jadwa@henry.cs.reading.ac.uk (James Anderson) (05/03/88)
In article 1380 of comp.ai: bwk@mitre-bedford.ARPA (Barry W. Kort) says: > Suppose I were able to inculcate a Value System into silicon. > And in the event of a tie among competing choices, I use a > random mechanism to force a decision. Would the behavior of > my system be very much different from a sentient being with > free will? Well ... I take, "free will" to mean that an agent can choose what action to take despite the choices that other agents might make and despite unchosen events in the world. There are a number of corollaries to this definition. 1) I am a strong minded person, so I often exercise free will, but you can deny me free will, say, by killing me. 2a) You might exercise your free will by making an oath with yourself never to deny me free will, say, by never applying irresistible force to me. 2b) Making an oath does not deny your own free will. You can chose to break the oath. 3a) Random events in the choice mechanism deny free will to the extent that they prevent the agent from determining the outcome of a decision. 3b) I know of no way to discover, by observing the behaviour of an agent, that exception (3a) applies to it. So the answer to your question is "yes": I can not tell apart the behaviour of a random system and one with free will. A free-will system could, for example, chose to behave randomly. 4) An event which is not of any agents choosing might deny me free will. The event may be random, as in losing a game of Russian roulette, or deterministic, like being being drowned in a rising tide. So far, so good. But here comes the free will paradox, with knobs on! If the world is deterministic I am denied free will because I can not determine the outcome of a decision. On the other hand, if the world is random, I am denied free will because I can not determine the outcome of a decision. Either element, determinancy or randomness, denies me free will, so no mixture of a deterministic world or a non-deterministic world will allow me free will. I am going to think about that for a little while before I post again. James (JANET) James.Anderson@reading.ac.uk
vu0112@bingvaxu.cc.binghamton.edu (vu0112) (05/05/88)
In article <770@onion.cs.reading.ac.uk> jadwa@henry.cs.reading.ac.uk (James Anderson) writes: >3a) Random events in the choice mechanism deny free will to the > extent that they prevent the agent from determining the > outcome of a decision. Interesting. So you understand freedom to be the ability for me to determine my own actions, as opposed to them being determined by external sources. Thus you're defining freedom in terms of a different kind of determinism, a stance which seems problematic. Why not instead freedom as simply the *lack* of *complete* external control? This way you allow *degrees of freedom*. That is, I am free if I am not determined. This in no ways implies that I in turn must determine something else. This also helps us sort out the difference between subjective and objective uncertainty (freedom). Objective freedom is understood as the (e.g. quantum) inherent uncertainty in processes, whereas subjective uncertainty is the premise that I lack information about a possibly determinate process. On my definition, both conditions indicate freedom to me, but to you in the latter case we are still determined. >So far, so good. But here comes the free will paradox, with knobs >on! I think my definition knocks your knobs off. >If the world is deterministic I am denied free will because I can >not determine the outcome of a decision. For me, if the world is deterministic I am granted freedom when I cannot determine the outcome of a decision. >On the other hand, if >the world is random, I am denied free will because I can not >determine the outcome of a decision. Yes, no matter if the world is determined or not, I can never determine the outcome of a decision. This is an inherent epistemic limitation, independent of the state of the world. I think that's obviously correct. Further, by showing us that determinism is impossible, I think you've just demonstrated that freedom is necessary. >I am going to think about that for a little while before I post >again. What I find fascinating is the implicit assumption that determinism is the "normal" general case, and that freedom is come kind of strange property rarely seen. I think that just the opposite is true, that deterministic processes of any kind are very rare. Demonstrating the freedom of *any* organism, let alone humans, is trivial. Demonstrating our determinism is a silly philosophical waste of time. -- O----------------------------------------------------------------------> | Cliff Joslyn, Cybernetician at Large | Systems Science, SUNY Binghamton, vu0112@bingvaxu.cc.binghamton.edu V All the world is biscuit shaped. . .
sher@sunybcs (David Sher) (05/05/88)
It seems that people are discussing free will and determinism by
trying to distinguish true free will from random behavior. There is a
fundamental problem with this topic. Randomness itself is not well
understood. If you could get a good definition of random behavior you
may have a better handle on free will.
Consider this definition of random behavior:
X is random iff its value is unknown.
This is I believe a valid definition of randomness. But in this case
free will may be a subset of random behaviors. Other more
sophisticated definitions may be proposed for randomness.
On a similar note to decide if we can allow machines to take
responsibility (which seems to be bothering our english contingent), we
must decide just what responsibility is. We already entrust machines
with our lives and have for thousands of years (since we invented
boats).
