Saturday, November 02, 2019

Have we really measured gravitational waves?


A few days ago I met a friend on the subway. He tells me he’s been at a conference and someone asked if he knows me. He says yes, and immediately people start complaining about me. One guy, apparently, told him to slap me.

What were they complaining about, you want to know? Well, one complaint came from a particle physicist, who was clearly dismayed that I think building a bigger particle collider is not a good way to invest $40 billion dollars. But it was true when I said it the first time and it is still true: There are better things we can do with this amount money. (Such as, for example, make better climate predictions, which can be done for as “little” as 1 billion dollars.)

Back to my friend on the subway. He told me that besides the grumpy particle physicist there were also several gravitational wave people who have issues with what I have written about the supposed gravitational wave detections by the LIGO collaboration. Most of the time if people have issues with what I’m saying it’s because they do not understand what I’m saying to begin with. So with this video, I hope to clear the situation up.

Let me start with the most important point. I do not doubt that the gravitational wave detections are real. But. I spend a lot of time on science communication, and I know that many of you doubt that these detections are real. And, to be honest, I cannot blame you for this doubt. So here’s my issue. I think that the gravitational wave community is doing a crappy job justifying the expenses for their research. They give science a bad reputation. And I do not approve of this.

Before I go on, a quick reminder what gravitational waves are. Gravitational waves are periodic deformations of space and time. These deformations can happen because Einstein’s theory of general relativity tells us that space and time are not rigid, but react to the presence of matter. If you have some distribution of matter that curves space a lot, such as a pair of black holes orbiting one another, these will cause space-time to wobble and the wobbles carry energy away. That’s what gravitational waves are.

We have had indirect evidence for gravitational waves since the 1970s because you can measure how much energy a system loses through gravitational waves without directly measuring the gravitational waves. Hulse and Taylor did this by closely monitoring the orbiting frequency of a pulsar binary. If the system loses energy, the two stars get closer and they orbit faster around each other. The predictions for the emission of gravitational waves fit exactly on the observations. Hulse and Taylor got a Nobel prize for that in 1993.

For the direct detection of gravitational waves you have to measure the deformation of space and time that they cause. You can do this by using very sensitive interferometers. An interferometer bounces laser light back and forth in two orthogonal directions and then combines the light.

Light is a wave and depending on whether the crests of the waves from the two directions lie on top of each other or not, the resulting signal is strong – that’s constructive interference – or washed out – that’s destructive interference. Just what happens depends very sensitively on the distance that the light travels. So you can use changes in the strength of the interference pattern to figure out whether one of the directions of the interferometer was temporarily shorter or longer.

A question that I frequently get is how can this interferometer detect anything if both the light and the interferometer itself deform with space-time? Wouldn’t the effect cancel out? No, it does not cancel out, because the interferometer is not made of light. It’s made of massive particles and therefore reacts differently to a periodic deformation of space-time than light does. That’s why one can use light to find out that something happened for real. For more details, please check these papers.

The first direct detection of gravitational waves was made by the LIGO collaboration in September 2015. LIGO consists of two separate interferometers. They are both located in the United States, some thousand kilometers apart. Gravitational waves travel at the speed of light, so if one comes through, it should trigger both detectors with a small delay that comes from the time it takes the wave to travel from one detector to the other. Looking for a signal that appears almost simultaneously in the two detectors helps to identify the signal in the noise.

This first signal measured by LIGO looks like a textbook example of a gravitational wave signal from a merger of two black holes. It’s a periodic signal that increases in frequency and amplitude, as the two black holes get closer to each other and their orbiting period gets shorter. When the horizons of the two black holes merge, the signal is suddenly cut off. After this follows a brief period in which the newly formed larger black hole settles in a new state, called the ringdown. A Nobel Prize was awarded for this measurement in 2017. If you plot the frequency distribution over time, you get this banana. Here it's the upward bend that tells you that the frequency increases before dying off entirely.

Now, what’s the problem? The first problem is that no one seems to actually know where the curve in the famous LIGO plot came from. You would think it was obtained by a calculation, but members of the collaboration are on record saying it was “not found using analysis algorithms” but partly done “by eye” and “hand-tuned for pedagogical purposes.” Both the collaboration and the journal in which the paper was published have refused to comment. This, people, is highly inappropriate. We should not hand out Nobel Prizes if we don’t know how the predictions were fitted to the data.

The other problem is that so far we do not have a confirmation that the signals which LIGO detects are in fact of astrophysical origin, and not misidentified signals that originated on Earth. The way that you could show this is with a LIGO detection that matches electromagnetic signals, such as gamma ray bursts, measured by telescopes.

The collaboration had, so far, one opportunity for this, which was an event in August 2017. The problem with this event is that the announcement from the collaboration about their detection came after the announcement of the incoming gamma ray. Therefore, the LIGO detection does not count as a confirmed prediction, because it was not a prediction in the first place – it was a postdiction.

It seems to offend people in the collaboration tremendously if I say this, so let me be clear. I have no reason to think that something fishy went on, and I know why the original detection did not result in an automatic alert. But this isn’t the point. The point is that no one knows what happened before the official announcement besides members of the collaboration. We are waiting for an independent confirmation. This one missed the mark.

Since 2017, the two LIGO detectors have been joined by a third detector called Virgo, located in Italy. In their third run, which started in April this year, the LIGO/Virgo collaboration has issued alerts for 41 events. From these 41 alerts, 8 were later retracted. Of the remaining gravitational wave events, 10 look like they are either neutron star mergers, or mergers of a neutron star with a black hole. In these cases, there should also be electromagnetic radiation emitted which telescopes can see. For black hole mergers, one does not expect this to be the case.

However, no telescope has so far seen a signal that fits to any of the gravitational wave events. This may simply mean that the signals have been too weak for the telescopes to see them. But whatever the reason, the consequence is that we still do not know that what LIGO and Virgo see are actually signals from outer space.

You may ask isn’t it enough that they have a signal in their detector that looks like it could be caused by gravitational waves? Well, if this was the only thing that could trigger the detectors, yes. But that is not the case. The LIGO detectors have about 10-100 “glitches” per day. The glitches are bright and shiny signals but do not look like gravitational wave events. The cause of some of these glitches is known. The cause of other glitches not. LIGO uses a citizen science project to classify these glitches and has given them funky names like “Koi Fish” or “Blip.”

What this means is that they do not really know what their detector detects. They just throw away data that don’t look like they want it to look. This is not a good scientific procedure. Here is why.

Think of an animal. Let me guess, it’s... an elephant. Right? Right for you, right for you, not right for you? Hmm, that’s a glitch in the data, so you don’t count.

Does this prove that I am psychic? No, of course it doesn’t. Because selectively throwing away data that’s inconvenient is a bad idea. Goes for me, goes for LIGO too. At least that’s what you would think.

If we had an independent confirmation that the good-looking signal is really of astrophysical origin, this wouldn’t matter. But we don’t have that either. So that’s the situation in summary. The signals that LIGO and Virgo see are well explained by gravitational wave events. But we cannot be sure that these are actually signals coming from outer space and not some unknown terrestrial effect.

Let me finish by saying once again that personally I do not actually doubt these signals are caused by gravitational waves. But in science, it’s evidence that counts, not opinion.

155 comments:

  1. Thank you very much for this post, I was not aware of that lack of confirmation of gravitational waves.

    I must say there is more to complain about this, beside the circumstancial fact that a specific finding was not confirmed independently. It is the lack of an upfront method to confirm the findings of any experiment in the first place.

    In my opinion, the whole protocol of how experiments should be carried on is incomplete without setting up a procedure of how check it, and this protocol should be shared so that the scientific community may approved it, or improve it.

    I hardly believe how this is not the standard procedure in experimental Science. We are in 2019, please.

    ReplyDelete
  2. I wrote this to the LIGO Team shortly after the first Gravitational Wave, the response was we cannot confuse Seismic with Gravitational Wave Detections

    "As an interested engineer and member of the public in regards to the claimed detection of gravitational waves, I was interested to know more about the effect of the magnitude 6.6 earthquake that occurred under the Gulf of California on 13th September 2015 08:14:12 UTC with much weaker aftershocks (around magnitude 3) occurring after that including on 14th September 2015 (the closest earthquake in time to the gravitational wave detection and above magnitude 3 occurred at Mon, 14 Sep 09:37 UTC). Since the Gulf of California is thin water body that is at certain points along its length appears to be almost equidistant in terms of surface distance between both LIGO detectors (Hanford and Livingston), I was wondering if these small aftershocks set of any extended noise and resonances in and around the Gulf of California that then becomes visible to both LIGO detectors in spurious synchronicity?

    Therefore it would be helpful if you could point me and others to whether or not and in what manner the Gulf of California earthquake on the 13th September 2015 and the subsequent aftershocks were seen by the LIGO detector as distinct events or spuriously correlated events across the two LIGO detectors even if they are outside the main frequency band used to detect gravitational waves.

    In your paper ”Environmental Influences on the LIGO Gravitational Wave Detectors during the 6th Science Run” from September 2014 you mention an earlier 5.9 magnitude earthquake in Peru that is cited as part of the evidence that you know how to characterise the environmental sources of noise that disturb the detectors ability to observe gravitational waves. Is a more recent paper examining the effect of the Gulf of California 6.6 magnitude earthquake and the subsequent aftershocks on the IGO detectors, that would be much more relevant to the time window most relevant to the gravitational wave detection claim?"

    ReplyDelete
    Replies
    1. James Arathoon: >we cannot confuse Seismic with Gravitational Wave Detections

      That's like saying "You can't fertilize your lawn too much." Does it mean it is impossible to over-fertilize your lawn, or does it mean be careful you don't over-fertilize your lawn?

