sure about. It seems in some places you affirm it and in other places you seem to suggest it’s probably not true. So if you could clarify that for me, I’d be interested.
Another idea would be sort of the opposite, that instead of the smaller parts causing behavior of the the whole that something of the whole could cause behavior of the whole or the smaller parts. It’s the idea that the behavior of the whole could not even theoretically be understood just by understanding what is going on the smaller parts and adding it all up. It seems to me that you reject this idea, although I am still not entirely sure.
Maybe only one of these two ideas could be true together, or maybe they could both be true at the same time. Or maybe neither of them is true and there is something else.
Also since I’m talking about the mind here, do you make a distiction between the mind and the brain?
Because you can never know enough about the system to be able to treat in any other way. So in the case of weather models, most models themselves run deterministically. But a single run is never considered useful by itself. So you run your model over and over making slight changes in the starting data each time. If ¾ of the runs show rain in the mountains on Friday, then you predict a 75% chance of rain in the mountains on Friday (to simplify a bit).
So the things that we treat as probabilistic should be done so only because to treat them that way yeilds results that are useful and practical to our needs? You still did not mention what the things are that you insinuated earlier were not probabilistic.
This is very much the case with what goes on in brains. A lot of neural processes work on the basis of a threshold: a large number of inputs are going up and down, but nothing happens till they collectively cross some threshold level. Since there is also a lot of noise, the best you can do is say that the probability of crossing the threshold is going up or down. This is very different from how we design electronic circuits where we put together large numbers of small units that all behave in a very predictable manner.
The best you may be able to do practically from the point of view of trying to understand the ways the brain works physically from the way we study how small processes might affect the whole is from this probabalistic point of view, but I’m asking about the theoretical assumptions of probability behind the practical which inform it.
Yeah, so back to free will. Sam Harris offers this: “The popular conception of free will seems to rest on two assumptions: (1) that each of us could have behaved differently than we did in the past, and (2) that we are the conscious source of most of our thoughts and actions in the present.”
Does this capture your idea of free will?
Moreso the first item than the second item. I think we do a lot more uncionciously than conciously.
Does it get at what House was talking about?
I have a guess that based on what House said before he would agree with me, but you’d have to ask him to comment.
And does it distinguish, say, between people with Tourette’s and people without—what would be “acts of free will” for most of us are not such for Tourette’s tics?
I’ve never been diagnosed with Tourrette’s but I believe I actually have a mild form of it. I certainly have a lot of tics. And almost every aspect of it that is described in that video or in the Wikipedia article on it I identify with. All of these tactile urges, a sense of doing something impulsively I know is dangerous or that I shouldn’t, feeling an urge to repeat certain words that sound certain ways, sometimes and urge to really burst out with vulgar outbursts. The most prevalent is certian eye, face, hand, or neck ticks. They usually come more unonciously in certain senarios. Certain ticks come up more often with more certain reliable scenarios or when certain stimuli is present. So I identify.
A lot of things we tend to do unconciously we can have a concious control over if we put it in the forefront of our concious mind and focus on it. Obviously for some people it may take more mental energy to control certain things over others. And it may wear you out more and sometimes eventually overtake you. But you still have control over whether or not to make an effort in a certian direction regardless of how successful you are.
This is all my impression of things.
Has anyone read this book? Amazon reviews seem to say it’s pretty dry read, but that doesn’t mean to me that it’s bad:
wrong thread;)- ignore
Another idea would be sort of the opposite, that instead of the smaller parts causing behavior of the the whole that something of the whole could cause behavior of the whole or the smaller parts.
Except that that only the smaller parts existed at the beginning of the universe. The egg defined (and existed before) the chicken.
Thanks for jumping in Jerry. It sounds like that’s what Downbylaw is saying as well, although he said “at least not in the sense I think you are suggesting.” implying that maybe there was some other sense we weren’t addressing. I didn’t really get what he was saying about electrons had to do with the question though.
