Common Misconceptions within Physics and Astronomy

[Charlie!]

It can't- no one knows how to get a Born rule out of the theory. WIthout getting the Born rule, you either can't make predictions or you make wrong predictions.

The same is true for all interpretations. The square of the amplitude is just put in. Well... there actually IS an idea how it could be the result of deeper theory.

Looks like that just hides the extra premises under some jibber-jabber (as Mr T would say) about anthropics. It's still putting in rules by hand.

if you measure something symmetrical with amplitudes 1/sqrt(2), you know that you should assign both measurements probability 1/2 because you don't have any evidence that could point one way or the other. Then if you postulate the existence of many independent such qubits, you know that the amplitudes multiply and the probabilities multiply, so you can show that for all amplitudes that are a binary fraction, the Born rule holds. I'm not sure how to do the same thing for non binary-fractions though. This argument does sneak in some extra information, but only a little, and possibly the minimum.

[SU3SU2U1]

The same is true for all interpretations. The square of the amplitude is just put in.

I don't think you understand your own interpretation- many worlds does NOT have a collapse postulate, and hence doesn't have any idea of probability. Everett's original paper was an attempt to derive Born probabilities from unitary evolution, but unfortunately its lacking. More recent work by Deutsch, Wallace, etc attempt to derive the Born amplitudes from game-theory type considerations, but that approach hasn't yet proved successful (its not obvious to me that its even doable in principle).

The problem with many world's is that no one knows how to use the many world's formalism to make any prediction. Even physicists who believe many worlds is the correct interpretation can't use it to make calculations. They always back in to something like many-minds because they aren't careful about their formalism. Many minds is a god-awful interpretation, but at least it gives results.

No one even knows what "probability" even MEANS in the context of many worlds.

As long as you define by hand what counts as "measurement" for each experiment.

Right, but you can actually do that. If you sit down with the Von Neumann postulates and an experiment you can use the formalism to calculate a result and get an answer. Experiments can be done to test that answer.

However, if you sit down the many worlds postulates you cannot get an answer unless your experiment involves an infinite ensemble (which is impossible). This is the only interpretation with this problem- consistent histories gives an answer, transactional theories give an answer, Bohm gives an answer.

Remember- interpretations CHANGE THE POSTULATES. There is so much confusion over this that I feel like I should write a book.

[mfb]

It can't- no one knows how to get a Born rule out of the theory. WIthout getting the Born rule, you either can't make predictions or you make wrong predictions.

The same is true for all interpretations. The square of the amplitude is just put in. Well... there actually IS an idea how it could be the result of deeper theory.

Looks like that just hides the extra premises under some jibber-jabber (as Mr T would say) about anthropics. It's still putting in rules by hand.

Please do not remove the context of statements you quote.

But it is too early to tell whether that could work.

I don't think you understand your own interpretation- many worlds does NOT have a collapse postulate, and hence doesn't have any idea of probability.

My statement referred to the Copenhagen interpretation. Which I assumed to be clear, as it is directly under the quote of your statement about predictions from this interpretation.

You can still interpret probability as a statement about past events. For future events, it is a bit more complicated, that is right - roll a dice once, and all sides will show up for some versions of "you".
But using the amplitude-squared of the different worlds as probability to "get/be into this world" still gives consistent results.


As a side-remark:

if you measure something symmetrical with amplitudes 1/sqrt(2), you know that you should assign both measurements probability 1/2 because you don't have any evidence that could point one way or the other. Then if you postulate the existence of many independent such qubits, you know that the amplitudes multiply and the probabilities multiply, so you can show that for all amplitudes that are a binary fraction, the Born rule holds.

A split into parts with equal amplitude is exactly the point where the MWI interpretation of probability is very natural as well.

[SU3SU2U1]

For future events, it is a bit more complicated, that is right - roll a dice once, and all sides will show up for some versions of "you".
But using the amplitude-squared of the different worlds as probability to "get/be into this world" still gives consistent results.

And now you've started backing into many minds, not many worlds. If you don't understand the importance of deriving the Born rule or something very close within the context of many-worlds, you don't understand the interpretation.

