LESSWRONG
LW

Comment Permalink

Beliefs should pay rent. Explanations should cause anticipated experiences -- that is, make predictions.

An explanation is only as good as its predictive power.

Yes, explanations are associated with predictions and it is often a bad sign when an explanation does not recommend a prediction. But no, an explanation is definitely not only as good as its predictive power.

What we're (or at lease I'm) talking about is referred to in the literature as Hempel's symmetry thesis: that every adequate explanation is potentially predictive and every adequate prediction is potentially explanatory. It may well be the case that every explanation is could have been a prediction but the second part, that every adequate prediction co... (read more)

See in context

14 Examples of mysteries explained away

by RolfAndreassen

30th Sep 2011

1 min read

14

I am looking for examples of mysterious answers that were eventually explained *away* by science. I can think of two: One is the belief that the behaviour of living things was explained by the mysterious force of elan vital, and not by mere chemistry; which was destroyed by the synthetisation of urea. The other is the special (and mysterious) role of the conscious observer in quantum mechanics, which was explained away by demonstrating that rocks can get entangled with electrons just as much as brains can. Can anyone furnish me with other examples?

I observe in passing that phlogiston is *not* such a mysterious answer. Eliezer is down on it, but I think unjustly so; for people did in fact perform experiments on phlogiston, including the final experiment to find the weight of the phlogiston that had passed out of the burning material and into the byproducts. It turned out that the phlogiston had negative mass... in other words, that the direction of the transfer had been misidentified. But if you think of phlogiston as `negative oxygen', it makes the same predictions as modern chemical theory. This is no worse a mistake than mistaking the direction of the current, a mistake which is *still* enshrined in our sign conventions; it is not a mysterious answer of the form "X->Y" with no details of X given and any value allowed for Y.

However, I digress. Mysterious answers blown away by experiments, anyone?

Personal Blog

14

New Comment

73 comments, sorted by

top scoring

Click to highlight new comments since: Today at 2:42 AM

Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

[-]Jack14y130

No significant fraction of physicists ever believed consciousness had a special role in quantum mechanics.

7Paul Crowley14y

This sounds plausible, but [citation needed].

3lessdazed14y

What does the picture look like if we (appropriately) think of belief in terms of subjective probabilities? In other words, at the height of the consciousness folly, how many would have offered 999 to 1 odds against it? 99 to 1? 9 to 1?

5Jack14y

It's a good question, I have no idea. What is disturbing is that the few who endorsed the idea weren't crackpots but very, very smart people. Most notably.

0Shmi14y

I suspect that no theorist in the field, except maybe for von Neumann and Penrose, would bet any significant amount of money at any odds you quote on consciousness being shown to be a necessary part of entanglement. Many would, of course, be happy to quote you some "reasonable" odds, as long it was your money at stake and not theirs.

2NancyLebovitz14y

It's the other way around for Penrose-- he thinks that quantum effects are necessary for consciousness.

0lessdazed14y

Renormalize to take account of the diminishing returns of money contributing to utility and rephrase the bet in terms of utility, whatever the dollar amounts. Scary examples nonetheless, no?

[-]Jack14y110

This treats mysterious answers as categorically distinct from scientific answers which I think is wrong. Some answers explain less than other answers and the less an answer explains the more mysterious it is. Science replaces answers that explain less with answers that explain more. That is, it replaces mysterious answers with less mysterious answers. Phlogiston is less mysterious than elan vital but more mysterious than modern chemistry. The Standard Model of quantum mechanics explains far, far more than the pantheon of Greek gods- but it is not categorically different.

Adding: I upvoted. It is still a good discussion topic, it's just that answers should be given the way scminux did.

0khafra14y

Did you mean "the less an answer explains, the less mysterious it is"? The hypothesis "the Greek Gods did it" can explain anything; the Standard Model only explains things which follow the rules of quantum mechanics.

