Schneier on Security
A blog covering security and security technology.
« EU Bans X-Ray Body Scanners |
| Friday Squid Blogging: Squid Camouflage »
November 18, 2011
A Link between Altruism and Fairness
I write a lot about altruism, fairness, and cooperation in my new book (out in February!), and this sort of thing interests me a lot:
In a new study, researchers had 15-month old babies watch movies of a person distributing crackers or milk to two others, either evenly or unevenly. Babies look at things longer when they're surprised, so measuring looking time can be used to gain insight into what babies expect to happen. In the study, the infants looked longer when the person in the video distributed the foods unevenly, suggesting surprise, and perhaps even an early perception of fairness.
But the team also say they established a link between fairness and altruism. In a second part of the experiment, the babies chose between two toys, and were then asked to share one of the toys with an experimenter. About a third of the babies were "selfish sharers": they shared the toy they hadn't chosen. Another third were "altruistic sharers": they shared their chosen toy. (The rest chose not to share. They may have been inhibited by the unfamiliarity of the experimenter, or maybe they just weren't that into sharing.)
What's interesting about the second half of the study was that by and large it was the babies who had previously been surprised by the unfair cracker and milk distribution who tended to share the preferred toy with the experimenter (the altruistic sharers). The babies who shared the rejected toy hadn't expressed much surprise over unequal distribution. This led the researchers to suggest that there's a fundamental link between altruism and a sense of equity.
Both psychology and neuroscience have a lot to say about these topics, and the resulting debate reads like a subset of the "Is there such a thing as free will?" debate. I think those who believe there is no free will are misdefining the term.
What does this have to do with security? Everything. It's not until we understand the natural human tendencies of fairness and altruism that we can really understand people who take advantage of those tendencies, and build systems to prevent them from taking advantage.
EDITED TO ADD (12/14): Related research with dogs.
Posted on November 18, 2011 at 5:50 AM
• 49 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
"Is there such a thing as free will?"
As long as the voices tell me there is, I will believe it.
There have been similar tests carried out on large primates and they show similar traits to adult humans.
But perhaps of more interest are those experiments which are very similar but carried out on dogs with food in discreet pieces such as dog biscuits.
The results showed that with a releativly small number a dog could tell if there were more or less biscuits added to a bowl not just by comparison but by memory from a previous occasion.
I just wish I could find the link to it.
Outstanding. They can punish the altruistic/sharing kids to break them of that habit; otherwise life will shit all over them for it as adults.
The kids who didn't share any toys are future corporate executives ...
As usual in this kind of discussion there is an allusion to the idea that if there is no free will then there is no responsibility and no basis for punishing socially undesirable actions. Not so. The only firm basis for punishment that I see is deterrence, either of the individual or of others who know of the punishment. Whether that works by external determination or by free will, however defined, strikes me as irrelevant. If it works, it serves the necessary social purpose.
We do have inhibitions against punishing people who cannot for some reason learn from it (idiots, children, psychopaths), even though the example of their punishment could deter others. That is our compassion working, which is admirable and important. That is, I think, the interesting part of the issue that remains, and that may be what is often cast in terms of free will.
"I think those who believe there is no free will are misdefining the term."
Seriously Bruce? The alternative is a wishy-washy definition that according to the article, seems to boil down to "the human brain is pretty darn complex, so voila! We can call its output 'free will'".
Now I'll agree that his wishy-washy definition is probably going to be fine for practical purposes of understanding human behavior.
That, and keeping the masses convinced that they have free will is probably good for societal security.
So go ahead and use "free will" however you wish, but I'd really rather you picked one of the "it's magical" or "it's a delusion" choices that we so-called mis-definers believe it to be.
I think those who believe there is no free will are misdefining the term
I'm surprised you consider the linked opinion a good argument. Aside from the strawman positions and non sequiturs, it commits the sin it accuses of others: redefining free will.
