31

Rupert Sheldrake recently had a TEDx presentation removed from the TEDx web-site, for making various controversial claims, basically calling all of science into doubt. I am highly skeptical about nearly all of them, most seem particularly cranky and mystical.

The most interesting to me is the claim that the gravitational constant (G) changes over time, or space. In particular, he claims that the estimate is obtained by different labs making measurements on different days, and taking the average, and then the International Committee on Metrology takes an average of averages to decide the "value of big G" (I imagine that they actually just make an estimate of the value). Sheldrake argues that if we looked at the raw data, we might begin seeing patterns (e.g. correlations between changes at different labs indicating that the errors in G are not actually measurement errors, but real, meaningful fluctuations (on a daily or annual scale), but that "no one has done this ... because G is a constant".

So the question is: is there any strong evidence for or against G being a difficult-to-measure constant, rather than being a fluctuating value?

naught101
  • 421
  • 4
  • 9
  • @RobZ: ***I*** have no objection. I think Sheldrake's claim is probably wrong. But his claim is that G is changing (significantly) on a daily or annual basis. I have no evidence with which to argue against that. Clarified the question. – naught101 May 29 '13 at 00:09
  • 6
    Is he saying "what if that isn't a measurement error..." or is he claiming that it isn't measurement error? –  May 29 '13 at 00:26
  • 2
    @Sancho When I watched the video my understanding is he thinks that G could be variable, but his bigger issue is that the labs aren't reporting their daily measurements so the data set can't be examined to if there are meaningful patterns in the data that could indicate a variable G or some other interesting environmental effect that causes our measurement of G to change (i.e. "going through a patch of dark matter"). – rjzii May 29 '13 at 00:27
  • At the start of the video, he says that thinking that universal constants are constant is one of the 10 dogmas of science, which he claims are able to be shown to be problematic. For me, it seems his big issue is that science has no room for his hypotheses, and he appears to be implying that the fact that the raw data for G is not analysed properly is an indication that the science is trying to hide something. – naught101 May 29 '13 at 00:59
  • 3
    It reads very much as if he is simply asking a huge 'what if' question, and pointing out that because of our assumptions we aren't following up on the possibility that G might vary. I don't think there is a real skeptical claim to debunk. – DJClayworth May 29 '13 at 01:55
  • @DJClayworth: my understanding is that skepticism works in both directions. Sheldrake is being skeptical of a widely held scientific viewpoint. I'm asking if his question has any validity, evidence, or sense. – naught101 May 29 '13 at 01:58
  • 6
    While this is a good skeptical question, not sure if we have the requisite expertise to answer it here. I'd be interested in an answer, but maybe you would have better luck at Physics.SE? – Larian LeQuella May 29 '13 at 02:32
  • 2
    Let's use a little sense here, guys. A variation in the value of G at the 1% level on time scales similar to a day would show up as a *huge* effect on the period of satellites around our fair planet. Even if you assume the variation happens slowly enough that the orbits don't get re-shaped, the periods would vary. Moreover this would cause a global non-conservation of energy unless the Earth itself expanded and contracted as G changed. Try figuring the gravitational binding energy of the planet. Now figure out what to do with 1% of that for a few hours.. – dmckee --- ex-moderator kitten May 29 '13 at 03:14
  • 1
    @LarianLeQuella Not on physics, please. We don't allow questions about "fictional physics" nor do we allow personal or non-mainstream theories, and as I've outlined in my previous comment this idea has huge, glaring holes in it. My best guess: Mr Sheldrake simply doesn't understand the analysis of experiments to measure G and therefore fails to understand the scale of the expected variation in the results. These experiments are not easy. – dmckee --- ex-moderator kitten May 29 '13 at 03:16
  • 2
    @RobZ The more I think about it the more the binding energy argument matters. You see, less force should mean the Earth expands, right. But that reduces the binding energy *more*. You don't need to figure out to do with 1% of the binding energy you need to figure out what to do with 2% of the binding energy (which in case you haven't worked the figures is equal to about 3 hours of the sun's *total* output). If, instead, you look at the value growing you have to find that kind of energy somewhere. – dmckee --- ex-moderator kitten May 29 '13 at 03:36
  • @dmckee: thanks, that looks like a good enough answer for me at least. Feel like posting it as one? – naught101 May 29 '13 at 03:56
  • 1
    "[No](http://meta.skeptics.stackexchange.com/questions/80/should-we-expect-original-research-to-occur-on-this-site) [original](http://meta.skeptics.stackexchange.com/questions/2170/how-should-references-be-handled-vs-original-research) [research](http://meta.skeptics.stackexchange.com/questions/617/under-what-conditions-is-original-research-problematic-on-community-mediated-s)". – dmckee --- ex-moderator kitten May 29 '13 at 04:14
  • 1
    it wouldn't surprise me at all if many of the things we perceive as constants actually aren't, but their fluctuations being so minute they fall inside the inaccuracy of our means to determine them so they average out to zero. For G this may well mean it fluctuates on say the 20th decimal. No conspiracy, just physics :) – jwenting May 29 '13 at 05:45
  • 5
    With regards to the question, saying that he "basically calling all of science into doubt." is a bit of an exaggeration. When I saw the talk and from what I have read of his recent work he's basically beating the drum to remind people not to allow the scientific community to become too dogmatic in their thinking not that science or the scientific method is wrong. In other-words, we should avoid rejecting something that goes against the current dogma, for example [quasicrylstals](http://www.popsci.com/science/article/2011-10/chemist-accused-quasi-science-wins-nobel-quasicrystal-discovery) – rjzii May 29 '13 at 12:52
  • 8
    Sheldrake is not claiming that scientists are trying to hide anything. Most skeptics don't claim that the pope is hiding evidence that God doesn't exist. Sheldrake is claiming that those scientists hold certain beliefs as dogma and therefore don't really investigate them with the scientific method because they think they already know the answer. Thy guy had a blog titled "Science is a method, not a position." – Christian May 29 '13 at 18:08
  • There have been other cases of physical constants changing (or more precisely being re-estimated and producing new estimates outside the previously estimated margins of error), with one notorious case being the gradual change in the value for the charge of an electron [pointed out by Richard Feynman](http://en.wikipedia.org/wiki/Oil-drop_experiment#Millikan.27s_experiment_as_an_example_of_psychological_effects_in_scientific_methodology), which did not bring credit to the physics community – Henry May 30 '13 at 23:14
  • This is a classic case of modus ponens with unproven proposition and therefore impossible to prove true or false as long as the proposition is not fully known. – MrDosu Oct 30 '14 at 10:57
  • Well that would explain why my diet isn't working. I lose weight but the gravitational constant increases by the same amount – GordonM Aug 10 '17 at 13:45