      :-) I guess the follow-up is, "Is your certainty founded on the single example you described in your 2014 paper?"

      Delete
    2. Wrote vs. Said (minus immaterial changes):

      Back to my friend in the subway ->
      Back to my friend on the subway

      The first direct detection of gravitational wave ->
      The first direct detection of gravitational waves

      This first signal measured by LIGO looks ->
      The first signal measured by LIGO looks

      For a reference on these quotes. ->
      Could be a link "(Quote references)".
      Audio says check the links, but you deleted
      that part without providing a link.

      Nobelprizes -> Nobel prizes

      Delete
    3. Dr Castaldo,

      Thanks, I have fixed this. The reference for the quotes is linked to in the paragraph, I just forgot to delete the sentence saying to check the info below the video.

      Delete
    4. James Arathoon,

      Since you are an engineer, I'd like your opinion on what I think of as an engineering objection to the LIGO claims:

      1) The scale of the individual mirror pairs of the interferometers is 10^3 meters.

      2) The wavelength of the employed laser is of order 10^-7m.

      3)The atoms of the mirrors are of order 10^-12m and are chaotically vibrating on a scale @ 10^-11m.

      4)From 3 it follows that there is no determinable fixed length separation between the mirrors at the atomic scale.

      5) How then can the system detect a fixed-length variation in the separation between the mirrors at the claimed scale of 10^-20m, approximately 100 million times smaller than the atomic scale of the mirror surfaces? In other words, the system as built should not be sensitive to the scale of the claimed signal detection.

      So, from an engineering point of view, do you think that is a valid objection and if not, why not? Thanks.

      Delete
    5. "
      5) How then can the system detect a fixed-length variation in the separation between the mirrors at the claimed scale of 10^-20m, approximately 100 million times smaller than the atomic scale of the mirror surfaces? In other words, the system as built should not be sensitive to the scale of the claimed signal detection."


      First, read the literature. This is a FAQ and it is easy to find answers to your questions.

      Second, do you really think that no-one in the entire LIGO team ever thought of your elementary objection, and they were just too stupid to notice it or just built the detector anyway to prevent losing face? Really?

      Delete
    6. Phillip,

      First, I've read the literature, probably more of it than you have. If you have an answer to the question, spit it out. If you don't have an answer but you think one is out there somewhere, go find it. I'm not your research assistant.

      Second, I think the LIGO team consists mostly of scientists like you, primarily mathematicians who, when it comes to physics, can't think their way out of a paper bag without an equation to guide them.

      Delete
    7. @bud rap: "I think the LIGO team consists mostly of scientists like you, primarily mathematicians who, when it comes to physics ..."

      And I think the LIGO team has a lot of scientists who are very good at the experimental/observational side of physics. Scientists who are very hard-headed, and skeptical of theoretical paper bags.

      Delete
    8. Jean Tate,

      ...the LIGO team has a lot of scientists who are very good at the experimental/observational side of physics.

      Maybe. You're certainly entitled to that opinion. But on the merits, I don't see it.

      Delete
    9. "First, I've read the literature, probably more of it than you have. If you have an answer to the question, spit it out. If you don't have an answer but you think one is out there somewhere, go find it. I'm not your research assistant."

      First, while answering good questions is fine, answering FAQs is a waste of time. I'm not your research assistant.

      "Second, I think the LIGO team consists mostly of scientists like you, primarily mathematicians who, when it comes to physics, can't think their way out of a paper bag without an equation to guide them."

      Second, I have a degree in physics, not mathematics. What are your qualifications?

      Delete
  3. Now have we actually found the Higgs boson? I mean have we confirmed its existence?

    ReplyDelete
  4. The observation of the neutron star merger in the electromagnetic spectrum was a clear independent verification of LIGO and VIRGO, because the GW signal pointed to the patch in the sky where the astromers observed the whole range of electromagnetic signals consistent with a the merger that were sent after the GW were sent out.

    >> The problem with this event is that the announcement from the collaboration about their detection came after the announcement of the incoming gamma ray. Therefore, the LIGO detection does not count as a confirmed prediction, because it was not a prediction in the first place – it was a postdiction.

    No, it is not a postdiction, because the GW teams determined the patch in the sky WITHOUT using the information from the gamma ray observation and according to rules fixed BEFORE the gammay ray observation. Maybe the gamma ray observation helped significantly to narrow down the patch of the sky (together with the GW signal or not) and the astronomers were able to catch the whole EM range, but that is irrelevant.

    But as I told a friend of mine, I only fully believe LIGO / VIRGO if they find a second and third such event.

    ReplyDelete
    Replies
    1. "No, it is not a postdiction, because the GW teams determined the patch in the sky WITHOUT using the information from the gamma ray observation and according to rules fixed BEFORE the gammay ray observation."

      You are missing the point. I know full well that that is what they say. But since no one besides members of the collaboration actually knows what went on, this is not independent confirmation. The only thing we know is that they had the information and they knew where to look and what to look for.

      Delete
    2. Hi Tom and Sabine,

      The 170817 detection timeline is indeed somewhat unusual. LIGO's internal single-detector trigger came 6 minutes after Fermi, instead of real-time (due to a network outage according to LIGO; GW data was buffered at Caltech).

      LIGO's first skymap came 4.5 hours after Fermi and Integral (due to a major glitch at the exact moment of the merger according to LIGO. How the signal was recovered beneath the 1000x stronger glitch is baffling even experts.)

      Interestingly, ESA also had a simultaneous network outage, and Integral data was buffered at Kiruna ground station in Sweden.

      VIRGO actually had no signal, but because just three days earlier VIRGO claimed its first black hole signal, LIGO assumed the 170817 source had to be within VIRGO's blindspot.

      We also know NASA had better data than they published, because a NASA scientist uploaded an improved GRB skymap that got strangely deleted within 30 minutes. Later the first LIGO skymap was also uploaded by a NASA scientist.

      The strongest arguments in favor of LIGO is their distance estimate and the smaller size of their skymap (inside NASA-ESA triangulation), though critics say these could be due to LIGO's cosmographic optimization algorithms (doubtful).

      Keep in mind that if powerful telescopes such as SWIFT detect a GRB they can usually locate its exact source within just 5 minutes. But they couldn't catch the signal on that day (too weak or out of view).

      So as you say, it would be really important to see at least one successful automatic (i.e. real-time) counterpart prediction by LIGO, to exclude any and all doubts.

      Delete
  5. Gravity Spy is the name of the citizen science project.

    The Hulse and Taylor observations have been independently confirmed, at least in the sense that other binary neutron stars/pulsars have been observed and their estimated orbits (etc) are decaying at a rate consistent with gravitational wave radiation (GWR). There is one double pulsar for which many more effects have been observed (i.e. not just orbital decay), effects consistent General Relativity.

    KAGRA, in Japan, is a third GWR observatory of similar design to LIGO and Virgo. It has had many delays but should be online by the end of the year (at least for engineering trials). There is one with a similar design being planned/built in India, IndIGO.

    There have been a number of resonant mass GWR detectors, but none detected anything, by the independent validation/replication measure.

    LISA is a space-based GWR observatory; a lot of work has been done, and it could well go "live" by 2034.

    Finally, the International Pulsar Timing Array has been running for some years now; one of its goals is to detect GWR. So far, nothing.

    Important: each (type of) GWR detector/observatory/observation is sensitive to a certain range of frequencies.

    ReplyDelete
  6. “ We are unable to solve Einstein’s equations exactly for the case of two black holes merging,” says Neil Cornish at Montana State University, a senior figure among LIGO’s data analysts. Instead, the analysts use several methods to approximate the signals they expect to see." I found it interesting that wavelet-analysis is being utilized. A blanket statement such as "I think that the gravitational wave community is doing a crappy job justifying the expenses for their research. They give science a bad reputation" is surely unfounded. LIGO has done an exemplary job educating the public. Science gets a 'bad reputation' (if it does so at all) when myths are promulgated by reputable physicists, myths such as their own variant interpretation of Hugh Everett's 1957 paper. Reading the "Review of Particle Physics" we find: "Sometimes large changes occur. These usually reflect the introduction of significant new data or the discarding of old data." (Phy Rev D, page 18, August 2018).

    ReplyDelete
    Replies
    1. The Einstein field equations are only exactly solvable for one body. Unlike Newtonian mechanics, the two body problem is not generally integrable. This is because for two masses in a mutual orbit the generation of gravitational radiation acts as a 3rd body. For two bodies in an orbit far apart in a near-Newtonian orbit the perturbation of gravitational waves is small and so perturbation methods can be used. For a system that is losing energy in the emission of gravitational waves this becomes more difficult. Then up to the moment of actual merging the equations become impossible. Then with ring down things become integrable again.

      This "by eye and hand curve fitting" is just a way of making a physically plausible statement that gravity wave signatures at the exact merger should not be so extraordinary as to negate everything. In other words there should not be some extreme catastrophe that say destroys the black holes into energy or a naked singularity or some other thing. The idea is that nature is not malevolent, should we say.

      Delete
  7. Sabine, you wrote: "...an event in August 2017. The problem with this event is that the announcement from the collaboration about their detection came after the announcement of the incoming gamma ray".
    How is this possible?. Wikipedia says that gamma ray (GRB170817A) lasted only 2 seconds, while GW170817 lasted around 100 seconds.
    About particle physics, I agree with you that 40 billion dollars are too much to be spent in the new collider; given that SUSY particles might exist much far away from that future collider range. Alternate solutions must be found.
    I do not agree that would be a good idea to spend 1 billion euros in climate predictions. Extreme weather is, so far, not linked to climate change (nor their models). It is better, for example, not building in risky places for floods.