Neuroscientists aren't defining free will out of existence. Instead, they're using the traditional notion of free will and showing it doesn't work.
Nahmias tries to save free will by redefining it, not as being freely able to choose, but as the appearance of being freely able to choose. Free will is no more.
Here's a good example of the BS that Nahmias has to do:
People are threatened by a possibility I call “bypassing” — the idea that our actions are caused in ways that bypass our conscious deliberations and decisions. So, if people mistakenly take causal determinism to mean that everything that happens is inevitable no matter what you think or try to do, then they conclude that we have no free will. Or if determinism is presented in a way that suggests all our decisions are just chemical reactions, they take that to mean that our conscious thinking is bypassed in such a way that we lack free will.
There's a clear (and extremely basic) error here. What you think is based on the physical brain and is just as determined as everything else. Concious minds aren't bypassed, they're just not free. Nahmias is trying to have conciousness both ways: It's based on physical phenomena (no soul), but it's not based on physcial phenomena (independent of everything with causal links).
Is it even theoretically possible to devise a repeatable scientific experiment that can tell the difference between free will and determinism? I rather doubt it. Without that, does it even make any sense to ask the question? Free will is like God -- it's something that people want to believe in despite a complete absence of evidence for its existence.
@ Oliver Holloway,
The New Scientist article is not what I saw originally but it certainly describes the experiment so it's more than close enough grab a cigar.
+1 to you.
One thing that annoys me about the news reports is that the researchers aren't really researching "free will" they are researching decision making. The free will aspect is added by journalists to make a headline.
Is it even theoretically possible to devise a repeatable scientific experiment that can tell the difference between free will and determinism?
Probably not. One of the things I found interesting when I was learning about Finite Automata is that any NonDeterministic Finite Automata can be emulated by a Deterministic Finite Automata and vice versa. That implies to me that you can't prove from the output that a complex process is nondeterministic. [Note that this is not a particularly strong argument in respect to free will as the word determinism is ambiguous, also there is no reason to assume that we are finite state machines]
Free will is like God -- it's something that people want to believe in despite a complete absence of evidence for its existence.
Free will is more like consciousness. Descartes' statement, "I think, therefore I am." is powerful because we all have the experience of being a being that thinks. Likewise we all have the experience of being a being that chooses. Even those who adamantly deny free will are aware of making choices. They just discount the validity of the choice after the fact. I don't see any way in which neuroscience can completely discredit the notion of consciousness, even a complete description of the processes of consciousness would not negate the experience of it. There is at least one way to discredit free will--predict 100% of the actions of an arbitrary person.
Let me offer up yet a different viewpoint. In the long run, is it really healthy for a society or an organism to "build systems to prevent them from taking advantage." Look at the law with it's fundamental understanding of truth as the result of an adversarial system. There is a cogent argument that we need the bad guys, the liars, the outliers. If there were no crooks there would be no defense attorneys and if there was no defense there would be no case or controversy and thus no judging. The court system would cease to exist.
Put another way, is security a battle the white hats really want to win. Maybe not. With no stimulus that the bad guys provide there is no growth, just stagnation. Is a society in which there was perfect and total cooperation a evolutionary healthy society? Evolution itself would suggest not.
Even if the entire universe was deterministic, that does not necessarily mean humans can't have "free will".
Even if every action and decision I will take in my entire lifetime is a deterministic consequence of the state of the universe at that moment, *I don't know* that complete state (nor could I ever know the complete state of the universe, nor could I ever know the complete ruleset being used to evolve one state into the next). In effect, I have to make decisions and choices on the basis of seriously incomplete information about what the consequences will be. To me that's "free will". (I'm also a believer in the multiverse theory that every possible outcome does happen in some multiverse.. I think of the universe as a giant NFA that contains every possible state that is permitted by the rules. So I have "free will" in the sense that I can remember a linear history of what was happening in my little corner of the universe, corresponding to a path backward through various states of the NFA.)
