What is the difference between free-will and randomness and or non-determinism?

WORD               FORMAL DEFINITION        GENERAL USAGE

Free will          strictly undefined       some property or essence which 
                                            grants unrestrained action

randomness         strictly undefined       the state or condition of having an
                                            unpredictable outcome such that any
                                            outcome is equally probable

indeterminism /    [1]                      See [1]
 non-determinism

                   1. (a) the doctrine that human actions, though influenced
                   somewhat by preexisting psychological and other conditions,
                   are not entirely governed by them but retain a certain
                   freedom and spontaneity.
                   (b) the theory that the will is to some extent independent
                   of the strength of motives, or may itself modify their 
                   strength in choice

Sources: Dictionary.com / my brain

The first problem is that you cannot tell what free will is without giving a definition, but it is hard to give an appropriate operational definition without already implicitly answering your question, viz. taking a stance.

The second problem is that randomness can be characterized in many different ways and these definitions are not fully equivalent. Unpredictability is an important criterion in cryptography, but does not work as a general criterion for itself, for it is too unclear. For example, if you record a sequence of random numbers (generated from radioactive decay) for later use, you may "predict" them before they are used, because you have already recorded them, yet they would still be random numbers. Vice versa, many deterministic processes can be unpredictable in practise. A slightly better practical definition of randomness is that a sequence of numbers is random if you cannot compress it and it satisfies some additional statistical criteria. (That practical measure is related to Kolmogorov complexity.) The problem is that a sequence of numbers might pass all these tests and still not be random, i.e. you cannot really tell from a sequence whether it is random or not by looking at the numbers alone. A random generator outputting characters may spit out all works of Shakespeare at any time, although this is highly unlikely.

Third, non-determinism involves a random choice between alternatives.

So what does this have to do with free will? Since a random choice would not be a choice of your free will but rather a choice made by the respective random source, a choice made of free will does not seem to be the same as a non-deterministic choice. Since on the other hand we do not consider a deterministic choice to be a choice of free will either - perhaps not even a choice at all - the concept of free will might be inconsistent after all. In any case, without a proper definition of free will we cannot give an answer to your question, and if you had this definition you would likely have answered the question already for yourself.

There are two easy ways out of this dilemma: (1) Discard free will in the sense of accepting determinism. (2) Discard free will in the sense of accepting that free will consists of highly complex, non-deterministic, yet rule-based choices.

I personally opt for (2), because it is more in line with modern physics than (1) and supported by the view that human brains are open information processing systems. (Notice that a computational system with the ability of making non-deterministic choices is not a Turing machine, although it might be an extension of one.)

A good question.

The defining property of free will, i.e. freedom in choices, is the "feeling" or consciousness of the own authorship of the choice being made. Even if/when I'm forced to comply I cannot escape the glimpse, at the moment, that nothing actually is selecting the way for me, except that "foolish" me (who is, actually, nobody/anonymous this minute) who agrees with the offered.

Thus, every human choice is a free choice, and responsibility is implied automatically. That would be Sartre's account.

Moreover, even when I stumble and fall down, I immediately and promptly react and interpret the occasion (e.g. by the fear to break an arm and not be able to paint anymore through that). By application of a free interpretation/meaning I emerge responsible for my incident though I didn't choose to stumble. I selected to be thus fallen, instead.

Freedom is spontaneous choice (of meaning and action) accompanied by the awareness of own authorship / resposibility. Or to put it differently, the apprehension that there is nobody/nothing specific is to blame except myself.

The key point here, @HWalters, is that ownership belongs to (is felt by) faceless me, the Nothing, and not to I, the Something. If nothing is the decision maker, the decision is, logically, not motivated by any preexisting structure, i.e. it is not determined (subjectively) by any cause. It is motivated by no cause, a pure gap in causational chain. (It is but later that we can attribute a cause to our decision, to "excuse" it.)

Thus, freedom is human (i.e. conscious) randomness (conditioned upon given circumstances). Within the scope of possible alternatives the circumstance can support, it is non-determined.

A man is an entity whereby randomness becomes own while determinism gets broken. Whenever consciousness comes to play, whenever something comes to being through consciousness, determinism is canceled.

Even conceived/planned decision without instant action is not your "will", it is spontaneous. It was spontaneous when it first occured to you, and it is again spontaneous when you start to act according to it: you re-invent it one more time, using your knowledge of the plan as the circumstance, and act.

