Subcultural engagement for the 21st Century...
Barbelith is a new kind of community (find out more)...
You can login or register.

Underground > Laboratory >

Scientific Analysis of Ritual Effects

16:50 / 17.10.06

A recent comment in the Science and Politcs thread got me to thinking about what sort of rigorous experimental framework could be set up to determine the validity of the claim that ritual has verifiable effects. Sn00p said:

Ritual magic is verifiable. You can do a ritual and get an effect. The variables involved may be different, but if what you're trying to prove is whether your actions have an effect, this is verifiable, i.e do a ritual, record the results. While the basic structure of all magic ritual is objective, the actual mechanics is subjective, so you would only ever be able to prove magic to yourself using rigorous experimentation in the form of a ‘subjective science’.

The distinction between "objective structure" and "subjective mechanics" is something that I didn't understand very well in this post and would like to have clarified, but I didn't want to rot the politics thread. The whole notion of the mechanics of something with observable effects being subjective doesn't really sit well with me; if the effects are extant and reproducible, then there must be some way of decomposing the causal chain - to discover an objective physics of magic, as it were.

One framework for testing the validity of "paranormal" claims already in existence is the one James Randi set up for his Million Dollar Challenge which has been discussed here previously on a couple of occasions. Some other relevant Temple threads that may lend something to the discussion are here, here and I'm sure I recall a couple of others, though I haven't been able to find them at the moment.

So, here's the central question that's been answered many times in the Temple but never, I think, in the Lab: what constitutes proof of the efficacy of magic? As scientists, if there's an undiscovered causal mechanism out there that magic exploits, then describing it in a scientific framework and developing a falsifiable theory for it would be incredibly valuable contributions to human knowledge about how the universe works.

17:02 / 17.10.06

All i meant was that everybody is an individual, so the exact same rituals actions that work for one person probably wont work for another. So i'd say that part of it is subjective, (perhaps i'm using the word subjective in the wrong context). All i meant is it's an internal individual process.

But on the large scale the magical act is objective: certain action, certain response.

17:07 / 17.10.06

(sorry for the double post)

I'm actualy quite embaressed about how confusing my first stement is, it's not something which is educated very well so it can be hard to talk about.

Just to throw the idea out there, isn't there expeirments with quantum mechanics to show the perception of an event somehow alters it? Infomation traveling through time? I have a book on it at home. I'll post the gist of the chapter when i get back in a few days.

17:48 / 17.10.06

...isn't there expeirments with quantum mechanics to show the perception of an event somehow alters it?

There's some confusion about this: basically, it is impossible to measure a particle without affecting it because whatever means you use to measure a particle must necessarily interfere with (bounce off, or interfere with the waveforme) it and hence change its properties.

There's also good evidence that a particle/wave is in a superposition of states - multiple potential states at once - until it is measured, at which point the state becomes established. This has some really interesting implications in terms of causality, and there's actually an experiment being done right now to take a look at the possibility of retrocausality using entangled particle pairs: see this NewScientist article (unfortunately requires a subscription, but I can PM the text to anyone who wants).

What any of this means with respect to magic is questionable, however. I'm a little concerned about the tendency for quantum weirdness to be a "last refuge" for explanation of paranormal events by people who don't really understand quantum theory. In short, you can't just say "quantum uncertainty!!!" to explain the mechanism of magic - if you want to appeal to quantum theory to explain paranormal phenomena, then it needs to be done by making specific predictions within the framework of quantum mechanics and testing whether they're borne out in experiment.

18:10 / 17.10.06

Yeah i'm sorry, i may be wrong, the book im quoting is The Dilbert Principal

18:43 / 17.10.06

what constitutes proof of the efficacy of magic?

I'm nearly certain it'd have to be something like you'd see for treatment outcomes for mental health therapy -- large study groups using, umm, what's the word... it's a term for diagnostic tools measuring subjective things like "self-efficacy" and "satisfaction." I can't think of what they're called, but they're a kind of questionnaire. An "inventory"? Argh.

That I think this is the way to study 'em probably reveals a lot about how I think rituals work.

18:47 / 17.10.06

Oh, and my understanding of this: it is impossible to measure a particle without affecting it because whatever means you use to measure a particle must necessarily interfere with (bounce off, or interfere with the waveforme) it and hence change its properties.