If "every action [you] take in [your] entire lifetime is a deterministic consequence of the state of the universe" then You DON'T "make decisions and choices on the basis of seriously incomplete information" because your don't make choices at all. That is, fundamentally, what determinism means. You may think you make choices and you may (deterministically) rationalize them afterward, but the choice was an illusion.
Re: Multiverse theory. If every outcome happens then there are also no decisions being made and the particular universe that we both happen to be occupying in this exchange is basically random (or probabilistic).
Randomness does not solve the free will problem because it ignores the "will" part.
Mr. Schneier, with all due respect, I have to disagree with you on this issue.
Even if one was to use definition provided by Mr. Nahmias, which is "a set of capacities for imagining future courses of action, deliberating about one’s reasons for choosing them, planning one’s actions in light of this deliberation and controlling actions in the face of competing desires", neither moral culpability nor any other "practically important" implication of free will is salvaged.
The reason for that is simple: algorithms in accordance with which aforementioned deliberations and assessments take place are not affected by individual, but rather produced by interactions between the individual's innate properties and external conditions, both of which are outside of individual's control.
Free will, as defined by Mr. Nahmias, does not allow one to claim moral responsibility since the individual is clearly incapable of managing the factors that guide formation of algorithms in accordance to which his "Nahmiasian free will" operates.
Legal responsibility is an entirely different matter, however (I must say I feel sincere pity towards legal systems that bootstrap legal responsibility from moral responsibility)
As you are aware (if you remember ;) I don't particularly believe in accidents (only lack of forsight / poor risk analysis).
But to play the right hand man of he who hath horns and a forked tail ;)
"Randomness does not solve the free will problem because it ignores the "will" part"
A thought experiment.
I write down in a book a whole load of day to day events that require choice. To each one I append a list of six responses that I can take.
When I have to make a choice I trow a dice and then look up the choices in the book and take the action indicated by the dice.
Now In one respect I have free will in that I've chosen to make up the book, however at the same time the free will choice to do this compleatly robs me of free will as what I do is based entirely on the random fall of the dice.
Can you as an observer of my actions, not being privy to the actual rules I'm using or that they are in place say that it is I or the dice that decide and where or not the free will exists?
JimFive over here mentioned Decartes' notion "cogito ergo sum".
I did not intend to write a long text here so I will want to merely point out something here, by mentioning that the psychoanalyst Jaques Lacan found Decartes' notion of "cogito ergo sum" problematic.
I no longer remember the extact problem and would rather not guess to try making a point here. I probably recognized it as something I already knew and forgot about it, heh.
It is discussed in his paper Seminar XI.
I allowed myself to do a quick google search here just now; a possible problem could be, this argument that subjectivity is not found in the consciousness, which would undermine the idea of a "free will", insofar as thinking about things constitues you being free.
I think you misunderstand me. I am not arguing for or against free will as such. In a deterministic world, however, your entire procedure is not "free" because you were caused to do that by the initial conditions of the beginning of the universe. Even adding a non-deterministic element (the dice roll) does not create free will, however, because it leaves out the "will" part. The dice roll may create "freedom" in the sense of a non-predictable outcome, but that outcome is not your "will".
I was not saying that free will is a result of consciousness, nor that thinking about things makes you free. I was making an analogy comparing people's thoughts about consciousness to their thoughts about free will.
I have not read Lacan and have only spent scant minutes reading internet summaries of him. From that scant reading I will say that it appears to me that Lacan is concerned about the "I" which I think is missing the point. Descartes was attempting to find something that existed that he could not doubt. What he came up with is that something is thinking about a question, to do that, that something must exist. He experiences that something as himself and labels it "I". Whether that is an objective or subjective "I" is irrelevant to the existance of that "I". It is certainly, however, a conscious "I" because if it wasn't it would not know that it was asking a question.
"I was not saying that free will is a result of consciousness, nor that thinking about things makes you free."
What is your point here? :) I never accused you of anything like this.