-David Sher
ARPA: sher@cs.buffalo.edu BITNET: sher@sunybcs
UUCP: {rutgers,ames,boulder,decvax}!sunybcs!sherjeff@aiva.ed.ac.uk (Jeff Dalton) (05/06/88)
In article <770@onion.cs.reading.ac.uk> jadwa@henry.cs.reading.ac.uk (James Anderson) writes: >If the world is deterministic I am denied free will because I can >not determine the outcome of a decision. On the other hand, if >the world is random, I am denied free will because I can not >determine the outcome of a decision. Either element, determinancy >or randomness, denies me free will, so no mixture of a >deterministic world or a non-deterministic world will allow me >free will. Just so. Having one's actions determined randomly isn't much help. One of the problems with discussing free will here is that it's too easy to simply rehash arguments that have been handled in the philosophical literature. I thought it best to make this point in response to an article that I agreed with, though, because I'm not claiming that no one ever says anything valuable or that I am some kind of expert in these matters with no time to listen to the rest of you. Nonetheless, anyone who is seriously interested in such topics should be willing to do some reading. I would recommend Dennet's Elbow Room: The Varieties of Free Will Worth Wanting for its discussion of free will, for its relevance to AI, and for the interesting things that come up along the way. Jeff Dalton, JANET: J.Dalton@uk.ac.ed AI Applications Institute, ARPA: J.Dalton%uk.ac.ed@nss.cs.ucl.ac.uk Edinburgh University. UUCP: ...!ukc!ed.ac.uk!J.Dalton
bwk@mitre-bedford.ARPA (Barry W. Kort) (05/06/88)
I was intrigued by David Sher's comments about "machine responsibility". It is not uncommon for a child to "spank" a machine which misbehaves. But as adults, we know that when a machine fails to carry out its function, it needs to be repaired or possibly redesigned. But we do not punish the machine or incarcerate it. Why then, when a human engages in undesirable behavior, do we resort to such unenlightened corrective measures as yelling, hitting, or deprivation of life-affirming resources? --Barry Kort
lloyd@eniac.seas.upenn.edu (Lloyd Greenwald) (05/06/88)
In article <10942@sunybcs.UUCP> sher@wolf.UUCP (David Sher) writes: >It seems that people are discussing free will and determinism by >trying to distinguish true free will from random behavior. There is a >fundamental problem with this topic. Randomness itself is not well >understood. If you could get a good definition of random behavior you >may have a better handle on free will. > This is a good point. It seems that some people are associating free will closely with randomness. To me true randomness is as difficult to comprehend as true free will. We can't demonstrate true randomness in present day computers; the closest we can come (to my knowledge) is to generate a string of numbers which does not repeat itself. Can anyone give us a better view of randomness then this? I've heard some mention of true randomness at the quantum level. Does anyone have any information on this? Given that current theories of free will tie it so closely to randomness, it seems necessary to get a handle on true randomness. Lloyd Greenwald lloyd@eniac.seas.upenn.edu
vu0112@bingvaxu.cc.binghamton.edu (Cliff Joslyn) (05/07/88)
In article <4543@super.upenn.edu> lloyd@eniac.seas.upenn.edu.UUCP (Lloyd Greenwald) writes: >This is a good point. It seems that some people are associating free will >closely with randomness. Yes, I do so. I think this is a necessary definition. Consider the concept of Freedom in the most general sense. It is opposed by the concept of Determinism. We can say of anything, either it is absolutely determined (it will always do one thing and only one thing), or it is somwhat free (sometimes it will do one thing, other times another). This is so whether we talk of molecules in a box or the actions of an organism. >To me true randomness is as difficult to comprehend >as true free will. I agree. That's because both psychological free will and randomness are cases of my general sense of Freedom. Freedom is a very difficult things to understand. >We can't demonstrate true randomness in present day >computers; von Neumann machines are highly Determined systems. They posess so little Freedom that it is essentially null. This is what they have been designed to do. However, it is easy to demonstrate that von Neumann machines are slightly free. Consider the distribution of bit errors in a cpu or RAM, or of read errors on a disk drive. These are random events. To that extent the computer is Free. This is not especially useful or interesting Freedom, nevertheless it is there. >the closest we can come (to my knowledge) is to generate a string >of numbers which does not repeat itself. This is not possible in a von Neumann machine. >I've heard some mention of true randomness at the >quantum level. See recent (last two years) articles in _Scientific American_ concerning hidden variables theories in QM. As I described in a brevious article, we can think of two cases of randomness, subjective and objective. Subjective randomness is usually equated with ignorance. For example, in Newtonian physics if I had sufficient information about initial conditions I could predict the roll of a die. Objective randomness is your "true", or irreducible, or inherent, or unavoidable randomness. There has been a great debate as to whether quantum uncertainty was subjective or objective. The subjectivists espoused "hidden variables" theories (i.e.: there are determining factors going on, we just don't know them yet, the variables are hidden). These theories can be tested. Recently they have been shown to be false. >Given that current >theories of free will tie it so closely to randomness, it seems necessary to >get a handle on true randomness. In my mind, the critical thing to understand about Freedom is that Freedom is always relative; Determinism is always absolute. What I mean is that when we talk about something being Free, we can always talk about degrees of freedom. A six sided die is more Free than a four sided, a twelve than a six. Or consider a probability distribution: it's Freedom is generally measured by it's entropy, which takes values in the interval [ 0, inf ). In order for the distribution to reach the infinite limit, it must be uniformly distributed over the whole positive real interval. This distribution is not well defined. In other words, we know what it means for something to be completely Determined. I submit that it is not possible for somethings to be completely Free. Absolute Freedom is an infinite limit; absolute determinism is a zero limit. This is obviously true in the realm of human affairs as well. It is easy for me to completely determine your actions: put you in a Skinner box, or straight jacket, or just kill you. And while I espouse Free Will, I do so only in this relative way. In no way can you tell me that you are absolutely free: drug delusions, dreams, illness, epilepsy, all kinds of physical/biological factors come into play which somewhat limit the Freedom of your mind. -- O----------------------------------------------------------------------> | Cliff Joslyn, Cybernetician at Large | Systems Science, SUNY Binghamton, vu0112@bingvaxu.cc.binghamton.edu V All the world is biscuit shaped. . .
bwk@mitre-bedford.ARPA (Barry W. Kort) (05/07/88)
James Anderson writes: >If the world is deterministic I am denied free will because I can >not determine the outcome of a decision. On the other hand, if >the world is random, I am denied free will because I can not >determine the outcome of a decision. Either element, determinancy >or randomness, denies me free will, so no mixture of a >deterministic world or a non-deterministic world will allow me >free will. It is not clear to me that a mixture of determinism and randomness could not jointly create free will. A Thermostat with no Furnace cannot control the room temperature. A Furnace with no Thermostat cannot control the room temperature. But join the two in a feedback loop, and together they give rise to an emergent property: the ability to control the room temperature to a desired value, notwithstanding unpredicted changes in the outside weather. Similarly, could it not be the case that Free Will emerges from a balanced mixture of determinism (which permits us to predict the likely outcome of our choices) and freedom (which allows us to make arbitrary choices)? Just as the Furnace+Thermostat can drive the room temperature to a desired value, Cause+Chance gives us the power to drive the future state-of-affairs toward a desired goal. If you buy this line of reasoning, then perhaps we can get on to the next level, which is: How do we select goal states which we imagine to be desirable? --Barry Kort
baba@calvin.uucp (Duane Hentrich) (05/07/88)
In article <31024@linus.UUCP> bwk@mbunix (Barry Kort) writes: >It is not uncommon for a child to "spank" a machine which misbehaves. >But as adults, we know that when a machine fails to carry out its >function, it needs to be repaired or possibly redesigned. But we >do not punish the machine or incarcerate it. Have you not bashed or kicked a vending machine until it gave up the junk food you paid for. It is my experience that swatting a non-solid state TV tuner sometimes results in clearing up the picture. Indeed we do "punish" our machines. Sometimes with results which clearly outweigh the damage done by such punishment. >Why then, when a human engages in undesirable behavior, do we resort >to such unenlightened corrective measures as yelling, hitting, or >deprivation of life-affirming resources? For the same reason that the Enter/Carriage Return key on many keyboards is hit repeatedly and with great force, i.e. frustration with an inefficient/ineffective interface which doesn't produce the desired results. No value judgements re results or punishments here. If this doesn't belong here, please forgive. d'baba Duane Hentrich ...!hplabs!oliveb!tymix!baba Claimer: These are only opinions since everything I know is wrong. Copyright notice: If you're going to copy it, copy it right.
