William M. Briggs

Statistician to the Stars!

Gravitational Waves And Discovering Cause

History_of_the_Universe.svg

You’ll have heard by now that the existence of gravitational waves have been “confirmed.” The scare quotes are intentional, but not in the sense that I (necessarily) disbelieve in the confirmation; instead, they’re used to highlight that confirmation-as-decision is a dicey subject.

The gist on the discovery is this: deep in space, a pair of black holes decided to mate, a process which sent ripples of spacetime cascading our way; these waves hit the Laser Interferometer Gravitational-Wave Observatory, or LIGO, and the machine that goes Bing! sprang into action.

The LIGO itself is in two parts, one in Washington (not DC, which is probably why it worked) and the other in Louisiana. Precisely:

Each detector is shaped like a giant L, with legs four kilometers long. Laser light bounces back and forth through the legs, reflecting off mirrors, and amazingly precise atomic clocks measure how long it takes to make the journey. Normally, the two legs are exactly the same length, and so the light takes exactly the same amount of time to traverse each. If a gravitational wave passes through, however, the detector and the ground beneath it will expand and contract infinitesimally in one direction, and the two perpendicular legs will no longer be the same size. One of the lasers will arrive a fraction of a second later than the other.

LIGO must be unbelievably sensitive to measure this change in the length of the legs, which is smaller than one ten-thousandth the diameter of a proton, or less than the size of a soccer ball compared with the span of the Milky Way…”There are so many knobs to turn, so many things to align, to achieve that [sensitivity].” In fact, the experiment is so delicate that unrelated events such as an airplane flying overhead, wind buffeting the building or tiny seismic shifts in the ground beneath the detector can disturb the lasers in ways that mimic gravitational signals. “If I clap in the control room, you will see a blip,” says Imre Bartos, another member of the LIGO team at Columbia.

This is even more sensitive than a teenage female is to Facebook postings. The extreme touchiness is why “researchers carefully weed out such contaminating signals and also take advantage of the fact that the detectors in Washington and Louisiana are highly unlikely to be affected by the same contamination at the same time.”

Highly unlikely is not, of course, equivalent to impossible. It—highly unlikely—isn’t even definable on its own, and is in this case stated in circular terms. Why? Because all probability is defined only with respect to certain evidence. Here that evidence is tacit, and includes the idea that either gravitational waves are real or that nothing else exists that can cause matched measurements at the two detectors. I am in no way claiming that such other causes exists, but I have no proof that one doesn’t.

Nancy Cartwright recently made the same point about current tests in space, using Gravity Probe B, for the predicted geodetic and frame-dragging effects caused by the mass of the earth as it flies through space.

Consider the Stanford Gravity Probe Experiment, which put four gyroscopes into space to test the prediction of the general theory of relativity that gyroscopes should precess due to coupling with space-time curvature. The Gravity Probe prediction about its gyroscopes was about as free of condition as any claim in physics about the real world could be. That’s because the experimenters spent a vast amount of time — over twenty years — and exploited a vast amount of knowledge from across physics and engineering. They tried to fix it so that all other causes of precession were missing; hence all the other causes would be, ipso facto, describable in the language of physics. Moreover if they had not succeeded and other causes occurred, then any that they couldn’t describe would make precise prediction impossible. If you can’t describe it, you can’t put it into your equations.

Same situation. There may be some other cause, or causes, that knock the gyroscopes akimbo that are unrelated to space-time curvature. Again, I make no claims that any such cause exists. I certainly haven’t evidence one does. But the point remains: everything witnessed at LIGO or Gravity Probe B is seen through the filter of theory, which is to say, on certain premises accepted as true.

There have been in science “discoveries”—and here the scare quotes fulfill their usual sardonic purpose—that were later proven to have been signals caused by something other than the proposed cause. Cold fusion immediately comes to mind. Before everybody understood what was happening, cold fusion was not well supported by non-observational evidence, which is to say, by theory. Yet some scientists believed what theory was there was true, others didn’t. The ones that didn’t held to other theories (about what was causing excess heat in some test tubes). Either way, theory was there.

Again either way, those theories might not have been as universal as proponents thought. Cartwright emphasizes that many theories, perhaps most, are only held ceteris paribus, meaning they are not universal and without exception. And that is just another way of saying the theories are incomplete.

Short way to think of this: observations and the equations (or theory) said to produce the observations can’t prove cause. Proving cause comes when we understand the true nature or essence of things involved. True or full understanding of thing’s essence does not come easy.

Update This article by Feser is apropos: ‘“Brute Fact View” according to which the universe simply exists without explanation, and that’s that.

50 Comments

  1. Proving cause comes when we understand the true nature or essence of things involved. True or full understanding of thing’s essence does not come easy.

    Can it really ever come at all?

    Future observations of gravity waves detection coupled with observed phenomena such as gamma-ray bursts or visible light from supernovae will add some confidence to this “confirmation.” But even then how can we say we will have a true of full understanding of gravity waves?

  2. I worked Dr Webber (Prof at the University of Maryland College Park) in the early ’70 looking for gravity waves. He had two large aluminum disks which collected all Electro-magnetic signals (light). We then focused on the very low frequency band, tossing out noise. Most of the signals were noise. Then we looked for pattern matching with the remaining signals from the two disks. Well it didn’t work. But we did demonstrate the utility of long baseline interferometry.

  3. Gary,

    I’m still stuck on what “spacetime” is, and how it can warp (relative to what?). I can repeat all the right sentences like an IFL-Science congregant but in the end I don’t understand it. Sometimes I wonder if it’s just a prettier, repacked aether (Perri-aether?).