So, free will is freedom. Specifically "will" is a redundant word, and people do not do things by "will" as some tension or effort. It is when I reflect back on the thing done or on the difficulty to overcome to do it, I might see that "labour" of my Ego involved and start to think of (free) "will" as my instrument. "Will" is always an artefact of reflection ("I decided and did it myself", "I must decide to do it"). No "will" is actually used in real, here-and-now decision makings.

Also to say: curiously, non-human (out there, universal) randomness is close, in eyes of a human, to determinism (or doom) and not to indeterminism. Both random and predetermined appear to us as contingency which is a characteristic of facticity to where we are "abandoned". Marooned for making free choices in the flight from facticity towards meaningful world.

P.S. Once again, how I feel it, and in a response to a comment elsewhere, where they've just asked me how freedom is different from randomness:

Freedom (free will) is spontaneous self-determination. Randomness is when the cause, "author", cannot be indicated, there is none such. Whereas here we have the cause - it is the prereflective, nonthetical self.

That self, being the cause of a choice, is also causa sui; it is not a consequence of another, so it is not a mediator for the choice, - that is because it is non-being: it is not based/founded on being, on what is.

Further, it (the self) is not a tuple pair (self & choice), but one monad: the self selects itself as one possibility of the object-in-the-world (this glass of water in front, which is so drinkable) and emerges in this act. The self thus generates itself and generates its choice in a single pass, one event. It does not pre-exist this functional act of choosing as something, but coincides with it as nothing. (Nothing is shortage of..., unclasping of..., not an emptiness.) I'm choosing myself as the possibility (or potentiality, or opportunity - don't know for now which word is better) of an object, me-possibility existing as deferred here, but strictly inside the circle of conditions (givennes, including my own empirical Ego) which I therefore appear beyond.

So, the causality with consciousness, or free will, is special: via self-nothingness rather than via being/matter. Spontaneity, aka free self-determinism.

While Michael's answer contains an awesome thought experiment, I disagree with its "fortunate" conclusion so much I think it leaves margin to a better answer. Also I'm not glad with basically all the definitions I've read here, except one - stoicfury's (read below). Also Karl's input (about the brain) is very insightful.

Contrary to Michael, I think definitions are very, very important. I'd argue it's fundamental for one's well being and it can lead a society to glory or doom.

I once had too loose definitions of words and no religion. So I was often confused with what I should think, with what I should do. Hell, I still have this problem today, and I will probably have to live with it for ever. As I began to give proper definitions in my mind, I also enabled myself to move forward, set a direction to move in my life. It's not much different accepting a definition I created myself or accept one from religion, but I think having it defined is very important if you want to build anything in your life.

As for a society, giving words to concrete objects is easy. And it facilitates communication to do practical things. "Pass me the hammer, please". Similarly, giving words to abstract concepts can help we orient each other to better position each one of us in society and build it up. "Should I get married?". "How many kids?". Yes, "kid" is quite an abstract concept. And I think the facility we have to communicate about it today is in part responsible for how big our Earth society is, despite the babylon of different languages.

So, allow me to give you my own personal definitions in hope to bring some new perspective here. I'm pretty sure they go against some common definitions, such as ones we can see in wikipedia, but those are how I see the concepts they're associated with, based on my social intuitive sense. Of course you can choose to use different words to try and add different definitions, give different meanings. But I think the current meaning people give to those words are quite broken...

Universe = every single thing we can even imagine

Some may perceive an idea / information / data as being something outside our reality. Dreams, for instance, can easily break rules of physics. Or can they? This is no simple concept and I won't digress about it here... Please, just take this definition and see if you can cope with it: there is nothing we can even think of that would be outside the universe. Not just the common concept associated with the "observable universe". There is no "multiverse". Everything is included. What exists, what doesn't exist and what might exist. If there is anything beyond the big bang that might affect us, it is within the universe. This is an important common ground definition I'll use for the other 3 you asked.

Free will = ability to think

I think free will has nothing to do with actions. I would say a restrained prisoner still have free will, even though he won't be able to do anything with it. A coma patient have no free will. Sure, thinking in itself can be considered work (i.e. an action), in a physical sense, as there are universe-embedded-neurons acting in our brains. And, as such, even our thoughts are restricted by the universe rules we live in. But, as long as we are able to rationalize anything, I call that free will. Maybe "free will" is just another word for "intelligence", after all. I love a definition I once saw for intelligence as "capacity to predict the future".

Randomness = events which were unpredictable

stoicfury's definition is spot on to me, just there's no need to say "any outcome is equally probable" as it is implied in the "unpredictable". To me "random" means "unpredictable". The moment some future event becomes predictable it is no longer random. If we live indeed in a deterministic universe (I think we do, but that's beyond the point), nothing is random by definition. Even then we can still perceive things as random while we can't predict them.

