PSA: The unwritten code of ghost physics
Dec. 11th, 2006 12:28 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
jackMany science-fiction/fantasy stories have a ghost or two, and it's often unclear what they can do. I've attempted to research the basic laws which govern such beings so everyone can make use of them.
* Ghosts rest on flat surfaces. The most likely explanation for this is subconscious direction from the ghost. In the comic 1/0 (which has caught up Terry Pratchett as the story providing the most handy-metaphors) ghosts naturally assume the shape they had at death (or possibly the most representative point in their life) and hover over the ground, but with concentration can change their shape a bit or a lot and move about.
It also explains a ghost's body, which is invariably quite like a real body, but doesn't actually have (even ghost) chemistry and physics going on inside.
The theory is almost entirely consistent, the only problem being it doesn't feel like a satisfying, physics, explanation, as you can justify almost anything like that.
* Another hypothesis is that ghosts have inertia. If they remained stationary, it would be meaningless because stationary only has meaning relative to something. If they continued in a straight line they'd drift away, within 5m for about 10 minutes, and gone in an hour. If they followed gravity naturally, they'd fall into the earth. However, if they had a tinier attraction to gravity, it'd be basically undetectable, but they'd orbit along with the rotation of the earth.
I don't know how this would interact with artificial gravity, or warp drive -- Geordi La Forge going up in a Turbo-lift in StarTrek isn't particularly explained, nor even going up stairs.
* Or that other things have a normal effect on ghosts, but not ghosts on other things. This breaks "Every action has an equal and opposite reaction", which I'm not sure if it makes sense or not. But then ghosts would feel but not exert gravity. They could be supported by earth and car seats and things, and blown by winds, and hit by people, but not pick things up, nor walk through things.
* There is often some unreliable telepathy, with people picking up thoughts from the ghost. This sometimes has to do with emotional intensity, but basically, is at times unconnected apart from plot need.
* Ghosts can't affect other objects, except possibly as times of strong emotion. Sometimes time of day is important, too. They can generally only blow things about, often small things, but occasionally large things. Some ghosts can actually write, though normally only at the end of the film.
* However, they can't do this in the presence of anyone acquainted with scientific method or scepticism. In front of people who think it's the wind, or who see ghosts in every kitchen, they can do all sorts of things. In front of someone who says "Here are two candles. If you can blow out the one I designate first each time I light the pair, I'll believe in you with 99.9% confidence" they are entirely powerless, and merely gibber invisibly. (To be fair to Piers Anthony, he did treat this sensibly, if misogynistically.)
* There is normally an arbitrary cut off in sapience which decides who can become a ghost, of whom a random selection will. It's often dependent on state of mind, eg. unfinished business, etc.
To me this is evidence that sapient consciousness has a special place in that universe, but sometimes there is an unexplained rationalistic explanation.
In some canons, inanimate objects become ghosts but fade almost instantly (thank you, Terry Pratchett, for having more consistency in comic fantasy than in many 'serious' books :)) In Night's Dawn trilogy, it's interesting: you can have electronic copies of people, who count as separate people for the purposes of making ghosts!
* Go into the light!
* Ghosts rest on flat surfaces. The most likely explanation for this is subconscious direction from the ghost. In the comic 1/0 (which has caught up Terry Pratchett as the story providing the most handy-metaphors) ghosts naturally assume the shape they had at death (or possibly the most representative point in their life) and hover over the ground, but with concentration can change their shape a bit or a lot and move about.
It also explains a ghost's body, which is invariably quite like a real body, but doesn't actually have (even ghost) chemistry and physics going on inside.
The theory is almost entirely consistent, the only problem being it doesn't feel like a satisfying, physics, explanation, as you can justify almost anything like that.
* Another hypothesis is that ghosts have inertia. If they remained stationary, it would be meaningless because stationary only has meaning relative to something. If they continued in a straight line they'd drift away, within 5m for about 10 minutes, and gone in an hour. If they followed gravity naturally, they'd fall into the earth. However, if they had a tinier attraction to gravity, it'd be basically undetectable, but they'd orbit along with the rotation of the earth.