ok@quintus.UUCP (Richard A. O'Keefe) (05/07/88)
In article <10942@sunybcs.UUCP>, sher@sunybcs (David Sher) writes: > It seems that people are discussing free will and determinism by > trying to distinguish true free will from random behavior. There is a > fundamental problem with this topic. Randomness itself is not well > understood. If you could get a good definition of random behavior you > may have a better handle on free will. In particular, consider the difference between _random_ behaviour and _chaotic_ behaviour. A phsyical system may be completely described by simple deterministic laws and yet be unpredictable in principle (unpredictable by bounded computational mechanisms, that is). [Pseudo-random numbers are really pseudo-chaotic.] > Consider this definition of random behavior: > X is random iff its value is unknown. I do not known David Sher's telephone number, but I do not find it useful to regard it as random (nor as chaotic). Conversely, when listening to a Geiger counter, I am quite sure whether or not I have heard a click, but I believe that the clicks are random events.
ok@quintus.UUCP (Richard A. O'Keefe) (05/07/88)
In article <1179@bingvaxu.cc.binghamton.edu>, vu0112@bingvaxu.cc.binghamton.edu (Cliff Joslyn) writes: > In article <4543@super.upenn.edu> lloyd@eniac.seas.upenn.edu.UUCP (Lloyd Greenwald) writes: > >This is a good point. It seems that some people are associating free will > >closely with randomness. > > Yes, I do so. I think this is a necessary definition. > > Consider the concept of Freedom in the most general sense. It is > opposed by the concept of Determinism. For what it is worth, my "feeling" of "free will" is strongest when I act in accord with my own character / value system &c. That is, when I act in a relatively predictable way. There is a strong philosophical and theological tradition of regarding free will and some sort of determinism as compatible. If I find myself acting in "random" or unpredictable ways, I look for causes outside myself ("oh, I brought the wrong book because someone has shuffled the books on my shelf"). Randomness is *NOT* freedom, it is the antithesis of freedom. If something else controls my behaviour, the randomness of the something else cannot make _me_ free. I suppose the philosophical position I can accept most readily is the one which identifies "free will" with being SELF-determined. That is, an agent possesses free will to the extent that its actions are explicable in terms of the agent's own beliefs and values. For example, I give money to beggars. This is not at all random; it is quite predictable. But I don't do it because someone else makes me do it, but because my own values and beliefs make it appropriate to do so. A perfectly good person, someone who always does the morally appropriate thing because that's what he likes best, might well be both predictable and as free as it is possible for a human being to be. > There has been a great debate as to whether quantum uncertainty was > subjective or objective. The subjectivists espoused "hidden variables" > theories (i.e.: there are determining factors going on, we just don't > know them yet, the variables are hidden). These theories can be tested. > Recently they have been shown to be false. Hidden variables theories are not "subjectivist" in the usual meaning of the term; they ascribe quantum uncertainty to objective physical processes. They haven't been shown false. It has been shown that **LOCAL** hidden variables theories are not consistent with observation, but NON-local theories (I believe there are at least two current) have not been falsified.
vu0112@bingvaxu.cc.binghamton.edu (Cliff Joslyn) (05/07/88)
In article <940@cresswell.quintus.UUCP> ok@quintus.UUCP (Richard A. O'Keefe) writes: >In article <1179@bingvaxu.cc.binghamton.edu>, vu0112@bingvaxu.cc.binghamton.edu (Cliff Joslyn) writes: >> In article <4543@super.upenn.edu> lloyd@eniac.seas.upenn.edu.UUCP (Lloyd Greenwald) writes: >> >This is a good point. It seems that some people are associating free will >> >closely with randomness. >> >> Yes, I do so. I think this is a necessary definition. >> >> Consider the concept of Freedom in the most general sense. It is >> opposed by the concept of Determinism. > >For what it is worth, my "feeling" of "free will" is strongest when I >act in accord with my own character / value system &c. That is, when I >act in a relatively predictable way. Yes, it is more complicated, isn't it? What if, instead of having a "value system", you were rather in the grip of some hideous, controlling "ideology." Let's say you're Ronald Raegan, or Botha, or Gorbachev. I wouldn't want to deny them free will, but would say that their ideologies are highly determining (at least on political issues). On the other hand, let's say that I am an intelligent, impressionable child in an "ideas bazarre," say a super-progressive, highly integrated school in New York. Then my value system will be in constant flux. Also, don't confuse predictibility with determinism. There are degrees of predictibility. If I know the distribution of a random variable, I can make some degree of prediction. >Randomness is *NOT* freedom, it is the antithesis of freedom. >If something else controls my behaviour, the randomness of the >something else cannot make _me_ free. Yes, it is critical to keep the levels of analysis clear. If something external *determines* your behavior (for example, a value system), then your behavior is *determined* no matter what. The *cause* of your behavior being free in no way implies you are free. But we aren't talking about something controlling you, we are talking about whether you are controlled. My assertion is that if you are completely controlled, then you cannot act randomly. If you are free, than you can. Try this: freedom implies the possibility of randomness, not its necessity? >That is, >an agent possesses free will to the extent that its actions are >explicable in terms of the agent's own beliefs and values. >For example, I give money to beggars. This is not at all random; it is >quite predictable. But I don't do it because someone else makes me do it, >but because my own values and beliefs make it appropriate to do so. In order for something to be not at all random, it must not just be quite predictible, but rather completely predictible. To that extent, it is determined. Again, we're talking at different levels (probably a subjective/objective problem). Let's try this: if you are free, that means it is possible for you to make a choice. That is, you are free to scrap your value system. At each choice you make, there is a small chance that you will do something different, something unpredictible given your past behavior/current value system. If, on the other hand, you *always* adhere to that value system, then from my perspective, that value system (as an *external cause*) is determining your behavior, and you are not free. The problem here may be one of observation: if a coin "chooses" to come up heads each time, I will say that it's necessary that it does, as an inductive inference. There's a lot of issues here. I don't either of us have thought through it very clearly. >A perfectly good person, someone who always does the morally appropriate >thing because that's what he likes best, might well be both predictable >and as free as it is possible for a human being to be. He is not free so long as it is not possible for him to act imorally. I say it is then impossible to distinguish between someone who is free to act imorally, and chooses not to, and someone who is determined to act morally. >It has been shown that **LOCAL** hidden >variables theories are not consistent with observation, but NON-local >theories (I believe there are at least two current) have not been falsified. Thanks for the clarification. -- O----------------------------------------------------------------------> | Cliff Joslyn, Cybernetician at Large | Systems Science, SUNY Binghamton, vu0112@bingvaxu.cc.binghamton.edu V All the world is biscuit shaped. . .