    I also did feel some hesitation when I read that they needed to create all new analysis methods to separate the signal from the noise. Purpose-built statistical methods make me wary, because it seems like a good way to creep in biases and expectations.

  4. Beowulf, you have my appreciation and respect.

    As with all discoveries confirmation is needed. This result will need to be confirmed but based on the team involved and the improvements in sensitivity, it should hold up. Joe Weber, then of U Maryland, announced in the 1970s that he had detected gravitational waves in his “Weber bar” detector. An unnecessarily contentious review of his results was done over the next few years (in which I was a tangential participant on the theoretical side) and the conclusion finally was that he had not detected them. He was not accused of falsifying results, simply making a mistake. I have always felt a bit sorry he was subjected to the experience because he was a careful and talented experimentalist who pioneered work in this area for those to follow. It is an interesting example of how science works. The wiki page has a nice discussion and some references. He passed away some time ago but I believe he would be pleased with this result.

    https://en.wikipedia.org/wiki/Joseph_Weber

  5. James,

    Yes, most of us are Flatlanders with respect to grasping higher dimensions. I just try to stick with the standard analogies and not worry about the vagaries.
    From the popular press reports (to be taken with a ton of salt) I understand that they measured changes due to gravity waves in the interference patterns of the laser beams. What special statistical methods did that require?

  6. It is also pertinent that this is the first reported “direct” measurement of gravitational waves. There has been strong inferential evidence of gravitational waves from observations of objects such as close binary star systems in which the orbital decay strongly indicates that the energy being lost was in accord with predictions of gravitational radiation from such systems and not explainable by other known processes. That is not a direct measurement, only an indication of the explanatory capability of the theory. I was the theoretical hanger-on for a couple of those early observations. While my role was simply to calculate things, I did participate in the observations and sometimes wish I had chosen observational astronomy as a specialty. Working at an observatory was much like a wilderness experience, far from the maddening crowd and intensely demanding.

  7. Gary,

    I remember it being mentioned in passing in some articles before the news came out, but now I can’t find them. The news searches are all flooded with the post-discovery articles now, too.

    So I guess ignore what I said, since I can’t back it up!

  8. The experiment worked so well so fast, it should be repeated very soon and very often, or never again at all. How far along this moves our understanding of gravity we shall see. Confirm Einstein, confirm the lack of GUT. Maybe God’s in the gravity? I’d have egg on my face then, huh? 😉

    JMJ

  9. JMJ: You actually realized God might be in the gravity! Maybe God IS the gravity and we’re measuring God? There are so many possibilities with this!

  10. Question for the physicists here:
    “A team of scientists announced on Thursday that they had heard and recorded the sound of two black holes colliding a billion light-years away, a fleeting chirp that fulfilled the last prediction of Einstein’s general theory of relativity.” (from the New York Times) This means sound that occurred over a billion years ago made it across the galaxy and we recorded it?

    How could sound from a billion years ago be measured accurately when clapping hands in a control room causes a blip? Nothing interfered with the sound in a billion years? I have a difficult time understanding how everything can change but then when we look at quantum mechanics and space/time, nothing changes even over billions of years. Why is that so? What makes the sound immutable across time, but everything else changes?

  11. Sheri, the Gray Lady’s use of “sound” to describe a wave, with the accompanying term “chirp” is what might charitably be called poetic license. Sound is a longitudinal wave and requires a medium (i.e. other than vacuum) to propagate. And the point of the source being a billion years distant on our time line just means that the wave propagates with the speed of light–if you travelled on the wave, it would appear stationary to you, and the time would be infinitesimally small. Go to
    http://hyperphysics.phy-astr.gsu.edu/hbase/relativ/tdil.html
    and it will show you the effects of time dilation graphically.

  12. P.S. Briggs, you’ve written a fine article on this–stole my thunder; I was going to post and still may about this and Wigner’s “unreasonable effectiveness of mathematics in describing the natural world.”

    FAH and Beowulf–many thanks for your comments about the history of this. My physics was in magnetic resonance, so I didn’t get into the heavy stuff.

  13. Briggs,
    “Proving cause comes when we understand the true nature or essence of things involved.”
    Is this an insight or just a banal tautology? How do you determine the true nature or essence of things? I have this uncharitable image of you standing behind Becquerel as he puzzles over the fogged photographic plates and uranium salts and criticizing him for coming to unwarranted conclusions. Unworthy of me I know. A good review of the black hole chirping is here:

    http://motls.blogspot.ca/2016/02/ligo-discovers-black-hole-merger-12.html#more

    and here

    http://motls.blogspot.ca/2016/02/ligo-wows-bh-masses-3629-to-62-suns-51.html#more

    It explains the sound analogy as well.

  14. This article and the Nancy Cartwright’s were refreshing. Needless to say why.

  15. Joy, please tell me why. I can not get any useful content from the Nancy Cartwright article. I see only vague enuendo and breathless claims that lead to nowhere — no examples or evidence is presented.

  16. I will tomorrow.

  17. I would also like to thank you, you know, for the Grendel thing.

  18. … if you travelled on the wave, it would appear stationary to you, and the time would be infinitesimally small,

    which partially explains the popularity of surfing.

  19. I think I may be beginning to understand what Briggs’ discussions on cause and effect are trying to say but I am concerned about how useful a point it might be. As I understand the point, linking an effect to something that we can say “caused” it is more difficult than many lay people, scientists, and mathematicians may be using as a working definition of cause and effect. Further, technical procedures people use to allege cause and effect may not in fact be sufficient to maintain that a “cause” and effect relationship exists. Certainly there are many examples where a cause and effect relationship has been proposed when none actually existed.

