LESSWRONG
LW

Simulationist sympathetic magic

1 min read

Comment Permalink

DataPacRat13y20

I'm assuming much more limited computational resources than all that are being used in at least some of the simulations - something much closer to the minimal required to fool a simulated inhabitant into thinking that their universe runs on full physics, with plenty of shortcuts taken (such as deliberate falsification of experiments performed by simulated physicists).

I'm also assuming that at least one goal of those doing the simulating is to find a simulation that emulates their own past as closely as possible, given whatever information on their past they still have.

With these assumptions, then the 'sympathetic magic' thing would seem to increase the probability of the desired outcome occurring in at least a few such simulations, thus at least mildly increasing the probability that the person involved will experience the desired effect. It isn't going to happen in all simulations; but even increasing the probability from 50% to 51% could have some use.

abramdemski13y20

With these assumptions, then the 'sympathetic magic' thing would seem to increase the probability of the desired outcome occurring in at least a few such simulations, thus at least mildly increasing the probability that the person involved will experience the desired effect.

It would increase the probability of the outcome occurring in simulations in which the protagonist does not come up with the trick. (Or doesn't apply it to this occasion, as DanielLC mentions.) In simulations where the protagonist does come up with the trick (and applies it to this c... (read more)

5DanielLC13y

In that case, the simulation wouldn't be accurate enough that you'd have the same test. You write down that you saw a pigeon when you didn't in April. Emulated you writes down that he saw a family of ducks when he didn't in February. He decides his test failed. Several months later, he sees a pigeon and thinks nothing of it.

21

Simulationist sympathetic magic

13th Nov 2012

1 min read

21

While I've been wrestling with the inspiration needed to turn my fanfiction into actual fiction rather than just an author's tract... I have had an unrelated but fun thought, which I'm throwing into the breeze here for any improvements that can be suggested.

If our universe is more likely to be a simulation, a reconstruction of the past by our descendants than the base level of reality; then that reconstruction is likely to be imperfect, based mainly on surviving records (and memories of anyone whose brain survives intact long enough for upload-style scanning).

Therefore, if somebody precommits to only leave behind records which correspond to particular events... then it seems plausible (to within the bounds of 'the brain is a quantum computer' levels of plausibility) that those events become more likely to be experienced. For example, if a protagonist were to precommit to mentioning that during their walk, a bird landed right in their hand to eat a bread crust, whether or not such an event actually happened; then the probability that they will then experience a bird landing in their hand increases.

There are, of course, extreme limits on what can be accomplished with such trickery, even in theory. Violating the known laws of physics is right out, as are events dramatic enough to leave behind more traces than their own memories and journals. It also seems highly recommended for someone who wishes to try this to precommit to never leaving a trace that they are going to try it, as that would leave a record the future simulationists could use to discount their less-probable reports.

So - what additional thoughts could be added to the above to make it more plausible, at least to those who've heard of simulationism in the first place?

(And since it seems more likely than not that someone will ask: No I've never tried using such simulationist sympathetic magic myself, and since I still question the basic assumptions behind such mass-simulation in the first place, I have no intention of trying it in the future, either.)

Simulation Hypothesis

21

New Comment

20 comments, sorted by

Click to highlight new comments since: Today at 4:25 PM

[-]Eliezer Yudkowsky13y250

Blog each time you have sex. This makes you more likely to remember histories in which you had sex, and possibly even makes you more likely to experience future occasions where you had sex. If a friend gets sick, don't mention anything anywhere an electronic record can find it, or the whole thing might very well have been your fault! And for a different variant on the same principle, wait for a large lottery jackpot, buy a ticket using a quantum random number generator, use a webcam to record yourself watching the numbers being drawn, and precommit that if you win, you will donate 10% of the money to workshops on timeless decision theory providing that they agree to play the video at the opening session of that conference.

[-]DataPacRat13y10

If a friend gets sick, don't mention anything anywhere an electronic record can find it, or the whole thing might very well have been your fault!

On the other paw, if you have an enemy who you wouldn't mind experiencing a malady...

[-]Oligopsony13y250

And since it seems more likely than not that someone will ask: No I've never tried using such simulationist sympathetic magic myself, and since I still question the basic assumptions behind such mass-simulation in the first place, I have no intention of trying it in the future, either.

I see what you did there.

[-]DataPacRat13y110

Well, it did seem the thing to do...

[-]DanielLC13y120

I would expect that if their simulation is accurate enough for you to precommit to the same test, you're not going to fool it with a journal entry.

Also, if they simulate many worlds, you'll almost certainly end up in an Everett branch that precommitted to a different test, and if they simulate Copenhagen, you'll almost certainly end up with the wrong random numbers.

[-]DataPacRat13y20

I'm assuming much more limited computational resources than all that are being used in at least some of the simulations - something much closer to the minimal required to fool a simulated inhabitant into thinking that their universe runs on full physics, with plenty of shortcuts taken (such as deliberate falsification of experiments performed by simulated physicists).

I'm also assuming that at least one goal of those doing the simulating is to find a simulation that emulates their own past as closely as possible, given whatever information on their past they still have.

With these assumptions, then the 'sympathetic magic' thing would seem to increase the probability of the desired outcome occurring in at least a few such simulations, thus at least mildly increasing the probability that the person involved will experience the desired effect. It isn't going to happen in all simulations; but even increasing the probability from 50% to 51% could have some use.

[-]DanielLC13y50

I'm assuming much more limited computational resources than all that are being used in at least some of the simulations

In that case, the simulation wouldn't be accurate enough that you'd have the same test. You write down that you saw a pigeon when you didn't in April. Emulated you writes down that he saw a family of ducks when he didn't in February. He decides his test failed. Several months later, he sees a pigeon and thinks nothing of it.

