Oh my... another meditation on the issue of free will, and the existential angst we would experience if we learned that our choices are deterministic. Personally, however, I think that the whole question of free-will vs. determinism is a false dilemma. For me, another issue is more important:

Determinism vs. randomness.

If my choices are not deterministic, they must be random. I would rather be deterministic than random. In any case, if they are random, it means that I don't have free will either.

I like to think that all my choices are a result of the past; my genetics, my childhood upbringing (that I don't even remember, but which formed the neural pathways in my brain), my early memories, my environment, education, experiences, friends, knowledge that I have absorbed from the world... In each and every moment, I make a choice, which is the best choice I can make given my brain power/structure, my motivations and the external constraints given (is it raining? can I fly? when do I need to pay the rent). Even my motivations are largely determined by my genetics - avoid pain, strive for pleasure. I definitely hope that my choices are not random.

How does morality come into play, if our choices are deterministic? It doesn't - my morality is my internal concept that I use to make choices more quickly/easily. I don't impose my morality on other people and I don't really care how they make their choices, but I support different forms of punishment that modify the incentives of other people so that the society can function.

Finally, I don't think that the future is predictable, even though it is deterministic. Like you don't know what 1048936701349 * 13046871435 is before you calculate it, like even the computer cannot predict the result before calculating it (i.e. the fastest calculation algorithm is also the fastest prediction algorithm), the same way we cannot predict the future before it happens, i.e. before the universe "calculates" it.

This reminds me of the Isaac Asimov spoof article "The Endochronic Properties of Resublimated Thiotimoline" which describes "experiments" on a substance that reacts so aggressively with water that it reacts before the water is added. Can't find the text unfortunately, but Wikipedia has a summary: http://en.wikipedia.org/wiki/Thiotimoline

What I really hate about these sorts of stories and attending discussions is that we all pretend we know what "free will" is, when no one has ever come up with satisfying definition (just look at the Wikipedia page). My humble opinion: all hand-wringing about lack of free will should be put on hold until we can answer the following question:

In what way will an entity that has free will behave differently than both a fully deterministic entity and an entity that is deterministic excepting occasional random events?

I have yet to see either an answer to that question, or a definition of free will that leads to one. Until then, I consider all speculation and meditation on the "revelation" that we have none to be a tale full of sound and fury, signifying nothing.

The predictor already exists, there's a common FMRI experiment in neuroscience where with about 80% accuracy the scientists can predict you're about to push the button about one second before you decide to do it. Someone no doubt will post a link, I don't have time to Google it now. I'm guessing it's the inspiration for this story.

I'm more curious about what you guys think. I've long thought the arguments against free will are much stronger than those for it, but it's an unsettling idea to live with.

Anyone that liked this should definitely check this one as well by the same author: http://infinityplus.co.uk/stories/under.htm

The question of "Do we have free will?" always becomes a lot less confusing if you question the question itself, by asking what you mean by "free will" in the first place. There are some nice posts on Less Wrong that deal with this, they're worth a read if you're curious:

http://wiki.lesswrong.com/wiki/Free_will_(solution)

I'd say this touches more on the mechanics of time travel than free will.

So if this "predictor" is nothing more than a circuit that sends information to itself in the past, there must be a "me" that pressed the button without seeing the light, before the data is transmitted to the past to the "me" that then sees the light but at that exact point splits off into a separate timeline. And then I would have free will again.

Of course I am assuming time travel where travelling forward in your own timeline is impossible, something the article seem to not really go into.

Does such a detector imply that there is no freewill? Would determinism imply it?

Imagine the predictor works, what does that mean? One second prior to the action, the course you would take is set. This is actually probably a point in favor of the possibility of freewill; humans are course-taking machines, we have a whole life (and our evolution before that) to acquire dispositions that will one day help us survive. What kind of freewill do we want? Do we want to be "free to dodge a brick when thrown at us" or "free, one second before the brick arrives, to duck or not duck". Which of these is the important one? The kind of freewill we really want isn't the kind where we are free floating actors, it is the kind where our history, personal and evolutionary, dictates our present. The point is, real freedom is not about being without limits, it is about having sensible responses to ones environment; creatures evolve the freedom to avoid being eaten, the freedom to anticipate others' actions, the freedom to manipulate other agents with our clever words. Dan Dennett makes a better case for this than I, please look into his wonderful books on the topic.

Daniel Dennett: --Elbow Room --Freedom Evolves

Just because you can predict what simple decision I will make before I am aware of it consciously, does not imply I don't have free will. All it says is that my conscious awareness of some other part of my brain making the decision is delayed by x milliseconds i.e. we are not making decisions in the upper layers of consciousness, but below it.

When you think about it, it is what I would expect. Your awareness is the product of nerves firing, so they have to fire before it can be reflected in your consciousness. If you decide to recall some information, the neurons storing that information have to be activated (they have to fire). If you can create a device (and we have) to detect those nerves as they are firing you will know what person is trying to recall before (by some miniscule lead time) they are aware of it.

Sadly we're not informed what happens if you program a robot to press the button at fixed intervals unless it observes a flash.

Ted Chiang's Predictor would be quite the expansion pack for Hermann Hesse's Glass Bead Game.

Here's a similar appearance by Vernor Vinge: http://www.nature.com/nature/journal/v407/n6805/full/407679a...

I love pieces of writing like this.

Determinism isn't something I've explored quite satisfactorily enough just yet, and reading articles such as this one sends me straight back into the depths of the literature.

This is known in the dynamical systems literature as strong anticipation:

http://digital.csic.es/bitstream/10261/7493/2/Dynamical_mech...

http://www.haskins.yale.edu/Reprints/HL1521.pdf

My predictor says that some of you will want to read The Hundred-Light-Year Diary by Greg Egan (in Axiomatic [1]) ... if you like this kind of thing.

[1] http://en.wikipedia.org/wiki/Axiomatic_%28story_collection%2...

Why is greater than one second is a problem? You can chain many predictor together using robots that when seeing a flash from one predictor presses another predictor. That way you can move the message back in time many seconds.

Cool idea for creative writing class, would've helped in plausibility if they'd thrown a few nano and quantum randomly into the essay instead of "negative time delay circuit".

Relevant: http://en.wikipedia.org/wiki/Timequake

That reminds me of Feynman's Thesis and his (failed) attempt to build his Feynman's Radio...

Reminds me of Comed-Tea in "Harry Potter And The Methods Of Rationality":

"If you drink it, something surprising is bound to happen which makes you spill it on yourself or someone else. But it's charmed to vanish just a few seconds later"

Even if we were free thinking individuals, able to make rational decisions on our own, we base those decisions on past events and the current situation. So even if you get to choose, you'd probably make the same rational choices as long as your past experience and the environment is the same.

This would be true one second ago, and a second ago this would be true about the second before that.