smoliar@vaxa.isi.edu (Stephen Smoliar) (05/08/88)
In article <31024@linus.UUCP> bwk@mbunix (Barry Kort) writes: > >It is not uncommon for a child to "spank" a machine which misbehaves. >But as adults, we know that when a machine fails to carry out its >function, it needs to be repaired or possibly redesigned. But we >do not punish the machine or incarcerate it. > >Why then, when a human engages in undesirable behavior, do we resort >to such unenlightened corrective measures as yelling, hitting, or >deprivation of life-affirming resources? > This discussion seems to be drifting from the issue of intelligence to that of aggression. I do not know whether or not such theses have gone out of fashion, but I still subscribe to the hypothesis that aggression is "natural" to almost all animal life forms, including man. Is your adult self so rational and mature that you have not so much as banged your fist on the table when your software does something which particularly frustrates you (or do you feel that adults also transcend frustration)?
doug@feedme.UUCP (Doug Salot) (05/09/88)
In article <1179@bingvaxu.cc.binghamton.edu> Cliff Joslyn writes: >In article <4543@super.upenn.edu> Lloyd Greenwald writes: >>This is a good point. It seems that some people are associating free will >>closely with randomness. > >Yes, I do so. I think this is a necessary definition. > >[good points about QM vs Classical vs Ignorance as views of Freedom deleted] I don't believe randomness (in the quantum mechanical sense) is important to a sense of free will. The illusion of free will is what's important, and when dealing with a computing machine (the brain) which makes state changes on the time order of milliseconds, it is simply impossible for that machine, self-aware or not, to view its state changes as deterministic when its state changes are based on finer-grained state changes that occur at or near the speed of light. It seems to me that it would be straight-forward to give a computer program with the ability to monitor its *behavior* without giving it the ability to find causal relations between its holistic state and its behavior. -- Doug Salot || doug@feedme.UUCP || {trwrb,hplabs}!felix!dhw68k!feedme!doug Feedme Microsystems:Inventors of the Snarf->Grok->Munge Development Cycle
ok@quintus.UUCP (Richard A. O'Keefe) (05/09/88)
In article <1182@bingvaxu.cc.binghamton.edu>, vu0112@bingvaxu.cc.binghamton.edu (Cliff Joslyn) writes: > Again, we're talking at different levels (probably a > subjective/objective problem). Let's try this: if you are free, that > means it is possible for you to make a choice. That is, you are free to > scrap your value system. At each choice you make, there is a small > chance that you will do something different, something unpredictible > given your past behavior/current value system. If, on the other hand, > you *always* adhere to that value system, then from my perspective, that > value system (as an *external cause*) is determining your behavior, and > you are not free. I'm not sure that there is any point in continuing this, our basic presuppositions seem to be so alien. Orthodox Christianity holds that - God is able to do anything that is doable - God is not constrained by anything other than His own nature - it is impossible for God to sin From my perspective, such a God is maximally free. From Chris Joslyn's perspective, such a God is minimally free. I think the problem lies in my disagreement with Joslyn's definition quoted above. He defines freedom as the ability to make choices, and seems to regard unmotivated (random) ``choices'' as the freest kind. I also take exception with his view that a value system is an external cause. My "value system" is as much a part of me as my memories. Or are my memories to be regarded as an external cause too? Note that behaving consistently according to a particular value system does not mean that said value system is immune from revision. There are some interesting logical problems involved: in order to move _rationally_ from one state of your value system to another, you have to believe that the new state is _better_, which is to say that your existing value system has to endorse the new one. I would regard "scrapping" one's value system as irrational (and it is not clear to me that it is possible far a human being to do it), and if being able to do it is freedom, that's not a kind of freedom worth having. My objection to randomness as a significant component of freedom is that a random act is not an act that **I** have *willed*. If I were to randomly put my fist through this screen, it wouldn't be _my_ act any more than Chris Joslyn's or J.S.Bach's. It is sheer good luck if a random act happens to be in accord with my wishes. Unfortunately, we have living proof that randomness as such is not freedom: consider the people who suffer from Tourette's syndrome. It might be objected that behaving in a way consistent with one's goals, beliefs, and wishes is too much like behaving in a way consistent with a program to be counted as freedom. It would be, if we were not capable of revising said goals and beliefs. To be free, you have to _check_ your beliefs. I propose as a tentative definition that a robot can be said to possess free will provided that its actions are in accord with its own mental models and provided that it has sufficient learning capacity to be able to almost wholly replace the mental models it is initially provided with.