[-]abramdemski13y20

With these assumptions, then the 'sympathetic magic' thing would seem to increase the probability of the desired outcome occurring in at least a few such simulations, thus at least mildly increasing the probability that the person involved will experience the desired effect.

It would increase the probability of the outcome occurring in simulations in which the protagonist does not come up with the trick. (Or doesn't apply it to this occasion, as DanielLC mentions.) In simulations where the protagonist does come up with the trick (and applies it to this case), it's a sufficient explanation for the various notes which the protagonist leaves; the events don't need to happen. So leaving false notes will only have the desired effect in simulations in which the protagonist is sufficiently different so as to not leave those notes.

This could still be worthwhile to the protagonist.

For fun fiction, such a simulated protagonist could eventually understand the trick and what is happening. (The simulated protagonists actions would still have no supernatural correlation to other events in the simulation.)

If our universe is more likely to be a simulation, a reconstruction of the past by our descendants than the base level of reality; then that reconstruction is likely to be imperfect, based mainly on surviving records (and memories of anyone whose brain survives intact long enough for upload-style scanning). Therefore, if somebody precommits to only leave behind records which correspond to particular events... then it seems plausible (to within the bounds of 'the brain is a quantum computer' levels of plausibility) that those events become more likely to be experienced.

How does this cope with simulation dynamics? They may seed the simulation with lots of old real-world data, but once it starts running, there are going to be tons of micro and macro divergences which make data dated after the simulation start less than useful. (eg. imagine the simulation run starts at a point in 2000; by the time it reaches its version of today, it likely will have diverged significantly - it doesn't matter if you write that you had dinner with your girlfriend because in the simulation you just barely missed meeting her a year ago etc.)

[-]abramdemski13y80

Suppose we do rejection sampling. Then the universe restarts at 2000 every time a datapoint is violated. The subjective experience of entities inside such a simulation would be best described as random death with occasional survival if some arbitrary criteria are met. Writing (long-lasting) believable false reports will alter the criteria to include those reports somehow being written.

MCMC algorithm is more interesting. It introduces a strange contortion of time; we generate an initially random space-time with the known datapoints clamped, and then re-sample the unknown bits for a long time to get a good posterior distribution over the possibilities. Real time is a sometimes-nonsensical meandering through possibility-space. Time as experienced "in the simulation" is relatively normal, but it is interesting/boring to argue about whether entities living in such a simulation have experiences in a meaningful sense. Causality doesn't work at all as it should, but it will appear to work roughly as it should most of the time. Events will conspire to bring about a random assortment of facts which are the clamped values, but the better-quality samples will make the conspiracies look like true chance.

Metropolis-hastings is an even more advanced technique, but I don't think there is anything special about the subjective experience if the simulated entities in MH as compared to basic MCMC.

I assume that if a statistically significant number of people noticed that they were trying sympathetic magic and it was working, then the simulation would have to be restarted or tweaked since it could alter the history of the world in significant ways. So you might want to plan out that aspect of your strategy before collecting any data.

[-]thomblake13y50

I assume that if a statistically significant number of people noticed that they were trying sympathetic magic and it was working, then the simulation would have to be restarted or tweaked since it could alter the history of the world in significant ways.

We're already living in a world where huge numbers of people believe sympathetic magic is working.

How many of them wouldn't believe it if it wasn't working?

[-]CronoDAS13y60

Wow, those "Law of Attraction" dumbasses might actually have been on to something! ;)

[-]A1987dM13y40

Maybe that's what Will_Newsome meant by “generalized pseudo-anthropic selection effect”.

[-]thomblake13y40

I think Will_Newsome 'means' things that make sense, about as much as the Old Testament was meant to contain metaphors for quantum physics.

[-]ArisKatsaris13y20

Nice, you've figured out a way to make testable predictions regarding at least some varieties of the simulation argument.

Now let's use what you hypothesize about to disprove it, and finally move on already from all the simulationist idiocy...

Is the idea that some group has confounded the knowledge base used to maintain said simulation for the express purpose of bestowing superpowers upon their progeny or all of future humanity?

If that's what you're going for, then I'd refer to that group as either crazy or a bunch of really fun people that were having a laugh, knowing the probability of their efforts paying off to be extremely slim, who coincidentally were absolutely spot on.

I imagine the most effective approach would be propagating reports of small, odd phenomena that hint at the universe having (had) features not (yet) identified by the laws of physics. Then the descendants could exploit the resulting uncertainty of the simulation through heroic journeys of super-power discovery. The descendants would either be protagonists, or all of humanity; I think the latter more ambitious but also permissive of a larger scope and greater intrigue than the former.

[-]DataPacRat13y10

Well... that's not quite the idea I had in mind - but as an idea on its own, it's another one worth adding to the inspiration pile.

No I've never tried using such simulationist sympathetic magic myself, and since I still question the basic assumptions behind such mass-simulation in the first place, I have no intention of trying it in the future, either.

Ha! But are you saying that because you really never tried or because you tried and it worked? :p

However, about your thoughts on the whole mass-simluation affair: if we live in a faithful simulation and we discover we are living in it, then this raises the probability that even our ancestors found out that they were living in a simulation.

ETA: you could also try to outsmart your simulators by leaving uncorrelated notes about all your life, and producing false evidence about everything you do or that happens to you, so that future archaeologist will have a hard time reconstructing your life. That will either cause your death (due to the excessive complexity of your reconstruction) or they will need a reliable source of information... that yourself will have constructed!

More from DataPacRat

Curated and popular this week