    But it seems he goes further. So much so that I think he has essentially proposed that there is no such thing as cause and effect, unless one postulates a cause or causes that are impossible to determine physically.

    Let’s take an example. Suppose we have a lump of iron of 2 kg of mass and it is in deep space, very far from any other objects. Now suppose we measure its velocity and determine its velocity to be 10 meters per second in a direction we define as the X direction, and define the position at which we ended the measurement as x = 0. Now we arrange to zoom ahead of it, or have an assistant far out in front of its current position in the direction of its motion. In 10 seconds we have our assistant measure the position of the lump exactly (within some specified error bounds) 10 seconds from the time we measured the velocity originally. (We could worry about synchronizing clocks and passing information back and forth, but let’s just say we accept some specific level of measurement error.) Now our assistant will measure the position of the lump to be 100 meters in the x direction. There are a couple of cause and effect claims we might make. 1) The velocity of the lump at time zero “caused” its position to be different in the future or 2) its velocity of 10 m/s +- delta at time zero “caused” its position to be 100 m/s +-delta 10 seconds into the future. Both of these cause and effect statements would be based on our previous observations or assumptions about how object motion changes or not as a function of some list of known factors (eg. Newton).
    The question to ask Briggs here is does this demonstrate that the velocity at x = 0 and t = 0, coupled with complementary information about nearby matter (none), charge (none), etc. allow someone to say that the velocity “caused” the later position to be what it was? As I understand the Briggs position, his answer would be no. He may say that the Newtonian hypotheses and complementary information could “explain” the future position, but it would say nothing about what “caused” it. In this view, no physical theory (and possibly mathematical result) would ever be able to say anything about “cause” and effect. This seems to be because we can also think of other possible causes, such as the lump started at x =0 v=10, but then disappeared into an alternate universe where it spent an hour moving about, only to jump back at exactly the point in space and time at which we measured it (or the points at which we observed it along the way) and in the alternate universe it was manipulated by some principles or intelligence to make it look like it was explained by Newton et. al. Or it might be that it was interacting in some way, very hard to measure, with all the rest of the universe in just the right way to maintain its momentum constant. (There is a so-called Mach’s principle that hypothesizes this “cause” for local effects.) Under these conditions for ruling out possible “causes” and effects, alternative causes for any events we choose to connect in that way can never be ruled out.
    This seems to say that there is no way ever to determine cause and effect, except by hypothesis, essentially a supernatural hypothesis, or at least super-physical. This strikes me as ruling out any working standard for what we demand of an experiment or observation to establish cause and effect. Which essentially means there is no such thing as cause and effect in the physical world. Another way of saying that is that we will never know what “causes” things, we can only know what set of rules “explains” the results of some experiments.

    Another line of thought might maintain that the velocity we measured at x = 0 was actually “caused” by the ensemble of quantum mechanical states of the constituent particles of the lump and that ensemble spread the existence over all of space (or space-time if we wished) and when we measured at time t = 10, we simply collapsed all those individual possible states into the one that looked like it was “caused” by the initial lump velocity we measured (collapsed) at t = 0. Or more, that there might be a large number of alternate universes occupied by a large number of possible states for the lump and we just measured two of them, one at t = 0 the other at t = 10 and there was some chance, or probability, or likelihood, or whatever uncertainty about what we would measure.

    So the question to ask is: can anyone give one concrete example of a real physical experiment that has been or could be done that would demonstrate “cause” and effect? (Consistent with the Briggs position on cause and effect) Is there a way to articulate the probability that such an experiment demonstrated “cause” and effect, especially since we could likely generate as many possible other “causes” as we had time to spend? If the answer is no, then it may be that the restriction of what constitutes a “cause” is so severe as to rule out the notion of cause and effect for any practical purpose.

    The situation reminds me a bit of conversations I have had in the past with people who believe in a literal interpretation of (pick your favorite) creation account. Some people maintain that the age of the universe is some short period of time (say a few thousand or hundred thousand years). When one points out that observations of stars and galaxies indicate they are all moving away from each other at a certain rate and that extrapolating that backwards gives one a rough estimate of when they may have started out as on the order of a few billions of years, the reply is that the universe was created a short time ago, but the astronomical objects were created just so that they appear to have all emanated from some more central earlier position. Similarly, if one points out that isotopic dating of materials on earth suggests that the age of the earth is some large number of years, the reply is that the earth was created with an isotopic composition just so that it appears to be that old. In this way the cause of every observation can be hypothesized to be just so as to appear as if current states were “caused” by the physical phenomena. The end result is that there is literally no way to distinguish between possible “causes” simply in terms of explanatory power.

  20. Dear FAH,

    Something Matt wrote previously may help you understand him re cause. I’m going to go about citing Matt a bit differently (using contiguous parts of his own paragraph), as I also have my own pedagogical purpose.

    Here’s the first contiguous part of Matt’s paragraph.

    Knowledge of cause is the grasping of essence, of the natures and substantial forms of the objects under consideration. None of these things are material in themselves, but are universals above and beyond the material world.

    So, yes, Matt says that there is always some other, maybe even an infinite number, of explanations that fits the ‘materials’ (the facts as we have observed them). Matt has taken to using the ‘one banana’ idea (one group ate one more banana than the other group) to illustrate the point. Which you also see is correct. There always are alternative explanations to our observations.