bwk@mitre-bedford.ARPA (Barry W. Kort) (05/09/88)
I was doing fine reading Cliff's rejoinder to Lloyd's comments until I came to this part: >>We can't demonstrate true randomness in present day computers; >>the closest we can come (to my knowledge) is to generate a string >>of numbers which does not repeat itself. [Lloyd] >This is not possible in a von Neumann machine. [Cliff] I was under the impression that a simple recursion (or not-so-simple if one is a fan of Ramanujan) can emit the digits of pi (or e or SQRT(2)) and that such a string does not repeat itself. I think what Cliff meant is that a von Neumann machine cannot emit a string whose structure cannot be divined. If I wanted to give my von Neumann machine a *true* random number generator, I would connect it to an A/D converter driven by thermal noise (i.e. a toasty resister). --Barry Kort
geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) (05/09/88)
In article <31024@linus.UUCP> bwk@mbunix (Barry Kort) writes: >It is not uncommon for a child to "spank" a machine which misbehaves. >But as adults, we know that when a machine fails to carry out its >function, it needs to be repaired or possibly redesigned. But we >do not punish the machine or incarcerate it. > >Why then, when a human engages in undesirable behavior, do we resort >to such unenlightened corrective measures as yelling, hitting, or >deprivation of life-affirming resources? > Pray tell, how do you repair, or redesign a human? Is "Clockwork Orange" the model we want to strive for? If you had a machine which was running amok, and you did not know how to repair it or redesign it, would not destroying it or isolating it from the objects of its aggression be a prudent course? Punishment can serve to "redesign" the human machine. If you have children, you will probably know this. Unfortunately, it doesn't work with everyone.
vu0112@bingvaxu.cc.binghamton.edu (Cliff Joslyn) (05/10/88)
In article <948@cresswell.quintus.UUCP> ok@quintus.UUCP (Richard A. O'Keefe) writes: >I'm not sure that there is any point in continuing this, our basic >presuppositions seem to be so alien. All the more reason to keep talking, in my book! >Orthodox Christianity holds that >[1] - God is able to do anything that is doable >[2] - God is not constrained by anything other than His own nature >[3] - it is impossible for God to sin >From my perspective, such a God is maximally free. Hmm. [1] and [3] seem blatantly inconsistent - surely sinning is doable? [2] seems to be saying that there are no external causes acting on God - doesn't this beg the question as to the cause of God's nature? Perhaps you don't require God to hold to logical argument - I have no problem with that. I'm sure you're sensitive to the logical problems in making arguments about such absolute and self-sufficient entities as Gods. >From Chris Joslyn's perspective, such a God is minimally free. ^^^^^ Cliff >I think the problem lies in my disagreement with Joslyn's definition >quoted above. I think the problem is that I'm basically talking from an Objective perspective: how can I know whether something is free or not? Whereas you're speaking phenomenologically. >He defines freedom as the ability to make choices, and Surely this is the essence of the concept! What else could you possibly offer? Spite is the essence of Freedom. See, I believe, Kierkegaard. >seems to regard unmotivated (random) ``choices'' as the freest kind. *This* may be the real argument - and where you might have me. >I also take exception with his view that a value system is an >external cause. This is also sensitive - the question is: external to what? To your body? Obviously not: your value system is in your mind, which is in your body, so you're right that it's not external to you as a whole. On the other hand, it does seem to be external to your *will* your *deciding agent*, *that which is making the choice*. Otherwise, you would be *compelled* to act in accordance with your value system. On the contrary, I say that you *are* free, which means you are *free to choose* whether or not to adhere to your value system, whether or not to commit an evil act, say. Surely this is also traditional theology? OK, so you are free *not* to adhere to your value system. On what basis is this choice made? Can you tell me? If you can, then we can recurse the argument and say: well, then you are also *free* to ignore that basis. In the limit, we will arrive at chance. > in order to move >_rationally_ from one state of your value system to another, you have >to believe that the new state is _better_, which is to say that your >existing value system has to endorse the new one. I agree. But notice that you are now offering *rationality* as the basis for decision making - not free will. Are you free to act irrationally? If not, then you are not free. >I would regard >"scrapping" one's value system as irrational (and it is not clear to >me that it is possible far a human being to do it), and if being able >to do it is freedom, that's not a kind of freedom worth having. You're making my argument for me: we're not arguing whether freedom is *worth having*, but whether you have it! If, as you suggest, it is impossible for people to scrap their value systems, then surely we can say that they are not free to scrap their value systems. To that extent, their value systems constrain their freedom, and their values systems partially determine (that is control) their behavior. >My objection to randomness as a significant component of freedom is >that a random act is not an act that **I** have *willed*. Of course. But you're getting confused: your will determines your behavior, true. What determines your will? >Unfortunately, we >have living proof that randomness as such is not freedom: consider the >people who suffer from Tourette's syndrome. Yes, but reflexes are not acts of whole people, rather acts of parts of people. Something in the Tourette's victim's *body* is controlling their behavior - here the behavior itself is free from mental control, in fact the mind (and that is what we are arguing about: mental freedom) is not involved. That's the problem: they desparately *want* to determine their behavior. -- O----------------------------------------------------------------------> | Cliff Joslyn, Cybernetician at Large | Systems Science, SUNY Binghamton, vu0112@bingvaxu.cc.binghamton.edu V All the world is biscuit shaped. . .
vu0112@bingvaxu.cc.binghamton.edu (Cliff Joslyn) (05/10/88)
In article <31337@linus.UUCP> bwk@mbunix (Kort) writes: >I was doing fine reading Cliff's rejoinder to Lloyd's comments until >I came to this part: > >>>We can't demonstrate true randomness in present day computers; >>>the closest we can come (to my knowledge) is to generate a string >>>of numbers which does not repeat itself. [Lloyd] > >>This is not possible in a von Neumann machine. [Cliff] > >I was under the impression that a simple recursion (or not-so-simple >if one is a fan of Ramanujan) can emit the digits of pi (or e or >SQRT(2)) and that such a string does not repeat itself. > >I think what Cliff meant is that a von Neumann machine cannot emit >a string whose structure cannot be divined. Hmm, I suppose you're right. I was thinking of your typical psuedo-random process whose cycle length was a function of the size of the seed. I forget the impact on the argument at this point: it seems it would rest on the epistemic grounds of determining a truly random string from a simply chaotic one. My impressions is that this is not always possible. Food for mailer Food for mailer Food for mailer Food for mailer Food for mailer -- O----------------------------------------------------------------------> | Cliff Joslyn, Cybernetician at Large | Systems Science, SUNY Binghamton, vu0112@bingvaxu.cc.binghamton.edu V All the world is biscuit shaped. . .