..is that part of the problem is that once you get small enough, you're actually studying the things that perception is made of -- you need photons to see, and if what you're looking at is a photon (or smaller), and you're not actually using the photon to trigger your optic nerve, then you're somehow going to be jostling it with photons (or things connected to photons) that ARE going to make it into your optic nerve.

Is that correct?

19:42 / 17.10.06

...and you're not actually using the photon to trigger your optic nerve, then you're somehow going to be jostling it with photons (or things connected to photons) that ARE going to make it into your optic nerve.

Essentially, yes. In fact, even when the photon *does* trigger the optic nerve (or detector, etc.) it does so by being absorbed and converted to an electrical impulse. Basically, you can't measure anything without "touching" it, and touching a particle or wave alters it, even if you're only touching it with light.

Now, this doesn't go all the way to explain the fundamental weirdness of the single photon double-slit experiment (see the "When observed emission by emission" section at the bottom.)

22:14 / 17.10.06

Just to add that observation needn't be done by a human or even sentient observer -- a purely mechanical device counts as an observer.

Kiltartan Cross

09:18 / 18.10.06

An approach that might work would be to enlist a very large number of magical practitioners, each to attempt their own ritual to influence the outcome of a single series of random events.

In other words, we might say that "at such-and-such a time we will be rolling a hundred dice. We'd like you all to attempt to ensure that we get a lot of sixes" to a thousand different magical practitioners. If we subsequently rolled those dice and came up with a load of sixes, we could suspect that at least one of those magical rituals had had an effect. We could then repeat the experiment with only half of the original participants, and see if we still got an effect, then trying the other half, and so on - essentially narrowing it down until we ended up with one (or more) rituals which definitely seemed to be having an effect.

This would allow great freedom of scope, as the magicians would be able to try any and all methods of magic; instead of strictly testing one method at a time, we'd be looking for something, anything that had an effect, and then working backwards to determine what it might be.

There would still be limitations, for instance, if magic requires the intervention of some mysterious metaphysical entities to work, those entities might decide not to allow it be 'proven' to exist.

Still, we could give it a whirl and see what we get.

ONLY NICE THINGS

09:53 / 18.10.06

Possibly of interest - there's a paper by the Cochrane Group on intercessory prayer:

Roberts L, Ahmed I, Hall S, Sargent C, Adams CE. Intercessory prayer for the alleviation of ill health: a systematic review. Forschende Komplementarmedizin 1998;5(1):82-86.

I don't have access to the text, but I think that the review group appeared to demonstrate that intercessory prayer does appear to have some effect on recovery rates. Does this prove the existence of God? Not really. But it does suggest that there might be at least a psychological or physiological effect from ritual.

12:33 / 18.10.06

NewScientist recently covered a multi-year clinical study on intercessory prayer here
which found no effects, so I'll be interested to see if I can find the text of the study you're citing there.

16:57 / 18.10.06

it is impossible to measure a particle without affecting it because whatever means you use to measure a particle must necessarily interfere with (bounce off, or interfere with the waveforme) it and hence change its properties.

Well, that's kinda true, but there's also the observer principle which has nothing to do with that; wavefunction collapse. You can't know the position and momentum of an electron for example, not because the photons bounce it off course, but because if you want it's momentum you have to measure it as a wave, and if you want it's position you have to measure it as a particle.

Heisenberg uncertainty principle; Heisenberg showed that, even in theory with a hypothetical infinitely precise instrument, no measurement could be made to arbitrary accuracy of both the position and the momentum of a physical object.

note that The uncertainty principle in quantum mechanics is sometimes erroneously explained by claiming that the measurement of position necessarily disturbs a particle's momentum

See also Le Chat du Schroedinger.

18:57 / 18.10.06

...if there's an undiscovered causal mechanism out there that magic exploits... Mirror

I think the main problem you'd encounter would be the apparent acausality of magic. The effects of ritual are often psychological and almost always manifest as coincidence or luck, the events we'd consider 'results' are explicable without recourse to magic. A scientist would likely say they would have happened anyway, without the mumbo-jumbo.
Fundamentally though you run across a problem analogous to Hume's problem of miracles i.e. that when you get enough evidence that a violation of natural law is believable, it becomes a part of our conception of natural law and thus no longer 'magic'. Additionally, it's always going to be more likely that any evidence of magic is group hallucination or explicable in some other way.
"Which is more likely – that a man rose from the dead or that this testimony is mistaken in some way?"

