Scientific bias and the void-of-course moon


i-ccbc028bf567ec6e49f3b515a2c4c149-old_pharyngula.gif

Stuart Buck persists in claiming that scientists have a bias against the supernatural, and that we dismiss it out of hand. This isn’t true; the problem is that supernatural explanations are poorly framed and typically unaddressable, so we tend to avoid them as unproductive. What one would actually find, if one took the trouble to discuss the ideas with a scientist, is that they are perfectly willing to consider peculiar possibilities if they are clearly stated. We’ll even briefly consider something as insane and worthless as astrology, which is even less credible as a field of study than Intelligent Design.

Here’s an example from years ago on Usenet, in the newsgroup sci.skeptic. An astrologer, Thomas Seers, was insisting that his weird little pseudoscience was a suitable topic for a science course. One of the skeptics, Robert Grumbine, politely asks him for specifics:

Robert Grumbine: Let us say that I teach astronomy.  Let us suppose I’ve decided to spend an hour on astrology.  What would my presentation be?  Keep in mind that this is a science class, so part of my job would be to discuss what experiments have been done that demonstrate that it works, as well as to describe how it works (in the sense of how the students could make the predictions themselves).

Thomas Seers: Hello Robert,
You appear to be asking a serious question, so I will give you an experiment to try.  This will also give you an insight to what Alchemists did years ago.
On 10/20/99 from 2 AM to 8 AM EDT, mix a bowl of jello and you will find it won’t jel. My basic students have this as a homework assigmnet to learn of a void-of-course Moon period. Silly thing, huh. It can be repeated over and over again. Don’t spill it now :-).

I’m not an astrologer (surprise!) and I really don’t know what a void-of-course moon is—I think it means the moon is in a position where it really isn’t in one of the canonical zodiac signs, but I could be wrong—but Seers was a very serious astrologer who taught courses in the subject and cast horoscopes professionally. I think astrology is a lot of nonsense, but he at least had the credentials to represent his field, and gave us a testable operational definition that we could explore.

Apparently, during a void-of-course moon, you are supposed to experience lots of bad luck and things are just generally supposed to go wrong. Seers claimed that one simple, reliable test of this phenomenon that he and many of his students had routinely evaluated empirically was that gelatin wouldn’t set. Furthermore, he had used his astrological expertise to give us the specific date and time this phenomenon could be observed.

Now, if scientists (and that unruly mob of skeptics in the newsgroup) actually had an a priori bias against any supernatural explanation, we would have rejected this idea immediately, tossed a few metaphorical brickbats at the looney, and laughed long and hard. The moon in some place in the zodiac would stop cakes from rising, Jell-O from setting? Preposterous!

That’s not what happened. Instead, a number of us took the trouble to actually try it.

Bob Grumbine’s results:


Other messages followed this original post, where I (and others)

noted how we were proposing to make the jello and asking Mr. Seers

whether this would have any effect on the outcome.  We were assured not.

Experimental Results:

10/20/99 at 2:30 AM EDT my wife mixed a batch of consumer-grade jello

according to directions.  She split the jello to two containers, one

about 1.5 cups, and one about 4, and put into the fridge.  When I checked

at 6:50 AM EDT, both had firmly jelled.

10/21/99 at 3:45 AM EDT my wife again mixed a batch of consumer-grade

jello and split in to two containers as before.  At 8:00 AM EDT, both

had jelled, though the large was somewhat un-firm.

Conclusion:

We followed Mr. Seers procedure both at a time he predicted that the

jello would fail to jell, and on a ‘control’ day.  On neither day was

there any difficulty apparent in the jelling procedure.  His prediction

is falsified.

Going Further:

Mr. Seers needs to do further work on either his prediction method,

or on his students’ ability to make jello.  For the latter, notes

from our experience:

1) It is better to define what is meant by ‘jelling’. 

a) By ‘firmly jelled’ I mean that in tapping the surface with my finger,

the finger came away dry and there was no imprint of my finger on the

surface.

b) ‘somewhat un-firm’ (on the control day) means that my finger comes

away with some jello on it, and there is an imprint of my finger in the

jello, but that imprint remains for minutes thereafter.

— It is possible that Mr. Seers’ students are labelling things as ‘not jelled’

that I would call jelled.