1Jack14y

By 'explain more' I don't mean "explain a greater range of possible phenomena" I mean something like "explain better and in more detail various aspects and relevant causal variables involved in that phenomena"

3Shmi14y

There is a simple metric: a less mysterious answer has better predictive power.

3Jack14y

It's too simple by far. Being able to predict something is not the same as having causal knowledge.

0Shmi14y

Than please give an example of an answer that is less mysterious but has no better predictive power.

5Jack14y

There are trivially false explanations with tons of predictive power: "The storm happened because my barometer was at low". (Edited for clarity)

0Shmi14y

And it is a less mysterious answer than what? "Because was pissed"? Then yeah, your answer is less mysterious.

3Jack14y

I'm not saying predictive power isn't a good indicator of a less mysterious answer. I'm saying prediction isn't what explaining is really about, to begin with.

1Logos0114y

Beliefs should pay rent. Explanations should cause anticipated experiences -- that is, make predictions. An explanation is only as good as its predictive power.

1Jack14y

Yes, explanations are associated with predictions and it is often a bad sign when an explanation does not recommend a prediction. But no, an explanation is definitely not only as good as its predictive power. What we're (or at lease I'm) talking about is referred to in the literature as Hempel's symmetry thesis: that every adequate explanation is potentially predictive and every adequate prediction is potentially explanatory. It may well be the case that every explanation is could have been a prediction but the second part, that every adequate prediction could be a successful explanation is certainly false. I don't think any more than the above barometer example is needed to show this. But just in case: here is another. Say you see the shadow of a flag poll on the ground. You know what time it is and thus the location of the sun. Therefore, you can predict the height of the flag poll. But the facts about the shadow and the sun do not explain the height of the flag poll. Explanations just don't have the same logical form as a predictions. This is part of the reason why Hempel's once dominant theory of explanation is no longer accepted. Scriven is the figure most associated with the criticisms. This isn't (necessarily) anything to freak out about. I'm not arguing for any wishy washy nihilism about explanations. There are probably some who do, but personally I think the best going theory of explanation is the causal theory of explanation, particularly one involving a manipulationist theory of causation like that of Judea Pearl or James Woodward (I'm dropping names so you can google if you want). Under this theory as explanation of X tells you what sorts of things you would have had to manipulate to change X. Experimentation replaces prediction as the central activity of science. This does not mean we stop predicting or that predictions are no longer important. I do not at all disagree that beliefs should pay rent. I am simply not comfortable saying that the metric b

-4Logos0114y

I never once even hinted at a claim that they did. I said that an explanation is "only as good as its predictive power." I never once mentioned anything about there being symmetry between explanations and predictions -- my statements were entirely unilateral. Explannations that lack predictive power are not useful. I could explain that the Gods of Ysgard cause storms by going bowling in the clouds after getting drunk. You can't make any useful predictions from this but it's a perfectly simple explanation, far simpler and comprehensive than any actually useful explanation of thunderstorms. We throw it out precisely because it is so useless. Sir, experimentation without prediction is impossible. Experimentation is meant to falsify predictions in order to validate hypothesis into theoretical models. Then you shouldn't argue against it. Explanations are not predictions, this is true, but nobody was claiming that they were except you. Regardless, no explanation is any more valuable than its predictive power.

-1Shmi14y

And that is where we disagree. To me, without a better predictive power explaining is just a feel-good exercise and has little to do with science.

0Jack14y

You think causality is just a feel-good exercise and has little to do with science?

0Shmi14y

Please feel free to be more specific. Causality as a consequence of Newtonian mechanics or Special or General Relativity is not an explanation in itself, but just that a consequence. It is also a (mostly valid) observation. On the second thought, this is turning to be a dialog, which is more appropriate for a chat room, than for a public forum, so I will stop here.

4lessdazed14y

If this string of comments is much upvoted, then it would certainly be appropriate for here, despite there being only two people in the discussion.

1rysade14y

I have to say this discussion has me intrigued. Feel free to post the results of the discussion here. I am interested in hearing how it all turns out.