    ReplyDelete
    Replies
    1. Antonio,

      I don't understand the question, sorry. Maybe it helps if you look at Wikipedia again where it also says the gamma ray signal began 1.7 seconds after the GW merger signal.

      Delete
    2. Let's see if (using this) the link to LIGO's site appears: www.ligo.org /science/Publication-GW170817GRB/index.php
      Anyone interested can read true science in there.
      In here, you may believe Sabine when she tells you that "LIGO detection does not count as a confirmed prediction".

      Delete
  8. The problem you raise with curves showing increased frequency with increasing amplitude occurs right at the merger. The curve right up to the final fraction of a second can be computed and the later ring down can be computed, but the physics right at the merger is not known. I have been doing a lot of work with looking at quantum hair on two merging black holes. This is relevant when the horizons are extremely close. This runs into a lot of difficulties. Approximations have to be appealed to. Further, general relativity does not predict changes in topology. The point where horizons merge is where GR simply breaks down. The reason the LIGO collaboration has no algorithm for this is there is no known physics right at the merger.

    What the LIGO collaboration did with the predictions is to do some curve fitting and make some half-way reasonable guesses. This is in a sense what might be called “Afro-engineering,” but in some sense you have to respect Afro-engineering for it represents a willingness to press on with a problem with some “bet” the issues there are not so serious as to totally negate the results.

    How could we solve this problem with the merger? I think it will be quantum gravitation. Further, signatures of quantum hair physics, BMS symmetries and related physics could be found in the future.

    ReplyDelete
    Replies
    1. Dr. Sabine Hossenfelder:
      Lawrence Crowell:

      "Afro-engineering" is a racist and pejorative term; whether you claim you "admire" it or not. I will presume you were not aware of this. "Jury-rigged" is a suitable non-racist substitution meaning the same thing, a make-shift solution.

      See: https://en.wikipedia.org/wiki/Jury_rigging

      From the article, with [references], not my own words: Afro engineering (short for African engineering)[iv] or nigger rigging[v] are racist, pejorative terms for shoddy,[vi] second-rate workmanship,[vii][viii] with whatever materials happen to be available,[ix] or to describe a fix that is temporary, done quickly, technically improperly, or without attention to or care for detail.[x][xi][xii] "Nigger-rigging" originated in the 1950s;[iv] the term was euphemized as "afro engineering" in the 1970s.[v][xi]

      Delete
    2. I was not quite aware of this. It was sort of a term I conjured up.

      Delete
    3. I can't say I have ever come across the term afro-engineering. But it already sounds to me as suspect.

      Lawrence Crowell: I've heard that black holes have no hair, but what on earth is 'quantum hair'?

      Delete
    4. Quantum hair is just quantum information on a stretched horizon just above the mathematical horizon of GR. In a BPS black hole with extremal condition, it forms what content can potentially be observed.

      Delete
  9. Let's say we have a competent person with adequate physical knowledge and with absolutely now astronomy, astrophysics, cosmology etc knowledge.
    I can imagine such a person would still have some legitimate doubts before accepting LIGO reports without reservations.

    Existence of black holes
    Properties of black holes
    properties of gravitational waves
    measurement apparatus
    software algorithms
    interpretation of results
    psychological biases stemming from academic prestige
    psychological biases stemming from big cost (over $1 billion)
    why are no Ligo events from our galaxy?
    a lot of retracted events
    plain error in a long chain from detector to published article

    Amd then is John Ioannidis who famously found that 'There is increasing concern that most current published research findings are false'

    ReplyDelete
    Replies
    1. Would such a person have legitimate doubts accepting reports about FRBs (fast radio bursts)? GRBs (gamma ray bursts)?

      If not, why should the LIGO reports be treated differently?

      If no such doubts, what makes LIGO different?

      Delete
    2. Jean Tate,

      What makes LIGO different??? OK. For the record, FRBs and GRBs are widely observed types of electromagnetic radiation. EMR is itself ubiquitous throughout the cosmos. Everywhere scientists imagine spacetime to exist, observation tells us there is an omnidirectionally sourced flow of EMR.

      Gravitational waves, in contrast, are theoretical phenomena that no one has ever directly detected, with the possible exception of LIGO's dubious and increasingly beleaguered claims. Big difference.

      Delete
    3. @bud rap: um, no.

      GRBs were quite a mystery for some time, then it was found that they comprise at least two distinct sub-classes (and there continue to be papers postulating a third, or even fourth, sub-class). True, today they are widely observed.

      FRBs are, today, not yet widely observed. And getting to the point of unambiguously identifying them took a lot of effort (there are terrestrial sources of RFI - radio frequency interference - that look very much like FRBs).

      Yes, no one has yet directly detected GWR, other than the LIGO and LIGO+Virgo teams. However, it has most certainly been indirectly detected.

      The main difference, from an astronomer's point of view, is that GRBs and FRBs were first observed (astrophysical models came later), while GWR transients were expected, per a remarkable well-tested theory (General Relativity), and direct detection came later.

      Delete
    4. Jean Tate,

      The point was that there is a big difference between unambiguous, direct, empirical evidence and a model-dependent inference with extremely minimal empirical support. To be clear, I'm speaking only of the observational facts, not their subsequent interpretation as GRBs and FRBs. (I'm not disputing those interpretations, they are just irrelevant to the point being made.)

      You do raise the interesting question of whether General Relativity predicts gravitational waves. The answer to that question is yes - or no, depending on whether you interpret the spacetime of GR as a substantival or relational concept. The modern interpretation is substantival; spacetime is a physical substance that affects, and is affected by, matter.

      There is, unfortunately, no empirical evidence supporting this substantival interpretation. It is, in essence, an assumption of the model. If that assumption is wrong, as the negative empirical evidence suggests, then GR does not predict gravitational waves.

      That in turn, conforms well with the history of GW detection, in which the scale of the expected detection has been revised steadily downward to the extent that we are now attempting to measure GWs on a sub-atomic, quantum scale (10^-20), while employing a souped-up version of 19th century classical machinery. And the wheels are coming off that fiasco. It's just like OPERA and BICEP2 only in sickening slow motion.

      We cannot continue to do science this way, chasing theoretical rabbits down theoretical black holes. You cannot move the goalposts after every predictive failure because of an unshakable belief that your theoretical models cannot be wrong. That approach has failed in the past and is failing now - spectacularly in this case. It is probably impossible to overestimate the damage being done to science.

      Things have to change, but given the socio-economic imperatives of Big Science with its insatiable need for ever larger, more elaborate, and more expensive experimental setups chasing ever diminishing returns, there is a very real possibility that science, as a discipline, will be so tarnished in the public imagination, as to effectively usher in a new dark age.

      How to recover from this self-induced mess should be the most important topic in fundamental physics. As it is, with the notable exception of Sabine, Lee Smolin and a handful of others, the scientific community does not even acknowledge that the problem exists.

      Instead we get ongoing "research" into the nature of "dark matter", despite the fact that after 40 years of searching the empirical results are unequivocal - there is no "dark matter" in the cosmos. This widely deployed approach, wherein a consensus "model" is impervious to negative empirical evidence, is conceptually unscientific and at the heart of the crisis in physics.

      Delete
    5. Thanks bud rap, that's a cogent response.

      I'd like to focus on the astronomy first and astrophysics second.

      A rather large number of contemporary astronomical observatories (a blanket shorthand) rely for their operation on "GR Rules, OK?". A spectacular, recent example: GAIA ... the data pipeline includes "corrections" for apparent "displacements" due to gravitational lensing (or bending) by the Sun, over the whole sky. And for sight-lines close enough, all solar system planets and large moons. And even dwarf planets (Pluto, various TNOs, even Ceres; though these sight-lines are barely above the objects' surfaces). Yes, GAIA makes direct detections ... but these are converted to coded radio signals, and beamed back to receivers here on Earth. And these signals make sense only if you accept as valid the physics and engineering the mission publishes. Where to draw the "direct, empirical" line?

      "There is, unfortunately, no empirical evidence supporting this substantival interpretation. It is, in essence, an assumption of the model. If that assumption is wrong, as the negative empirical evidence suggests, then GR does not predict gravitational waves."

      I am unfamiliar with even the term "substantival", much less "substantival interpretation".

      With respect to GR, moving away from astronomy (to some extent), how valid are Pound-Rebka type experiments (done here on Earth, in a lab)? Atomic fountain clocks, used as exquisite gravimeters which incorporate GR time delays (ditto)? The APOLLO Collaboration's lunar ranging experiment (half ditto)?

      How reliable are the Hulse-Taylor observations? And all those of other binary pulsars?

      In short, if one questions the "substantival interpretation", what of the many different kinds of tests of GR become reliant on this questionable interpretation?

      Delete
    6. Jean Tate,

      And these signals make sense only if you accept as valid the physics and engineering the mission publishes. Where to draw the "direct, empirical" line?

      I don't understand the point. Radio transmission signaling has been a reliable and well understood technology for over 100 years. Do you question whether the friend you are talking to on your cell phone actually exists and is saying what they're saying because you know...technology is you know...???

      On the GAIA end, the system could be flawed and transmitting inaccurate data, but that's an engineering problem and if there is such a problem, when anomalous results are observed, they will be eventually be understood as such and likely be traced to the source and corrected. At the margins, there will always be room for error, that does not in any way undermine the general, foundational relevance of empiricism to science.

      Without empiricism there is no science. The empirical line needs to be drawn at the invocation of unobservable entities and events to account for empirical observations, not somewhere in a well understood causal chain that transmits empirical observations for our consideration.

      For the purposes of this discussion, substantival spacetime can be thought of as the spacetime of Wheeler (Spacetime tells matter how to move; matter tells spacetime how to curve.). In other words, spacetime can have a physical effect on observables and be affected by observables in turn.