2 Answers2

30

At this time the Gravitational Constant is regarded as a constant with problematic/low accuracy by physicists. From University of Washington Big G Measurement:

Since Cavendish first measured Newton's Gravitational constant 200 years ago, "Big G" remains one of the most elusive constants in physics. The value of big G tells us how much gravitational force acts between two masses separated by a known distance. In Einstein's language of general relativity, it tells us the amount of space-time curvature due to a given mass. Together with Planck's constant and the speed of light it is considered to be one of the most fundamental constants in nature. Big G is a necessary ingredient in determining the mass of the earth, the moon, the sun and the other planets.

Several measurements in the past decade did not succeed in improving our knowledge of big G's value. To the contrary, the variation between different measurements forced the CODATA committee, which determines the internationally accepted standard values, to increase the uncertainty from 0.013% for the value quoted in 1987 to the twelve times larger uncertainty of 0.15% for the 1998 "official" value. This situation is an embarrassment to modern physics, considering that the intrinsic strength of electromagnetism, for instance, is known 2.5 million times more precisely and is steadily being improved. (The situation of G becomes more understandable if one considers the weakness of gravity: the total gravitational force twisting on the pendulum of a typical Cavendish torsion balance is only equivalent to the weight of a bacteria and that small force must be measured very precisely.)

Since we are talking about physics, this is true to the best of our current understanding.

The question of how/why do we know that G (and other constants) are indeed constant was addressed in Physics.SE: What is the proof that the universal constants (G, ℏ, …) are really constant in time and space?

Regarding the claim that the data isn't made public. I couldn't find any evidence that labs are constantly remeasuring and updating G and I couldn't find that this data is hidden. The data is made public through journal articles. Also, the value is not updated constantly as the National Institute of Standards and Technology published a figure that was last updated at 2010.

Here are 4 different values for G from different sources:

When plotted in a graph with error bars, we have the following:

Graph of Gravitational Constant Measurements

So measurements are not constantly done, the "official" value is not updated regularly, and the results of the measurements are published.