You can't "by hand" insert the amplitude-squared rule without adding in a lot more structure. Go read Everett's paper, then move on to Deutsch, Zurek and Wallace.

As an exercise, write down the additional postulate you think you would need to
1. "use the square of the different worlds as probability to "get/be into this world""
2. Not be the Copenhagen/ensemble interpretation

I think you'll find its tougher than you think, but possible. However, it won't lead to the simple many-worlds interpretation.

Unfortunately, advocates of the many worlds interpretation generally don't like to point out that they still have an incomplete formalism.

[mfb]

>> Unfortunately, advocates of the many worlds interpretation generally don't like to point out that they still have an incomplete formalism.
In that case, you might like that I already said that.
Unfortunately, you think that the Copenhagen interpretation would be complete - or at least I think that you think that.

>> And now you've started backing into many minds, not many worlds.
Well, many worlds works fine without regarding minds. In fact, it just follows from applying quantum mechanics to all particles. You came up with the question of minds and probability.

[SU3SU2U1]

Unfortunately, you think that the Copenhagen interpretation would be complete

Yes, the copenhagen interpretation would be complete. Its also circular- which is a problem. You have a complete mathematical formulation, as laid out by Von Neumann.

Well, many worlds works fine without regarding minds. In fact, it just follows from applying quantum mechanics to all particles.

NO! It absolutely does not. As soon as you tried to define a probability, you invoked "probability to "get/be into this world." This statement is loaded with assumptions, but you are implicitly invoking many-minds.

Thats my whole point. Have you looked at Everett's original paper? Have you looked the work of Deutsch, Zurek, Wallace? I've suggested doing so a few times this thread.

Here are a few questions to guide discussion:
1. How are the postulates of many worlds different from the postulates of copenhagen quantum mechanics?
2. How are the postulates of many worlds different than the postulates of many minds?
3. Why is it crucial for the many worlds interpretation to derive something like the Born rule. Why is it NOT crucial for Copenhagen/Ensemble quantum interpretations?

You seem to have this high level idea of what you THINK many worlds is, but you've never actually sat down with the postulates.

[gmalivuk]

Let's not turn this thread into another argument about MWI, mkay?

[Whitebluur]

Let's not turn this thread into another argument about MWI, mkay?

Was about to say something, haha. Discussion is perfectly fine but things were getting a bit heated...

[Dopefish]

It's an interesting discussion, but not one that belongs in this particular thread. Also, it's a discussion that was largely already had on the previously linked thread, with largely the same arguments.

I think what we can take away from it as a misconception is "The common interpretation of MWI isn't technically MWI", but calling any particular one of the various QM interpretations misconceptions themselves calls for a lot more defence then is probably suitable to a thread of this nature.

[Whitebluur]

I think what we can take away from it as a misconception is "The common interpretation of MWI isn't technically MWI", but calling any particular one of the various QM interpretations misconceptions themselves calls for a lot more defence then is probably suitable to a thread of this nature.

Whenever I have an idea about something, I always assume that I am wrong... you learn a lot more that way.

[The Geoff]

When I was 6 I knew there were other colours we couldn't see, Infra Red and Ultra Violent.

Still not sure I was wrong

[gmalivuk]

Yeah, that kinda comes down to how you want to define "color".

[MHD]

I think that before we venture intro the unknown territory of discussing Quantum Mechanics, how about we actually, actively eliminate misconceptions?

http://lesswrong.com/lw/r5/the_quantum_ ... _sequence/

[Charlie!]

I think that before we venture into the unknown territory of discussing Quantum Mechanics, how about we actually, actively eliminate misconceptions?

http://www.amazon.com/Modern-Quantum-Me ... 0805382917



Or, on the other hand, we could always listen to the guy with the red text. That's usually good.

[doogly]

http://www.amazon.com/Modern-Quantum-Mechanics-2nd-Sakurai/dp/0805382917

So much this.

[++$_]

Yeah, that kinda comes down to how you want to define "color".

Unless the point is that "Ultra Violent" probably isn't a color by any definition.