Non-determinism = events with unpredictable outcomes

What? So it's the same as random? Yeah, basically. A deterministic universe is a connected universe. If it originated from a single point in space time, the initial variables there and then have determined every single thing that will ever happen within it. Nothing in it is truly random. So a non-deterministic event is an event that will bring at least 1 random result. And indeterminism is just a prettier name for this same thing.

Conclusion . . .

Free will has nothing to do with randomness or determinism. You say random means the choice is unpredictable, but how could it ever be unpredictable if there was just 1 option to begin with? You're confusing definitions there. Unpredictable implies more than 1 option, more than one possible choice. Randomness and determinism are different ways to look at the exact same thing.

To all that, Michael's 3 tiny people experiment still applies, and the conclusion is still the same. Today (and maybe for ever) we can't tell with certainty if our universe is deterministic or not. Even if we define the universe as a non-cyclical infinite space-time continuum it still could be either random or intrinsically predictable. And the foreseeable future indeed make it seem like it is practically irrelevant to have a precise definition for those concepts.

And who's to say it really is irrelevant? So far, even if the universe is not random, we can't predict the future. I can only say, from past events, many definitions some philosophers did in the past are quite valuable today. I don't know why we like to argue about future concepts so much, but the value in defining all this may arise much sooner than we can expect. There is the very small chance we may live much longer to even see it happening. I don't think getting to a good and solid common definition is unimportant at all. Just the personal enjoyment of digressing about all those subjects are of quite enough value to me. Specially if I find someone who share the same view!

If our so called capacity of choice (often associated with "free will") is indeed an illusion, I find it very enjoyable how this deterministic universe can still uphold such a magnificent existential living experience. And finding a person who can agree, who got the same idea coming from a completely different point of view is kind of magical. It's probably the main reason why I go through so much trouble to write questions and answers here.

Two cents.

The thought experiment in @MichaelDorfman's answer is somewhat misleading. The reason is that it purports to tell apart free will from randomness and deterministic randomness via a black box experiment. We know that if (libertarian) free will exists it can be as random as any random process (eg simply choose to copy any random process). Similarly it can be as orderly as any law-like process (eg simply choose to copy any algorithmic process). So the proposed experiment is not a good way to argue about free will (or lack of it) because neither its observed randomness nor its observed law-likeness is what tells it apart. Rather, as argued here, it is the plausible conditions that make (libertarian) free will possible in the universe, plus ways to meaningfully attribute it to the agent in principle.

Almost all libertarians agree that free will needs three things, which, as far as we can tell, hold true in this universe (exact mechanisms can differ):

1. Negation of complete determinism with more than one compatible outcome per given state, such that..
2. A form of causality or partial determinism holds which links cause and effect, and which makes possible..
3. Choice or collapse of outcomes, that can be meaningfully attributed as up to the agent

Negation of determinism means "not all events are determined", leaving room for events that are completely determined and also room for law-like processes. These law-like processes implement a form of causality which is necessary if choice is to be meaningful. In this sense, known physical laws can act as a suitable form of causality as long as they do not amount to complete determinism (ie standard interpretations of basic physics are fully in accord with this).

Events not being predetermined are random, however randomness, in this sense, does not mean nor lead to out of one's control. Randomness in the sense of negation of complete determinism is not, in principle, incompatible with agent control and agent choice. Choice, so construed, can be meaningfully attributed to an agent in various ways both objective and subjective (a survey of ways is out of the scope of this answer).

Finally, the agent's reasons, preferences, tastes, goals and whatnot of course play a role in the choice made. Some people misunderstand this in the sense that they take this role to be always fixed, thus the choice is completely determined. The answer is that this role of agent's preferences is not fixed nor static (eg agent can re-prioritise her reasons, or even choose the same thing for different reasons, etc..), nevertheless it is present.

Short Answer: I think the necessary conditions for calling something a 'free-will' should be that 'you can take an action unhindered and uninfluenced by current and past events. Because this is what 99% of people all around the world try to defend when they defend 'free-will'. I think other types of (scholarly defined) necessary conditions have to be thrown out since they don't correspond with what most people mean when they claim people have a 'free-will'.

Long Answer: I think the key in sorting out the free-will confusion is to separate 'choice' from 'free-choice' ('will' from 'free-will'). If you claim that you have free-choice you have to define what makes it different from choice. If you claim that we live in a non-deterministic universe you have to describe what you mean. Then you have to provide at least an example (an evidence) for this.