      The problem with this view is that spacetime itself is not an observable. Which is to say, there is no direct empirical evidence for such a substantive spacetime - it is only an assumption of the standard model.

      This means that spacetime better fits the description of a relational concept, one like temperature, that provides a descriptive framework for discussing relationships between observables but does not represent an observable "thing" in itself.

      If you have a tolerance for verbose philosophical discourse you can find a historically based elaboration here: https://plato.stanford.edu/entries/spacetime-theories/

      Questioning the substantival interpretation has no more bearing on the observed successful predictions of the GR formalisms than the various QM interpretations do on the successful QM predictions.

      In the case of the Hulse-Taylor observations, it is the gravitational wave interpretation of the results that is called in to question, not the results themselves. This situation applies across the board with respect to all GR predictions that are borne out by observation; the formalism works but the spacetime as causal-entity view is no longer valid when spacetime is understood as merely a relational concept.

      That, in turn, makes clear that the "mechanism" behind the gravitational effect remains unknown and therefore a proper subject for renewed, fundamental research. However, as long as the metaphysical curved, waving spacetime concept clutters the scientific imagination, little effort and even less progress can be expected in that area.

      Delete
    7. @bud rap: Thanks.

      Going back to this: "There is, unfortunately, no empirical evidence supporting this substantival interpretation."

      For GWR, could we realistically hope to have any empirical evidence supporting that interpretation, within the next century or so, say?

      I like this: "in a well understood causal chain that transmits empirical observations for our consideration" :)

      In LIGO, big chunks of matter, held in place by very thin wires, were observed (thus empirical, right?) to have moved. In a particular way. The motion can be modeled (analytic functions and digital analyses) and it seems consistent with what you'd expect a particular kind of GWR transient to be (yeah, error bars, caveats, etc apply).

      Presumably the motion of the big chunks of matter is empirical, right?

      Delete
    8. Jean Tate,

      For GWR, could we realistically hope to have any empirical evidence supporting that interpretation, within the next century or so, say?

      Since GWR depends on the substantival interpretation and theorists, because of predictive failures at higher scales, have pushed the predicted scale well-down into the quantum regime at 10^-20m, I don't see how there can be any unambiguous empirical evidence, given that direct observations are not currently available at such a scale and it is highly unlikely they ever will be. (See my question to James Arathoon in this blog post thread.) If things continue as they have for the last 50 years the predictions will no doubt be down at the Planck scale by the end of this century.

      It's long past time to stop. The efforts to detect gravitational waves have failed repeatedly, therefore it is a reasonable scientific conclusion to say that GWs do not exist in empirical reality, and that conclusion is consistent with all the non-observations of substantival spacetime.

      Presumably the motion of the big chunks of matter is empirical, right?

      Yes, it is, but at what scale is that motion actually taking place? That is where the problem lies. The actual physical detection takes place somewhere within a few orders of magnitude of 10^-7m, given that as the wavelength of the employed laser.

      We also know that a test signal can be injected into the causal chain well above the 10^-20 threshold that mimics the expected signal detection at the observed detection scale. Therefore, it is not possible to claim an unambiguous detection of a 10^-20m GW signal, given the machine sensitivity issue.

      Given all the other well-documented irregularities surrounding the LIGO claims and some unambiguous predictive failures, it seems LIGO continues on maily as a function of institutional inertia. According to @Guest in his 5Nov 8:47am post:

      LIGO had one other good opportunity to prove its functionality, and that is continuous GWs from fast-spinning pulsars. Almost everybody expected these, but again LIGO found nothing in all 221 cases.

      That represents an unqualified failure, of prediction or detection. Either the GWs aren't there or LIGO can't detect them. In either case, the results suggest that the whole LIGO enterprise is of dubious scientific value.

      Delete
    9. @bud rap Thanks.

      We may have reached the point of "agreeing to disagree", at least at the level we can continue discussion in the comments of this blog post.

      Just a few "tidy up" things though.

      Searches for GWR other than LIGO etc: do you think eLISA (and similar) are the same, in that direct detection of GWR is impossible?

      How about the International Pulsar Timing Array?

      Re: "That represents an unqualified failure, of prediction or detection. Either the GWs aren't there or LIGO can't detect them. In either case, the results suggest that the whole LIGO enterprise is of dubious scientific value." Such is astronomy!

      A seasoned astronomer might respond, re the expectation of detecting GWR from fast spinning pulsars, with "Well, duh! Theoretical models bite the dust. Again. Like that have never happened in the history of astronomy, right?"

      A cautious astrophysicist might respond a little differently, perhaps "Well, duh! So we don't know everything there is to know about neutron stars. Like that never happened before, right?"

      Delete
    10. Jean Tate,

      Searches for GWR other than LIGO etc: do you think eLISA (and similar) are the same, in that direct detection of GWR is impossible?

      It is a matter of basic logic that it is impossible to detect something that doesn't exist. Fifty years of negative empirical evidence strongly suggests that GWs don't exist, so yes, it will be impossible to detect them.

      That lets the seasoned astronomer and cautious astrophysicist off the hook; they needn't carry water for a failed interpretation of General Relativity. The reason we haven't detected GWs is simple - they aren't there. Building a bigger, better, more expensive detector won't solve that problem.

      Delete
    11. Gravity waves exist in general relativity for much the same reason electromagnetic waves exist in Maxwell's theory. In a post-post-Newtonian expansion general relativity gives a Maxwell-like equation. The main difference is a factor of 2 that means these waves are generated by quadrupole moments and not dipole. A dipole moment in GR would violate momentum conservation.

      Look on wikipedia for weak gravitational waves, or look in Weinberg's book on gravitation and GR. There you can find derivations of weak gravitational plane waves. This is a basic exercise in 1st year grad GR courses.

      This theory then predicts gravitational waves and LIGO has in fact detected them to within reasonable doubts. This is an empirical confirmation.

      Delete
    12. Lawrence Crowell: But if gravity is quantized, isn't there some minimum value or strength of gravity that can be realized?

      So if gravity is quantized wouldn't gravity waves dissipate to nothing at some horizon distant from their causation?

      Likewise if space is quantized; there would be a minimum distance of movement as a result of a GW, space could only be warped in integer values of some minimum.

      It would mean all the smooth functions describing such waves are just approximations of some underlying step function; which at some point decreases from one unit of "warpage" to zero units.

      Or perhaps become stochastic at that point, randomized zero's and ones that average out to the computed real-number value.

      Delete
    13. Lawrence,

      As usual you are proffering an argument based on your mathematicist beliefs. Mathematicism, however, has no basis in science, nor is it logically sound. It is an unscientific belief system based on a poorly thought-out pseudo-philosophy, which nonetheless has become the operating paradigm of the fundamental-physics community. (https://en.wikipedia.org/wiki/Mathematicism)

      It is mathematicism that prevents even the most absurd mathematical models from being discarded as scientific failures. (See string theory - and the big bang for that matter;)

      It is mathematicism that insists that empirical failures, no matter how numerous, cannot negate a mathematical prediction. Gravitational waves are a perfect example.

      Predictions of GW phenomena have failed repeatedly over the last 50 years and with each failure the prediction has been scaled lower, necessitating new and ever more sensitive equipment. Wash, rinse, and repeat with each predictive failure and you arrive at LIGO.

      Your belief that LIGO provides "empirical confirmation" of gravitational waves is illustrative of the way mathematicism has perverted science. Fifty years of predictive failures have not been counted as evidence against gravitational waves, but the merest wisp of a possible detection, beleaguered by procedural irregularities, dubious engineering claims, and a lack of independent confirmation, is promptly proclaimed a successful "empirical confirmation". That kind of selective, "empiricism of convenience" is antithetical to science.

      Science cannot proceed on that basis. The results have been, and will continue to be, endless, self-inflicted snipe hunts (https://en.m.wikipedia.org/wiki/Snipe_hunt) in which the gullible never get the joke, and tax-payers get to fund the ever more elaborate searches for imaginary mathematical entities.

      This theory then predicts gravitational waves...

      Gravitational waves are not a prediction of General Relativity per se, but of a mathematicist interpretation in which the mathematical spacetime of GR is reified into a physical (substantival) spacetime which, in turn, affects, and is affected by matter and energy. There is, however, absolutely no empirical evidence to support that peculiar view. It is simply another unscientific belief that arises in the context of mathematicism.

      Fundamental physics is in crisis because its conceptual incorporation of mathematicism has undermined and all but eliminated empiricism as a foundational basis for doing science. In allowing this to happen, the fundamental physics community has separated itself from the discipline of science. The situation is untenable; without empiricism, there can be no science.

      Delete
    14. Thanks @bud rap and @Lawrence Crowell! :)

      Seasoned astronomer's possible response: You guys who wrangle theories and equations have fun. Give me enough $$ and I'll build detectors/instruments/telescopes/observatories which will have a look-see if there's anything like GWR (transients, continuous, whatever) and let you know what I find. P.S. Check out the literature, there's quite a bit on pulsars whose orbits seem to be decaying. Oh, and it took just a small tweak to make the IPTA a GWR detector (so you don't have to give me $billions).

      Cautious astrophysicist: OK, so there is a big disagreement on what GR is, what it predicts, etc. Given that binary pulsar observations are consistent with the existence of GWR, I'm in favor spending some $$ to see if something else - observationally - is also consistent with the existence of GWR. When more observational results are in hand, you two guys can duke it out over the fundamental theoretical implications.

      Interested observer who has a lifespan of millions of years: I'm continuing to observe the binary pulsars. Well before I pass the second Chandrasekhar limit, I will certainly find out what happens when two neutron stars merge.

      Delete
    15. Jean Tate,

      Give me enough $$ and I'll build detectors/instruments/telescopes/observatories which will have a look-see if there's anything like GWR (transients, continuous, whatever) and let you know what I find.