Community
  • 1
SIMEL
  • 29,037
  • 14
  • 123
  • 139
  • Why the downvote? – SIMEL May 29 '13 at 08:55
  • 1
    I don't think this deserves downvote, but there are few issues: 1) whole "Can it be that the Gravitational Constant is not a constant" part is pure speculation 2) explaining why no original research belongs in comment, rather than answer, 3) I don't see how Einstein's quote about God playing dice has any relevance to the subject. – vartec May 29 '13 at 11:19
  • 4
    @vartec, what I meant to show is that the physicist community does change its mind, like shown with quasicrystals and with Einstein's famous quote against quantum mechanics (there are other example, but they are less world changing than quasicrystals and quantum mechanics). Or in other words, If Sheldrake has any scientific evidence to his hypothesis he should conduct the research and present the results to the scientific community. The answer is not "G is forever constant", but like all scientific truths it's "G is a constant to the best of our current knowledge". – SIMEL May 29 '13 at 11:30
  • @IlyaMelamed I agree with vartec's objections. I would have concluded this answer at the end of the quote "... must be measured very precisely.)" –  May 29 '13 at 11:42
  • 1
    @vartec, Sancho, "Ask and you shall receive" – SIMEL May 29 '13 at 12:03
  • Indeed, nothing in science is permanent, everything is up for change given enough and verifiable evidence that the current understanding is incorrect. But anyone claiming that there's some conspiracy to hide the fact that G isn't a constant isn't worth debunking. – jwenting May 29 '13 at 12:09
  • +1 from me, neat answer, but maybe I will reserve my acceptance for an answer with some citations from actual research about fluctuations in G, as per the whole no-original-research rule discussion. The highlighted parts of the first quote are especially useful though, so thank you :) – naught101 May 29 '13 at 12:15
  • So the answer is basically that Rupert Sheldrake is right? The physists just define big G as constant because of theoretical arguments instead of empirical evidence. If nobody in theoretical physics community looked at the numbers and analysed whether or not there are correlations, I do think it's quite right for Shaldrake to criticize the physics community. – Christian May 29 '13 at 16:37
  • 2
    No, he's right that the number is "decided" as an average of several independent measures that show slightly different numbers. What he is postulating is that there is a periodic change in the value of G. What physicists say is that Gravity is the weakest of the 4 fundamental forces, so it's harder to measure it, so we have greater error in its measurement. It's like trying to understand what a person standing right next to you is saying, vs. understanding someone who is 200 meters from you with his back turned. In order to prove he's right he needs to do the research, which he didn't do. – SIMEL May 29 '13 at 16:45
  • 1
    @IlyaMelamed : (1) It's not Shaldrake's role to do the research. I would expect the physicist community to actually do that research. If they don't, that worth critizising. (2) He asks the physicists to put their numbers online so he can do the research. According to his charge physicists don't put their data online and he therefore can't run the numbers. If the data is withheld from him you can't blame him for not running the numbers. – Christian May 29 '13 at 17:21
  • 1
    @Christian, physicists have a theory that explains why G is constant, see the answer in Physics. If he sees flaws in the theory, or has made contradictory experimental results, he should publish them. Otherwise, he's just like a creationist that shouts "The Theory of evolution is wrong" but doesn't provide the proof. – SIMEL May 29 '13 at 19:29
  • Egads, man! Don't connect the damn dots on a plot like that. The line means less than nothing. – dmckee --- ex-moderator kitten May 29 '13 at 20:52
  • Indeed, "connect the dots" is near enough to always wrong that you should just not do it, full stop. – dmckee --- ex-moderator kitten May 29 '13 at 21:01
  • 2
    @RobZ, something else is going on, it's called the 3 other fundamental forces. The Gravitational force is really weak compared to them, so it's very hard to measure it. Everyone is aware of this, and that's why the measures for this constant are so inaccurate. There is also a theory that backs up that G is a constant. At this point he needs to bring proofs to the table. – SIMEL May 29 '13 at 21:49
  • @RobZ, I addressed the issue of what I think a suffice answer here: http://meta.skeptics.stackexchange.com/questions/2388/how-to-answer-scientific-claims – SIMEL May 29 '13 at 22:52
  • @IlyaMelamed : If you ask a good biologists about his evidence for evolution he can point you to a variety of empirical evidence. The highest ranked answer in the physics stackexchange question doesn't contain any empirical evidence. It's just disgraceful if the physicists don't have any empirical evidence for their theory and can only rely on theory in which they believe because they consider it "beautiful" or "simple" or whatever standard theoretical physics people use these days to choose on theory over another. – Christian May 29 '13 at 23:46
  • 1
    @IlyaMelamed : The thing that's in dispute is whether all the data is published or just the average plus the confidence interval. The source that you link to doesn't give the raw data of the measurements. Claiming that the data of measurements is published is therefore highly misleading. – Christian May 29 '13 at 23:51
  • 9
    @Christian The stable shape and period of orbits *is* evidence. We have quality records of the planetary orbits going back thousands of years. Physicists don't need to talk about this stuff because it is bleeding obvious. Obvious enough that anyone who passed first semester college physics could check it, if they cared to. – dmckee --- ex-moderator kitten May 29 '13 at 23:58
  • @IlyaMelamed - But saying that "something else is going on, it's called the 3 other fundamental forces." kind of plays directly into his argument in the video (worth listening to if you haven't already) in that there might be issues with the measurement of G not because G is changing (although he postulates that it could be) but because something in the environment might be affecting the measurement in an interesting way. Also, from what I was reading if any of the three other forces is affecting the measurement wouldn't it be limited to electromagnetic? Or some variation in the local gravity? – rjzii May 30 '13 at 00:02
  • 2
    @dmckee : (3) No answer of physics stackexchange mentions either orbits or shapes. This one doesn't either. You are basically saying that both answers don't provide us with the real evidence because of which physicits to believe that G is constant. If that conclusion is so easy why don't you write up an answer that shows every one that G doesn't fluctuate by 0.01% using the data we have from shapes and orbits? – Christian May 30 '13 at 08:10
  • @RobZ, you are right, only the EM force can directly interfere with the measurement. I'm talking about something else, when we say that G has a high error, it needs to be defined high compared to what. That what is other physical constants e.g. c, h, me, e, etc. And it's true, we know the mass of an electron with a 4.4e-8 relative error, which is nothing compared to the 1.2e-4 relative error of G, the reason is that other constants are measured through interactions through the other (stronger) forces, sadly, G can't be measured through any other force but gravity. – SIMEL May 30 '13 at 11:50
  • 1
    After thinking about the issue a bit more and remembering back my physics education dmckee is probably right and phyicists have a real reason to assume that G doesn't fluctuate. Still it quite amazing how an answer that doesn't provide the real reason but only relies on authority get's upvoted and accepted on this website. Based on the information that Ilya listed, Sheldrake argument is reasonable. The sign of good skeptic reasoning is to be convinced by the right arguments instead of accepting an argument because you like it's conclusion. – Christian May 30 '13 at 12:16
  • 1
    Does this answer bury the lede? Ron Maimon's answer on Physics.SE argues that G is a constant *by definition*, and it there is a fluctuating force, it isn't G that is changing, but some other value (e.g. the speed of light). – Oddthinking Jun 06 '13 at 14:55
  • 1
    Any chance of getting and update to take into account the latest measurement of [G = 6.67191(99) × 10−11 m3 kg−1 s−2](http://www.nature.com/nature/journal/vaop/ncurrent/full/nature13433.html)? – rjzii Jun 24 '14 at 20:03
7