2) Shape of container likely matters.  Not because of the shape _per se_,

but in the degree to which it is easy to cool.  A hemispherical container,

as we used, is very good at preventing cooling (and hence jelling).

A large hemisphere with full recipe would have difficulty cooling (_did_

have trouble cooling) sufficiently in the time allowed.  A long flat

pan would be more effective, and an ice cube tray even more so.

— It is possible that Mr. Seers’ students are using hemispheres as well,

3) Cooling time matters.  We allowed 4.5 and 4 hours, respectively, on

the control and VCM day.  After placing the jello in the fridge, the

door was left closed for the rest of the period. 

— It is possible that Mr. Seers’ students a) did not allow as long a

period, and b) opened the fridge routinely in order to examine the

progress (or lack thereof) which kept the interior warmer than required

for prompt jelling.

Meta-conclusions:

The ball is back in Mr. Seers court as far as the VCM affecting the

jelling of jello is concerned.  I’ve offered some thoughts on experimental

controls, based on my experience.  His next round of experiments should

take these points in to account.

My original question still stands—what would my presentation be

for an hour presentation to an astronomy class? Keep in mind that this

is a science class, so part of my job would be to discuss what experiments

have been done that demonstrate that it works, as well as to describe

how it works (in the sense of how the students could make the predictions

themselves). 

Note, by the way, that one thing that I would be saying throughout the

length of the term is that the students should _not_ take my word on

anything.  Honest as I am, I can be wrong nevertheless.  If it is science,

there should be a way to test whatever it is I say (in principle at least,

granted some experiments get pretty expensive to run).

Wayne Hoxsie also took a shot at it.

My first attempt took 1.5 hours to gel, then I noticed that there was a

“quick” method that uses ice cubes to cool the mix faster.  I used this

for subsequent controls and for the actual experimental run.  All

gelled in about 30 minutes at 12 C.  The time/temperature plots for all

runs (except the first without ice) were essentially the same.

Some other precautions I took were to mix all the packages together

into a homogenous mixture and store it in a tupperware container and

all water used was sampled at the same time and stored (or frozen) for

use in the experiment.

I also tried the experiment, and did controls on different days when the moon was not void-of-course. I actually make gelatin and agar routinely in my work—we imbed embryonic and larval fish in it so they can’t wiggle while we’re observing them.

I did this ‘experiment’ 3 times, once on Monday, once on Wednesday, and

just this morning. Each time I made up 17% gelatin (Sigma brand) between

5:30 and 5:45 AM, and 17% and 10% gelatin between 7:00 and 7:10. In all

cases, it was made in distilled, RO-filtered water, heated in a microwave

oven. I tested for whether it would set by putting a large drop on a glass

slide, and setting that on ice, or by putting a large test tube with 10ml

of the gelatin in the refrigerator.

In every case, on every day, the gelatin set within 30 seconds of being put

on ice; within half an hour of being put in the refrigerator. The 10% gel

was slower to set and less firm than the 17%, but there was no difference

between days.

I also tried remelting and resetting gelatin from prior days on Wednesday

and this morning. No difference.

I also attempted (most charitably, I think) to come up with an explanation for why Seers and his students would have had this problem.

By the way, I wonder if this wouldn’t have anything to do with auto-defrosting refrigerators. Their temperature does cycle with a regular rhythm, and I wonder if the occasional failure to gel you’ve seen might correspond with such fluctuations in temperature.
I should also mention that the refrigerators I have at the lab do not have auto-defrost—it’s not a good idea in freezers intended for long term storage of fragile biological materials like antibodies.

Seers did not accept any of our results, of course. He invented some amazing explanations: the best, I thought, was that he disqualified Grumbine’s results because he mixed his box of jell-o in two bowls. That was “breaking up the substance”.

His results?

My students lacking the scientific expertise follow a simple path, 1

box 0f jello mixed into one bowl, put in a refrigerator to cool ( as

long as needed ), following the instructions on the box. This has been

given out over a period of 20 years to them with at least a 90% return

on the results being the same, lacking a firm jelling process.

The box I mixed did not jell firmly, more soupy.

Our next experimental date will have to be on 11/30/99, new class

begins. That way the class can do it with you, then there will be a

comparison of results. Seems none of the astrologers on line tried it.

Alas for Thomas Seers, in addition to making gelatin with scientific grade reagents and a precision scale in small aliquots (I suppose that was “breaking up the substance”, too) I’d also thrown together a box of the store-bought stuff at home, before I’d left to work, exactly as he described. It set just fine.