bwk@mitre-bedford.ARPA (Barry W. Kort) (05/10/88)
Perhaps it would help if I offered a straw proposal for invoking one's free will in a specific situation. Assume that I possess a value system which permits me to rank my personal preferences regarding the likely outcome of the courses of action open to me. Suppose, also, that I have a (possibly crude) estimate of your value system. If I were myopic (or maybe just stupid) I would choose my course of action to maximize my payoff without regard to you. But my knowledge of your value system creates an interesting opportunity for me. I can use my imagination to conceive a course of action which increases both of our utility functions. Free will empowers me to choose a Win-Win alternative. Without free will, I am predestined to engage in acts that hurt others. Since I disvalue hurting others, I thank God that I am endowed with free will. Is there a flaw in the above line of reasoning? If so, I would be grateful to someone for pointing it out to me. --Barry Kort
ken@vedge.UUCP (Ken) (05/10/88)
In article <1177@bingvaxu.cc.binghamton.edu> vu0112@bingvaxu.cc.binghamton.edu (Cliff Joslyn) writes: >Why not instead (define) freedom as simply the *lack* of *complete* external >control? Although this definition intuitively concurs with our every-day notion of freedom (e.g. free to leave my country), I would question whether it is meaningful to our discussion of free will. I am concerned with the term "external". In our every-day lives we have a concept of that which comes from us (e.g. our speech) and that which comes from sources exterior to ourselves (e.g. a postcard from a friend). But in the realm of the infinitesimal, the realm of uncertainty, "mystery"; the place where quantum phenomenon live... how do we distinguish the "external" from the "internal"? If the laws which govern the motion of the partical/waves which compose my brain are external to me, then are also their individual motions? If we believe that the physical brain is wholly responsible for all the goings on in the mind, then are we left with any "internal" at all? In short, I question the value of your definition for its understanding of freedom in terms of "external" control. float ken(); /* fourth year student of Pure Math/Philosophy at U of Waterloo (Ontario) */
bwk@mitre-bedford.ARPA (Barry W. Kort) (05/11/88)
I appreciated Richard O'Keefe's suggestion that free will is intimately related to the freedom to learn. This idea is consistent with the notion that one cannot create a sentient being without free will. Moreover, it is evidently unpredictable what a sentient being will in fact discover and learn in his/her/its lifetime. --Barry Kort
bwk@mitre-bedford.ARPA (Barry W. Kort) (05/11/88)
Stephen Smoliar seems frustrated at the drift of the discussion. He writes: >This discussion seems to be drifting from the issue of intelligence to that >of aggression. I do not know whether or not such theses have gone out of >fashion, but I still subscribe to the hypothesis that aggression is "natural" >to almost all animal life forms, including man. Is your adult self so >rational and mature that you have not so much as banged your fist on the >table when your software does something which particularly frustrates you >(or do you feel that adults also transcend frustration)? It is not clear to me whether aggression is instinctive (wired-in) behavior or learned behavior. I think the pschological jury is still out on this question. It is certainly true that aggressive and non-aggressive behaviors can be learned. (Personally, I feel that assertive behavior is preferrable to aggressive, and tactful behavior is preferable to assertive. But tactful behavior is harder to learn.) As to your question about my personal habits when frustrated, I do not bang my fist on the table. Rather, I clench my teeth. And I do believe it is possible (though difficult) to transcend frustration. --Barry Kort
fink@mist.UUCP (05/12/88)
I'm new to this group and would just like to ask,
IS anything useful ever talk about here, you know,
like AI?
Don't reply, I'm 'U'ing out.
BYegilbert@cs.glasgow.ac.uk (Gilbert Cockton) (05/18/88)
In article <1176@cadre.dsl.PITTSBURGH.EDU> geb@cadre.dsl.pittsburgh.edu.UUCP (Gordon E. Banks) writes: >Punishment can serve to "redesign" the human machine. If you have children, >you will probably know this. Unfortunately, it doesn't work with everyone. How can an artificial intelligence ever rise above recidivism? Are there any serious examples of re-programming systems, i.e. a system that redesigns itself in response to punishment. Don't generalise from crude 'learning' systems either, I don't think these would scale up. That reminds me, what was it the behaviourists said about punishment?
bwk@mitre-bedford.ARPA (Barry W. Kort) (05/19/88)
In article <1176@cadre.dsl.PITTSBURGH.EDU> geb@cadre.dsl.pittsburgh.edu.UUCP (Gordon E. Banks) writes: >In article <31024@linus.UUCP> bwk@mbunix (Barry Kort) writes: >>It is not uncommon for a child to "spank" a machine which misbehaves. >>But as adults, we know that when a machine fails to carry out its >>function, it needs to be repaired or possibly redesigned. But we >>do not punish the machine or incarcerate it. >> >>Why then, when a human engages in undesirable behavior, do we resort >>to such unenlightened corrective measures as yelling, hitting, or >>deprivation of life-affirming resources? > >Pray tell, how do you repair, or redesign >a human? Is "Clockwork Orange" the model we want to strive for? >If you had a machine which was running amok, and you did not know >how to repair it or redesign it, would not destroying it or isolating >it from the objects of its aggression be a prudent course? Punishment >can serve to "redesign" the human machine. If you have children, you >will probably know this. Unfortunately, it doesn't work with everyone. We don't call it repair, we call it healing. But your question is a good one. Unfortunately, I must confess to a lack of expertise on the subject. I am convinced that there is hope for healing the human spirit, but I feel we have a long way to go before we get very good at it. In Clockwork Orange, there was a curious mix of well-intentioned healers (the psychiatric staff), a doubtful theory (the Ludovico technique), and some opportunist politicians. I don't recommend the Ludovico technique as a model of healing. It seems to me that destroying something that one does not understand is not a good policy. Any agent who destroyed that which he did not understand could plausibly be classified as an agent who was "running amok". By the Symmetry Principle, such an agent is inviting his own destruction or isolation. Could it not be that the "acts of aggression" are really diagnostic messages, which when properly interpreted, allow us to "repair/redesign/heal" the offending system? --Barry Kort
geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) (05/20/88)
In article <32390@linus.UUCP> bwk@mbunix (Barry Kort) writes: >>>But as adults, we know that when a machine fails to carry out its >>>function, it needs to be repaired or possibly redesigned. But we >>>do not punish the machine or incarcerate it. >>> >>Pray tell, how do you repair, or redesign >>a human? >>If you had a machine which was running amok, and you did not know >>how to repair it or redesign it, would not destroying it or isolating >>it from the objects of its aggression be a prudent course? > >We don't call it repair, we call it healing. > >It seems to me that destroying something that one does not understand >is not a good policy. Alex in the Clockwork Orange was a fairly good representation of a psychopath--a person with no conscience or moral feelings whatsoever. These persons are much more common in the general population than most people believe. They aren't all murderers and rapists, but are successful politicians, deans, chairmen of departments, lawyers, doctors, etc. One could well argue that an antisocial, amoral variant could have definite evolutionary advantages when mixed in relatively small numbers among a primarily social species. Evidence is that their brains are actually wired differently than "normal" peoples'. We "understand" them on a certain level: their behavior is characterizable and predictable. For example, it can be predicted that they will not change. If incarcerated they will continue to be amoral in prison, when (if) released they will continue to be amoral afterward and will not have "learned their lesson". No therapy of any kind has been found to be effective. The tendancy to psychopathy seems to be hereditary. I would suggest that we understand them well enough that we are justified in destroying those psychopaths who are convicted of murders until such time (if ever) we find a way of "healing" them, simply for self-preservation, if nothing else.
geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) (05/23/88)
In article <1187@crete.cs.glasgow.ac.uk> gilbert@cs.glasgow.ac.uk (Gilbert Cockton) writes: >In article <1176@cadre.dsl.PITTSBURGH.EDU> geb@cadre.dsl.pittsburgh.edu.UUCP (Gordon E. Banks) writes: >>Punishment can serve to "redesign" the human machine. If you have children, >>you will probably know this. Unfortunately, it doesn't work with everyone. >How can an artificial intelligence ever rise above recidivism? Are >there any serious examples of re-programming systems, i.e. a system >that redesigns itself in response to punishment. Certainly! The back-propagation connectionist systems are "punished" for giving an incorrect response to their input by having the weight strengths leading to the wrong answer decreased. In most connectionist learning systems such "punishment" is used exclusively, not rewards. You may consider this a "crude" learning system, but it probably isn't much cruder than that actual neural apparatus underlying most organic brains.
bwk@mitre-bedford.ARPA (Barry W. Kort) (05/25/88)
I appreciated Gordon Banks' commentary on the psychopathic personality. His article reminded me of the provocative (but depressing) book by M. Scott Peck, _People of the Lie: The Hope for Healing Human Evil_. I'm actually more in favor of banishment than execution. But I do believe that healing is a possibility for the future. But given the difficulty of driving bugs out of software, I can appreciate the challenge of expunging demons from the human psyche. I agree that prison is not an effective environment for learning morality. It's hard to learn morality and ethics in an environment devoid of role models. --Barry Kort
wet@xanth.cs.odu.edu (Warren E. Taylor) (05/25/88)
In article <1176@cadre.dsl.PITTSBURGH.EDU>, geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) writes: In article <31024@linus.UUCP> bwk@mbunix (Barry Kort) writes: >It is not uncommon for a child to "spank" a machine which misbehaves. >But as adults, we know that when a machine fails to carry out its >function, it needs to be repaired or possibly redesigned. But we >do not punish the machine or incarcerate it. > >Why then, when a human engages in undesirable behavior, do we resort >to such unenlightened corrective measures as yelling, hitting, or >deprivation of life-affirming resources? > "Spanking" IS, I repeat, IS a form of redesigning the behavior of a child. Many children listen to you only when they are feeling pain or are anticipating the feeling of pain if they do not listen. Many "modern" types do not agree with this. I am certainly not for spanking a child every time he/she turns around, but there are also times when it is the only appropriate action. This occurs when a child repeatedly misbehaves even after being repeatedly informed of the consequences of his behavior. I have an extremely hard-headed nephew who "deserves" a spanking quite often because he is doing something that is dangerous or cruel or simply socially unacceptable. He is also usually maddeningly defiant. You only need to observe a baby for a short while to see a very nearly unadulterated human behavior. Young children have not yet been "socialized". They are very self-centered, and most of the time care only for themselves. This is why they scream "mine" and pitch fits at the slightest resistance to their will. Many people do not realize that a child has learned to manipulate their parents pretty well about the time they have learned to toddle. They know just how far they can push a parent. Haven't you ever seen a child keep doing something you are telling him "no" for until you give the slightest indication that you are going to get up to deliver that spanking you've promised? They then promptly quit. Lots of parents then sit back down. This has allowed the child to learn your tolerance point. There is a 100% chance he will do it again soon. In such a case, you need to get up and spank the child, even after he has stopped because you got up. I remember similar occurrences with my nephew when I babysat him. His "no-no" was trying to play with an electrical outlet. I solved the problem by giving him 2 warnings, and if that did not do it, an unavoidable paddling followed. He learned quickly that it was not worth the pain to defy me. Adults understand what a child needs. A child, on his own, would quickly kill himself. Also, pain is often the only teacher a child will listen to. He learns to associate a certain action with undesirable consequences. I am not the least bit religious, but the old Biblical saying of "spare the rod..." is a very valid and important piece of ancient wisdom. Flame away. Warren.