[Diadem]

No, they aren't. I think we are talking past each other. Pick up any grade school science book (literally every book in my state approved this year for K-8 science has a diagram that looks very close to this) and you'll probably see something like this: http://tinyurl.com/6oay98q

My main problem with that picture is how sadly correct it is for how science is often done. Specifically the part about only reporting your results if you confirmed your hypothesis. This is sadly spot on. Noone ever publishes 0-results. The damage this does science, especially the softer fields, can hardly be overstated.

As for the "The hypothesis is true" part. That too is exactly how most scientists think about results. The only thing that picture misses is that where it says "analyze data" it neglects to mention the analysis should be done in such a way as to optimize the chance of accepting your hypothesis!

[mfb]

Noone ever publishes 0-results.

Don't say that in general, please.
"No new resonance has been found and limits on several new physics models are set."
"Theoretical predictions, [...], are found to be in agreement with the measurements."
"The single-jet inclusive forward jet spectrum is well described by all models, but not all predictions are consistent with the spectra observed for the forward-central dijet events." <- this usually implies that the calculations are not good enough, so it is similar to a 0-result.
SUSY search at Tevatron
"No excess over the Standard Model is observed [...]"
All preprints are from today, and particle physics only. It is the usual style of LHC papers, as long as there is a reasonable theory prediction available. The other entries from today are prospects for future measurements or hardware-related.

[eSOANEM]

Also, RE:0-results, what about the Michelson-Morley experiment?

That said, despite there being high profile negative results papers, there is a problem with publication bias, particularly in clinical trials. Ben Goldacre has some statistics on it somewhere, he even managed to find evidence of publication bias in papers about publication bias which is rather ironic.

[doogly]

Also, RE:0-results, what about the Michelson-Morley experiment?

Not implying you are guilty of it, but one great misconception surrounding Michelson-Morley is that it motivated Einstein's development of special relativity at all. Some folks do make an effort to set things straight.

[legend]

one great misconception surrounding Michelson-Morley is that it motivated Einstein's development of special relativity at all. Some folks do make an effort to set things straight.

Not directly, but from my understanding of the subject Michelson-Morley experiment did motivate Lorentz & Co. to investigate transformations that left Maxwell's equations invariant, which in turn did influence Einstein's work on SR.

[eSOANEM]

one great misconception surrounding Michelson-Morley is that it motivated Einstein's development of special relativity at all. Some folks do make an effort to set things straight.

Not directly, but from my understanding of the subject Michelson-Morley experiment did motivate Lorentz & Co. to investigate transformations that left Maxwell's equations invariant, which in turn did influence Einstein's work on SR.

I was under the impression that Lorentz and Poincaré's models, whilst providing some of the same predictions of SR (e.g. time dilation), still required an ether.

[AvatarIII]

Misconception: melting ice caps contributes significantly to rising sea levels.
Truth: The Arctic ice cap, and Icebergs have no effect on rising sea levels, and the effect of melting glaciers and land locked ice is negligible next to thermal expansion of ocean water. the reason melting ice caps is bad for global warming is actually due to the fact that ice and snow is the greatest reflector of solar radiation back into space, and the ocean is the greatest absorber of solar radiation.

[yurell]

On a note related to that:
Misconception: in the '70's, all the scientists were telling us we were heading for an ice age.
Truth: The media was telling 'us' (I wasn't born yet), not scientists. A few papers suggested global cooling due to aerosols in the atmosphere, but none of them suggested an ice age. Furthermore, six times as many papers were published predicting global warming in the same period (1965-1979).

[Diadem]

Noone ever publishes 0-results.

Don't say that in general, please.
"No new resonance has been found and limits on several new physics models are set."
"Theoretical predictions, [...], are found to be in agreement with the measurements."
"The single-jet inclusive forward jet spectrum is well described by all models, but not all predictions are consistent with the spectra observed for the forward-central dijet events." <- this usually implies that the calculations are not good enough, so it is similar to a 0-result.
SUSY search at Tevatron
"No excess over the Standard Model is observed [...]"
All preprints are from today, and particle physics only. It is the usual style of LHC papers, as long as there is a reasonable theory prediction available. The other entries from today are prospects for future measurements or hardware-related.