If anyone else is tempted to try the experiment, there is a void of course lunar calendar online—these dangerously confusing astrological episodes seem to occur frequently. If I’m reading it right, we’re going to be in a VOC moon tomorrow (Monday) from 4:47 AM GMT until 5:35 in the evening! And dear gob, but Mercury is currently retrograde! I hope no one has any big plans for the day.

This trivial and silly episode illustrates my point, though. Scientific thinking is ruled by an interest in evidence. If the Intelligent Design creationists were actually to propose something specific, there’d be biology labs all over the world where grad students and post-docs and PIs would take a quick stab at testing their claims…and if they actually panned out, they’d be jumping ship for the ID camp en masse, proposing new experiments of their own.

I know that if my jell-o hadn’t set, I would have been completely baffled. My next step would have been to get in touch with one of my chemistry colleagues, ask for his insight, and arrange a replication, preferably double-blind, to see if it still worked. And if I’d gotten consistent, repeatable results that others could also duplicate, I’d probably be proselytizing for at least some aspects of astrology right here right now. Wouldn’t that be embarrassing?

Comments

  1. says

    It’s cool that you bothered to debunk this so carefully but surely science has, and should have, some kind of bias against a priori unlikely hypotheses. Even if they aren’t consciously doing Bayesian analyses most scientists would see that deciding between the carefully built theories of people who have properly learnt the field takes priority over testing some magical nonsense pulled out of an astrologers arse and requiring extra, inexplicable physical forces.

    There are just too many possible extraordinary claims for it to be worth trying to find even ordinary evidence for or against them.

  2. TomS says

    …something as insane and worthless as astrology, which is even less credible as a field of study than Intelligent Design.

    I beg to differ.

    After all, astrology does, on occasion, make positive assertions. This “void-of-course moon” is an example. They may fail when tested, but that means that they can be tested.

    If astrology were content with saying “something is wrong with astronomy”, then it woud be in the same league with ID.

  3. says

    Actually, Matt, if you think about it, in some ways scientists would benefit from doing unlikely experiments of a certain type.

    For example, one that comes up every so often is an experiment to find out of inertial and gravitational masses might not be different.

    That’s an “unlikely” outcome, but that is exactly why it is useful to test it– if it turns out that they are not the same, it tells you that there is something there to explore.

    A nice side-effect is that we now know that they are the same to the twelth significant figure. If it’s found that they’re different on the thirteenth– now THAT would be exciting.

    It’s kind of like the way it’s going to be really cool whether the LHC finds the Higgs boson or not.

  4. EDZook says

    Well, I thought science was supposed to reject out of hand supernatural explanation…something about testability, right?
    Silly me…

  5. G.D. says

    “Now, if scientists (and that unruly mob of skeptics in the newsgroup) actually had an a priori bias against any supernatural explanation […]”

    Don’t we? Wouldn’t an explanation almost by definition be a natural explanation? Supernatural claims are generally unfalsifiable and unfalsifiable claims, being consistent with anything, are notoriously poor explanations (insofar as explaining something involves telling not only why A happened, but why A happened rather than B – whereas an unfalsifiable claim would be equally consistent with both and thus unable to provide such a story).

    Part of the problem, then, is that appealing to the supernatural usually seems to involve new “unexplained explainers” (hence just moving the need for an explanation one step back) – and if it were explained, that explanation would count as natural.

    In fact, consider some event E. Suppose that E was shown to be unexplainable by science and, for the sake of argument, provably unexplainable by science. What would be the most economical step to take in terms of theoretical adjustment? We have two options: 1) Add E to the basic axioms (“E just is, and its a basic and irreducible fact of the universe”) 2) Invoke some supernatural explanation, e.g. goddidit. Which option would be preferrable for explanatory purposes? It seems that invoking some supernatural being just posits some entity that need an explanation itself (where did it come from? why does it work the way it does?) leading to regress, or that supernatural being would itself just have to be added to the basic axioms – and in that case, we could just have added E to begin with. (Bear in mind that nothing in science so far even remotely suggests that we’ll ever encounter anything like basic unexplainable events like the E of this example, of course).

    Now this seems to provide what looks very much like an argument that rules out appeal to the supernatural a priori – showing that belief in the supernatural CANNOT be justified by observation.