    As an aside, this is why Matt says that ‘hypothesis’ ‘testing’ is silly and stupid. Come on, he says, ONE hypothesis — ONE??? — which we just ‘know’ is the real and true possible cause over against all the others, which is then compared via ‘statistics’ (with the scare quotes) to a ‘hypothesis’ that — and this may even be an even more meaningful distinction that Matt makes — compared to a ‘null hypothesis’ that says that our observed differences, which we clearly see, are literally, and I mean literally, due to NOTHING AT ALL.

    Since SOMETHING clearly caused the observed differences, the ‘null hypothesis’ is ridiculous and fallacious on its face, on its back, and on its side.

    So how do we EVER come to know the cause of a thing? Here is where, dear FAH, you may come to understand what even Matt does not: BOTH Plato, and then his pupil Aristotle, saw, as Matt writes:

    Thus to come to knowledge of cause is to understand universals

    And here we get to an absolutely fundamental difference between Plato and Aristotle. Plato DENIED that we could somehow ‘touch’, access, these universals. We CANNOT “come to knowledge of cause,” because we cannot, EVER, “understand universals.”

    Plato’s ‘Forms’ are FOREVER beyond our ken. We only have some small intuition of the Forms because, somehow, we ‘fell’ from the world of universals, and so, we yet retain a forever-inarticulate (but actual) sort of ‘memory’, an intuition, of their existence.

    So for Plato, all we could ever arrive at is ‘likely accounts.’ Plato also asserted that tying our accounts to Number was a good way to judge which accounts were more ‘likely’ — because Number seems to have the greatest similarity — is ‘closest’, as it were — to the Forms.

    So: for Plato, an individual scientist can confidently presume the existence of the Forms, the real causes of things, can accept that there can be such a thing as likely accounts, and accept that tying these accounts to Number is desirable where possible, but — here’s the key point — HE COULD NOT REMAIN A PLATONIST if he ever, ever, thought that he could have any more than a completely in-articulable, forever vague ‘intuition’ of the Forms themselves.

    In short, a scientist, it seems to me, can be fully a scientist, and a Platonist. But to be a Platonist, he MUST deny even the possibility of Aristotle’s fundamental assertion — that we CAN have access to the Forms themselves, and thus truly understand cause.

    However (which is perennially confusing to me) Matt persistently represents Aristotle’s position on this most fundamental matter as the only possible one:

    Thus to come to knowledge of cause is to understand universals, which we get through a form of induction. Induction is the immaterial “movement” from finite particularities to an infinite generality and is such that only rational creatures can accomplish it.

    To say it again, Plato would say that it is obvious that rational creatures — or any creatures — can NEVER accomplish an “immaterial ‘movement’ from finite particularities to an infinite generality.”

    I’m not saying that Matt is incorrect, exactly. I’m merely saying that Matt’s discussion of Cause is that of a confirmed Aristotelian.

    And I wish you, dear FAH, to understand that Plato, as well as Aristotle, has an account of these metaphysical matters.

  21. Dear William:
    I liked the first post of yours I saw, and I was intrigued by (and ordered) your book on Statistics, since the interpretation and pedagogy of statistics is one of my hobby-horses. I subscribed to your blog, but I was flooded by posts that were all over the place. I was about to discontinue my subscription.

    Then I saw your post on the LIGO experiment (detection of gravitational waves), and I was impressed by its accuracy and honesty. That is, whatever
    your social, political, religious, or ideological ideas might be (eg, creationist vs anti-creationist—this is only a shot in the dark), you were very careful not to prostitute scientific explanation to ideological agenda, as many unfortunately do, and turn it into patronizing or dishonest “scientific explanation.” Bravo! If the two of us were in a lifeboat with few provisions, I wouldn’t have to look at my back!

  22. John K, thank you very much for your incisive comments, with which, for the most part, I agree. Roger Penrose is a Platonist (perhaps because he is more a mathematician than a physicist) and I tend to follow him.

    You’re saying, if I understand you correctly, that there is a reality (i.e. universal forms?) underlying what we can quantitate and frame in science, and that we will never be able to understand completely this reality. This is very much what Bernard d’Espagnat says about quantum mechanics–there it overlays a “veiled reality” that is not penetrable.

    What is interesting is that general relativity and quantum mechanics have both been tested to the full experimentally, without any violations of the general theories being found. Yet relativity and quantum mechanics are not consonant–there are different assumptions in each about what measurements are all about. So that is a puzzlement.

  23. Ye Olde Statistician

    February 12, 2016 at 10:15 pm

    One way of determining whether you have a causal factor is to turn that factor on and off and see if the effect turns on and off. Not always possible, of course. Sometimes you can tune it up and down and see if the effect increases or decreases. Again, not always. But it is usually a useful marker of causation vs. correlation.

    Cartwright’s observations are basic: most scientific laws, even in physics are based on idealizations: perfectly elastic collisions, perfect vacuums, ideal gasses, point-masses, and the like. So two bodies fall at the same rate if we imagine them falling in a perfect vacuum, rather than an atmosphere or a swimming pool. Otherwise, rocks and leaves do not fall at the same rate.

    All this works well enough in physics and chemistry, although the uncertainties due to all the “ceteris” not actually being “paribus.” Ivar Ekeland noted in Mathematics and the unexpected that in calculating the path of a seven ball cannon in billiards would require taking into account the gravitational attractions of the spectators.

    But it also accounts for why this works less well in biology (where the matter is animate) and hardly at all in the “soft sciences” (where the subject matter can talk back. Two sodium atoms are essentially alike; two petunias never are.