1Jack14y

We seem to have fans so I'll trying to go into it later today. But you and anyone else interested in scientific explanation should start by reading or at least browsing the SEP article.

[-]Shmi14y80

The constant free-fall acceleration used to be "explained" by the inertial mass being equal to the gravitational mass, which was, of course, a mystery in itself. It took General Relativity to show that only one type of mass is needed. Specifically, gravity is not really a force (but a spacetime curvature), and the Newton's second law Fgrav=ma for gravity is only an approximation, valid for not very heavy and not very fast objects.

"Because GR" is a less mysterious answer (thanks, Jack) than the previous model (Newton's universal gravita... (read more)

[-]DavidPlumpton14y40

Spontaneous Human Combustion. Somebody living alone gets drunk/has stroke/heart attack spills alcohol/perfume on themselves and a cigarette ignites a fire. The body slumps onto carpet and the body fat together with clothing and carpet form a candle wick effect and a small high temperature fire burns for some hours. Parts of the body with low fat levels (e.g. lower legs) often remain unburned. It's a simple experiment to do with a pig carcass.

1wedrifid14y

Ignition via an accellerant and an existing fire seems to make a lie of the 'Spontaneous' part of the name! Which isn't to say "Human Candles' don't sound kind of awesome in their own right.

1Luke_A_Somers14y

The mystery was its spontaneity. Once you realize that all the cases had an accelerant and ignition source, it's no longer mysterious.

[-]Manfred14y40

Oxygen != heat, or even fire. Fluorine can set things on fire, for example. And of course the wonderful thing about energy is that it can change forms. So phlogiston theory is thoroughly useless, though I agree it's not a really mysterious answer.

"Emergence" or "quantum effects" to explain how the brain works might fall into this category.

5Logos0114y

Depends on what is meant by 'emergence'. If you mean to say that 'emergence' itself is "mysterious" then I would disagree. Chemistry is 'emergent' from physics -- molecules possess attributes that their constituent atoms do not -- but no one would describe this as "mysterious". It's simply a question of scale. Consciousness described as being 'emergent' from our neuroanatomy is hardly a 'mysterious' statement: it's simply a claim of "scale". One would not examine the exact behaviors of every atom in a molecule to predict its characteristics: no matter how much oxygen or hydrogen you have at room temperature the characteristic of "wet" would never be noted. Only dihydrogen oxide at room temperature (and in sufficiently significant quantities) has that characteristic. So too, potentially, is it with consciousness: while yes it is necessary to understand the workings of the constituent parts it is only by observing how they interact as a whole that consciousness can be comprehended -- in much the same way we would not discuss the phenotype of E.coli in terms of their spacetime tensor coordinates. But I digress.

3DanielLC14y

I've never heard of phlogiston being heat. True, but the vast majority of fire people normally deal with is oxygen-related. Only knowing about oxygen being an oxidizer would be almost as useful as knowing about all of them, at least when you're thinking about fire. Only knowing about phlogiston would be exactly as useful.

2Manfred14y

Whoops! I seem to have gotten my phlogiston theory mixed up with my caloric theory, which stated that fire was in fact made of particles of caloric, a near-weightless fluid that flowed between bodies in contact, explaining heat and cold. Sheesh, alchemists loved conjecturing fluids.

0DanielLC14y

I think you could call that a fluid without stretching the definition too much. Still, heat is an excellent example of something being explained away. They thought it was its own substance, like static electricity, but it turned out to be something that can easily be predicted by Newton's laws.

3AlexSchell14y

Surely you mean "easily in hindsight"?

0rysade14y

I agree that phlogiston was not likely thought of as a mysterious answer at the time. I think that what justifies calling it a mysterious answer today is that we could justifiably notice that we are confused. Whether it's confusing quality is a good reason to categorize it as a mysterious answer is a different issue, however.