  6. says

    I notice that his response did not include any sort of control… do his students ever try the same brand of Jello in the same bowl on a different day? Maybe it is, as the first fellow suggested, simply a result of an inappropriately large bowl.

  7. Jello says

    Astrology is an eternal testament to the ability of the human mind to imagine convoluted connections between phenomena that could not possibly be related.

  8. says

    Notkieran, I don’t know what physicists would typically estimate the probability of one day constructing an experiment that will show the types of mass as being different but is it were high it doesn’t seem to make the experiment less interesting than if it were low. I agree that either result would be cool but if the probability was held to be about 50-50 both results would cool and kind of surprising.

    IANA physicist but I am told there is no apparent reason why they should be equal, right? It just seems that they are?

    In any case, I think “supernatural” phenomena typically have much much lower prior probabilities than you are talking about, to the point where “It would be amazing for science if it were shown to be true” can safely be answered “Don’t worry, it won’t be”

  9. MPG says

    I also embraced, in my amateurish but enthusiastic way, an experiment in the “supernatural” when I read some god-clod claiming scientific proof of god in frozen water. The experiment required you to pour two glasses of water, one of which you froze straight from the tap/bottle, and the other you put in front of a stereo playing “devotional music” for two hours before freezing. Apparently, the water that “listens” to the hymns or whatever is supposed to freeze clearer than the water that gets frozen straight away, proving that god-songs (of any religion or denomination, apparently) purify the essence of the water or somesuch. I was intrigued, though the first flaw I spotted was the lack of a proper control – there was no glass that had been left standing in silence for two hours, and no glass that had been subjected to other kinds of sounds. Which is why I introduced precisely those conditions into my own rendition – eight glasses, two of which I froze straight away, two of which I subjected to a god-cacophany, two of which I subjected to the din of rock music and video games, and two of which were left in a quiet room.

    Sure enough, the water that was put in the freezer straight away froze with the characteristic air bubble-filled cloudiness one finds in home-made ice cubes. I was expecting the same to happen to the other glasses, but strangely enough they DID freeze a bit clearer than the first glasses. The thing was, they ALL froze a bit clearer, whether they had been subjected to hymns, a noisy evening of playing Mario Kart, or behind closed doors in a quiet room. Clearly, it wasn’t the hymns making the ice clearer, but the mere action of letting the water sit and reach room temperature for a couple of hours.

    I did a bit of research on why home-made ice gets cloudy, while store-bought ice, or ice from an ice machine, is usually quite clear. Air gets dissolved in the water, and when you freeze a substantial amount of it together, the surface freezes first and traps any dissolved air. With nowhere to go, the dissolved air forms into tiny bubbles between the ice crystals, which is why the ice appears cloudy. Commercial ice makers, however, make ice by pouring a continuous flow of water over chilled plates. The ice builds up gradually, meaning less air gets trapped and the result is a clearer cube. I looked into how to make clearer ice cubes in a freezer, and found many people had had success by boiling and cooling the water before freezing, releasing a lot of the dissolved air and reducing the amount of bubbles in the ice. I guess that just letting the water stand and reach room temperature (or a bit higher even, if it’s sitting next to a warm stereo) was sufficient to release at least a little of the dissolved air and make a visible difference to the appearance of the ice. Well, that was my totally unprofessional assessment, anyway. I’ve no idea how you could accurately measure the amount of dissolved gases in a sample of water, but I daresay there’s some more scientific method than freezing it. Still, this little experiment was exciting enough to warrant a bellowed cry of “SCIENCE! IT WORKS BITCHES!”.

  10. Oldfart says

    At the risk of being blasted by all you scientists, it seems to me that you have any number of things in physics that are simply there without cause and unexplainable except by the statement: “Because, if it was any other way we wouldn’t exist.” There are any number of “E”‘s in Physics and Math that are just accepted and never explained, probably because they cannot be explained. Pi, for instance. Or the Fine Structure Constant. You should celebrate all those things and their mystery. Someday, you or your children may be able to answer that WHY. Or maybe not.

  11. mas528 says

    I think it would take an extremely patient scientist to sort though all of the out-and-out fraudulent claims, the honest but credulous claims and the honest claims of unexplained events. Then you would have to gather evidence, not over days, but weeks, months and years.