19:21 / 18.10.06

Well, causality is a tricky business even for standard science. But I don't think that this is a real barrier to testing the paranormal. I mean, all you really need to do is demonstrate that you can reliably get effects that aren't just down to luck - I wouldn't be terribly impressed by the magician who could influence an unbiased coin to come up heads about half the time over long runs, say. More interestingly, most people have a poor grasp of statistics and probability and so see non-randomness incorrectly (I teach statistics and do a well known parlour trick: get one group of students to toss 100 coins and write down the results, while you get another to make up 100 random results. The fakes are easy to spot, since the really random sequences contain lots more unlikely behaviour, whereas the fake sequence is far too regular.)

Now, clearly, there are filters in place that obstuct the acceptance of new ideas, but this is true for any new science as well as the paranormal. The key is to get results that aren't down to perceptive credulity - iirc, there are a few experiments that demonstrate this phenomenon of believing something is going on, when nothing much is.

21:48 / 18.10.06

Yeah, confirmational bias is often a confounding factor, looking for the results you want.
It's much easier to test for specific things like psychic powers, telekinesis and clairvoyance and so on because the results are more concrete. Can this dude levitate a hanky, is this person correct when they say the lost keys are behind the sofa etc.
I'm quite interested in instances of respectable institutions giving credence to magic powers such as police getting clairvoyants in to help with investigations and oil companies employing dowsers. There may not be watertight proof of these things but they must seem reliable enough for organisations to pay people to do them. I can't see an oil company shelling out (haha) for something that isn't reliable enough to turn a profit.

00:47 / 19.10.06

Having worked for mining companies before, you'd be surprised at what kinds of things money will get spent on, particularly in highly profitable companies where a few tens of thousands of dollars doesn't even amount to loose change... like oil companies, for instance. Also, the people in charge of making business decisions are as likely to be ignorant of science, and capable of being taken in by quackery, as anyone.

11:38 / 19.10.06

For sure, quacks get rich every day. I'm more interested in the people who believe in what they're doing, and get endorsed by people like the police because they get results. If the dowsers do find oil, that's evidence there's something at work even though it's not proof.
Proving magic is probably easier approached backward via falsifiability, like most things. What experiments would *disprove* the power of ritual? Similarly, what would undeniable proof of magic look like?

13:40 / 19.10.06

If the dowsers do find oil, that's evidence there's something at work even though it's not proof.

At the risk of being pedantic, I'm going to correct you here, since what you've precisely said isn't quite right. In order for there to be evidence that there is something at work, the dowsers would need to be finding oil at a rate which was significantly better than chance ("significant" here in some loose but technical sense involving statistics). The only reason I'm flagging this is because the distance between what you said, and what I said is one that I *think* sometimes separates sceptics from believers.

17:16 / 19.10.06

The problem is sample size really (see links at the bottom, non-statisticians). Even if we set a low significance of five or ten per cent, we still have loads of confounding factors and have to do a double blind comparison at least twenty times and preferably a whole load more to get a reliable significance, not easy in most circumstances. And even then there are more likely explanations for the results, it would just show that something was happening, not necessarily what the participants *say* is happening. I mean to say, you are right.
F'rexample, if our hypothetical oil dowsers get it right half the time, and our sample of random fake dowsers don't get it right more than one in a hundred times, would that constitute proof? No, of course not, it's still more likely the dowsers were sent to areas more likely to have oil, or they had automated landscape recognition of oil-indicating scenery and subconsciously twitched the rods, or that the rods the professionals were using were slightly magnetised and are influenced by the subterranean iron deposits hitherto unconnected to oil etc. etc.
The explanation offered by the practitioners (magic) is less likely than hypothesis X (other factors we are unaware of that don't violate current scientific understanding).

Statistical significance article, particularly note the signal to noise ratio.

17:54 / 19.10.06

One thing I was hoping to discuss in this thread, perhaps with some input from the Temple side, is the potential gains to science and society that could be had if magic were to be shown effective and a mechanism determined (the "is it worthwhile" part of the summary)?

An important part of this would be to catalog what sorts of effects magic can produce. What should we be looking at? According to those who currently practice, what are the limitations? If there are none, what would be a guess at the probability of being able to cause any given event to occur?