      This has been done for 50 years. Nothing of significance has been found with the possible exception of the shaky, and therefore hardly conclusive, LIGO claims.

      Check out the literature, there's quite a bit on pulsars whose orbits seem to be decaying.

      That's right, and the number of GW detections from those binary pulsars = 0. If GWR is the mechanism for the observed decay, why can't LIGO detect it?

      Oh, and it took just a small tweak to make the IPTA a GWR detector...

      And the number of unambiguous GW detections so far? Exactly 0.

      When more observational results are in hand, you two guys can duke it out over the fundamental theoretical implications.

      Now that's funny. GWR is only a theoretical implication and the extensive observational results so far are overwhelmingly negative. The problem at the heart of the crisis in physics is the institutional inertia that renders fundamental physics incapable of acknowledging and dealing with negative empirical results. Your comments here are illustrative of just that problem.

      Delete
    16. @bud rap: I think you are missing something pretty big.

      "That's right, and the number of GW detections from those binary pulsars = 0. If GWR is the mechanism for the observed decay, why can't LIGO detect it?" I don't think anyone would, or could, seriously expect LIGO to detect any GWR, transient or not, from any of the binary pulsars (shorthand) that have been found so far. For starters, the frequency of any GWR would be far, far, far too low for LIGO to detect. And none of these binary pulsars are anywhere near approaching their last few pre-merger days. Or weeks. Or months. Or years. Or decades. Or ...

      Re the IPTA: you'd have to ask the astrophysicist about what expectations on detection of GWR are, or have been. The astronomer could care less.

      Perhaps the biggest part you missed, or are talking past, is this: "The problem at the heart of the crisis in physics is the institutional inertia that renders fundamental physics incapable of acknowledging and dealing with negative empirical results."

      GR is barely a century old. The GWR-caused inspiral merger of a binary neutron star would likely take many thousands of years, perhaps even millions of years. Astronomers don't live long enough to be able to observe super-rare events in our own galaxy (unless they're very lucky or unlucky), and faint ones beyond the Local Group must await another century or three of technical development and $billions or $trillions in funding before they could be detected.

      The good news, for you, is that astronomers love watching the sky, no matter what fundamental theories say or don't say, nor what philosophers think. So, just keep giving us some $$ and we promise to keep you informed of what we see happening in the sky! :)

      Delete
    17. Jean Tate,

      There is no compelling scientific logic evident in your remarks; it is all hand waving and special pleading. Sure we haven't found GWs despite 50 years of effort, you say, but give us more time and $$$ and someday we will deliver direct, empirical evidence for their existence.

      The problem being, of course, that you are chasing a dubious, objectively-failed theory, one that apparently cannot be dislodged from the collective imagination of the scientific community no matter how many times attempted detections fail to produce anything but negative results. Tomorrow is yet another day to carry on, and so we are told to,

      await another century or three of technical development and $billions or $trillions in funding before they could be detected.

      Or maybe not detected, who can say! Maybe it will take a millennium and $$$gazillions! That's a prescription for careerism maybe, but not science. And then there is this:

      The good news, for you, is that astronomers love watching the sky, no matter what fundamental theories say or don't say, nor what philosophers think. So, just keep giving us some $$ and we promise to keep you informed of what we see happening in the sky!

      Were he still alive, I'm sure Halton Arp would have a good laugh at that smug affectation. Astronomers see (or seek) those things that theorists tell them to see (dark matter, dark energy, and gravitational waves), and do not see those things theorists tell them not to see (discordant redshifts).

      Delete
    18. bud rap: Just two things:

      Assuming you have no problems with observations that neutron star binary orbits are decaying, what - if anything - do you propose (or think) is the cause? And how do any such causes relate to General Relativity?

      Re Halton Arp: I can't do a "Senator, you're no Halton Arp" (we never met as far as I know), but I can decry your poor understanding of astronomy. In particular, you seem unaware of the existence of surveys (there are hundreds, and they generally cost a lot of $$). As I think I noted in a post elsewhere, Arp would be over the Moon at the vast amount of high quality survey data that's available today, for free. He'd be rather less happy that his "discordant redshift" ideas have totally bitten the dust, in the light (ha!) of survey data.

      Delete
    19. ...his "discordant redshift" ideas have totally bitten the dust, in the light (ha!) of survey data.

      Spare me the hand waving - cite something that you consider a definitive refutation of the observed discordant redshifts. I've never seen anything more than lazy statistical arguments or gravitational lensing invocations that are overly generalized from individual instances. Show me some well developed analysis like this:

      https://arxiv.org/pdf/astro-ph/0509630.pdf

      The problem you're going to have, of course, is that contriving a definitive refutation of the discordant redshift interpretation requires just that - a contrivance, because the fact that there is a large volume of observational evidence strongly supports discordance.

      While as the referenced paper makes clear, there are individual cases where the redshift discordance is only a matter of coincidental alignment or gravitational lensing, there are many other cases where the relationship cannot be so dismissed. And if it isn't apparent to you by now, I don't accept arguments from authority as scientifically valid. - I do not care what the contrived consensus of the cognoscenti is.

      No matter how you slice it, there is a large body of empirical evidence for discordant redshifts while there is none for say, "dark matter". But there is plenty of telescope time allotted for "dark matter" searches but research into the observed discordant redshifts is denied telescope time. You can't do science that way.

      I understand perfectly well how the modern astronomical community functions, and it is the way it functions that I am criticizing. You seem to be trying very hard not to understand that criticism.

      Delete
  10. Very good job - your opinion piece in the New York times!

    If a convincing model of climate change - convincing that is to politicians – could be achieved, this would be worth more than a billion. even more than the 40 billion for the new particle collider.

    And Palmer’s idea of a CERN for climate change – wonderful. How can we not have that?

    ReplyDelete
  11. The Laser Interferometer Space Antenna (LISA) is a worthy candidate to merit funding for science. This project is a space mission led by the European Space Agency to detect and accurately measure gravitational waves using multiple zero-drag spacecraft that are linked by laser beams.

    Such a space based zero-drag satellite project would eliminate the ground based noise produced by gravity wave detectors anchored here on earth.

    Similar to LISA, The Deci-Hertz Interferometer Gravitational wave Observatory (or DECIGO) is a proposed Japanese, space-based, zero-drag gravitational wave observatory.

    This highly sensitive emerging detection technology can explore many more widely varied gravity centric astronomical processes in fine detail is a noise free medium and for that reason in a good place to spend limited research money.

    LIGO technology has shown that gravity wave detection is possible but it is time for that coarse technology to be replaced with LISA something far more scientifically valuable.

    ReplyDelete
    Replies
    1. The LISA, or eLISA as it is now called since NASA backed out, is a great program. It consists of three spacecrafts that permit a free floating optical laboratory within it. This is so the optical lab part, sort of a spacecraft in a spacecraft, is shielded from solar wind and tiny impacts with micrometeoroids that perturb the path away from a geodesic. These three spacecrafts are put on Earth chasing solar orbits with different inclinations and periapsis so the spacecraft maintain a constant triangular distance from each other.

      The advantage of having the optics free floating is that if there are translations of these masses after a gravitational wave passes this will result in a persistent change in the laser interference. In the Penrose diagram for the Schwarzschild metric i^0 contains the Poincare symmetries, and the I^+, positive infinity, are Poincare symmetries and I^+ is extended to P⋉T where the ⋉ means a semijoin or right product of the Poincare symmetries with a set of abelian translations T. For every element of the Poincare group there is a set of translations so the Poincare symmetries are a normal subgroup that splits G = P⋉T. We observers, or the eLISA are essentially at T^+ and we can detect these translations or BMS symmetries.

      the much longer baseline of the eLISA means the detected frequencies are much lower. We can potentially detect signals from black hole mergers where signals are more highly red shifted by the tortoise coordinates. Information from close to the near merging horizons could be resolved, and these will be as BMS symmetries.

      Delete
  12. It seems to me that gravitational waves, if real, do imply that gravitation also has push characteristics.

    ReplyDelete
    Replies
    1. Not in the sense in which this often appears in crackpot literature.

      Delete
  13. I'm absolutely sure LIGO didn't measure real gravitational waves. Sad news is we'll never see retractions of LIGO papers that claimed detection of extraterrestrial gravitational waves. There are reasons to believe LIGO team are fooling themselves by means of their own GW templates, which are capable of finding spurious GW signals where there is just random noise.

    ReplyDelete
  14. Filter out some kind of signals from a noise-stream is no good idea, since noise contains any kind of signals - if you wait long enough. You may find lyrics of William Blake (or others) if you wait long enough. And if you wait much longer, you will find it also in two - or three - independent noise-streams with a fixed time delay.

    The same way spiritists detect "Paranormale Tonbandstimmen" (nice song from Laurie Anderson) or detect voices and messages in the radio between the frequencies of the radio stations.

    If the signals are with the same amplitude as the noise signal itself, that's pure expensive spiritismus, glossed over with some math.

    ReplyDelete
    Replies
    1. GPS relies on filtering out some kind of signal from noise. And it demonstrably works. If you know what a signal should look like then you can delve deep into the noise to find it.

      This is the principle behind many of the LIGO results, and it is perfectly reasonable to say that we know from theory what these starquake signatures look like so let's see if we can find them. Howeever, one has to be very strict with the rules of correlation to avoid fooling oneself, and I think this is where the LIGO team have failed in presenting results clearly enough the the healthy skeptics.

      Delete
    2. MikeS6:48 AM, November 04, 2019

      Nonsense to place GPS besides LIGO.

      And the sentence "that we know from theory what these starquake signatures look like so let's see if we can find them." is debunking:

      You KNOW that from theory? Exactly? And for sure? What a naivety!