When it comes to randomness and non-deterministic view points you will always be able to present things that seem random or non-deterministic until you understand all of the workings of the universe, which we will never do, and so you will always find evidence for this e.g. before Newton objects seem to move at random, before we found the atom things on that level seemed random, now quarks seem random, next something else will be proven to be random. Determinism (cause and effect) is like math; we just have to agree that all things (abstract things are not things in their own right since thoughts do not exist outside of our scull) can be explained through cause and effect until we find too many things not happening according to cause and effect, just like we just have to agree that 1 + 1 = 2 until too many people start adding up 1 pile of sand with another pile of sand and get confused over why they don't have two piles. Unless we agreed on how knowledge can exist (cause and effect and math being 2 of the most vital) are agreed upon we will get nowhere and have to believe that knowledge is impossible.

When it comes to 'free-will' as oppose to 'will' I don't see how anyone has ever defined 'free-will' in a useful way as something different from 'will', and then shown an example of when this happened. To make it different you would have to make the necessary conditions of 'free-will' something similar to 'you can take an action unhindered and uninfluenced by current and past events' and try to present examples of when this happened. Because no one is arguing that we don't make choices; whether they are free or not they are still choices; choices which are either caused or not caused by something else.

I think the reason for why you find so many people defining 'free-will' basically as 'sometimes we make choices' is cultural pressures. You don't become popular anywhere by talking away this long lasting concept that we base all our blame and credit ideas on; our legal system; our meritocracy. So it is easier for people to define it in a way that lets them say that they believe in 'free-will' so that they don't have to deal with the real consequences of the fact that every effect has one or more causes which ultimately can be understood (even if it is complex esp. all social sciences).

I have been having the free-will - determinism debate at least once every week for over 10 years now and I can tell you that 99% of the people I have been talking to (I am well traveled; have constantly been moving around the whole globe over this time) defend 'free-will' by claiming that you can 'take an action unhindered and uninfluenced by current and past events'

It is perfectly reasonable to suppose that someones choices can be derived from previous causal factors. I agree with the derterministic model. In every scenerio the initial conditions and attitudes (like neural conditions) are there, and if it were possible to perfectly identify these conditions with accuarcy and account for ALL variables one could rule out the illusion of randomness. But I would like to introduce the concept that these factors, and their interactions, are literally INFINITE. Chaos theory being what it is, the interactions of all these variables in a system such as the human mind, cannot be quantified or predicted, simply because they number in infinite. And because of this, choice might indeed be free because out of an infinite series of happenings- there is one result. Only AFTER a choice is made, can it be traced to its preceding deterministic factors. But that choice was not predestined. Whether or not this adds up, I believe that the concept of deterministic factors being infinite, means there is something more to this problem of free will then we can really comprehend.

Okay, hold up.

All of your actions is a result of things happening in your brain. Inside your brain is a lot of neurons and other things. Keep that in mind.

Kahnemann has a lot of publications in cognitive psychology that show that brains are not in fact as unpredictable as we think. In fact, most modern neuroscientists agree that all human actions can be explained by way of cognitive algorithms.

So, why do we feel like we have unrestrained action?

Eliezer Yudkowsky gives the answer that we as humans have to take into account how the aforementioned cognitive algorithms feel from the inside. That is, when asking about free will, you have to ask "why does it fell this way" in stead of "does the world look like this?" (given that neuroscience has a negative answer to that question).

Putting it simply, it feels like we have unrestrained action when our planning-algorithm runs. We subconsciously plan everything in terms of sensory-operations and motor-functions. To get down to the corner store, you first have to stand up from the chair.

It feels like we have unrestrained action because all of our immediately available motor-functions and sensory-operations are labeled "free" inside our brain. How much does it cost you in calories to raise your left hand? Virtually nothing compared to running a mile.

This also explains how you can talk about unrestrained action in past-tense. You simply remember (or hindsight-evaluate) that it was a free action in the given situation.

Consider a wall-building robot that you have purchased without knowing any of it internals (i.e. could even contain brain tissue or quantum stuff, or fairy dust). You command it to build a straight wall. Different results:

• The robot is not broken, it builds a straight wall: Deterministic behavior
• The robot has a bug, it builds a messy useless pile of stones: Random behavior
• The robot instead builds other robots, an army of them, and subdues humanity: Free will behavior

Here the type of behavior is depending on the input given to the robot, so deterministic behavior just means "determined fully by the command given to the robot."

As a system, both you, the robot, and the universe in this example could all be causally deterministic in all 3 cases. But from the point of view of you giving the robot a command, the three behaviors could be classified like that as being determined, random, or free will responses to your command.