I was already aware of you not having expressed such an opinion above (from what I read above anyway), but I saw an interesting problem in thinking of, so called "free will" as consciousness; and thought, partly that Lacans discussion would be interesting for others and perhaps to you, if you thought that a though alone or the act of thinking could be free will.
Did you perhaps mean to write "Free will is more like believing in your own consciousness"?
If so, the point you made to me, seem unfortunate. :)
I see that you also wrote to me "(...)nor that thinking about things makes you free." This objection here puts your opinion here at odds with what I would argue to be commonsensical. Because I think I would rather rely on doubt than knowledge.
"(...)I don't see any way in which neuroscience can completely discredit the notion of consciousness."
I agree wholehartedly with this, because I believe it makes sense to believe in ones personality (btw, having nothing to do with personality types), being the accumulation of experiences and a life lived, consciously and unconsciously, and that one somehow is relating to it, because it simply is a necessary, being alone with your own brain.
I am reminded of a recent comic by "Cyanide and Happiness", that makes fun of a guy that at first tells himself that "Today is the first day of the rest of my life", then falling in doubt as the image is depicting his brain talking to him from beneath the skull saying "so is tomorrow", leading up to the punchline frame of him drooling at a computer in the dark with a soda can next to him.
The brain in this comic looks like the meme "forever alone" face, with a tear in his eye.
(I haven't looked at the original paper, but...) It seems as though the author is trying to push the idea that the novel contribution here is that altruism and fairness concerns are linked. However, both experiments seem to me to be tests of fairness concerns -- at early stages of development, children can't understand that other people may have different preferences over the toys. So the equivalent of choosing an even split of a cake in the toy experiment is to share the "best" toy. Sharing the lesser toy is akin to giving away a quarter of the cake. So it seems to me that what the authors show is: children who are concerned about fairness also tend to be fair. Which is interesting, but not quite the story they seem to be after.
"I think you misunderstand me. I am not arguing for or against free will as such."
I think perhaps we are misunderstanding each other.
The point I was driving at is that it is not possible to tell if free will exists or not.
In a world where you can "only view" from a distance an "effect" and hazard a guess at it's "cause" you cannot run an experiment to verify your guesses. Because by so doing in part you become part of the experiment and thus will effect the outcome of the experiment. And also you have random input over which you have no control.
Thus discussing "free will" is rather like talking about the state of a cat in a box. It was if I remember correctly an issue Albert Einstein got badly hung up about. Because he believed in "determinism" in all things physical and "free will" in all things human accepting neither quantum mechanics nor ordained by god.
Personaly I hold no greate faith in either determinism or free will because both are a little trite when all is said and done, And I suspect things are in some form of reality a lot lot more interesting, but then I always was the eternal optomist ;)
This is reminding me of the Monty Python's soccer game between philosopher's ... :-)
I'm a "compatibilist" according to wikipedia, and JimFive is probably an "incompatibilist". There are also labels for various subsets of these two camps, but none of the common compatibilist positions seem to match my views. "Free will as a pragmatically useful concept" might be the closest.
I had typed up a huge explanation of how I think the universe works, but I just gave up and erased it because this blog is a lousy place for discussing things like determinism and free will. Neither of which seems to have anything to do with the article, anyway -- Bruce just threw that canard in there to send us all off into philosobabble-land!
I just noticed now your other reply to me, which I missed earlier, and I wanted to say I think you are right about the issue of the "I", however this is already something I sort of intended to problematize when I mentioned "subjectivity" earlier.
I guess this raises the sensible question; why doubt that one can be sure one is thinking? My answer to such a question (and therein lies a problem), how did this knowledge about this supposed "I" come about? I think it was the philosopher Wittgenstein that pointed out that one does not really think a thought, the thought is just there, without "a priori" knowledge (Kant). I believe Wittgenstein also made a point, asking if when sensing or thinking that one is wanting a banana, how could you really know you wanted a banana and not an apple instead?
