I am a bit skeptical of the conclusion given the methods. Here, there's no observable phenomena independent of the test apparatus that corresponds to the proposed cause. The conclusion is circular.
1. Theory predicts gravitational waves when massive objects collide and that the gravitational waves would have an effect that could be measured by the experimental instruments.
2. The experimental instruments measure something.
3. This is considered proof that massive objects collided.
4. Therefore gravitational waves exist.
To reframe my skepticism, the experiment measures something. The conclusion as to what it measures, however, is unsupported by statistical inference or direct experience of a causal phenomenon. That's not to say that what the phenomena measured -- the earth resonating -- is uninteresting or unimportant or even inconsistent with the theory of gravitational waves.
Yet, I don't find the possibility of a geophysical cause -- i.e. that the earth maintains consistent dimensions at a sub-atomic scale -- the many orders of magnitude less likely than gravitational waves necessary to reach a conclusion. In particular, I find natural fluctuation to be more likely because the experiment acknowledges its existence.
For a point of comparison, consider the Perihelion precession of Mercury that provides evidence in support of general relativity. The theory was used to predict the results of an observable event. The experimenters trained their telescopes at a particular location and particular time and observed phenomena consistent with a prediction based on the theory. The same is true of the Higgs. In both cases the experiment is of the form "when X, I will observe Y."
The reasoning here is:
If X, then Y.
Y, therefore X.
It treats an ordinary implication as mutual implication.
(1) It simply fails to understand the scientific method, which is empiricism, not mathematical/logical proof. Scientific evidence is essentially failed disproof, not logical proof.
(2) It mischaracterizes the nature of the prediction, which includes not merely that something will be measured, but that a particular pattern will be measured.
(3) It proposes unspecified "geophysical causes" as an alternate explanation, but there was no pre-existing geophysical model which predicted the pattern observed. (Any after-the-fact geophysical -- or other -- alternative explanation which explains the observed pattern would also need to perform differently on some other test to be verifiably different, and then we could do the test to distinguish the source.)
(4) It misstates the reasoning to contrast it with other experiments, this is exactly "when X, I will observe Y" (where X is "I construct detectors of a particular type in more than one location" and Y is "I will periodically detect particular patterns of signals on those detectors -- not just one of them alone -- which the model predicts will be produced by collisions of massive, distant objects.")
Under the theory of science as the study of what is falsifiable, there's nothing here to falsify because there is no way to disprove that a conjectured but unobserved collision of two massive bodies was something else or didn't occur. Which is to say that it is impossible to falsify that a conjecture is a conjecture.
That there is not a geophysical theory, doesn't have a bearing on the correctness of the gravitational wave theory one way or the other...anymore than the absence of a helio-centric model for the solar system made the geocentric model more correct or the absence of a theory of oxygen made the theory of pholgiston more correct. More importantly, both these incorrect theories had reasonable explanatory power to the point that they were useful.
The reason they were useful theories is because they were predictive, pholgiston allowed a person to calculate the weight of ashes after burning and the geocentric solar model made the prediction of the location of stars possible with reasonable precision. On the other hand, theories that offer conclusions about unfalsifiable propositions are what Carnap and the Vienna circle termed "metaphysics".
The conclusion that the experiment justifies is that the Earth resonates. There is no external event to which the measurements can be correlated to establish causality. There's no confidence interval. It's a case where the observations confirm a pre-existing world view under the same human cognitive structures by which seashells on mountain tops confirm a world-wide flood. It assumes that because we live on the Earth we know everything about it.
Anyway, it's a case of over-reaching with the conclusions. It's an argument from design.
> Under the theory of science as the study of what is falsifiable, there's nothing here to falsify because there is no way to disprove that a conjectured but unobserved collision of two massive bodies was something else or didn't occur.
Yes, there is: if the predicted kind of observations did not occur, it would imply one of two things:
(1) the model of gravity waves and their generation and propagation on which the prediction was based was incorrect, or
(2) the expectation of large-object collisions on which the prediction was also based is incorrect.
Now, were that the case, distinguishing which of those assumptions was false would require coming up with a new set of experiments that would have different results if the first was correct and the second false than if those were flipped, and yet a different set of results if both were false.
> That there is not a geophysical theory, doesn't have a bearing on the correctness of the gravitational wave theory one way or the other
Science isn't about correctness, its about continuous refinement of models which better predict observations. The absence of a better alternative model doesn't "prove" that a given model is "correct", but science deals with neither proof (except in the negative sense) nor correctness. (Further, the model of gravity waves being tested here is an implication of broader models whose other implications have also withstood attempts to falsify them.)
Lower two graphs are predicted waveform in event of binary system coalescence. Upper waveform left and right are observed events at each LIGO facility.