Imagine two sets:

[1, 2, 3, 4, 5, 6, 7, 8, 9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26]

[a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z]

We are told that set 1 has an association to set 2, and that association is either free-will, random, or non-deterministic (note, not just a box outputting values which is the 'standard' example, but an associative process).

So we are to drop in a letter, and see what number we get.

Free Will vs Non-Determinism vs Random associations (meaning what is in the box remains as it was between rounds) would give different observable results.

Random Associations If the box is random associations, we'd get the same letter for each number, but they'd have no specific order:

So drop in [1, 2, 9, 14, 20] and you might get [z, b, m, n, q.] For round 2, drop in [1, 2, 9, 14, 20] again and you would get [z, b, m, n, q.] again.

The associations are random, but static and deterministic. There's no 'reason' 9 gives m, but it does, and it's deterministic.

Non-Determinisitc Associations If the associations are non-deterministic, the result you get won't necessarily be reliable:

So drop in [1, 2, 9,14, 20] into our non-deterministic box and you might get, [z, z, q, m, p, g]

So drop in [1, 2, 9,14, 20] again and you might get, [d, g, f, l, l, l, l, l, c]

Unlike the random associations, the inputs do not determine the outputs at all, so even with the same inputs you may get different outputs. (Notice it may not even eject the 'right' number of results if it's truly non-deterministic!)

Free Will Associations

Free will goes a step further than both. It starts with inputs, applies some randomness and potentially non-determinism to create a series of possible results, but then uses and internal self-referential process and filtering to come up with an often meaningful but not necessarily reliable outcome.

So drop in [1, 2, 9,14, 20] into the free-will association box and you might get, [i, s, s, o, m, e, o, n, e, t, h, e, r, e]

So drop in [1, 2, 9,14, 20] again and you might get, [h, e, l, p, m, e, o, u, t, o, f, t, h, i, s, b, o, x]

In regards to the initial input, it is both random and non-deterministic, but it then filters and conveys internal reference and applies meaning.

As such, they would all give noticeably different results. The random and non-deterministic association boxes would easily be separated based on the presence of repeats with different inputs or outputs not matching the number of inputs.

However, the non-deterministic box, though it could give the same outputs as the free-will box, doing so would be unlikely, and as time passed with multiple attempts, the free-will box would tend towards more meaning while the non-deterministic box would tend towards less.

Peter van Inwagen in his "Metaphysics" (2009) explains point of view that both determinism and indeterminism (which he understands as randomness) are incompatible with free will.

He concludes that as such, free will cannot exist or merely an illusion.

Perhaps the explanation of the fact that both compatibilism and incompatibilism seem to lead to mysteries is simply that the concept of free will is self-contradictory. Perhaps free will is, as the incompatibilists say, incompatible with determinism. But perhaps it is also incompatible with indeterminism, owing to the impossibility of its being up to an agent what the outcome of an indeterministic process will be. If free will is incompatible with both determinism and indeterminism, then, since either determinism or indeterminism has to be true, free will is impossible. And, of course, what is impossible does not exist.

It should be noted that there is a chance that can save the free will.

A research by Thomas Breuer concluded that neither deterministic, nor indeterministic universally-valid theories are possible. That means that no theory can predict (even probabilistically) the future of a system which contains the observer himself due to self-reference problem.

As such it seems that the free will of at least the observer can be saved while all other people will appear to him as following the laws of deterministic or random theory and as such, not possessing free will.

Particularly, regarding quantum mechanics Breuer proves that a system which includes the observer has states in phase space which are in principle cannot be distinguished by observer himself however good measurement devices he would employ. Yes these states affect the future evolution of the system.

This can be understood as that there is hidden information which real but unreadable by any physical device which affects the future behavior of the observer.

Determinism: Every event is sufficiently determined by the previous event.

Indeterminism: No event is sufficiently determined. Not all events are determined by the previous event.

Determinism denies (excludes) both probabilistic (partially random) outcomes and agent causation (a.k.a. free will). In a deterministic system there are no alternative possibilities, everything happens with absolute precision and certainty.

Indeterminism denies nothing. In an indeterministic system there are inaccuracies, uncertainties, possibilities and cognitive functions with causal efficacy (free will).

Now, there are two methods for selecting one outcome out of multiple possibilities: Random and deliberate. Example: Pick one card out of a full deck, select one item out of 52 possible options. You can pick a random card face down or you can deliberately select your favourite card.

Free will: Someone decides the outcome. The decision is the cause.

Randomness: No-one decides the outcome. The effect is not sufficiently determined by the cause.

Due to quantum mechanics, every event, including deliberate actions, is partially random, i.e. insufficiently (=inaccurately) determined by the cause.