Publishing 0-results in a lot easier in physics, but results are much more exact, and statistical errors are generally much smaller. Theories also make much stronger, testable claims, so not finding these is often very important. In the softer sciences predictions are less strong, and statistical errors are much bigger. A result is that 0-results often are genuinely uninteresting, because they often just mean your experiment was junk. However sometimes they aren't, and there is a strong selection bias against 0-results that makes it very hard to publish any of them.

This selection bias is still there in physics and astronomy, but less strongly. At least I think it's less strongly, that may also be selection bias though

[The Geoff]

Yeah, that kinda comes down to how you want to define "color".

Unless the point is that "Ultra Violent" probably isn't a color by any definition.

It's certainly more violent than fuchsia though, for your average sunbather at least

[Quizatzhaderac]

<wrong>Work is a measurement of moving an object perpendicular to gravity.</wrong> Actually antiparallel to gravity's pull. The "foot-pound" doesn't say which direction.

<wrong>An electric charge (regardless of motion) produces a magnetic field. A magnet can be split into positive and negative monopoles.</wrong> An electric change produces an electric field. A moving electric change produces a magnetic field. Magnetic flux is zero for any given volume (discounting monopole conjecture). Given magnets, charged objects and ignorance, a bad scientist could even conclude they're two distinct forces.

<wrong>A gram is a unit of weight</wrong> A think this is so prevalent for linguistic reasons. We have verbs for "Weigh" but not for "determine the mass of". We also almost always infer mass from weight. In all my chemistry classes I never once weighted out 50 milliNewtons of calcium carbonate.

<wrong>An event horizon is produced by a gravitational field of sufficient intensity</wrong> An event horizon is produced by a sufficiently extensive gravitational field. I got this from a books using "rubber sheet" visuals where black holes made wells with vertical slopes. A better image would be event horizons being like the water table below the sheet.

<wrong>Dark matter is matter that hasn't been observed electromagnetically, but has been gravitationally.</wrong> Dark matter is matter that can't been observed electromagnetically, but has been gravitationally. Or maybe I was right the first time. In either case I rarely hear about the other group in contrast to dark matter. It might just be my ignorance, but I find it hard to be accept being certain of how exactly how much of what hasn't been seen can't be seen.

[gmalivuk]

<wrong>Work is a measurement of moving an object perpendicular to gravity.</wrong>

Who has this misconception? I can't recall ever hearing work defined that way.

An event horizon is produced by a sufficiently extensive gravitational field. I got this from a books using "rubber sheet" visuals where black holes made wells with vertical slopes. A better image would be event horizons being like the water table below the sheet.

The problem with the "rubber sheet" visual is that, while it can approximately illustrate gravitational potential, it's not the same thing as spatial curvature under general relativity.

[doogly]

Dark Matter generally refers to anything dark. Models which do not interact with EM (WIMPs -- weakly interacting massive particles) are favored over MACHOs (massive cold heavy objects), mostly due to the need to hang out in halos rather than fall in, which cold things would still do. But although WIMPs are favored, dark matter is not defined as being made out of them.

[Diadem]

<wrong>A gram is a unit of weight</wrong> A think this is so prevalent for linguistic reasons. We have verbs for "Weigh" but not for "determine the mass of". We also almost always infer mass from weight. In all my chemistry classes I never once weighted out 50 milliNewtons of calcium carbonate.

I'd actually say it's a common misconception that we can't use gram as a unit of weight. Because there's absolutely nothing wrong with this usage.

<wrong>An event horizon is produced by a gravitational field of sufficient intensity</wrong> An event horizon is produced by a sufficiently extensive gravitational field. I got this from a books using "rubber sheet" visuals where black holes made wells with vertical slopes. A better image would be event horizons being like the water table below the sheet.

An event horizon is formed by a sufficiently compact gravitational source. I'm not quite sure what you were trying to say with your example, but I don't think that was it, so I thought I'd point out.