    I don’t know about scientists, but the words “supernatural”, “paranormal”, and “extra-sensory” really bother me.

    If it were real, it would all be perfectly natural, and if it were real, there would be an explanation for it.

    Posted by: Oldfart | October 6, 2008 11:31 AM

    ” There are any number of “E”‘s in Physics and Math that are just accepted and never explained, probably because they cannot be explained. Pi, for instance.

    Umm. What is it that you don’t understand about PI? Even the egyptians understood it 6000 years ago.

  12. says

    Oldfart: Pi is really bad choice of example there. Pi is forced by it’s definition and logic to have the value… pi. You may as well say that there are only anthropic arguments for why the number 6 takes the value it does.

    As for physical constants there are attempts at non-anthropic explanations that will yield testable predictions: multiverse models with the values varying (so there is nothing special about the values we see except that we can and do live in the neighbourhood with those values) and even broadly “Darwinian” where universes produce daughter universes in black holes (and the conditions for producing more black holes those producing the same heavy elements that make life likely).

    Really it’s a puzzle rather than a mystery, but that’s all the more reason to celebrate ;).

  13. says

    Matt:

    The whole question of the inertial and gravitational masses is interesting, because it’s been found that they do appear to be the same.

    If they are the same, then there are certain (provisional) theories that should probably be trusted. If they are not, then these theories have been falsified.

    Look at it this way:

    Every time a scientist does an experiment, he is looking at two possibilties: the high-odds probability that it will go according to theoretical prediction, and the low-odds probability that it will not.

    Investigating the claim made merely inverted the odds spread, took the serious mick out of the claimant, and usefully, as a side-effect, made some tasty dessert.

    Now that, to me, says useful science, as soon as someone finds a spoon.

  14. spence-bob says

    Someday, you or your children may be able to answer that WHY. Or maybe not.

    I’d put my money on science being able to answer that question before religion can.

  15. Bad Albert says

    I know that if my jell-o hadn’t set, I would have been completely baffled. My next step would have been to get in touch with one of my chemistry colleagues, ask for his insight…

    Well, you see, that’s your problem. You always consult with some other scientist who is also biased against the supernatural. Forget all this logic and testing stuff. Life is much simpler that way.

  16. Karl Withakay says

    + 1 to Mozglubov

    If his students did not reproduce the exact same procedure on a different day/night, they cannot claim the results are different on the special day/night in question: they were assuming the control that the Jello would gel on any other day/night with the exact same procedure. Actually performing the control might have highlighted various flaws in the procedure.

  17. says

    There are any number of “E”‘s in Physics and Math that are just accepted and never explained, probably because they cannot be explained. Pi, for instance.

    Could you make your complaint a bit more specific? There’s nothing remotely weird about pi, as far as mathematicians are concerned. It’s not at all clear what you’re claiming about it.

  18. says

    Notkieren:
    I should be clear about this: I think both he jelly thing (I’m too British to call it Jell-o) and the inertial v gravitational mass are (for different reasons) worth doing.

    My point is only that if scientists NEVER say “that theory, while it explains the current data, is too far fetched to deserve us putting resources into testing it” science can’t get anywhere. Hence, there is a bias against the supernatural because supernatural explanations have a record of sucking.

  19. Pablo says

    Can someone explain something to me?

    Why the FUCK should we think the phase of the moon has any effect on whether jello gels or not?

    Is it in the horoscope or something?

    “Do not make jello today, it will not set properly.”

    One of the most important things in science is to actually do experiments to test a hypothesis. And hypotheses originate from observation, as well.

    Who in the blazes has their jello not set up properly and thinks, “Gee, maybe it has something to do with the phase of the moon?”

    Although this is a scientific test, it is a terrible example of how to do science.

  20. Sam C says

    Well, now you understand why that great visionary in biology, that pioneer of crea… sorry… intelligent design, Dr Michael Behe, told us at the Dover Trial that astrology can be considered a science.

    Of course, he could be wrong. He usually is. His next book might be called Post-Darwian Theobiology: Insights from Ignorance.

  21. tsg says

    At the risk of being blasted by all you scientists, it seems to me that you have any number of things in physics that are simply there without cause and unexplainable except by the statement: “Because, if it was any other way we wouldn’t exist.”

    Bah, strong anthropic principle. If those things (whatever those things are) were different, life would be different and they’d be saying the same thing. Or there would be no life and there wouldn’t be anyone to comment on how wrong they are.