I don't know how this would interact with artificial gravity, or warp drive -- Geordi La Forge going up in a Turbo-lift in StarTrek isn't particularly explained, nor even going up stairs.
* Or that other things have a normal effect on ghosts, but not ghosts on other things. This breaks "Every action has an equal and opposite reaction", which I'm not sure if it makes sense or not. But then ghosts would feel but not exert gravity. They could be supported by earth and car seats and things, and blown by winds, and hit by people, but not pick things up, nor walk through things.
* There is often some unreliable telepathy, with people picking up thoughts from the ghost. This sometimes has to do with emotional intensity, but basically, is at times unconnected apart from plot need.
* Ghosts can't affect other objects, except possibly as times of strong emotion. Sometimes time of day is important, too. They can generally only blow things about, often small things, but occasionally large things. Some ghosts can actually write, though normally only at the end of the film.
* However, they can't do this in the presence of anyone acquainted with scientific method or scepticism. In front of people who think it's the wind, or who see ghosts in every kitchen, they can do all sorts of things. In front of someone who says "Here are two candles. If you can blow out the one I designate first each time I light the pair, I'll believe in you with 99.9% confidence" they are entirely powerless, and merely gibber invisibly. (To be fair to Piers Anthony, he did treat this sensibly, if misogynistically.)
* There is normally an arbitrary cut off in sapience which decides who can become a ghost, of whom a random selection will. It's often dependent on state of mind, eg. unfinished business, etc.
To me this is evidence that sapient consciousness has a special place in that universe, but sometimes there is an unexplained rationalistic explanation.
In some canons, inanimate objects become ghosts but fade almost instantly (thank you, Terry Pratchett, for having more consistency in comic fantasy than in many 'serious' books :)) In Night's Dawn trilogy, it's interesting: you can have electronic copies of people, who count as separate people for the purposes of making ghosts!
* Go into the light!
no subject
Date: 2006-12-11 08:43 am (UTC)(I could work it out, given Bayes theorem, some sensible assumptions and a cop-out prior of 50%, but it's too early in the morning for that...)
no subject
Date: 2006-12-11 02:13 pm (UTC)Set up two candles a distance apart. Choose one at random. Stand still and request the ghost try to blow that out. Wait for as long as necessary (to give the ghost a chance to get mad, presumably staying by designated candle) until one goes out. Repeat ten times.
H0: There is no ghost, and the first candle to be blown out is random.
H1: There is a ghost, who blows out the candles you request each time.
Event: Ten times, the designated candle goes out first.
P(E|H0) = (1/2)^10 = ~ 1/1000 = ~ 0.1%
So we reject H0 with 99.9% confidence. It's been a while, but am I doing this wrong?
(Of course, we should probably alternate the candles and only count half the results, to take into account the hypothesis that one is somewhere windy. That's still 1/2^5~3%.)
(And it assumes the ghost gets it right every time. If the ghost isn't that precise, it'd be harder to tell. But in most films, getting angry within half an hour, within a metre radius seems well within capabilities.)
no subject
Date: 2006-12-11 02:58 pm (UTC)You said "believe in you with 99.9% confidence". This is NOT disbelieve in the null hypothesis with 99.9% confidence. I did a calculation assuming that the only models were H0 and H1 (with a prior probability of 50% each), and for ten candles P(H1|E) (the probability that's actually meaningful, dammit) comes out as 0.999024 - not bad. With a more reasonable prior, and some modeling of extra alternative hypotheses, then you'd need a lot more candles.
no subject
Date: 2006-12-11 03:42 pm (UTC)If you rephrased the null hypothesis to be "What I say aloud has no effect on which candle goes out first," wouldn't rejecting it be the same as admitting some supernatural influence was at work? OK, you can never show it's a ghost as opposed to PK without a double blind complicated test, but it'd be enough for me :)
I'm also vague on Bayesian reasoning. Is it that H0 should have a large prior, since I *am* previously almost convinced of its truth? Obviously there are lots of such experiments, so I expect some to "succeed", but if I sit down and design one, it feels more meaningful.