A quick search shows that: on page 36 in Lacan's Seminar XI, Lacan makes a point about how the fact that Decartes is expressing his notion to "us". Something which is problematic, probably because one can ask why such an insight is to be deemed as important and relevant as one probably would want it to be.
The notion by objectively knowing you simply exist for certain, is here thought of as being made irrelevant by the notion of doubt. I eh am a little unprepared here, so I will dub this problem of instant knowledge, a problem about a priori knowledge (which is not possible, or so is my point anyway). It's like there is this extra problem of actualization, which would be the difficulty in affirming the status of knowledge, for it to be relevant and not an excuse for something else (certainty, relevance, truth, logic, inherent meaning, etc).
My simple point here, could perhaps be explained most succinctly by attributing this eh ontic phenomenon as something "autistic", to help on this notion of a thought being something spontanious and not controlled mental process. I suppose counting numbers could be thought of as being a more controllable mental process with no apparant spontaneity, however it seem pertinent to then ask, where one learned to count or count in integers, or even if starting to count from the number one or zero. Btw, the way I see it, math is all about pattern making, with patterns understood as being predictable.
Some wiki on the internets: Autism "Coined in 1912 by Swiss psychiatrist Paul Bleuler (1857-1939) from Latin autismus from Ancient Greek αὐτός (autos, “self”)."
The word "autism" has of course a different meaning to it, with regard to what I wanted to make a point out of here where I used the word "autistic". I could perhaps have used the word "idiosyncratic", but thought "autistic" sounded more startling and interesting.
@bob, haha nice so true...
Interesting though, they should also see if the amounts of altruism & fairness change over time or remain the same...
Logically, where would classical free will fit in the "randomness" experiment ?
The experimenter's decision to introduce randomness into daily routine has been caused by interactions between his genetics and his environment (in regards to both of those, the researcher never was and never will be "free")
"Classical free will" is an oxymoron.
"Nahmiasian free will" is rather irrelevant since it does not have the moral and legal implications ascribed to free will and is not free in any meaningful way.
"The experimenter's decision to introduce randomness into daily routine has been caused by interactions between his genetics and his environment (in regards to both of those, the researcher never was and never will be "free")"
Actually not quite.
Person 1 the experimenter, makes their decision to introduce randomness due to three causes,
1, Nature (genetics),
2, Nurture ( what they learn),
3, Environment ( what they learn in).
Likewise Person 2 the observer of Person 1 has chosen to observe due to the same three causes.
In each case the causes are unique to the individual person. Even if they were conjoined twins, the fact that they see every thing slightly differently means that the results of the three causes are different.
We know this from a pair of conjoined twins from over a hundred years ago who lived quite long lives they both did all the normal things including having their own seperate (but together) family lives. From contempory reports we know that their personalities were different as often were what they chose to do. This has been confirmed by observations on other conjoined twins.
Interestingly it appears non conjoined twins can appear to be more "in phase" with each other than conjoined twins.
Any way back to the experiment within an experiment.
With regards "free will" there are two persons involved each of whom may or may not be exhibiting free will, so on the assumption they are 50/50 choices there are four equally likely outcomes.
Thus for Person 2 the observer they have to determine without being aware of Person 1's "personal experiment" if Person 1 is exhibiting free will.
After observing Person 1 for sufficient time and without being aware of Person 1's personal experiment the only conclusion they can accuratly come to is that person 1 is exhibiting "random behaviour".
However the purpose of the observation experiment is for person 2 to determin if person 1 is "exhibiting free will by observation alone".
As I have shown because of Person 1's personal experiment into exhibiting random behaviour, it is not possible for person 2 to make any valid conclusion as to if person 1 is or is not showing free will.
Unfortunatly the observation experiment by simply having a yes/no question is going to produce an answer along the lines of,
"I have observed person 1 for X time and they have not exhibited consistent behaviour on choices to the same or similar stimuli in the same environment on successive occasions. I therefor conclude that they are not showing signs of determinism in their choices, thus the opposite behaviour would indicate the excersice of free will".