rargyle@wsccs.UUCP (Bob Argyle) (05/29/88)
In article <5323@xanth.cs.odu.edu>, wet@xanth.cs.odu.edu (Warren E. Taylor) writes: [stuff deleted] > Adults understand what a child needs. A child, on his own, would quickly kill > himself. ... > Flame away. > Warren. so stop interferring with that child's free will! [W.C.Fields] :-) We genetically are programmed to protect that child (it may be a relative...); not so programmed however for protecting any computers running an AI program. AI seems the perfect place to test the freewill doctrine without the observer interferring with the 'experiment.' At least one contributor to the discussion has called for an end to AI because of the effects on impressionable undergraduates being told that there isn't any free will. Send Columbus out and if he falls off the edge, so much the better. IF we get some data on what 'free will' actually is out of AI, then let us discuss what it means. It seems we either have free will or we don't; finding out seems indicated after is it 3000 years of talk.vague. So is the sun orbitting around the earth? this impressionable undergraduate wants to see some hard data. Bob @ WSCCS
gilbert@cs.glasgow.ac.uk (Gilbert Cockton) (05/30/88)
In article <1209@cadre.dsl.PITTSBURGH.EDU> geb@cadre.dsl.pittsburgh.edu.UUCP (Gordon E. Banks) writes: >>Are there any serious examples of re-programming systems, i.e. a system >>that redesigns itself in response to punishment. > >Certainly! The back-propagation connectionist systems are "punished" >for giving an incorrect response to their input by having the weight >strengths leading to the wrong answer decreased. I was expecting this one. What a marvellous way with words has mens technica. To restore at least one of the many subtle connotations of the word "punishment", I would ask Are there any serious examples of resistant re-programming systems, i.e. a system that redesigns itself only in response to JUST punishment, but resists torture and other attempts to coerce it into falsehood? I suspect the next AI pastime will be lying to the Connectionist machine (Hey, I've got this one to blow raspberries when it sees the word "Reagan", and that one now adds 2 and 2 to get 5!). Who gets the key to these machines? I'm already having visions of Luddite workers corrupting their connectionist robots :-) Will the first machine simulation be of some fundamentalist religious fanatic? -- Gilbert Cockton, Department of Computing Science, The University, Glasgow gilbert@uk.ac.glasgow.cs <europe>!ukc!glasgow!gilbert The proper object of the study of humanity is humans, not machines
cfh6r@uvacs.CS.VIRGINIA.EDU (Carl F. Huber) (05/31/88)
In article <5323@xanth.cs.odu.edu> wet@xanth.cs.odu.edu (Warren E. Taylor) writes: >In article <1176@cadre.dsl.PITTSBURGH.EDU>, geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) writes: > >"Spanking" IS, I repeat, IS a form of redesigning the behavior of a child. >Many children listen to you only when they are feeling pain or are anticipating >the feeling of pain if they do not listen. > Also, pain is often the only teacher a child will listen to. He From what basis do you make this extraordinary claim? Experience? or do you have some reputable publications to mention - I would like to see the studies. I also assume that "some" refers to the 'average' child - not pathological exceptions. > I have an extremely hard-headed nephew >who "deserves" a spanking quite often because he is doing something that is >dangerous or cruel or simply socially unacceptable. He is also usually >maddeningly defiant. ^^^^^^^^^^^^^^^^^^^ Most two to six year olds are. How old is this child? 17? What other methods have been tried? Spanking most generally results from frustrated parents who beleive they have "tried everything", while they actually haven't begun to scratch the surface of what works. >learns to associate a certain action with undesirable consequences. I am not >the least bit religious, but the old Biblical saying of "spare the rod..." is right, so let's not start something we'll regret, ala .psychology. > >Flame away. > Warren. voila. cheers. -carl
terry@wsccs.UUCP (Every system needs one) (06/11/88)
In article <566@wsccs.UUCP>, dharvey@wsccs.UUCP (David Harvey) writes: > In article <5323@xanth.cs.odu.edu>, wet@xanth.cs.odu.edu (Warren E. Taylor) writes: >> Adults understand what a child needs. A child, on his own, would quickly kill >> himself. Also, pain is often the only teacher a child will listen to. He >> learns to associate a certain action with undesirable consequences. > > Spanking certainly is a form of behavior alteration, although it might > not be the best one in all circumstances. It has been demonstrated in > experiment after experiment that positive reinforcement of desired > behaviors works much better than negative reinforcement of undesirable > behavior patterns. David: But what about th "pseudo-observer effect" (my pseudo-terminology)? If you beat your child, and then the child proceeds to behave in the manner you desired him (or her, if she's your daughter and not your son :-) to behave, then the beating worked (produced the desired effect). In this fashion, the parent (or total stranger who beats children) has positive reinforcement of the beating effecting the behavior. Given that the child has responded to being beaten once, it is logical to assume that he would do so again... this, coupled with the prior positive reinforcement to the parent (or stranger), makes it more likely that they will beat the child in the future, given a similar situation. Consistent reinforcement is more effective than inconsistent reinforcement, be it positive or negative. Besides, you always have your hands; how often do you happen to have ice-cream immediately available? | Terry Lambert UUCP: ...{ decvax, ihnp4 } ...utah-cs!century!terry | | @ Century Software OR: ...utah-cs!uplherc!sp7040!obie!wsccs!terry | | SLC, Utah | | These opinions are not my companies, but if you find them | | useful, send a $20.00 donation to Brisbane Australia... | | 'Signatures; it's not how long you make them, it's how you make them long!' |
geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) (06/14/88)
In article <2436@uvacs.CS.VIRGINIA.EDU> cfh6r@uvacs.cs.virginia.edu.UUCP (Carl F. Huber) writes: >In article <5323@xanth.cs.odu.edu> wet@xanth.cs.odu.edu (Warren E. Taylor) writes: >>In article <1176@cadre.dsl.PITTSBURGH.EDU>, geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) writes: >> >>"Spanking" IS, I repeat, IS a form of redesigning the behavior of a child. >>Many children listen to you only when they are feeling pain or are anticipating >>the feeling of pain if they do not listen. > Whoa! This is not a quote from me! Myself, I would prefer non-violent forms of punishment, since I think kids learn the legitamacy of violence from being spanked. But, I should mention, I don't have kids, so I may not be the one to ask about it.
bill@proxftl.UUCP (T. William Wells) (06/16/88)
In article <558@wsccs.UUCP>, rargyle@wsccs.UUCP (Bob Argyle) writes: > We genetically are programmed to protect that child (it may be a > relative...); Please avoid expressing your opinions as fact. There is insufficient evidence that we are genetically programmed for ANY adult behavior to allow that proposition to be used as if it were an uncontestable fact. (Keep in mind that this does NOT mean that we are not structured to have certain capabilities, nor does it deny phenomena like first-time learning.) > IF we get some data on what 'free will' actually is out of AI, then let > us discuss what it means. It seems we either have free will or we > don't; finding out seems indicated after is it 3000 years of talk.vague. I hate to call you philosophically naive, but this remark seems to indicate that this is so. The real question debated by philosophers is not "do we have free will", but "what does `free will' mean". The question is not one of data, but rather of interpretation. Generally speaking, once the latter question is answered to a philosopher's satisfaction, the answer to the former question is obvious. Given this, one can see that it is not possible to test the hypothesis with AI.