  24. JohnK
    Thank you for your discussion. It clarifies for me the context in which cause is considered here. In particular, the focus seems to be on metaphysics rather than the world in which most of my thought resides, which is doggedly physical. Physics has evolved a great deal from the time of Aristotle and Plato, I think. The view I am content to spend my thoughts within is that a thing within the mental construct is defined by its measurement and is only defined if it can be measured. For something to be a valid (which really means useful within the discipline, not intrinsically good, bad, or indifferent) concept, it must be able to be measured quantitatively by any observer and any observer who performs the same measurement of that concept under the same conditions should get a quantitatively identical result. At least any observer who is not perversely trying to mis-define the measurement prescription. So it seems that I was looking for a connection of the discussion of “causes” to strictly physical considerations, which was not intended.

    I looked at the links you provided and they helped a great deal. First it seems to me that the notion that Briggs discusses of “cause” is intrinsically methaphysical, not physical. So that dealing with concepts that are “immaterial”, “universal”, “forms”, “particularities”, etc. immediately places the conceptual framework outside of the realm of the measureable physical world. The concept of “form” for example seems impossible to describe in terms of a quantitative measurement that multiple observers could perform and achieve the same result for the same “form”. That is perfectly reasonable I suppose within the realm of metaphysics, although I have no real background in metaphysics.

    If one deliberately removes metaphysics from consideration and focuses strictly on physics, I thought that Briggs’s discussions might relate to a practical framework for assessing cause and effect within the realm of physics, or chemistry or biology. It seems to me that rather than the words “cause and effect” in this context something like explanatory power might work, but that seems to have problems also. For example, physical scientists might have the results of some measurements, say they measure y as a function of some number of xsubi variables. By brute force, they can (and often do) simply perform a multivariate regression of the y’s on the x’s including perhaps first order interaction terms and get an expression with some number of regression constants. It may turn out that the residuals of the fit are small in some sense compared to the magnitudes of the quantities concerned so that any measurement that was a part of the set is quite close to the value calculated from the fit parameters. Not so much in physics, but in other disciplines people frequently say that the variance of the measurements is “explained” by the model (the model being the regressed expression), which if one wanted to one could call a theory. The problem with cause and effect comes in when someone tries to use those words with respect to the x’s and the y’s based on such a simplistic analysis. Many math techniques address taking sets of points out, examining subsets, fiddling with transforms, etc. but it seems nothing helps except more measurements, preferably with greater control of conditions, as Ye Olde suggests.

    It is a bit clearer if one considers tests of a theory. From what I gather about the notion of cause and effect considered here, meaning the metaphysical focus, cause and effect as discussed here is irrelevant to a physical theory. For example, relativity was proposed around the turn of the last century. It is based on just a couple of hypotheses about how the distribution of matter and energy results in (I hesitate to use the word “cause”) dynamic measurements. As time progressed various testing experiments were devised and then performed, generally with increasing accuracy over time as techniques improved. The measurement of gravitational waves discussed here is one such. As the results built up, the difference between the predictions of the theory and the results of experiments shrank considerably so that now many of the tests agree with the theory to parts in 10^10 or more and all but the most trivial consistent alternative physical (but not metaphysical) theories are ruled out. What it seems Briggs is saying is that such agreement of a theory with experiments is irrelevant to the notion of cause, because cause is intrinsically concerned with concepts for which measurement are impossible. That would mean the statement that some experimental result is “caused” by one of the aspects of the theory, for example the equivalence principle, would not be a useful statement. Not wrong, just not useful or applicable. It seems to suggest that the more apt expression would be that the theory “explains” the measurements.

    Then the practical question becomes, what is different about the way relativity “explains” the relevant measurements versus the way the brute force multivariate regression “explains” those hypothetical x and y above measurements. If the notion of “cause and effect” as discussed in terms of forms and particularities is ruled out because it is intrinsically about non-measureable quantities, then what is the most useful way to describe the relationship between a theory and experimental results.

    The martian box example seems simply a case in which the most effective measurements had not yet been devised and performed to find a sufficiently explanatory theory, from a physical perspective. Much like atomic structure was not understood until Rutherford did his scattering experiments. An incomplete or inadequate theory will often fail to explain experimental results. But from the physical perspective, it is not attributed to the “cause” being something immaterial or ethereal (i.e. intrinsically not measureable) but rather simply not yet measured or defined.

    The other place cause and effect appears in physical theories is what is usually called causality. The notion is that an event has associated with it a set of other events which can (but not necessarily will) cause it and a complement of that set, which is the set of events that cannot cause it. The measurement of the caused event is somehow, via a theory, explained by summing in some well-defined way the interactions between the events that can cause it and the caused event, quantitatively. In this context the causality propagates from the causing events to the caused event by a set of rules that comprise whatever theory is under consideration. The effects of the various forces are described and measured in this way. For example, if the period of a pulsed signal from an astronomical source is some magnitude, say 0.04 seconds, then it is common to consider that the physical size of the matter and energy distribution of the source has to be commensurate to the maximum speed at which causality can propagate, which so far seems to be the speed of light. If the size was larger than that, then physical effects could not propagate fast enough to “cause” the measured variation of the signal. For example, the first pulsar discovered, PSR B1919+21, had a period of 0.04 seconds and was at first nicknamed LGM-1 (for Little Green Men 1) entertaining the notion that it was an alien signal because it seemed hard to envision a stellar source small enough to produce such a short signal. Our sun has a radius of about 2.3 light seconds and the energy coming from the pulsar indicated much more output than our sun produces from an object about 1/50 the size. The causality argument forced the conclusion that the object had to be about much smaller than our sun but putting out much more energy than the standard stars of that size that we knew about would put out. It was a quandary how to explain the signal without invoking LGMs. But this difficulty was based on the notion of cause and effect and trying to determine how so much energy could be produced from so small an object. The causality argument led to the identification of candidate objects to “explain” the measurements, the result being the neutron star and a Nobel for Hewish
    .
    So it seems cause and effect is not absent from physical considerations, but it is strictly confined to the relationships between measureable quantities. This appears to be a little different and somewhat more restricted than the notions entertained here.