[-]RolfAndreassen14y30

I might mention the context of my question: On a different website I got into a debate with someone who thinks that the mind and rationality are mysterious, non-physical things, about which science can never say anything useful. I wanted to give the historical examples of such views being demolished, and found that I couldn't call to mind more than two or three. My hope was that, for future iterations of that debate, I could refer back to a long list of cases; but it seems to be hard to come up with really strong examples of subjects being declared beyond ... (read more)

[-]Morendil14y30

The germ theory of disease cleared away a number of mysterious answers, like humorism, of which terms like "phlegmatic" are a vestige.

[-]lessdazed14y30

rocks can get entangled with electrons just as much as brains can.

Rocks are conscious?!

[-]pedanterrific14y190

Kinda puts a new spin on "I had a pet rock once. It died.", huh?

Mental image: Petunia looking on, half in concern and half in amusement, as five-year-old HJPEV works determinedly to bury a pebble in the backyard. "I know there's no such thing as 'Rock Heaven', Mom, I'm not stupid!"

7lessdazed14y

Ba-dum PAH!

[-]pedanterrific14y130

Say, rather, 'kinda colors my impression of...' or maybe 'kinda gives a different flavor to...'

1Solvent14y

I love this website.

8Manfred14y

Getting entangled with electrons seems a pretty silly definition of consciousness :P

[-]MichaelHoward14y30

God?

1lessdazed14y

Substantially but not entirely. If it's important to you that there is historical evidence for a belief in a god, such as a tradition related to a revelation, then the specific versions posited in the past have have been explained away, and that undermines any present reason to believe. If philosophical arguments about what must be, such as the cosmological argument, are more important to you, then less so because the god believed in is constructed to not be falsified by present known things - but that constructed thing isn't necessarily unfalsifiable.

0[anonymous]14y

God[s] used to be responsible for lots of mysterious lightning bolts, earthquakes, plagues, meteors, biogenesis, schizophrenic episodes, etc. These mysteries were blown away by understanding what they are and what actually causes them, but God itself isn't a mysterious phenomenon - it's nothing - so it can't be explained away by understanding what it actually is.

3MichaelHoward14y

I didn't call God a mysterious phenomenon. The post asked for examples of mysterious answers.

[-][anonymous]14y30

Is Spontaneous Generation an example of what you are looking for, or an example of something you aren't looking for?

On the first read, it appears to be a mysterious answer, because it doesn't seem to explain anything, but people also performed experiments on it well before it was finally rebutted, which would seem to put it into the phlogiston category.

[-]MichaelHoward14y20

Relevant sequence posts to save others from searching...

Mysterious Answers to Mysterious Questions

Explaining vs. Explaining Away

[-]Logos0114y20

The other is the special (and mysterious) role of the conscious observer in quantum mechanics, which was explained away by demonstrating that rocks can get entangled with electrons just as much as brains can.

I'm curious as to why, given that the Copenhagen Interpretation allowed for this to occur, this demonstration would 'explain away' the CI. The exact mechanism of that reduction is what I'm after: how does that demonstration 'destroy' the mystery which allowed for that demonstration to occur without invalidating 'the mysterious model'?

I find myself confused, is what I'm saying.

0Jack14y

Can you rephrase?

0Logos0114y

Well, the CI allowed for all forms of entanglement. So I'm curious as to how experimental results that were consistent with the model somehow "explained away" said model. That doesn't seem to be how the science I am familiar with works. EDIT: (Side note: I do not have a horse in this race; to me both the MWI and the CI are equally fallacious. My personal belief is that no interpretations are necessary at all.)

-1DanielLC14y

Experiments don't explain stuff away. Simpler models do. Also, they have to be consistent with previous observation. Otherwise it's not explaining why it happens; it's just showing that it was wrong in the first place. Explaining away is when you give a simpler model that has the same results as the more complex ones. You've just explained away the complexities. I think you don't understand the question with this. Do alternate universes exist or not? Does the past exist. Do different places exist? They're all related to the calculations in the same basic way.