Consider furthermore that the event was observed at each facility with the appropriate lightspeed time offset, and that the wave directionality of the event was lateral and not radial from the center of the earth or some other point as would be expected from a "geophysical" cause.
Furthermore I find it highly unlikely that multiple theoretical and applied physicists would come to apparently total agreement on the significance of these findings and simply overlook the gaping logical fallacy you imply.
It's not a gaping logical fallacy. It's human nature to tell interesting stories. Black holes colliding is sexy. "We measured a 10^-18 variation in the diameter of the Earth and we are uncertain regarding its cause" is not sexy.
As Jack Friday used to say, "Just the facts, mam."
Unfortunately not-entirely-unreasonable skepticism can be mistaken for hostility by people excited about a potentially groundbreaking result.
That said, it sounds like they did a lot more work to eliminate sources of error than you may be aware of. OTOH if seismic resonance was causing the correlation, there would probably be more time between the events at the two facilities.
Scientists will try to poke holes in these results while further experiments will try to corroborate them. Meanwhile laypeople will be introduced to more oversimplified and counterproductive ways to think about this stuff. Business as usual.
I'm sure I was not clear, since that was the first pass.
What the experiment indicates is that the Earth varies in size. Roughly:
measured distance of 10^-20 meters
4km is 10^-5 of earth circumference.
delta Earth's circumference 10^-25
total distance change in earth's diameter = 10^-18 meters
Given the non-intuitive nature of geology[1], I am saying that the possibility that the Earth varies that much in dimension due to it's internal structure is not so vanishingly remote as to be left unaddressed. Saying it's ten or a thousand times less likely doesn't move the needle much at that scale...if such a thing were said.
[1]: I'm old enough to have been introduced to platetechonics as a distuptive theory.
The LIGO detectors were very carefully constructed to rule out this sort of noise; they are the result of a lot of human engineering ingenuity. It turns out that before making this investment in instrumentation specifically for the purpose of detecting gravitational waves, the scientists in question also came up with the objections that you spent presumably less than fifteen minutes thinking of. They then spent time and substantial engineering effort to avoid those objections.
My tone here is rather curt as it's clear your skepticism is unfounded and that you didn't look at even the layman explanations put out by the scientists who made these claims. For example, the video announcement talks about the construction of the equipment in question: https://www.youtube.com/watch?v=aEPIwEJmZyE&feature=youtu.be... and specifically t=3310 talks directly about how they worked to avoid detecting the motion of the earth.
I'm old enough to remember Pons-Fleischmann. But not so old as to forget the more recent faster than light neutrinos social media storm. I've been thinking about the nature of scientific claims for more than twenty years. Few people are intentionally wrong: that doesn't mean astronomers as men of science didn't sware by the Ptolemaic cosmology.
To be clear, I am not denying the possibility of the earth expanding in and contracting in what passes for space time. I am skeptical of the proposition that the instruments are measuring the collisions of black holes is methodologically sound.
The careful preparation and engineering that occurred before LIGO was constructed resulted in an instrument that ran from 2002 to 2010 without detecting gravitational waves. The consequence of this $400 million experiment was not reexamining the theory, but sinking another $200 million into an instrument that created good tweetable data. That's the way careers and politics and human nature works.
There are three components to the theory. Spatial change, gravitational waves, and colliding massive bodies. The reason I am skeptical is that the scientists are inferring two of them: gravitational waves and colliding black holes from the component most likely to have other causes. If I had the mountaintop shells and the flood, god would be more plausible. If I have the shells and god, the flood is. Two outta three is my threshold for reasonable scientific inference in this case.
If you're not denying the possibility, then what exactly are you trying to say here? That you're skeptical that the experiment detected anything, or that the "anything" it detected is what they are saying it is?
Clearly the first can and will be found out over time as other detectors are being built to replicate these findings. However, I believe there is enough scrutiny of their claims that this kind of skepticism can likely be ruled out.
The second, that their story doesn't fit the data, is kind of odd to me. Gravity is so weak and the detectors they are creating are still so new that it seems likely that colliding massive bodies would be the first kinds of things they would pick up. Just as with early telescopes where objects that were very close or very bright were the first ones to yield useful data.
If you have another explanation that you believe fits the data better, put it forward and try and find a way to test it. That's how science works. But this is not the kind of thing that is going to collapse in a heap of logic. Data and the scientific method doesn't work like that.
Maybe it would help to re-frame the relevant part of the (apparent) argument:
Oscillations show up everywhere in nature, and even a pattern as specific as a frequency sweep with ringdown could be the result of many different phenomena. Even if in this case many possible sources have been ruled out, there must be others that we do not know about. Since the only observation we have to go on is the signal (so far), we should remember that the cause implied by the model is contingent on the signal not being one of these unknown sources.