Any amount of mass has a so called Schwarzschild radius. Compress the mass into that radius, and it forms a black hole. Doesn't matter how much mass you have. However overcoming electromagnetic and quantum-mechanic forces resisting such compression isn't easy, so that the only way we know to do it is to take several solar masses of matter.

<wrong>Dark matter is matter that hasn't been observed electromagnetically, but has been gravitationally.</wrong> Dark matter is matter that can't been observed electromagnetically, but has been gravitationally. Or maybe I was right the first time. In either case I rarely hear about the other group in contrast to dark matter. It might just be my ignorance, but I find it hard to be accept being certain of how exactly how much of what hasn't been seen can't be seen.

Well this is a matter of definition, of course. But the (original) standard definition of dark matter is simply "any matter that is dark", ie: "any matter that doesn't give off significant amounts of electromagnetic radiation". The earth is dark matter.

However as more and more theories of what dark matter can be get discarded, some people shrink the definition of dark matter to exclude discarded possibilities.

[mfb]

<wrong>An electric charge (even if not moving) produces a magnetic field. A magnet can be split into positive and negative monopoles.</wrong>

- An electric charge, even if not moving in our system, produces a magnetic field in all inertial systems which move relative to the electric charge.
- You need some strange definition of "moving" to handle particle spins here.

<wrong>The gravity of Black Holes outside their Schwarzschild radius is different from other collections of masses in some way.</wrong>
If you would replace the sun by a BH with the same mass, the orbit of the earth would not change in any significant way.

[Quizatzhaderac]

<wrong>Work is a measurement of moving an object perpendicular to gravity.</wrong>

Who has this misconception? I can't recall ever hearing work defined that way.

Me before high school physics. Also this webcomic. Movers might do positive or negative work. They'd average out near zero since they mostly move objects across the earth's surface, not perpendicular. I've never seen anyone explicitly say "perpendicular to gravity", but the definitions usually fail to mention the distance must be against gravity (or other force).

<wrong>An event horizon is produced by a gravitational field of sufficient intensity</wrong> An event horizon is produced by a sufficiently extensive gravitational field. I got this from a books using "rubber sheet" visuals where black holes made wells with vertical slopes. A better image would be event horizons being like the water table below the sheet.

An event horizon is formed by a sufficiently compact gravitational source. I'm not quite sure what you were trying to say with your example, but I don't think that was it, so I thought I'd point out.

Any amount of mass has a so called Schwarzschild radius. Compress the mass into that radius, and it forms a black hole. Doesn't matter how much mass you have. However overcoming electromagnetic and quantum-mechanic forces resisting such compression isn't easy, so that the only way we know to do it is to take several solar masses of matter.

If you integrate the gravitational field along a path from the observer to an event horizon you'll get c². With a sufficiently large mass a black hole could have no more than 1 g (or any value) gravity at the event horizon. An event horizon does not, for instance, occur once g = 9 X 10¹⁶ m/s².

<wrong>An electric charge (even without motion) produces a magnetic field. A magnet can be split into positive and negative monopoles.</wrong>

- An electric charge, even if not moving in our system, produces a magnetic field in all inertial systems which move relative to the electric charge.
- You need some strange definition of "moving" to handle particle spins here.

Fix in italics. The main point was that the electric field is not the same thing as the magnetic field.

As for my strange definition of "moving" I'll use "has non zero momentum".

[scottmsul]

I have one!

Big misconception in cosmology is that the universe expanded in 3 dimensions from a point, like an explosion going off. Any astronomer will tell you that's completely wrong and start talking about raisin bread.

The way I like to think of it is the universe started infinitely dense everywhere and just gets less dense over time.

[gmalivuk]

Misconception:

With a sufficiently large mass a black hole could have no more than 1 g (or any value) gravity at the event horizon.

Gravity at the event horizon is always infinite, in the sense that no finite force could prevent you from falling in once you're there.

As for my strange definition of "moving" I'll use "has non zero momentum".

This is still frame-dependent.

[mfb]

As for my strange definition of "moving" I'll use "has non zero momentum".

This is still frame-dependent.