1Logos0114y

There's a simpler model available for discourse here? Well, fascinating. Please explain. This, by the way, is a very dangerous mentality. Occam's Razor is only a heuristic. New evidence may favor greater complexity over less. General Relativity is more complicated than Newtonian gravitation, for example. It would seem not. I never saw any connection between entanglement with rocks and alternative universes. Please explain.

4DanielLC14y

Simpler than what? Newton's laws are simpler than Newton's laws plus elan vital, and once you understand enough, you realize they make much the same predictions. Ergo, you just explained away elan vital. General Relativity is simpler than Newtonian gravity plus time dilation. Once you get General Relativity, you've explained away time dilation. No. Evidence shows that the simpler models that don't explain this are outright false. The simplest one remaining is the most likely. Newtonian gravitation is simpler than General Relativity, but it does not match what we see. Postulating Newtonian gravitation with time dilation added in makes it more complex. Adding in the non-euclidean geometry stuff makes it still more complex. At this point, General Relativity is the simplest explanation we can think of. "Entanglement" means that each combination of states exists separately. If every atom of a rock is entangled, there is a separate amplitude for every configuration of the atoms. If every particle in the universe is entangled, this means that there is an amplitude for each combination of particles. For example, there's the configuration corresponding to the "universe" we see. There's also one that looks exactly like how it would have ended up if Germany won WWII. Technically, a lot more than one. Infinitely many, in fact.

-3Logos0114y

... sir, you have a very... 'slippery' manner of discourse. It makes it difficult to figure out exactly what it is you're trying to say and when. You suggested that there was a simpler model than either CI or MWI. So what is it? Yes but that's not the topic at this point. In other subthreads I'm sure elan vital may be relevant, but not this subthread. What in the hell? Excuse me, but... I'm not aware of any theories of 'time dilation' that predate General/Special Relativity. Your methods of 'explanation' leave me mightily confused, sir. You're missing the point -- that being that a simpler model that doesn't explain for the available evidence is wrong -- and yet your previous comment suggests otherwise. Huh? This isn't even grokking for me. ... that isn't even remotely possible under the known mechanisms of quantum entanglement, sir. Entanglement between two pairs only lasts until some 'measuring' event occurs upon one of the pairs. Actually, after reading: I can only say that... you have a very bad understanding of quantum mechanics if you think this is intelligible under the topic at hand. Quantum. Mechanics. Does. Not. Work. This. Way. I suggest, sir, that before you attempt to explain this any further you get a deeper understanding of the phenomena at hand.

1DanielLC14y

I seem to have lost track of the conversation. I'm sorry. MWI is a simpler model than CI. It does not have wave-form collapse. It has decoherence, which is functionally similar, but is an emergent phenomena of the actual laws of physics, rather than a law in of itself. Looking at your second comment: They didn't have experiments supporting it yet. Given what we knew then, Newtonian physics may have been the simplest explanation. Now, it of itself, is not even an explanation, in that it doesn't match the experiment. If we added a law to force it, it would no longer be simpler than general relativity. According to CI. There is no mathematical formalism as to what "measuring" is. They say it's when the system gets entangled with something "macroscopic". There has never been an experiment that showed wave-form collapse that couldn't be explained by decoherence, which arises from the same laws if you don't postulate wave-form collapse. CI doesn't. MWI does. It's possible that there are variations known as MWI that don't include things like this, but the one Eliezer is a proponent of does. In fact, the specific variation he's a proponent of (timeless physics), postulates that the past and future are exactly the same kind of alternate universe. I'd suggest reading the (quantum physics sequence)[http://lesswrong.com/lw/r5/the_quantum_physics_sequence/]. I'm not sure if Eliezer is as good at explaining it as he hopes he is, but I doubt he's worse than me.