Which as we know is totaly incorrect.
This or similar issues is why by far the majority of "rats in a maze" experiments produce compleatly false results.
The observer is not omnipotent and thus can not know all the rules of the game the rat is playing.
I can not remember the name of the experimenter that "blew the gaff" on this, but they found that in almost every case the experimenter setting up a maze did not take sufficient care with eliminating consistent smell and sound from the maze between experiments. The rats in effect were following either scent or sound trails (caused by their foot falls) to find their way back to the position that produced the reward.
Thus it is not possible to sett up an experiment whereby you can determine if "free will" exists or not because of "hiden rules/knowledge".
Thus rather than saying free will is or is not X, you can only say it is pointless trying to determine if it exists because you will never have sufficient knowledge to be able to say.
But we know this from other work in mathmatics and logic from before WWII, the fact that quite intelligent people have chosen to disbelieve it and waste their and others time and resources probably says more about the notion of "free will" than anything else...
Atleast most mathematicians except the fact that "random" is a concept not an actuality, that is they except the fact that it is saying "we don't know if a number sequence is deterministic or not" simply because we cannot look forward in time to the next bit in the sequence.
With regards the existance or non existance of free will or not, unfortunatly it is not irrelevant to our ordinary everyday lives. Because of crime and punishment. If free will does not exist then anybodies crime nomater how horrific is not their fault because it was "pre-ordained" thus to punish them in any way is wrong. I will leave it to others to follow the consiquences of that thought, but remember it caused Albert Einstein to have significant problems in later life.
Although the experiment is worthwhile, and I believe infants are moral and altruistic, the study contains enough methodological flaws for the results to be questionable.
No report as to 1) gender or 2) right-hand left-hand distribution which can influence the results.
The greater mean looking time for uneven resource distribution can indicate surprise, unfairness, greed, or cognitive strain.
The sharing task results can be caused by the clever hans effect, altruism, or the infant changes his/her preference.
The story about rats was reported in one of Feynman's books (probably "surely you must be joking Mr Feynman). I think the saddest thing was that after the guy had found out how rats navigate mazes, he went of to do something different. Most of the people researching rats just went on happily doing it wrong.
BTW. As a counterpoint to your argument, surely it can only be morally wrong to punish if you have free will. Since your decision to punish someone is just as determined as someone's decision to do something, there's no problem whatsoever. In which case the existence or not of freewill becomes irrelevant.
I don't believe in free will, however I choose to act as if I had it.
@JimFive: As I completely agree with your elaborations: Could you maybe suggest some further readings on that topic (talking about the off-topic "free will", not the "altruism and fairness")?
Any guesses which group CEOs and Lawyers fall into?
I'm vastly out of my comfort zone here, but what if the sharing behaviour of the babies is not determined by free will, fairness, altruism and the like but simply by identification. The baby identifying with the one getting the better deal may be less inclined to share stuff than another one identifying with the infant being disadvantaged. This identification process may of course also be influenced by genetics, nurture and environment as pointed out earlier by Clive.
It's been a long time since I did serious reading regarding free will so I don't have anything up to date.
I would, however, suggest avoiding the "ancient philosophers" suggested by Willie.
My previous local library has had a series of college level essays on various topics in philosophy including free will. I don't recall the name of the series right now, and as I've moved I don't have them available. My current local library is much more sporadic.
If I remember, I'll dig out some old books and post some references.
You seem to be arguing for agnosticism with regard to free will. That's fine. I can accept I don't know, or even, we can't know, or even, it isn't possible for any to know. However, free will either exists or it doesn't, regardless of whether we can ever know.
Disregarding what Clive has written above, it seem pertinent here to point out that if "will" equals to being "free" then the proposition or rather, the idea of "free will" in this manner has become an instance of something being a tautology, a circular reference.