  25. Where is the matching GRB? And no neutrinos? Also, how does a black hole of those masses form? Let alone two together?

  26. Andyd,
    Very astute observation. Right now the identification of the source of the signal detected is based on 1) its coincident detection by widely separated detectors which indicates it is not a local spurious noise source and 2) the shape of the signal agrees with (is well “explained” by) numerical calculations of the output of a black hole collision with very specific size limits on the objects. Based on the sensitivity of LIGO there should be more such things collected over time, which will help scope the frequency of occurrence of black hole pairs and collisions, which have only been based on theory to date. One thing lacking so far, that will be very nice when it occurs, is matching up the source of the GW signal with something detected with other means, such as an optical, x-ray, or gamma telescope. That will ice the cake. The most studied object in the sky, SN 1054, otherwise known as the Crab Nebula, is the primo example of that icing. Measuring the velocities of the expanding gas clouds via Doppler shifts gives an expansion velocity and evolving that back in time says it should have exploded about 1000 AD and bingo! Chinese observers recorded a bright object in 1054 that appeared on July 4 of that year at just the right place in the sky and lasted for a couple of years. Even more cool, the object at the center was identified via radio astronomy as a 33 millisecond pulsar with characteristics well explained by a neutron star of a particular mass which fit with theoretical calculations of supernovae from stars of an appropriate mass. The combination of multiple observations allowed a detailed and consistent understanding of supernova. One could say it linked a lot of causes and effects, but maybe it is safer to say here that the theory just had a lot of ‘splaining to do.

  27. John K,

    As an aside, this is why Matt says that ‘hypothesis’ ‘testing’ is silly and stupid. Come on, he says, ONE hypothesis — ONE??? — which we just ‘know’ is the real and true possible cause over against all the other

    Actually, the impetus behind Hypothesis Testing wasn’t a bad idea. Yes, it DOES test the hypothesis but in this fashion: if X cause Y then there must be a correlation between X and Y so if there is NO correlation then the hypothesis that there is a causal relationship between X and Y can be eliminated. IOW: a quick check.

    One of the problem is that, over time, this has been perverted to imply the reverse, namely, that any correlation, proves cause (despite saying the opposite). Compounding the problem is there is almost always some correlation in any given data set. Anymore, no one seems to go beyond the QAD test if the press releases are any indication.

  28. As for cause and effect, all that “cause” really means is that (for X causes Y), a change in X implies a change in Y but not the reverse outside of mutual causation. Other than that, “cause” really has no meaning except perhaps as a way to organize an analysis.

  29. When I toured the Hanford LIGO I learned that at the end of each arm of the detector there is a large (several ton) carefully suspended rest mass to which a mirror is affixed. The mono laser splits and travels down each of the orthogonal arms. The rest masses move albeit ever so slightly when a gravity wave passes through them. The back up detector on the east coast allows triangulation. Simple. Right?

  30. It only took 100 years since the prediction to detect an event which occurred over a billion years ago around the time that muticellular organisims began to evolve. It’s taken a long long time for that miniscule ripple to arrive here. For sure it’s not alone.

  31. Nancy Cartwright is worth listening to.

    So is Ernst Mach:

    The goal which it (physical science) has set itself is the simplest and most economical abstract expression of facts.

    When the human mind, with its limited powers, attempts to mirror in itself the rich life of the world, of which it itself is only a small part, and which it can never hope to exhaust, it has every reason for proceeding economically.

    In reality, the law always contains less than the fact itself, because it does not reproduce the fact as a whole but only in that aspect of it which is important for us, the rest being intentionally or from necessity omitted.

    In mentally separating a body from the changeable environment in which it moves, what we really do is to extricate a group of sensations on which our thoughts are fastened and which is of relatively greater stability than the others, from the stream of all our sensations.

    Suppose we were to attribute to nature the property of producing like effects in like circumstances; just these like circumstances we should not know how to find. Nature exists once only. Our schematic mental imitation alone produces like events.

  32. Scotian, The (September 15) Article was refreshing given the current state, in my view, of physics and science with respect to how it is normally represented by repetitive regurgitation of the subject by the likes of Brian Cox. Science reporters normally boast certainty and fact where there is theory.
    YOS,
    “Two sodium atoms are essentially alike; two petunias never are.” for now anyway.

    Cartwright’s observations are not new. She happens to be reminding people who don’t want to be reminded about the state of science and the nature of theory. Perhaps she ought to publish in the mainstream media YOS I don’t think she was writing for you.

    JohnK, There’s a lot to be said for black and white, measurement, definition if you can get it.
    “…a thing within the mental construct is defined by its measurement and is only defined if it can be measured. For something to be a valid (which really means useful within the discipline, not intrinsically good, bad, or indifferent) concept, it must be able to be measured quantitatively by any observer and any observer who performs the same measurement of that concept under the same conditions should get a quantitatively identical result.”

    ‘would’ get a quantitively identical result” if they could hold everything els still and if they could measure.
    How would you measure pain?
    You can’t.
    “intrinsically good, bad or indifferent”? what else is there.