      Today theories are stacked on hypotheses which are stacked on hypotheses which are stacked on hypotheses ...

      And experiments are designed to confirm wishful thinking.

      Delete
  15. Peter Galison wrote an interesting book, How Experiments End (1987, University of Chicago Press). Here we read: "An experimentalist will often design an apparatus precisely to exclude a background and, just as in the choice of where to look, may exclude phenomena later considered to be vital...once recorded, data selection again cuts between the foreground and the background as 'good' events are split from the 'bad'." and "sometimes an event can be thrown out simply because it does not look right." (page 256).

    ReplyDelete
  16. Up to now I've naively thought that the reason why we can (mostly) ignore changes of (frequency of) photons, is that wavelengths of (measured) GWs are muuuuuch bigger than lengths of interferometer arms, and (supposedly) without any inner structure.
    Notice that interferometer arms are (in principle, and will be for eLISA) based on vacuum, i.e. not on massive particles.

    PS I would bet that the first GW was injected (too much pressure), the other mostly true, and that independent analysis of the glitches is quite important to do; they may contain hints on quantum gravity.

    ReplyDelete
  17. Back when J. B. Rhine was testing "psychics" at Duke University, he routinely threw out what he considered to be bad data. A test subject would have a hot run, guessing hidden symbols at a better than chance level, and then have a number of misses. Rhine assumed that ESP varied in sensitivity, so only the "good" runs really counted. In this way he was able to provide evidence of extra-sensory perception. Something like circular reasoning, and your comments on LIGO procedures reminded me of this-- seems similar.

    ReplyDelete
    Replies
    1. Do you have a reference for that statement?

      Delete
    2. https://psycnet.apa.org/record/1937-01276-001

      This links to an abstract describing Rhine's statistical procedures. There is a link to the paper.

      Delete
  18. I think it's worth spending some time clarifying - and hopefully agreeing - on the terms "confirm", "validate", "verify", and "independent".

    LIGO is an immensely complicated piece of equipment intended to observe transients "in the sky". It is not an "experiment" like the double slit, or what happens at the LHC. So perhaps we should be using the language of astronomers? But given its history, a whole lot of people use the language of (experimental) physicists. And those who write theoretical papers.

    "Independent" should be easy: whoever is doing the confirming/validating/etc, they should not include members of the LIGO team (if you're a conspiracy addict, I suspect independence is impossible, because common funding agencies, for example, would be excluded).

    "Confirm", "verify", and "validate" are easily mixed up, for astronomical observations of transients, but two different things are easy to define:
    - you get hold of the raw data and run your own analyses
    - you observed the transient with your own equipment.

    When you start considering something other than a detection of a gravitational wave radiation (GWR) transient, it can get messy very quickly.

    These days all observatories have good clocks and data will have reliable time-stamps (it wasn't always so). Matching a GWR transient to a gamma-ray transient (a "GRB"), say, or an optical one involves much more than time-stamps. Localization (where are the transients, on the sky?) is very hard, because LIGO on its own gives very poor localization, so it's ~irrelevant that the Hubble Space Telescope nailed down the optical location of a transient to ~0.1".

    Perhaps, within a year, LIGO, Virgo, and KAGRA will each independently detect a transient (candidate GWR event), triangulation will give its position to ~1', and large numbers of teams of astronomers will report on what they saw there, independently. And perhaps some of the neutrino observatories will too.

    Maybe those who love models will report consistency with a neutron star/black hole merger? Perhaps it'll be something wild, and ideas about the destruction of alien battle forts will start to emerge ... ;-)

    ReplyDelete
    Replies
    1. One constantly reads how dark matter was *discovered* by observing the rotation of galaxies and the existence of dark energy has been *verified* by observing the 'acceleration rate of the universe'.

      This is a sloppy and irresponsible use of language.

      These "things" have been inferred within the framework of GR, which might, just possibly, be wrong.

      Language matters.

      Delete
    2. So, other than me, no one has used the word "validate" (so far). And only one other person used "verify" (Tom Weidig: "a clear independent verification").

      And, up to now, other than the comment of mine I'm replying to:
      1) Sabine uses "confirmation" (no modifier) once, and it's used just twice like this by others
      2) "confirmed prediction" is used by Sabine once, and once by someone else
      3) "independent confirmation"? twice by others (one of them me), three times by Sabine.

      I'll leave 2) for later.

      Except for my own use of "independently confirmed", I think there is considerable ambiguity in all the uses of 1) and 3). In particular, how many mean (something like) "I analyzed the raw LIGO data myself"? And how many (something like) "I found ~the same thing as LIGO did using my own equipment"?

      When the only working (interferometic) GWR detectors were the two LIGO ones (two sets of observations? or one? later), the only kind of independent confirmation of any GWR transient they reported is "I got the raw data and did my own analysis" kind, right? We cannot somehow go back in time and observe any of these transients with a different detector, can we?

      So, has any "I got the raw LIGO data and did my own analysis" independent confirmation been reported? For any of those transient GWR events (i.e. when LIGO was the only game in town)?

      When Virgo came online, the "I found ~the same GWR transient using my own equipment" kind of independent confirmation became possible.

      Did Virgo report any such independent confirmation?

      (to be continued)

      Delete
    3. Hi Jean,

      The only useful "confirmation" is electromagnetic confirmation of a real-time counterpart prediction by LIGO. Something which LIGO hoped to achieve many times per year, but hasn't yet achieved even once. But hopefully they soon will.

      All other types of "confirmation", even with multiple detectors, cannot confirm astrophysical origin and predictive power.

      Delete
    4. (continued)

      Astronomy has had transients since before there was even astronomy: guest stars and meteors, for example. For the latter, there is an interesting historical story: impacts on the Moon and Jupiter. These transient events were reported, by amateur astronomers, long before they were accepted by (most) professionals. In the last decade or so, a small handful have been independently confirmed in the sense of "I found ~the same transient using my own equipment". Before video recording, there was no way to do any "I got the raw data and did my own analysis". Yet impacts on the Moon and Jupiter are now accepted a transients ... how did that happen?

      In EMR (electromagnetic radiation) astronomy, "beyond the solar system", there is a taxonomy, which goes (roughly):

      class:variables, sub-class:transients.

      The most frustrating/curious transients are those which do not (appear to) repeat. And are observed by just one team/instrument/observatory. And which last mere ~seconds. GRBs (gamma ray bursts) and FRBs (fast radio bursts) are well known, recent examples; less well known are certain flare stars (in their non-flaring state they are too faint to detect). For most of these, but not all, the only possible kind of independent confirmation is "I got the raw data and did my own analysis".

      No one - as far as I know - today asks for "I found ~the same transient using my own equipment" for GRBs. Nor FRBs. Yet Sabine, in this blogpost, is asking for that for LIGO's GWR transients. Why?

      (170817: let's leave discussion of the GWR transient, and any possibly associated EMR transient(s) until later).

      (to be continued)

      Delete
    5. @JeanTate:

      "No one - as far as I know - today asks for "I found ~the same transient using my own equipment" for GRBs. Nor FRBs. Yet Sabine, in this blogpost, is asking for that for LIGO's GWR transients. Why?"

      I presume the reason is because GRBs and FRBs are direct detections of phenomena whereas gravitational waves are detected indirectly with potentially more than one phenomena giving rise to the signals observed (despite what anyone might claim to the contrary).

      Compare this to particle physics where theory is used to predict potential decay pathways which are then inferred from events observed in the detectors - essentially an identical process to that employed in the interferometer detections. A single event is not proof that the theory was correct... there's that whole pesky 5 sigma thing.

      However, you can't really get a 5 sigma certainty in gravitational wave detection because each event is unrepeatable, therefore corroboration must be performed through a secondary detection using, ideally, a different type of measurement (i.e. more Earth-based detectors are not a different detection method).

      Delete
    6. @Duosae: I'll turn the contrast waaay up.

      Imagine a world where there are as many, independent, observatories as capable/sensitive as Advanced LIGO as there are such for GRBs today (of the order of a dozen, say). Imagine they regularly report independent detections of GWR transients, and quite a handful of simultaneous cases (not unlike not so many years ago for GRBs). Wouldn't our discussion then be quite different? And this despite the fact that GWR has indirectly detected (binary pulsars, Hulse-Taylor, etc, even 1a SNe!).

      This may come as a surprise: it takes but seconds to find crackpot/alternative astrophysics ideas. In some of these, GRBs cannot be directly detected! :O Why not? Because gamma rays can be directly detected above the Earth's atmosphere only (observatories like H.E.S.S. can do so only indirectly). Which is impossible (e.g. Flat Earthers), or because the detectors cannot work (they rely on Relativity, which a great many crackpots deny; or modifications to Maxwell's equations; or ...). So when, and how, do we admit multi-messenger astronomy as full members of the club? And do we use different criteria for GWR and neutrino astronomy?

      Delete
    7. @JeanTate:

      I think you're confusing the point. GRB do not need independent corroboration for their detection - nobody doubts that a high intensity EM pulse is observed, the multiple observations are required to pin down where and what phenomenon is causing the observations.

      Contrast this with a particle decay or a GW, you are not certain what you are detecting what you think you are, even with multiple detectors.

      Let me give an example using GRBs - since you seem to be hooked on them. It's been proven that GRBs emanate from multiple source types. If we'd only happened to observe a single source type, a single phenomenon, we'd not know there were more possible.

      The question with GW and particle physics isn't that the signal observed is being questioned, it's the source type that is being questioned. There is no different criteria for GWs or GRBs or particle decay, it's just that the point of application of that questioning that you're percieving is different (it's not).

      One last thing - don't confuse indirect detection with knowledge of existence. We know gravitational waves exist.. but we don't know if we're actually detecting them in our instruments.