"If free will does not exist then anybodies crime nomater how horrific is not their fault because it was 'pre-ordained' thus to punish them in any way is wrong."
As a determinist and incompatibilist, I disagree. We still have desirable and undesirable states, and would still like to shape the world toward the desirable and away from the undesirable. Do we believe dogs to have free will? Do we try to alter behaviour on their part toward what we would desire it to be?
I look at all crime this way, that whether or not you can be "blamed" for it, you (or your surroundings, I suppose) need to be altered so as not to continue to act in this fashion; punishment is to rehabilitate. Also, laws and law enforcement alter the environment of actors and thus influence their behaviour - deterrence.
@JimFive: Thanks for getting back at me, but no need for digging out old stuff because of me. I'm sure I will be able to (and already did) find interesting books and papers on this topic myself, too.
@Zith: You are talking about "desirable and undesirable" states, which sounds politically correct. But would you agree that in fact a society (or parts of it) decides whether a state is desirable or not? Meaning what's (un)desirable is subject to changes and, well, subjective. (I'm not arguing here, my view on crime is the same as yours, I'm just curious about the terminology you used.)
"Since your decision to punish someone is just as determined as someone's decision to do something, there's no problem whatsoever. In which case the existence or not of freewil becomes irrelevant."
Yes and no (yup I'm getting a sore bottom sitting on the fence and my forked tail does not help ;)
I've shown how someone even without free will can behave in a non pre-ordained way, and also that someone with free will can behave likewise by introducing a "random" element into the choice process. Either way the result is the same free will or not the choice can be non determanistic to those within the system of the experiment.
So no even without free will the choice to punish is not pre-ordained in the experiment. But this is dificult without free will, the use of a random element and rule book would have to be pre-ordained. But that is not possible (see further down)
And I'm curious as to why nobody had used the T word (tautology) up to that point but more of that later.
"However, free will either exists or it doesn't, regardless of whether we can ever know"
Does it? The fact it has a name magics up a notion, but does it have a testable definition? If not then it is arguably just a meaningless term that is effectivly nebulous to the point of irrelevance.
Or to be a little more formal without a testable definitio it is something beyond the scientific method, thus like a belife in deities it is a matter of personal faith, it is not "a self-evident truth" it is mearly an assumption but it is certainly not a logical axiom.
But is it a non-logical axiom? those who have faith in free will would say yes those who do not have faith in free will might say no.
So can it be used to define a problem in such a way that it is testable, the answer is the awkward, no it if free will does not exist yes if it does exist...
Now if you read back before I described the experiment I indicated that I was playing a devils (he of the horns and forked tail) advocate, so I did not declare my viewpoint, just a counter to others expressed view point. But read on,
"As a determinist and incompatibilist, I disagree We still have desirable and undesirable states, and would still like to shape the world toward the desirable and away from the undesirable"
And I have used "pre-ordained" and "determanistic" apparently more or less interchangably (and nobody picked me up on it), but are they equivalent? Well no they are not.
A determanistic process is dependant on it's input conditions, and thus can have many different outcomes. Something that is pre-ordained is something that is not dependent on either a determanistic process or the selection of input conditions, it will happen without question when it is supposed to happen and nothing can change the outcome ie it is 100% certain in time, place or any other measure you care to use.
For pre-ordained to be possible it presupposes a number of things the most important of which is nothing can be random as everything is certain. For me this is far to greater a constraint on the physical world to accept
However pre-ordained also means that we cannot have desireable or undesirable states we can move towards. The choice cannot be made it is pre-ordained the out come is certain. Having the ability to change the future pre-supposes an ability to chose ergo free will.
So for the curious who are still with me at this point I will take off the horns and forked tail and return them to their owner and say,
For me I cannot belive in a pre-ordained universe,although I can accept a physical universe with a myriad of independent determanistic processes with random starting conditions for each determanistic process.