Some commentary: Imagining alternative explanations is only half of the work that comes next. Once more of these alternatives are found, we also need new experiments that will be designed to rule them out. It sounds like the space-based interferometers will go a long way toward ruling out potential "terrestrial" factors. And if the same signal is detected in both ground-based and space-based systems that is an even greater step forward.
I was also wondering how they successfully rule out other possibilities. Many people seem very eager to believe. Some skepticism here can help aid a more thorough understanding.
I suppose the prediction didn't include a scale (it was determined from the observation+model) so there is a little bit of room for model fitting there. And sure, other stuff could have generated a frequency sweep. Maybe if this is replicated the newer methods will narrow down the possibilities.
[0. A long history of theoretical and experimental discoveries lead to a specific, well-motivated theory (General Relativity) describing gravitational phenomena.]
1. Theory predicts a very specific and recognizable class of gravitational wave patterns for likely sources, whose details are determined by a small number of parameters that correspond to meaningful physical quantities (specifically, two masses and a distance; I don't know if there are any more than that).
2. Two independent experimental instruments measure a wave pattern that (apart from some low-level noise) is an excellent match to one of the predictions. Fitting the model to match the specific data, the resulting parameters have entirely plausible astrophysical values.
3. This is considered an example of a theory making a specific, novel prediction that is later confirmed by experiment: pretty close to the classic definition of the scientific method.
4. Therefore, the experimental data has provided strong support for the premises of the theory being tested: both its prediction of gravitational waves and its detailed dynamical predictions for the source scenario matching the observations.
Maybe I'm misunderstanding something in your argument, but I don't see any circularity here. What about this process isn't precisely the way that science is supposed to work?
Except that if two detectors, preferably three, detect the wave at the appropriate times then it can be said that whatever phenomena is the cause exists at a particular vector. A bunch of waves all coming from a direction pointing to the center of a huge galaxy would fit the wave model more than any earthbound cause.
Someone comes up with a theory that's mathematically simple, beautiful, and consistent. That's great, but we don't consider it "correct" until it actually predicts something novel and we verify the prediction holds experimentally.
The core idea is plausible, if circular at present. That's OK. Now, we take that and run with it to expand on what we think we understand and hopefully, confirm and even more hopefully, crack open some new physics.
Seems like Einstein nailed it.
But, questions remain. Any of those could yield insights of similar potency and, ahem... gravity.
In this case I don't know if your skepticism is warranted.
1st: Paper said that the waves are redshifted to a certain degree (it's in the abstract) [1], ergo 1.3 billion light-years away in space-time. I'm not thinking you want to be so fundamentally skeptical.
Fine, but I find comments like "I don't really grasp what's going on, but I can show it's crap using high school logic" to be extremely low-quality, and I'm sure I'm not the only one.
I read the article. I am in the habit of calling periodic changes in dimension "resonance." I consider it rather consistent with the use of "wave" in the discussion and hence a handy way of describing what the instruments measured rather than what the theory suggests as the first cause.
Resonance is quite a bad term if that's what you use it for. Resonance would require some amount of positive feedback. Vibration or oscillation would be a much better terms for simple periodic changes.
It's consistent with vibration caused by a wave which is why it seemed appropriate to me in this context...even if I think the claim of waves is an over-reach, I am not being deliberately argumentative.
1. Theory predicts gravitational waves when massive objects collide and that the gravitational waves would have an effect that could be measured by the experimental instruments.
2. The experimental instruments measure something.
3. This is considered proof that massive objects collided.
4. Therefore gravitational waves exist.
To reframe my skepticism, the experiment measures something. The conclusion as to what it measures, however, is unsupported by statistical inference or direct experience of a causal phenomenon. That's not to say that what the phenomena measured -- the earth resonating -- is uninteresting or unimportant or even inconsistent with the theory of gravitational waves.
Yet, I don't find the possibility of a geophysical cause -- i.e. that the earth maintains consistent dimensions at a sub-atomic scale -- the many orders of magnitude less likely than gravitational waves necessary to reach a conclusion. In particular, I find natural fluctuation to be more likely because the experiment acknowledges its existence.
For a point of comparison, consider the Perihelion precession of Mercury that provides evidence in support of general relativity. The theory was used to predict the results of an observable event. The experimenters trained their telescopes at a particular location and particular time and observed phenomena consistent with a prediction based on the theory. The same is true of the Higgs. In both cases the experiment is of the form "when X, I will observe Y."
The reasoning here is:
It treats an ordinary implication as mutual implication.