      Quite frankly, having more detectors on the Earth will only really improve resolution, not remove doubts. Only eLISA or an orthogonal confirmation can remove those doubts.

      Delete
    8. @Duoae: How about supernovae (SNe) then? How many "source types" are there? Or just plain stars, how many source types? And how did we go from just one (of each kind, sorta) to dozens or hundreds?

      Sometimes predictions work - e.g. microlensing searches for MACHOs (some found, but so far nothing but boring baryonic matter) - but very often in astronomy there are surprises - e.g. exoplanets (yeah, they exist, but what a menagerie!).

      LIGO, and other interferometric detectors, observe "things", a mix of glitches and signals that look like GWR transients. As time passes, more and more are detected - data is collected, we have "better statistics". Physicists and engineers get a better handle on the glitches, and may even discover some obscure terrestrial phenomena, along with some weird couplings of mechanical and electronic systems.

      And different classes of GWR transients are found. Along with more apparent associations with EMR transients, and perhaps some neutrino transients. All followed by flurries of astrophysics papers, attempting to "pin down where and what phenomenon is causing the observations."

      In short, yet another chapter in rich entanglement of observation and theory that is astronomy and astrophysics.

      Delete
    9. @JeanTate:

      Yes, I understand your point, and I am sympathetic to it however:

      "LIGO, and other interferometric detectors, observe "things""

      Yes, and until we definitively reach the point where we can say they are what we think they are then your argument is moot. Hence the reason why people are saying "we need to reach that point by getting the unequivocal data". You don't have better statistics from different events. That's not how statistics work: You have better knowledge of signal/noise ratio statistics, not event types.

      If I have a million runs on a TGA watching mass loss as the heat rises, it doesn't tell me anything about the processes that are leading to that mass loss without the orthogonal data giving me the information to understand the mass loss.

      Yes, I can see on a DSC thermograph there's an event that looks like phase change or oxidation or sublimation or decomposition... I can't definitively define what each thermal event is without having the knowledge of the system being observed... and that is what we call indirect observation and that is why we need orthogonal information.

      Delete
    10. @JeanTate:

      I should have addressed this in my previous comment so I apologise for that omission:

      "How about supernovae (SNe) then? How many "source types" are there?"

      I don't even understand the point of this question. What do you mean "How many source types are there"? How many atoms are in the universe? How many oxidation states of iron are there? How many covalent bonds can carbon form?

      Answer: As many as we observe and relate to theory.

      The point is, we work to understand all those questions though some are easier than others...

      To answer your linked question: "How did we go from one to dozens or hundreds?"

      We observed them directly through the visual EM spectrum and the non-visual EM spectrum. We tied multiple observations together (in multiple EM bands) and linked them through predictions in theory... and where theory was not good enough to understand the observations, theory was revised.

      That's my entire point. An interferometer is not direct observation. We need orthogonal results to confirm what we observe. This is very basic scientific practice.

      Delete
    11. @Duoae: Thanks. I think we are stumbling over something which may as simple as a definition or two, or may be as profound as what astronomy is (beyond the solar system anyway).

      Let me try this: "An interferometer is not direct observation." Plenty of EMR astronomy is based on interferometers, e.g. in radio astronomy (the VLA, or LOFAR, say). Why is LIGO different?

      Delete
  19. Sabine,

    Let me finish by saying once again that personally I do not actually doubt these signals are caused by gravitational waves. But in science, it’s evidence that counts, not opinion.

    I'm curious as to why you would have a strong opinion in this matter, given the lack of evidence. Are you trying to forestall any possible "slap" or worse. And I don't mean that as a joke - I thoroughly appreciate the precariousness of your position, professionally and personally.

    ReplyDelete
    Replies
    1. bud rap,

      Well, I guess have somewhat overstated the strength of my opinion. But the reason is (a) I cannot think of anything else that could produce such a signal, and (b) we know that the signals have to be there. But it's hard to quantify that. And just because I have not been able to think of a physical process that would have a similar spectral distribution doesn't mean there is none.

      The comment about the slap I am sure was joke, but nevertheless it illustrates how "seriously" these people take criticism.

      Delete
    2. Sabine: There is a saying in English, "People say in jest what they mean in earnest."

      I don't know if there is an equivalent in German, but basically if we say something as a joke, and anybody gets offended, we can always say "It was just a joke, I didn't mean it." But if they laugh, we have found a like-minded compatriot.

      Delete
    3. @Dr. Castaldo: There is a saying in German: "Die größten Kritiker der Elche waren früher selber welche".

      When I read the term "Afro-engineering" here for the first time, my immediate connotation was: being able to keep a system running even without access to resources to do it the "right" way, say for instance Scotty in the Enterprise. Prohibited resources -> Africa -> Imperialism etc.. I have no clue at all were racism has a place in this language game. In the same sense JFK didn't mean he was a doughnut.
      If your mind's in the gutter just don't make others responsible for that.

      Delete
    4. sshawnuff: So you are accusing me of racism because I recognized a racist term?

      It makes zero difference if you failed to recognize the term and inferred your own private definition. In the USA, Blacks will recognize this term as pejorative and racist and feel offended by it. I provided you the research to prove it; on this particular page Wikipedia's references are correct.

      If a Black man recognizes a racist term, does that make him a racist? No. Nor does it mean his mind is 'in the gutter'.

      Recognizing racism does make me a racist, just like recognizing crime does not make me a criminal, and recognizing bad logic does not make me guilty of it.

      In fact, the more one recognizes bad logic, the less one will be guilty of it. Hope that helps clarify things for you.

      Delete
  20. This comment has been removed by the author.

    ReplyDelete
  21. after all, it's symptomatic if experimentalists arrange their setup strictly in anticipation of theory predictions. In this issue I hit just like Sabine.

    LIGO collaboration should have taken equally all "glitches" under the testing if they are astrophysical origin. The black hole might still be a mathematical artefact!

    ReplyDelete
    Replies
    1. What about the folks at Virgo?

      A key test, which as far as I know was actually done, is the time delay between seeing a similar glitch in both LIGO set-ups ... if it's greater than the time it would take light to travel between the two locations (even in a vacuum), then it can't be astrophysical, can it?

      Delete
    2. @JeanTate: If we assume GWs travel at the speed of light, no. But in general, it is quite certain LIGO/VIRGO are detecting GW-like signals that are slower or even faster than the speed of light. They just don't get published.

      Delete
    3. @Guest: "it is quite certain LIGO/VIRGO are detecting GW-like signals that are slower or even faster than the speed of light. They just don't get published."

      May I ask, how do you know this?

      Delete
    4. But in general, it is quite certain LIGO/VIRGO are detecting GW-like signals that are slower or even faster than the speed of light. They just don't get published.

      Well that's certainly interesting. If LIGO/VIRGO are detecting GW-like signals that aren't behaving like GW signals then that suggests that they aren't detecting GW signals at all - just something that sometimes fits a GW profile. No wonder they're not publishing those results.

      Delete
    5. @Jean: LIGO has dozens of false positives every day even within the 10ms constraint. Imagine the false positives outside of the 10ms constraint. LIGO may not even know them, because their search algorithms don't flag them. Remember the superluminal neutrino? :-)

      Delete
    6. Consider general relativity in the weak limit, impose a perturbation on a flat metric g_{μν} = η_{uv} + δg_{μν}, for η_{uv} a flat Minkowski spacetime and δg_{μν} a perturbing small part. We take the transverse traceless part of this perturbation, which is why you might see δg^TT_{μν}, run this through the equations for connections and curvatures you get a linear waves equation similar to that for an electromagnetic wave. Similarly if we take the post-post-Newtonian expansion of general relativity these give a form of Maxwell's equations. These weak gravitational waves are very similar to electromagnetic waves. They move at the speed of light. The time delay between the various interferometers bears this out as well.

      These sprites or noise that are glitches in the device are some form of noise. There are engineers who built entire careers on understanding noise in electronic circuits, lasers and other systems. These represent some sort of fluctuation in this interferometer that could be associated with the mirrors, lasers or some other thing. You can be sure there are people working hard to understand these.

      These noise events or the signals associated with gravitational waves are not seismic events. Seismic waves travel several kilometers per second, s waves about 5km/s and p waves about 12km/s. These would not appear at two or more interferometers at a delay time expected for light speed.

      The issue of some importance here is the matter of rounding out the signal between the ramp up of higher frequency and amplitude and the ring down. General relativity is not forgiving there, and in a perturbation series the terms become large and numerous --- in fact infinite. The exact moment of the merger of the horizons is not something general relativity predicts well. It is in some ways similar to the ordinary event of a drop of water. Water at the faucet opening may increase, the bulge of water grows and the this breaks into two to form a drop that falls. That exact point where the surface of the water breaks is not an easy problem in hydrodynamics. Right at that point where the drop is connected to the source above, appearing as a typical teardrop shape, and it then breaks is not easy to understand. BTW, falling rain drops etc are more bowl shaped. General relativity also has analogues to hydrodynamics. This merging of two horizons has some comparison to this more ordinary process of forming a water drop. Maybe we should be studying that as well.

      Delete
    7. Thanks Guest.

      I have too many questions to be answered within these blogpost comments; would you please point me to some literature on these "dozens of false positives every day even within the 10ms constraint"?

      As far as I know, Gravity Spy has no glitch class that even remotely resembles any expected GWR signal. And I do not recall reading about such a high rate of false positives (yeah, my memory is much worse than it used to be, and I gave up trying to keep up with literature a while ago, so many papers!).

      What about Virgo, do they also get dozens of false positives a day? What about LIGO+Virgo?