According to Measurements of Newton's gravitational constant and the length of day Europhysics Letters (2015) vol. 110:

About a dozen measurements of Newton's gravitational constant, G, since 1962 have yielded values that differ by far more than their reported random plus systematic errors. We find that these values for G are oscillatory in nature, with a period of [P = 5.899 +/- 0.062 yr]

An article in response, Recent measurements of the gravitational constant as a function of time Physical Review D (2015) vol. 91 adds and corrects some data, but still finds a better fit to the time-varying model with 5.9 year period. It concludes:

The situation is disturbing—clearly either some strange influence is affecting most G measurements or, probably more likely, the measurements have unrecognized large systematic errors. The need for new measurements is clear.

DavePhD
  • 103,432
  • 24
  • 436
  • 464
  • 1
    This result from Anderson is not well regarded. Wikipedia has a reasonable summary: "This response notes that Anderson et al. not only omitted measurements, they also used the time of publication not the time the experiments were performed. A plot with estimated time of measurement from contacting original authors seriously degrades the length of day correlation. Also taking the data collected over a decade by Karagioz and Izmailov shows no correlation with length of day measurements..." You can read more in the article on G. – KAI Apr 18 '16 at 22:19
  • 1
    @KAI That whole passage from Wikipedia "This response notes..." is referring to the second article in my answer. – DavePhD Apr 18 '16 at 23:43