  33. A comment on Jim S.
    I too believe in the simplest explanation of physical phenomena. Or even bird watching. Ockham’s razor?
    Now why would anyone think that gravity waves would be manifest in electro-magnetic waves? When the first two protons were created and repulsed from each other (a strong force) at that same time gravity waves were created (a weak force). We will never detect them with two stationary disks. The correct experiment would be three disks: Arranged so that two on the extremities move in some fashion relative to the fixed disk between them. Then we could measure (the art of physics) the various consequences on the middle disk: Contraction, heat, movement(?), emissions (?). Who knows? Perhaps the emission and absorption of neutrinos?
    Much appreciation to FAH on the metaphysics of experience.

  34. Joy, I’m not familiar with Brian Cox but this statement by Cartwright “This 400-year-old image of the governance of nature is today being undermined by exciting new modes of understanding across the sciences …” is not exactly humble or lacking in certainty.

    “… physics and science with respect to how it is normally represented by repetitive regurgitation of the subject …”
    It is what it is. Should it be different each time?

    “Cartwright’s observations are not new.”
    She claims that they are. See my opening quotation above.

  35. James: “I’m still stuck on what “spacetime” is, and how it can warp (relative to what?). I can repeat all the right sentences like an IFL-Science congregant but in the end I don’t understand it. Sometimes I wonder if it’s just a prettier, repacked aether (Perri-aether?).”

    If you are still following this thread, I looked around at what I had but didn’t find anything at first that did not roll with heavy math. By coincidence my daughter is away at university and in her room I found a book I gave her once when she asked about relativity. It is by the man himself, Albert Einstein, a little book titled “Relativity the Special and General Theory”. It is available on Amazon in various forms and costs little. It has perhaps 5 equations at most and those mostly Lorentz transformations. The great thing about this book is that he carefully goes through his relentlessly physical thought process and displays his view on the link between math quantities and physical ones in everyday language.

  36. Joy: “intrinsically good, bad, or indifferent”

    I was trying to distinguish between a kind of moral or metaphysical notion of “goodness” versus a more workaday notion of utility as a tool. For example I think it is “good” to think about metaphysical and even spiritual things in whatever way fits your consciousness, but as a method of measuring things and making physical apparatuses work those thought processes may be hard to demonstrate as being useful in terms of quantitative improvement. Sometimes the two seem inextricable but I think there is a difference.

    An example of something that could be useful but still “bad” is what is going on at Mary Mount University right now. The board seems to have gotten into its head that they need to operate in a more businesslike fashion now. They brought in a president from the business world who is all about improving their “numbers” on things like retention and matriculation percentages. The new president thought to institute a survey for freshman and based on the results of that survey alone, dismiss some fraction of the respondents as being “not likely to succeed.” He also used the unfortunate analogy that “sometimes you have to drown the bunnies” for the good of the whole. He also fired two tenured profs who resisted his thrust. Well, while it may be “useful’ in terms of improving the numbers for the school, almost everyone thinks that what he has proposed is “bad” in terms of the traditionally nurturing, compassionate climate of the school as a Catholic university. It looks like the pushback is going to get rid of the president and re-instate the fired profs but it is not settled yet.

  37. Scotian: I didn’t notice the quote you picked out although I didn’t find her complacent. As for my remarks about Brian Cox, you won’t have been bombarded by the BBC’s chosen science man. I don’t think he’s up to the job. Check him out.

    FAH: I now see why JohnK included the phrase.
    When it comes to running public sector, companies/management I don’t think there’s a substitute for human minds acting without reference to “studies”. I don’t mean to rule out critical thinking or objectivity but it seems “studies” “surveys” in the wrong hands are misleading. In my field and depending upon the patient sometimes, physical science won’t help you one bit. Yet it’s considered heresy to say so by many. Discovering cause, or attempting to do so is an art not a science. You need the science of course the mix and it IS a MiX.
    I would also like to second what Beowulf said above about your contribution.

  38. Gravity waves would be standing waves. Thus perturbations would move along the wave faster than the speed of light. It might be challenge to measure those fluctuations. Ifn the wavelength stays the same wouldn’t that cause ripples in the Gravity field? Oh well, it is all above my pay grade.

  39. swordfishtrombone

    February 14, 2016 at 3:02 pm

    This comes across as really mean-spirited considering you’re dismissing a genuinely impressive, probably Nobel Prize-winning measurement of something predicted sixty years ago, just because it could possibly somehow maybe arguably be explained by something else.

    I read the linked Nancy Cartwright and Ed Feser articles. Cartwright’s weak nonsense comes across as a sort of feminist critique of physics’ exactitude while Feser’s is his usual reality-free pompous windbaggery. I don’t think Einstein would be impressed with either of these lightweights if he were still alive.

    This comment on Ed Feser’s blog made me laugh:

    “If you write at such length attacking scientists on the grounds of how science cannot prove things with absolute certainty, then unconditionally embrace ideas, which are many orders of magnitude less well supported, you cannot expect reasonable people to view you positively.”

  40. Ye Olde Statistician

    February 14, 2016 at 3:25 pm

    When the first two protons were created and repulsed from each other (a strong force)

    That two positively charged particles would repel each other is straight-up electromagnetism. E/M is indeed a strong force, but the strong force vis a vis protons is the one that binds them together in the nucleus of an atom, and this is simply another name for “electromagnetically, these boogers ought to fly apart like nobody’s business, but they don’t, so something must be holding them together.”