-3Logos0114y

Handling this first: No, it doesn't. MWI does not and cannot postulate universal entanglement. Entanglement is a phenomenon of paired particles sharing quantum states. That is all it is. You're abusing the term very severely to mean something quite radically different from what it actually means. Handling this second. I cannot accept the notion that CI is "more complex" than MWI as it has been explained thus far. CI does, yes, have the problem of wave-form collapse. But MWI has two problems of equivalent scope: 1) It fundamentally violates the Law of Conservation of Energy. 2) It requires information to be conveyable through quantum entanglement to essentially the entire universe simultaneously - and that's just so absurd I can't really take it seriously. I do not judge ideas by the names associated with them. Finally: This is not an acceptable response to the portion of the converstaion it is applicable to. It's like discussing colors of paint and responding "paint is dry once it has dried." Do you even remember what my original point here was? It was: Occam's Razor is only a heuristic. Be careful when proclaiming the simpler explanation is always right. It may not be in the face of new evidence.

3JoshuaZ14y

How does it do this? I'm not sure what you mean here. I'm guessing that you are thinking of this because you are "creating worlds" or something like this. But MWI doesn't really create worlds in any useful sense. New worlds are formed from smaller and smaller slices of the total amplitude of the wave function. I don't understand what you are getting at here.

-2Logos0114y

Took me a while to get back to you on this. I apologize for the delay. This, however (there being a 'total amplitude of the wave function) is the violation of the Conservation of Energy. For that to operate there would have to be a non-zero energy to be so divisible. And that just doesn't mesh with what I know of the origins of our universe, its current energy state, nor the remainder of physics. Students of nature have a long and troubled history of inventing new media or substances for the purpose of enabling their conceptions to be viable. Phlogiston. Luminiferous aether. And now 'the total amplitude of the wave function'. Until such time as there is a material reason to accept that concept under the standard falsificationist definitions of evidence -- I have no choice as a skeptic but to reject the notion. Because it's been a long while, I will remind you that not all rejections of MWI are created equal. My rejection of MWI is not and should never be considered an endorsement of the Copenhagen Interpretation. As I said previously, I do not believe there is any need whatsoever for any interpretation to occur. The mathematics of quantum mechanics as we have discerned them through experimental processes are elegant enough as is; there's no need to dress them up for Sunday, as it were.

1JoshuaZ14y

I'm not sure I understand this. Can you expand?

0DanielLC14y

Edit: Double post.

0Logos0114y

This post was duplicated. You might want to delete this one.

0Shmi14y

Given that MWI uses the mathematical formalism equivalent to CI, they are equivalent in the only respect that counts. For example, neither provides any testable predictions re alternate universes, except that they are not observable. Of course, the MWI model tends to make people feel good about Quantum Mechanics, but that is a purely psychological effect.

1DanielLC14y

CI has wave-form collapse. This isn't used in the mathematical formalism because there's no specific point at which it's agreed to happen. It's just assumed that it is when the system gets "macroscopic". They do show evidence of alternate universes, and they are not unobservable. For example, the double-slit experiment has two pasts. One for each slit. Also, according to timeless physics, the past and future is just alternate universes. If you believe the past exists, and you accept the basic idea, you believe that alternate universes exist.

-1Jack14y

The Copenhagen Interpretation does not say anything about consciousness collapsing the wave function.

0Logos0114y

And consciousness collapsing the wave function is entirely unrelated to entanglement in rocks.

[-]Lapsed_Lurker14y10

The Luminiferous Aether? I'm not too sure how mysterious it was, but it persisted a fair while before being squelched by the march of science.

http://en.wikipedia.org/wiki/Timeline_of_luminiferous_aether

[-]MinibearRex14y10

The obvious one nobody's mentioned: the motion of the stars and planets. Supposedly caused by Angels pushing objects in epicycles around the earth, at least until Newton came along.

3lessdazed14y

It predicted perfect circles, no? Was the problem with the theory, or with the scientists who held onto it as necessary epicycles accumulated?

1Jack14y

Thats a rather glib summary of the Copernican Revolution.

[-]Hyena14y00

This is simple: simultaneous creation blown away by evolution.

Moderation Log

14

Examples of mysteries explained *away*

14

14

Examples of mysteries explained away