    In short, to paraphrase Douglas Adams, you are a puddle marveling at how well the hole you are in fits you, so well that you conclude it must have been designed specifically for you.

  22. peter says

    Matt Heath @20 and others

    Isn’t the whole of general relativity based on the postulate that gravitational and inertial mass are indistinguishable? Its most famous prediction – the bending of light by gravitation – is derived very easily from this postulate (the man in the falling lift thought experiment).
    Of course, that’s not a reason for not testing the postulate time-and-time again – it sure would be fun to see a mighty house of cards collapse…..
    Peter

  23. Bouncing Bosons says

    Pablo – I think you’re a little to the side of the point here. The claim is not that saying the moon effects jello is science, but rather that scientists are willing and able to use the methods of science to investigate “supernatural” claims to see if there is any substance to them.

    It’s not supposed to be an example of “how to do science,” it’s an example of how Science is willing to listen to claims that are considered unscientific so long as they make specific, testable claims.

  24. says

    Peter:
    As I said, I’m not a physicist (I’m a mathematician) but I think it goes as follows. GR did indeed take a postulate the equivalence of gravitational and inertial mass, which was fair enough since experiments showed no difference (much like how special relativity takes Michelson and Morley’s negative result as a postulate).

    Now, as I understand it (and you should get this properly fact-checked) particle physics models don’t have this equivalence as a “forced move” – i.e. there is no obvious reason why things wouldn’t run fine with different values.

    They, are of course NEARLY the same so GR wouldn’t collapse catastrophically if they weren’t quite the same, it would just be a good approximation (which I understand is all it is currently held to be, anyway)

  25. says

    Thanks for bringing back the memories Paul. Fun experiment, as we got to eat the results.

    It happened that I was teaching astronomy at the time, so I probably used it as an example of science (our respective tests) vs non-science in class.

    Matt: In Newtonian physics there’s no reason for inertial and gravitational masses to be the same. In General Relativity, there is. Or, rather, the General Relativity Principle is that the two must be the same. Finding a discrepancy between the two would cast some serious question on GR. One sort of experiment for the equivalence is, iirc, the Eotvos (umlauts over the o’s) experiment. It and relatives have been repeated for a long time now, with increasing sensitivity, and never found a discrepancy.

    General: I have to second Paul’s original point — it isn’t that there’s an objection to testing things like ‘void of course moons’ or other supernatural things, it is that they’re so seldom stated clearly enough to have anything that could be tested in the first place. Beyond ‘is it testable’, the question is ‘would I learn anything interesting by doing this test’. Again, not ‘is something supernatural involved’.

  26. Peter Ashby says

    The thought process for the astrologers is the same as UFO nuts use. They see some lights in the sky and jump from there to ‘there is an alien spacecraft in the sky’. Which is how come UFO now means ‘alien spacecraft’ where it never used to. The relgious do it too of course, you found a parking place (who would have thought a parking lot would have a parking space in it?) so that means you DO have a personal angel.

    IT’s prescientific thinking, the cure is a good science education though it is not efficacious in every case.

  27. says

    Some attention should be given to jelling and not-jelling when there is a NOT-void-of-course moon. If the astrology is right, shouldn’t there be NO (or at least a whole lot fewer)failures to jell during such times? Anyone try that?

  28. Stuart Buck says

    Is there any particular reason you’re re-posting something you wrote 4.5 years ago?

  29. SteveM says

    (much like how special relativity takes Michelson and Morley’s negative result as a postulate).

    Actually, as I recall, Einstein was ignorant of the M-M null result when he formulated his theory. The constancy of the speed of light in all intertial frames he derived pretty much directly from Maxwell’s equations.

  30. abb3w says

    I believe the problem Mr. Buck has is due to a sloppiness in the demarcation (definition) of “supernatural”. In the interest of possibly helping to clarify the demarcation of “natural” from “supernatural” by coining an intermediate term “extranatural”.