Do you agree with Downbylaw in everything else he was saying?
Maybe we can discuss the issue of conciousness until he comes back. Unless you don’t agree with him in which case we may have to start the questioning over for me to see where you stand to better understand it.
I don’t want to abandon the conversations that had been going on this page regarding free will, conciousness, and related matters, but in the meantime until Downbylaw or someone else continues the conversation I have another question that is probably actually related to the questioning on this page depending on the different points of view people have.
For those who are athiests or who believe in naturalism or who don’t believe in God:
Why should we persue science?
I don’t know, Riss…I’m starting to lose interest in the God vs. Science debate. Today’s Catholic Church does not fundamentally disagree with a single component of advanced theoretical physics. Never the twain shall meet.
Instead of copy pasting a bunch of text, I’ll just put the link up. Of all the New Atheists (if that is even an appropriate description) I think Harris is the most interesting, and Sullivan is just good all around.
I’m glad I found this thread since I’ve read/watched a lot of the New Atheists and the popular exporters of science to the masses (Dawkins, Dennett, Harris, Hitchens, Krauss, Kaku, Greene, Degrasse Tyson, Pinker, …).
For those who are athiests or who believe in naturalism or who don’t believe in God:
Why should we persue science?
Science specifically or why do we pursue anything at all? I ask because it is a common challenge put to atheists – the question of what meaning life could possibly have without a God. The short answer for that is that it is up to us to make our own meaning through our relationships with others. As for science specifically, it’s that scientific knowledge can increase the quality of life for human civilization and this is something that even atheists are interested in. Why wouldn’t they be?
dp
But of course we do not really control everything about the decisions we make—we are acted upon by all sorts of external factors that influence (even determine) our decisions. The choices are essentially still ours, in that I can choose to go one way or another in a given situation, but that situation may have been brought about by circumstances entirely beyond my control, and so I am subject to making one choice from a set that was determined by factors outside of my will.
Free will in the sense that I first considered it really doesn’t exist. But I continue to stand behind the idea that “God” hasn’t predetermined everything.
House of Leaves, the problem with free will is the lack of a non-physical source of decision-making. If our decisions aren’t the results of combinations of external factors and biological events, then where are they coming from?A soul? What good evidence is there, or can there possibly be for the existence of souls?
Of course, we have the illusion of free will. I can be in the mood for a coffee and then go buy one. But that decision was the result of a desire which was the result of a physical stimulus which was the result of goings on in my body, etc. All our decisions wind up being spurred on by factors that were out of our control. I don’t think quantum improbability changes that.
DownByLaw
@HOUSE:
Ah, “external to your will” is an interesting way to put it. But we will have to be careful to not just be tautological: you have free will for those things that are not external to your will. So we should probably try to come up with a few things that seem like good ol’ ordinary exercises of free will.
Some of the items on your list of externals (“jobs, finances, governments, laws, marriage, friendships, time, hunger, self-preservation”) seem to be about the options available to you, others about shifts in consequences. So probably none of us here have a few million to drop on buying a big yacht, so that choice is not even available to us at all. On the other hand, we could all go out to singles bars every night. If we are single, that is no big deal. But if we are married, it might cause some serious problems with the spouse. So you might say the choice is there, but the outcome is too bad to consider.
Yeah, mental illness is something I want to look at. In particular I was thinking about Tourette Syndrome. A few years ago the neurologist Oliver Sacks did a really good program on PBS Nova that included a drummer, Matt Giordano, who has Tourette’s. Giordano has severe tics, but they disappear while he is drumming. I wanted to post the video, but PBS doesn’t have it online. But I did find this other video of Sacks about Tourette’s. This one features a man whose tics mostly disappear while he does martial arts. It’s a good video and a good introduction to Tourette’s. And I think it will be useful to our conversation. Here is part one:
And while I didn’t find a link to the PBS program, I did find this video of Matt Giordano playing the drums. Pretty cool: