As I said before, I consider it an offense against reason to explain it any further. If you want to be difficult, that’s your choice.
Clayton -
I cannot write a program which is smarter than I am.
I’m not so sure, I think much software dealing in expert domains, can solve problems in that domain better than a mind, even if we grant that mind rudimentary pens, papers, even calculators… but its a semantic question about ‘smarter’. and besides I agree with you in your Monism, and Methodological Dualasism exegisis, and also agree with your point about the infeasibility of modelling such complex systems as the brain to the point of being able to accurately predict precise future states of such systems.
I’ll take a stab, and I’m prepared to get shot from all sides but it seems to me that when we talk about a ‘will’ or ‘choosing’ we mean something like that there is an object that we can usefully abstract apart from its environment to which we ascribe more explanatory power to the systems inner state than to its environment when we consider how it transitions to a future state .
so this deals with the difference between a cockroach running towards or away from a wall, and a coin landing tail side when caused to fall.
The former cockroach ‘chose’ the direction to run, and the coin didnt decide which side to fall on. It also gives some meaning to a computer chess program choosing a move.
What then of ‘free will’ , or choosing in the specially human sense with which praxeology is concerned?
So heres my understanding of ‘free will’. This is an example of a continuum problem where at one end, we have primitive yet still complex systems, but they are relatively more coupled with their environments than systems like us, over here in an advanced state, that are more-decoupled, and that perhaps go so far as to have internal representations, and ‘illusions of choice’.
Thankfully seeing as how in economics people are mostly interested in human action, the notion of free will is therefore a lot less ambigious and tractable than if we were trying to understand animal behaviours in social contexts, or robot ‘choices’, in which case there would perhaps be a YET MORE vigorous debate about methodology (monism/dualism - mechanistics/teleological analysis).
Imagine that nightmare.
I mean it in a very well-defined sense. That is, I cannot write a program which can compress data better than I can. This is trivially true since whatever program I write is itself a device I can use to compress data.
There are very good reasons why data-compression is an “ultimate” measure of smartness/intelligence/whatever-you-want-to-call-it.
I want to underscore that the problem with predicting brain behavior is much deeper than other kinds of chaotic systems. Sure, it’s really difficult to predict the weather. But there is always the possibility that we will have a breakthrough which will vastly increase our ability to predict weather. But there is a qualitative difference to the human brain… the problem inherent in developing a “theory of everything” of the human brain is ineradicable, no matter what technological advances are made. It’s not just really hard, it’s impossible and provably so.
Clayton -
But is it ‘you’ that compresses data when you command your OS to tell WinRar to compress a file.
Your argument sounds like you are saying that no machine you can make is ever stronger than you, because whatever machine you design and build is itself a machine that you use to lift things.
@nir: Let us treat the brain as a computer program (say, by imagining we are running Blue Brain). Let us call this program B.
Let us further say that program X produces program Y and then we execute Y to produce output Z; the division between X and Y is wholly arbitrary, we may as well say that X outputs Z.
By the same token, if B writes a program Q that compresses a file F from 100KB to 10KB, then we can say B compressed F from 100KB to 10KB.
Just clicking on WinRAR is not the same as writing it. But a human brain wrote it, so it really doesn’t matter to the question of the capacity of the human brain whether you or someone else wrote it. Whoever wrote it was human.
Note that the form of compression that algorithmic information theory deals with is much more absolute than the kind of compression that WinRAR does… it is the smallest form to which each particular string can be compressed (relative to some reference machine, such as a Universal Turing Machine… and it turns out that the choice of reference machine doesn’t matter that much). Programs like Winzip or gzip or WinRAR are designed to compress “statistically”, that is, to perform well across many kinds of strings by looking for “statistical patterns” in them. But algorithmic compression results in the shortest possible string. It is an uncomputable problem, meaning, no one could ever write a program of finite size that does algorithmic compression on any given string.
Clayton -
Ok, so you are saying that you are as tall as the tallest thing you can build, as strong as the strongest thing you can build, and as fast as the fastest thing you can build. It strikes me as an odd way off looking at the world which I am not used to, but you have given your reasons for it (it hinges on the arbitrary distinction between the builder and the built), and who am I to argue with you on that.
Can two people write a program which is smarter than you are?
Do you think the Blue Brain project will simulate a human brain in the end? What will stop researchers from simulating something with 10x the neurons and 100x the synapses?
Human brains aren’t compression algorithms. I think your use of theorems from information theory is quite a stretch here.
I think this depends entirely on your definition for “comprehend”. I already know that my brain has a scale of values and it plans and causes actions so as to remove felt uneasiness. My brain also gathers external information to assist in making efficacious plans. Maybe predicting my own behavior to the minutest detail without the help of external technology is impossible, but certainly understanding gross properties of the brain is conceivable.
I can’t comprehend 325,796 objects all at once, but I can perform mathematical operations on that quantity or understand it as a small sequence of digits or possibly break it into prime factors.
I can’t comprehend a 100-line computer program all at once, and know which bytes are going where and when at which CPU cycle. But I can certainly break it into manageable conceptual chunks and manipulate it to achieve desired goals.
It seems full comprehension is overrated.
@nir: The obvious difference is that software is merely a pattern/abstraction. I am not the building I build. But in the same sense that tools merely extend the hand and notes extend the memory, so software is just an extension of the brain (in particular, its discursive logical capacity).
But the theoretical point stands in any case. The output of an intermediate program Y output by a program-writing program X can just as well be thought of as the output of X. Any distinction between X and Y is merely prejudicial.
Clayton -
Perhaps, though neither one of them alone will be able to comprehend it. This is a bit like how the market itself works. It is the sum of individual actions though no one individual could ever really comprehend it.
There is no reason I see why it won’t work. However, I don’t think it will help with the most fundamental questions of consciousness (namely, what exactly conscious awareness is or how it arises within the “gears-and-levers” physical world).
Well, I didn’t assert that the human brain is a compression algorithm. The arguments regarding compression have to do with setting bounds on what sorts of problems can be solved by a program of a certain size. It turns out that harder problems require bigger programs. This is an indirect consequence of Godel’s theorems; the harder the problem you want to solve, the bigger a program you will need to solve it (this is not a performance issue, either, I mean to solve it in any finite time).
I mean full comprehension as measured by the ability to find the most compact scientific theory (to “compress the brain program” to its smallest size). Again, I commend Chaitin, Calude, Hutter, Solomonoff, Li, Vitanyi, et. al. to your attention.
Clayton -
No one comprehends the motion of individual molecules in a 1 mililiter container of gas. And, really, no one cares. Rather, people speak of a few variables like temperature, density, pressure, mean free path, etc. Likewise, for some purposes, a greatly simplified approximation to a human brain could suffice. It’s another reason why your pedantic arguments from computation theory aren’t very compelling. You can’t prove the non-existence of a workable math formula that predicts Clayton’s brain with 99% accuracy.
This kind of thing is arguably an unanswerable metaphysical question. I guess, to me, “superhuman intelligence” doesn’t mean cracking these things. I’m thinking more along the lines of this: http://en.wikipedia.org/wiki/Holly_(Red_Dwarf “Holly prides himself on the fact he had an IQ of 6,000 [2] (apparently the same IQ as 6,000 PE teachers or 12,000 car park attendants)”
But there’s a crucial difference here. In the case of a gas, the macroscopic phenomena emerge from the statistical properties of the microscopic phenomena. There is clearly no such simple relationship between the microscopic state of the brain and macroscopic behavior.
Of course. And this is precisely what we operate with already. The human brain already contains an intuitive model of human behavior.
But that’s beside the point. Of course we can predict behavior (what the hell do you think praxeology is???) and of course we can analyze, reverse engineer and understand to a greater and greater degree the function of the brain. But we’ll never get a “theory of everything” with respect to the brain, and that was the original point. We’re so far from ever being able to surpass the human brain that the best we can hope to do is to understand the human brain better and better over time. We could also create brain-like machines and put them in an environment where they might evolve greater intelligence and see what happens. But we’ll not be designing a super-brain that will outwit us all, ala SkyNet or HAL 9000.
Clayton -
So just to be completely clear, you have no idea what you mean when you say, “a person has free will”. As a result, nobody else knows either. Fair enough, however typical of the free willy types. I won’t embarrass you any further.
You are lying. I don’t know what you mean. Ironically, neither do you. Anyway it’s a red herring to say you can’t “explain it any further”, since you’ve avoided ever explaining it to begin with. And that’s fine, I didn’t expect you to be able to even if you’d tried.
Again with the red herring and question dodging. You can’t begin to explain it further, since you haven’t explained it to begin with.
@nirgrahamUK, if you mean to describe “free will”, then what is it free relative to? Is there will that is separate from “free” will? And if you say “a person has free will”, what do mean by “a person”? An identifiable portion of his brain? Or the brain as a whole? Really, I’m just wondering from the perspective of a free willy, I’ve had trouble in the past trying to establish what any given person means when they say “a person has free will”.
the ‘free’ of free will is a modifier, meant to highlight that we are a long way on the continuum towards having complicated systems that we point to as choosers that are in this regard decoupled from their environments, than the other end where knees reflexively jerk when hit, and where rocks fall off cliff edges.
/Just my interpretation, others likely have their own ways of thinking about it.
My point in bringing all of this up is in objection to the theory that “a person” should be held accountable for his actions. The brain’s unconscious, uncontrollable, automatic processing and storing input, and constructing an ever growing algorith for cost-benefit, precedes action. It makes no sense to hold the the acting portion of the brain accountable for the instructions, which were out of its control.
That doesn’t mean the brain may not instruct the acting portion to influence the brain. Obviously (to me) that’s what we consider “a person changing his mind”. But it seems the acting portion isn’t responsible for the fact that it must act according to the instructions it recieves.
EDIT: So the brain may cause a knee-jerk reflex. But the brain also causes action. The acting portion can’t be held responsible for the unconscious, uncontrollable rest of the brain’s failure to send instructions to influence a change in the brain’s direction. Neither can an unconscious, automatic, and uncontrollable rest-of-the-brain be held responsible for failing to produce those instructions.
What are you doing? positing a homunculous that is the acting part of the brain, that just has to put up with the inputs of the robot part of the brain?
I don’t think it makes any sense to divy up the brain quite like that…
Your brain-body, is you, an object that I point to and say that you chose this or that action, not that the sky did it , or that your parents did it. Is that really so complicated and so contested? Are you pleading for the truth of a position of philosophical skepcticism, on whether ‘you’ exist to be held ‘accountable’ for ‘actions’, with or without further arguments that might explain why you don’t demonstrate your philosophy in your daily dealings ?
I’m not positing unconscious brain activity. It’s a given, nobody even denies unconscious brain activity. Neither does anyone deny that most brain activity is unconscious. When I say automatic and uncontrollable, I mean it doesn’t require conscious instruction, nor is there a route for it to respond to such. You can’t will that your hair grow faster.
Otherwise yes, it is a fairly accurate representation of my understanding to say that the automatic, uncontrollable, unconscious brain activity recieves, stores, accesses, and processes input, and produces output, some of which is conscious thought and action. I don’t think anyone challenges this. I think a lot of people have completely bypassed ever thinking about this, and have therefore jumped to conclusions about responsibility.
Do you deny that automatic, unconscious, uncontrollable brain activity precedes conscious action?
I deny a heirarchy. I’m sure there are automatic unconscious and ‘uncontrollable’ brain activity. (actually not sure what is meant by uncontrollable here), but that the activities, are happying simultaneously with the activities you classify as willed/conscious, and that there is feedback going both ways between these ‘sub-systems’.
but, anyway, all this talk about brains might just miss the point about moral responsibility. If you look instrumentally at a morally suspect action, if I can criticise you, and that is a way to mechanicaly influence your unconcious brain to reduce in the future the likelihood of you doing similar morally suspect actions, that would seem to be effacious. And looking at the situation askance, there is no basis to launch a claim against being held responsible for an aggregious act on the basis of a lack of free will, as any person who one might claim this against would by the premise of irresponsibility, not be responsible for holding you responsible for an eggregious act. (cool eh ?)
Ok well slow down, let’s see if I can get a straight answer the second time around. Do you deny that automatic, unconscious, uncontrollable brain activity precedes conscious action?
This has nothing to do with a hierarchy. But it’s silly to say you were conscious and acting before your brain ever functioned automatically, unconsciously, and uncontrollably.
A flock of birds is emergent of indvidual birds located proximate to each other in the sky. Do I deny that individuals bird activity precede flock activity? well, in the sense that individual birds have to gather together to get the flock started, they do proceed it. Is that the only sense in which you mean what you are challenging me with? My point was that whilst there is a flock, the birds and the flock system are simultaneous (the flock pretty much just is the birds right?) its not like the birds do stuff → then the flock does stuff. The birds do stuff and the flock does stuff simultaneously.
Conciousness is what, activity under a mode of self-awareness, but does anyone think that it is not emergent of a non-self aware substrate? But what does this have to tell us about moral responsibility anyway. Its not clear to me that you or I should even particularly care whether I am right or wrong about the details of how brains work, because if issues of moral responsibility are at the heart of our real disagreement, didn’t my other comments that weren’t directly related to this brain stuff address that more directly?