    For this, note I intend the word “infer” in the context of using rules constituting an Chomsky type-0 unrestricted grammar, meaning the resulting inferences are Recursively Enumerable, or “turing computable”. I would demarcate “natural” as pertaining to evidence which we have been able to infer relationships within; “extranatural” as pertaining to evidence which we have not yet been able to infer relationships but where the potential is within the RE domain; and limit “supernatural” to evidence where inference is simply not computably possible. (One might also redefine “unnatural” to formally describe things which are inherently paradoxical, such as a catalog of all catalogs which do not list themselves. However, the confusion from jargon use of an existing term would seem to overcome the benefit of parallel construction.) I would consider the definitive answer of the extranatural to be radioactivity at its first discovery: outside the known natural, but not utterly unknowable.

    So, to further illuminate and in an effort to piss off a billion or so folks who PZ would seem to have missed so far, let us suppose Ganesh (in the interest of compassion) deigned to wander up to PZ’s office to further by means of a visitation a poor atheist’s path toward enlightenment. Presumably PZ, after looking closely for the rubber mask line, witnessing a few minor miracles (say, by producing a window to remove the obstacle to the view), and checking to make sure his post-doc didn’t put LSD in the morning coffee, would eventually agree that Ganesh is real. However… Ganesh has some properties that are known and knowable; that is, his appearance and his association with removal of obstacles. His aspect is within the scope of cause-and-effect: that is, “if large obstacles vanish abruptly in apparent violation of laws of thermodynamics while in the vicinity of an elephant-headed dude, Ganesh is probably the cause”. This is similar to the basis that Richard Feynman’s father explained inertia to him: we know it’s there, and what it does, if not why. By the act of introducing himself, Ganesh would begin the transition from the “extranatural” to the natural (probably irritating string theorists, who doubtless would deserve it)… but by being vaguely conceivable as a semi-comprehensible entity, he cannot be completely supernatural.

    For something to be supernatural, its properties would have to preclude using it as a basis for any prediction. A voice from a burning bush saying “Go bug Pharaoh, ya shmuck” and then demonstrating the ability to inflict and remove Leprosy does not make the cut. Imagine sacrifices to a deity so unpredictable that they might result in smiting, being blessed with an abundance of cream pies, smiting with cream pies, or who knows what. Furthermore, the response would have to be something more subtle than that, or else it would readily be characterized as “when you do such-and-such, Loki gets mischievous”… and thus, allowing some potential of inference. Considering such an encounter as isolated anomaly in the usual computable background would be just as valid a basis for inference. We also do not seem to have many such “Miracle” anomalies in the data of science to worry about.

    Thus, to clarify the point for Mr. Buck: the natural is the clear current domain of science, the extranatural is within the potential scope of science, the supernatural is pointless to distinguish from the scope of science, and the paradoxical (unnatural) is just plain silly.

  31. steve_h says

    the problem is that supernatural explanations are poorly framed

    At last! this sounds like a job for …

    (Fanfare)

    (Loses interest)

  32. says

    Mas528 @ #13:

    I don’t know about scientists, but the words “supernatural”, “paranormal”, and “extra-sensory” really bother me.

    If it were real, it would all be perfectly natural, and if it were real, there would be an explanation for it.

    Agreed. If, for example, telepathy existed, it wouldn’t be “extra-sensory”, it would simply be another sense, perhaps one that only a small portion of the population has, but what of it? The five “normal” senses are also not present in every person, although admittedly their lack is rather less common than non-telepathy would be*. However, if you imagine a hypothetical society that was 99% deaf, the 1% who were hearing would seem to have “uncanny” abilities, such as being able to know that a person was approaching before they came into sight, from the sound of their footsteps. Scientists in this society could investigate these odd abilities. They could test the hearing, making sure that there was no way they could possibly detect the other person by the “normal” senses, e.g., no way they could see them, no possibility of hypersensitive smell (perhaps through testing for extra-sensitive smell). Once they confirmed that their “extra-sensory perception” was real, they could then begin to investigate how it worked. They would, for example, find that it worked best when the person walked on certain types of flooring. Or when the person coughed, the “psychic” would pick that up. They would begin to accumulate data about the types of situations where that sense worked. They would understand the concept of waves from observing waves in water, and pressure waves in solid ground, and someone might hypothesize that these individuals were somehow able to detect vibrations. After ruling out ground vibrations (while footsteps could presumably be detected that way, coughs could not. Also, being placed off the ground would not diminish their extra sense), they might propose that it was vibrations in the air, remarkable though that concept would seem. They could test this idea by placing different types of barriers between the sound source and the experimental subject, and find that certain types of barriers blocked the perception better (sound absorption).