In addition, it does not account for spin, which is non-zero in every frame.

>> Big misconception in cosmology is that the universe expanded in 3 dimensions from a point
In addition: GR and curvature of spacetime does not mean that our spacetime is located in some higher-dimensional space. For example: the model of an expanding balloon does not mean that there is space "inside".

[eSOANEM]

In addition: GR and curvature of spacetime does not mean that our spacetime is located in some higher-dimensional space. For example: the model of an expanding balloon does not mean that there is space "inside".

This definitely deserves its own point. The theorema egregium is not called the remarkable theorem for nothing.

[doogly]

Misconception:

With a sufficiently large mass a black hole could have no more than 1 g (or any value) gravity at the event horizon.

Gravity at the event horizon is always infinite, in the sense that no finite force could prevent you from falling in once you're there.

OK, but that is a silly sense. The sense for "strength of gravity" one would usually imagine being implied is if you let go of something, with what acceleration does it start to fall? This could be 9.8 m/s^2 at an event horizon quite easily.

[WarDaft]

<wrong>Work is a measurement of moving an object perpendicular to gravity.</wrong>

Who has this misconception? I can't recall ever hearing work defined that way.

Me before high school physics. Also this webcomic. Movers might do positive or negative work. They'd average out near zero since they mostly move objects across the earth's surface, not perpendicular. I've never seen anyone explicitly say "perpendicular to gravity", but the definitions usually fail to mention the distance must be against gravity (or other force).

The movers themselves are doing a great deal of work, the net work done on the things they move can be very low of course. This is because they (being not complete idiots I presume) realize that dragging things causes friction. Due to the inherent friction in the human body's movement however, they actually do a tremendous amount of work overall, they just do most of it to themselves. Movement against friction is still work. In fact, if you'll note, they say the further and faster they move something, the more work they do. This is still appropriate, as drag is proportional to speed (though I have yet to see movers go anywhere near fast enough for drag to be significant), and so the force involved greater, and thus the total work done higher. It's also conceivable that there is more internal friction in the human body at higher speeds, but that seems less less dependable.

In fact, I'd say the opposite misconception is presented more often, it was in Bill Nye the Science Guy for example. After moving a weight back and forth repeatedly, it was claimed Bill had done no work. Nonsense, he was doing work in each direction against friction. If lifting a block against gravity is work, despite the fact that it is stationary after being lifted (until you drop it anyway) then pushing a block against friction is equally work. The energy is just heat energy afterwards, instead of potential energy.


Oh, I've also heard people say that inclining a treadmill shouldn't make it any more work, that's also wrong.

[gmalivuk]

Misconception:

With a sufficiently large mass a black hole could have no more than 1 g (or any value) gravity at the event horizon.

Gravity at the event horizon is always infinite, in the sense that no finite force could prevent you from falling in once you're there.

OK, but that is a silly sense. The sense for "strength of gravity" one would usually imagine being implied is if you let go of something, with what acceleration does it start to fall? This could be 9.8 m/s^2 at an event horizon quite easily.

Are you sure? If it requires a trillion Newtons per kg to remain stationary just above the event horizon, then dropping anything into freefall should let it accelerate a trillion m/s^2, right?

[Diadem]

Misconception:

With a sufficiently large mass a black hole could have no more than 1 g (or any value) gravity at the event horizon.

Gravity at the event horizon is always infinite, in the sense that no finite force could prevent you from falling in once you're there.

OK, but that is a silly sense. The sense for "strength of gravity" one would usually imagine being implied is if you let go of something, with what acceleration does it start to fall? This could be 9.8 m/s^2 at an event horizon quite easily.

Are you sure? If it requires a trillion Newtons per kg to remain stationary just above the event horizon, then dropping anything into freefall should let it accelerate a trillion m/s^2, right?

What does acceleration even mean at the event horizon? Regular spacetime coordinates become singular, acceleration isn't even defined in regular Cartesian coordinates. I mean sure you could do a transform to another reference frame where your coordinates aren't singular, but but then what relation does the figure you get for acceleration bear with its everyday meaning?