    In the Beginning, there was just a single Superforce. As the energy levels dropped, the individual forces ‘froze’ out of the soup. Gravity became separate at Planck scale energy, 10^19 proton masses; the strong nuclear force at unification scale, or 10^14 proton masses; and the weak force at Weinberg-Salam scale, 90 proton masses, which is about 10^2. At present energy levels, electricity and magnetism are unified into a single ‘force,’ and experimentally we have achieved Weinberg-Salam scales and created the electroweak force.

  41. Swordfish trombone,
    What makes you say Cartwright is a feminist? I wouldn’t know.
    Her article was so innocuous I don’t know how you could find it so offensive.
    It might interest you to check the date it was published.
    Furthermore, I found her almost impossible to follow in her lecture, she talks so fast and flat I was quite surprised that the same person wrote the article.

  42. Ye Olde Statistician

    February 14, 2016 at 3:45 pm

    you’re dismissing a genuinely impressive, probably Nobel Prize-winning measurement of something predicted sixty years ago, just because it could possibly somehow maybe arguably be explained by something else.

    Yet the old-time scientists went to great lengths to consider all other possible explanations for the body of facts, not just the one that they believe in. This was called by Galileo, “the work of the intellect” or negotiatio intellectus. We saw this in a minor key when that particle at CERN apparently traveled faster than light, but there it was a case of violating an already-accepted theory. Eventually, it came down to measurement error. In fact, a fair number of late modern discoveries have been laid to measurement error.

    Also, we know with the certitude of logic that any finite set of facts will always have more than one coherent explanation, so it is natural not to allow enthusiasms to run away. The gravity waves were detected some while back and those involved sought diligently for an explanation other than gravity waves to account for the readings. A number of people I know had to keep their lips zipped until they were at least pretty-darned-sure. The main problem now is that it is difficult to duplicate the experiment without some pretty heavy-duty experiment, and the publications process is biased against “mere confirmation” studies.

    Cartwright’s weak nonsense comes across as a sort of feminist critique of physics’ exactitude

    A more careful reading is in order. “Feminist,” forsooth.

  43. Above in a note to Andyd I mentioned how cool it would be if observations of EM sources coincided with GW sources. It looks like we may not have long to wait. Fermi reported a gamma ray burst that looks like it may (here cause and effect raises its ugly head again) come from the same event. The link to the Fermi preprint is here

    gammaray.nsstc.nasa.gov/gbm/publications/preprints/gbm_ligo_preprint.pdf
    Lubos Motl has a nice discussion on his blog here

    motls.blogspot.com/2016/02/fermi-saw-gamma-rays-04-seconds-after.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+LuboMotlsReferenceFrame+%28Lubos+Motl%27s+reference+frame%29

    where he points out that this could be the first confirmation (or not) that two zero rest mass fields (EM and gravity) travel at the same speed, heretofore known as the speed of light. Confirmation (or not) of that would help us think about our previous notions of causality.

    Note: I deleted the http:// from the front of the links. It seems the comment filter thinks it is spam.

  44. Scotian,
    ““… physics and science with respect to how it is normally represented by repetitive regurgitation of the subject …”
    It is what it is. Should it be different each time?”

    It is different each time. Similarity is epistemic, not onitc.

  45. YOS,
    “Also, we know with the certitude of logic that any finite set of facts will always have more than one coherent explanation, so it is natural not to allow enthusiasms to run away.”

    I 100% agree with this. But how does this square with the Thomist deductive-reasoning-from-premises that are posted continually on this site?

  46. Briggs

    February 15, 2016 at 10:06 am

    Jim S,

    Spend a moment and think about this: “I 100% agree with this.” You answered your own question to YOS.

  47. Ye Olde Statistician

    February 15, 2016 at 10:22 am

    A science article in the March 2016 issue of Analog Science Fiction and Fact by John Cramer entitled “Gravity with 4-vector potentials: a theory revolution?” discusses the then-anticipated LIGO experiments as possibly deciding between the Standard Model general relativity (GR) and the Gravity with 4-vector potentials model (G4v). Both models have hitherto made the same predictions relative to gravitational lensing, the perihelion of Mercury, etc. But they apparently differ on certain aspects of the expected gravity wave.

    The intriguing thing is that if G4v holds as its electromagnetic counterpart appears to hold, then there are no singularities in black holes, no need to postulate ‘dark energy,’ no need for a nonzero cosmological constant.

    http://www.npl.washington.edu/AV/altvw180.html

  48. Ye Olde Statistician

    February 15, 2016 at 10:32 am

    how does this square with the Thomist deductive-reasoning-from-premises that are posted continually on this site?

    Simple. It applies only to theories (propter quid) induced from data as explanatory stories. There is always more than one theory that can connect all the points. (An analogy: facts are the stars; theories are the constellations.)

    However, this does not apply to deductive conclusions, such as the theorems proved in mathematics.

  49. YOS, that’s a very interesting link to G4V theory. Many thanks! I’ve scoured the web to see if an “official” decision between “conventional” GR and G4V theories was forthcoming, and it seems to be implicit. That is to say, the official Caltech announcement (see
    https://www.ligo.caltech.edu/news/ligo20160211 )
    announces confirmation of Einstein’s prediction, so that would seem implicitly to shoot down the G4V theory, but we’ll see.

  50. I’ve found in the caption accompanying the outfit’s image gallery, the following: “As the plots reveal, the LIGO data very closely match Einstein’s predictions.”

    (see
    https://www.ligo.caltech.edu/image/ligo20160211a )

    So that seems to shoot down the G4V theory. An excellent example of how science works!

Leave a Reply

Your email address will not be published.

*

© 2016 William M. Briggs

Theme by Anders NorenUp ↑