    Finally, they could perform autopsies on deceased hearing individuals, and notice differences in their inner ear structure, and, by examining them, and performing computer simulations, work out how these vibrations were perceived.

    *Incidentally, it seems to me that telepathy, if it existed, would be a strong argument against Darwinian evolution. All the other senses are functional without any intentional production of the phenomena they detect. One does not need the ability to make sound for hearing to be useful, even if it has since been co-opted for communicative purposes by various animals (including, of course, humans). However, what purpose would the ability to detect telepathic emissions have without a source of said transmissions? And what purpose would a transmitter be without a receiver? It would seem to be a genuine case of “irreducible complexity”, unless one could find a valid passive function for telepathy.

  33. Pikemann Urge says

    Hm, I don’t know. I’ve never had a negative experience with astrology. Mind you I never go looking for readings or anything, I just know people who are into it. They seem pretty much on the ball.

    Still, I don’t care to convince anyone of it – my experiences are not proofs of anything outside of themselves. In any case, decisions you make in life have to be yours, whether it’s your intuition or simple preferences or whatever that are guiding you.

  34. says

    Matt:

    Ah, I see what you mean. I think we pretty much agree on _what_ science should do, although not neccessarily why.

    I think that when someone says “If I am correct, then these following things should happen”, and it is not much of a cost of resources to go ahead and show that he’s wrong, then it is definitely worth our time to show that he’s wrong, because it adds to our arsenal and toolkit.

    From that point on, whenever he brings out the same argument again, we can just point him to the relevant experiment and say “Oh, you mean this one? We did it years go. Do try to keep up, dear boy.”

  35. says

    Oh, forgot to add:

    I’ll have to refer to Bouncing Bosons’ comment above: the reason why it’s not worth the time to test woo claims is that normally the woo claims are too vague and by nature unfalsifiable– all and any results are taken as proof of the theory.

    In this case, however, the predicted results were stated quite clearly.

  36. Samantha Vimes says

    I wonder if he has his students put pineapple in the Jello during the void-of-course moon. Enzemes, not astrology.

  37. Dunc says

    Given the vast numbers of people preparing vast amounts of gelatine or agar every single day, I’m pretty damn sure that if it didn’t set on certain days, people would have noticed. For one thing, there would be dessert factories ditching entire production runs, and that’s the sort of thing you notice pretty quickly.

  38. Ian says

    Takin’ on the jellies! You’ve got serious thrill issues, dude. Awesome. Next you’ll be looking for the East Australian Current!

  39. SteveM says

    Incidentally, it seems to me that telepathy, if it existed, would be a strong argument against Darwinian evolution. All the other senses are functional without any intentional production of the phenomena they detect. One does not need the ability to make sound for hearing to be useful, even if it has since been co-opted for communicative purposes by various animals (including, of course, humans). However, what purpose would the ability to detect telepathic emissions have without a source of said transmissions? And what purpose would a transmitter be without a receiver? It would seem to be a genuine case of “irreducible complexity”, unless one could find a valid passive function for telepathy.

    Purely for the sake of argument, it is conceivable that a single organ could be both transmitter and receiver. Much like a tuning fork can generate a tone, but will also vibrate in response to another tuning fork vibrating nearby.

  40. says

    Hmm … that’s a good point. And, I suppose I must acknowledge that it’s conceivable that an organ that evolved for one function could, incidentally, radiate a certain kind of “telepathic energy”. The ability to detect said energy could be useful for finding others of one species, particularly of the opposite sex, if the energy differed between males and females. Once that existed, then a feedback cycle could develop that would steadily enhance both control over radiation and ability to detect it.

  41. djd says

    Once that existed, then a feedback cycle could develop that would steadily enhance both control over radiation and ability to detect it.

    And there is an existing parallel in nature: the generation of nerve impulses and the contraction of (most) muscles create electric field impulses. This is a byproduct of their mechanism of operation (unequal charge exchange across cell membranes). However, some organisms have evolved the ability to use this phenomonon.

    Various aquatic species can detect electric fields (with sharks being a particularly well known example). Electric fish have musclelike organs (electroplaques) which can produce a powerful electric pulse capable of stunning other fishes.

  42. Sili says

    Well — I can easily see myself fscking up the instructions on the pack if I had to make anything at that ungodly hour. Perhaps the poor fella’s ‘student’s were the same.