That said and it is a matter of faith because it is not possible to test, my choice is for the existance of free will, because to many things just cannot work in an acceptable way without it. The primary reason for this is "random selection" is to slow to be where we are today. We and to a significant extent the Universe exhibit either a run of luck of astronomical preportions or the refining ability to emphasise or select desirable outcomes both at the physical level and mental level.
As an engineer, my first question is always "do you actually want to do that?" So when told that we're looking at whether there is freewill, I stab further to ask "what do you intend to do with the yes or no result?" The answer seems to usually be framed in the question of "can we blame this person for their actions."
In this case, you have two options. You can accept that your definition of the "self" has to encapsulate the entirety of the brain/mind, in which case you are looking to blame the brain/mind, and thus chemical activities of the brain are blamable.
The alternative is to take a religious tack, and declare there to be a soul. If you do, then freewill is purely defined by the religion you use to define the soul. If anything, then, the neuroscience starts to point more and more to your theory of a soul being wrong. If you insist in continuing to define the "self" as merely the soul, then yes... neuroscience is actively seeking to destroy freewill, by planting it firmly in the brain/mind and less in the soul.
The greater question, which comes after you are done deciding if someone is to blame, is whether or not that blame implies that it is acceptable to punish them. That is a MUCH more nuanced question, to which I would be more than content to spend a lifetime exploring.
" I think those who believe there is no free will are misdefining the term."
Finally I found someone other than me saying it!
The way I see it, we will do what we do because of who we are. To have acted in any other way than what we do would be to act against ourselves. I don't believe this robs us of anything, and in fact I feel it bestows great, awesome significance to what we do. When I do something, it is an extension of myself, my identity.
The same thing comes up in mathematics, and in my opinion lends to its beauty. Everything about a number is inherent to it, and everything it performs in a function is a direct result of its nature. Our various formulas are beautiful interactions of these solid, deterministic pieces that depend upon this.
Of course, the nature of a number is immutable, but that's just another area we're able to manipulate in our societal functions.
So it's not a matter of whether the future is set and cannot be altered. The present and future are occurring the only way they can, and our specific nature is required in making things turn out that way. If any of it were different than it is, it'd go in different directions. It's not important that we can choose what our desired and undesired states are; it's important that we have desired and undesired states and act accordingly.
@Tom: I used that particular phrase to be as general as possible. And to be clear, I'm using the word "state" as it's used in computer science.
I'd say that yes, society at large comes up with the idea of what is good and bad, what to enforce and not, and I would agree that morality is very much subjective; I'm a moral relativist. I also look at cultural systems as something like meta-organisms, appearing as if struggling for continued existence because only those that do persist. Morality plays a big part in that for a society, and inability to adapt it will make the whole less fit for survival. (Not to say that survival is more important than holding fast to principles.)
You should investigate oxytocin ("the moral molecule"), interesting vid over on TED by Paul Zak (www.ted.com/talks/lang/en/paul_zak_trust_morality_and_oxytocin.html) also have a look at Annie Murphy Paul's talk "What we learn before we're born" where the biochemical environment created during pregnancy has... well give it a watch also (www.ted.com/talks/annie_murphy_paul_what_we_learn_before_we_re_born.html), very reminesent of Dr Weston Price's pre-WWII research where primitive tribes would have booth the mother and father on special diets before conception because the tribe couldn't afford to be supporting members that couldn't pull their own weight.
Let me offer up yet a different viewpoint. In the long run, is it really healthy for a society or an organism to "build systems to prevent them from taking advantage." Look at the law with it's fundamental understanding of truth as the result of an adversarial system. There is a cogent argument that we need the bad guys, the liars, the outliers. If there were no crooks there would be no defense attorneys and if there was no defense there would be no case or controversy and thus no judging. The court system would cease to exist. Put another way, is security a battle the white hats really want to win. Maybe not. With no stimulus that the bad guys provide there is no growth, just stagnation. Is a society in which there was perfect and total cooperation a evolutionary healthy society? Evolution itself would suggest not.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.