betazed
Seeking...
- Joined
- May 9, 2003
- Messages
- 5,224
heh, we have been thru this one in endless circles before (posts from Gothmog and Mark1031 and others come to mind). Let me add my 2 cents (again)!
What do we need to have a free will (such a loosely defined concept - hopefully we will harden it later). Seems to me that we need an actor which acts out the free will and a free will machine (for want of a better name, still pretty apt though) which implements the axiom of choice in the precise mathematical sense. Now it should be obvious that the actor only plays a passive part in this play. take out the machine from a actor + machine combine and put it on another actor and we should reasonable say that the new actor now has free will. So for all our future discussion we will ignore the actor and concentrate on the machine.
So what are the key characteristics of this machine. I can think of the following :-
1. Unpredicatbility - that this should be a key characteristic can be proved easily by reductio ab absurdum. Assume that the machine is predictable. So let us create another machine - call it the free-will predictor - that can predict this machine. If such is the case then we can say it is really the predictor machine which is acting out the free will and our original free will machine is really an actor. if the predictor machine is also predictable we can follow the chain of logic till we find a unpredictable machine. We will call this machine the free will machine by definition.
2. partially Non-random - While individual responses from the free will machine is random, statistically there still should be a pattern which should show and underlying rhyme and reason (free will or no free will given a choice between taking $1000 and and $10000 most individuals will always take $10000).
3. Impossible to simulate exactly in functionality by anything other than a free-will machine - because if it can be simulated by anything then whatever that is simulating it becomes a free will machine and this machine stops becoming a free will machine.
4. Assignment of responsibility - this actually follows from 3 above but i wanted to state it separately. Once an action has been taken it should be possible to state that the sole cause of the action was the free-will machine. Because if it was not then something else was simulating it and that which was simulating it is the actual free will machine.
I cannot think of any other requirement. can anyone else?
Given these requirements it should be trivial to build a free will machine don't you think? and if it is trivial then the statement that free-will exists is also trivially true (it exists because we can create it). Of course whether humans are a free-will machine is another question altogether.
What do we need to have a free will (such a loosely defined concept - hopefully we will harden it later). Seems to me that we need an actor which acts out the free will and a free will machine (for want of a better name, still pretty apt though) which implements the axiom of choice in the precise mathematical sense. Now it should be obvious that the actor only plays a passive part in this play. take out the machine from a actor + machine combine and put it on another actor and we should reasonable say that the new actor now has free will. So for all our future discussion we will ignore the actor and concentrate on the machine.
So what are the key characteristics of this machine. I can think of the following :-
1. Unpredicatbility - that this should be a key characteristic can be proved easily by reductio ab absurdum. Assume that the machine is predictable. So let us create another machine - call it the free-will predictor - that can predict this machine. If such is the case then we can say it is really the predictor machine which is acting out the free will and our original free will machine is really an actor. if the predictor machine is also predictable we can follow the chain of logic till we find a unpredictable machine. We will call this machine the free will machine by definition.
2. partially Non-random - While individual responses from the free will machine is random, statistically there still should be a pattern which should show and underlying rhyme and reason (free will or no free will given a choice between taking $1000 and and $10000 most individuals will always take $10000).
3. Impossible to simulate exactly in functionality by anything other than a free-will machine - because if it can be simulated by anything then whatever that is simulating it becomes a free will machine and this machine stops becoming a free will machine.
4. Assignment of responsibility - this actually follows from 3 above but i wanted to state it separately. Once an action has been taken it should be possible to state that the sole cause of the action was the free-will machine. Because if it was not then something else was simulating it and that which was simulating it is the actual free will machine.
I cannot think of any other requirement. can anyone else?
Given these requirements it should be trivial to build a free will machine don't you think? and if it is trivial then the statement that free-will exists is also trivially true (it exists because we can create it). Of course whether humans are a free-will machine is another question altogether.