      Delete
  22. Sabine, you say:

    “Wouldn’t the effect cancel out? No, it does not cancel out, because the interferometer is not made of light. It’s made of massive particles and therefore reacts differently to a periodic deformation of space-time than light does.”

    The interferometer gets its shape from the fields, which bind the atoms and the molecules building this apparatus. So, according to your explanation, these fields – or the path of the exchange particles which realize these fields – have to be differently deformed by space-time than light. Is that what Einstein’s GRT is saying?

    ReplyDelete
    Replies
    1. Sabine,

      this was meant as a serious question. The issue behind it is whether Einstein's GRT is only a well working formalism or it is physics. The different reaction of different phenomena in a deformed space-time does not feel like comprehensible physics. It seems to me that it even violated Einstein’s strong equivalence principle.

      Delete
    2. I would translate "a periodic deformation" so that massive particles consist causal feedbacks but lightlike particles do not.

      Delete
    3. The strong equivalence principle says that in a gravitational field an observer who resides in a sufficiently small box will, when performing physical experiments, not notice to be in a gravitational field.

      Applied to the LIGO interferometer this can be understood in the way that the space where the interference process goes on can be split into such little boxes. And an observer in any of these boxes will not see the change caused by a deformed space-time. So, also a passing photon will not see it. So the question is in my view justified why this interferometer will make an interference caused by the influence of the deformed space-time visible.

      Delete
    4. The curvature of spacetime and information signaling of gravitational changes are different phenomena.

      Delete
    5. What do you mean by "information signaling of gravitational changes"? The assumed functionality of LIGO is that spacetime is deformed at the location of the photon paths inside the LIGO. What else do you mean?

      Delete
    6. I answered but Sabine didn't publish.

      If my explanation was unorthodox, please Sabine give yours.

      Delete
    7. Eusa,

      Your answer was not "unorthodox" is was garbled nonsense as, I am afraid, are most of your comments. You clearly have more opinions about physics than knowledge and I would appreciate if you would discontinue submitting ill-informed comments. Thank you.

      Delete
  23. Is slapping physicists a novel physical way to do physics?!

    It's only within the last five years that we have gotten gravitational wave detection events, as opposed to many decades of collider technology, so it's not surprising that there may well be problems in interpreting the data. Still, I would have supposed that handing out a nobel prize ought to indicate that such problems have all been ironed out! Like you, I have no doubt that events have been detected, but honesty as well as evidence of honesty is paramount to doing good physics and keeping physics honest and trust-worthy.

    ReplyDelete
  24. Looking at the location where the two neutron stars collided after the event, it became clear from non-LIGO observations that the gamma ray had indeed been caused by the merger or two neutron stars. Was it clear from the gamma ray itself in the first hour that it was caused by the merger of two neutron stars? I suspect not.

    If this is true, you can say that there was a successful prediction. LIGO predicted that the observed gamma ray had been caused by the merger of two neutron stars. That would be enough confirmation for me.

    ReplyDelete
    Replies
    1. Some people have pointed out that actually the electromagnetic signal does not fit all that well to the expectations for a neutron-star merger. I have written about this here.

      Delete
    2. So if the electromagnetic signal did not fit all that well for neutron stars, and subsequent observations show that it was indeed a neutron star merger (namely, the identification of heavy elements created in the event, which theory said might be formed in neutron star collisions) LIGO did indeed make a valid prediction. Unless new elements can be created in white dwarf mergers, which I don't believe people have claimsed.

      Delete
    3. Hi Peter,

      Just in October there was a paper in Nature that pointed out that no heavy element has actually been detected so far (unlike the media reported). The first "heavy" element now identified is Strontium, but this is mostly an s-process element, not r-process. So it's still somewhat ambiguous. (https://arxiv.org/abs/1910.10510)

      Even so, a binary neutron star merger certainly remains the most likely scenario. It could of course be argued that the type of short gamma-ray burst already pointed to this type of event, based on existing GRB models.

      There are other arguments in LIGO's favor: they got the distance right (cannot be derived from GRB signal) and their skymap was much smaller than NASA/ESA combined.

      Bottom line: I personally also wouldn't call it a "postdiction". It was a type of delayed and partial prediction. It's the delay and pre-existing data that leaves room for questions/speculation.

      LIGO can only end these doubts by finally (after 4 years!) achieving a first successful real-time counterpart prediction. It's a basic requirement, really.

      Delete
    4. One more consideration: was the EMR transient utterly unlike any (up till then) known transient? I think it's now called a kilonova ... how many kilonovae have been independently confirmed (other than that on 170817)?

      Delete
    5. Peter,

      You are not making sense because the creation of strontium was not inferred from the gravitational wave data and does not remove the existing tension. And for all I know no one has claimed that this information demonstrates it was a neutron star merger. I don't know if a white dwarf merger would do (probably not), but a white dwarf core collapse or white dwarf neutron star merger might. Let me ask you this: If further analysis of the em signal shows it cannot have been a neutron-star merger, will you consider LIGOs claim to be falsified?

      Delete
    6. @JeanTate: There are currently about two similar transients (of unknown origin) in the public domain. This lack of good datasets is one of the difficulties in judging 170817.

      Delete
    7. @Guest: Thanks. Can you point us to those two please?

      Delete
    8. @Jean: GRB 150101B and 160821B are considered possible kilonova candidates.

      Delete
  25. I find it unfair not count the event from August 2017 as an independent confirmation. To demand that they need to process their data faster then any other telescope so that they may send out their alert first does not make sense.

    With three or more detectors ruling out earthly origins should be easy so I do not see that as a problem any more.

    ReplyDelete
    Replies
    1. What is it with you people that you think it is "unfair" that I ask for what is perfectly normal scientific procedure? They wouldn't need to process their data faster than anyone else if we actually knew how they process their data. We know that LVC knew of the GRB event as they were doing their data analysis. Now no one has any way to tell whether that did or did not affect their analysis. Hence we cannot draw reliable conclusions. That's the situation as a matter of fact. I don't know how it's "unfair" to mention this -- this is all public information.

      "With three or more detectors ruling out earthly origins should be easy so I do not see that as a problem any more."

      Except this hasn't happened, Matti. We are still waiting.

      Delete
    2. I did pull up the neutron star merger paper. It was found by the online analysis in Hanford and Livingston missed it because of a coincident glitch. It was also seen at Virgo but because of the angle and sensitivity it did not pass the threshhold. The alert was sent out automaticaly and then a manual re-analysis was done which found corresponding signal at Livingston and Virgo. So I still think it is unfair to dismiss the detectetion because it took LIGO 6 minutes longer to process data.

      I'm specifically refering to "The collaboration had, so far, one opportunity for this, which was an event in August 2017. The problem with this event is that the announcement from the collaboration about their detection came after the announcement of the incoming gamma ray. Therefore, the LIGO detection does not count as a confirmed prediction, because it was not a prediction in the first place – it was a postdiction. " where you demand that LIGO process data faster than Fermi. Regarding glitches I hope that is something they are working on but the join detection is very strong evidence.

      But there is a second piece of evidence. LIGO/Virgo managed to predict the distance to the source which was later confirmed by measuring redshift.

      Delete
    3. That's right, Matti. There was a glitch in the detector. The supposed signal had to be extracted by removing the glitch, which is the same as saying they removed the unwanted part of the data. Saying that the corresponding signal was "found" at Virgo is somewhat of a stretch with a S/N of 2.

      Once again, no, I am not saying they have to do their data analysis faster than everyone else. If we actually knew what they are doing, this wouldn't be necessary. But we don't know. And in that situation I think it is merely good scientific practice to ask for an actual prediction that would remove any doubt. It is beyond me why so many are willing to settle on believing.

      Delete
    4. Even given that we have coincident and a correct prediction of distance.

      In what way is not the distance prediction good enough?

      Delete
    5. @Matti: The distance estimate is good but it too came only long after the event (about 2h) and after both Fermi and Integral data was already available.

      If the value did indeed come from a genuine GW, as we all assume, then it should be very easy to achieve a first real-time prediction that by design is done automatically within less than a minute.

      LIGO still hasn't achieved this after four years (even though it expected lots of such events), but hopefully it will do so before the end of the current observation run.

      By the way, as mentioned in a comment above, the distance to the transient could never be confirmed directly (by spectral analysis), so a foreground/background object was still possible. Only this October has a first indirect measurement apparently been achieved (via Strontium).

      Delete
    6. But the distance was not in the data from Fermi/Integral. Fermi/Integral pointed to a volume containing 1 billion galaxies and LIGO/Virgo manage to constrain it to a volume with a handfull. Improving the prediction by a factor 100 million. I just don't understand how this is not evidence that it is working.

      Delete
    7. @Matti: There's an assumption that certain stellar-mass GWR transients should be accompanied by EMR transients. GRB-type transients seem good choices, despite the fact that - as far as I know - no such had been thought to have generated detectable GWR before LIGO. And unless there is at least one sub-class (other than two "long" and "short" ones) which can be tied to an NS-NS merger, the existence (or non-existence) of a possibly associated GRB is weak tea indeed.

      Transients in the UV, optical/visual, and/or (near) IR are generally regarded as more desirable, if only because they are highly localizable, and galaxies in which they likely arose far more more reliably identified. And - as far as I know - the EMR signature of an NS-NS merger is expected to be radically different from any other class (yes, they'll be "novae", but of what kind?).

      The shortcoming is that a featureless (optical) spectrum cannot give you a redshift (and SED fitting is quite ambiguous for an unknown class of new object!), only one with "lines" in it. So, unlikely as it may seem, neither foreground nor background (to the tentatively identified galaxy) sources cannot be ruled out. As in ruled out to firm adherents of the scientific method.

      Delete
    8. Correct Matti. So let's just hope LIGO will soon achieve a first successful real-time counterpart prediction, thus excluding any other source of information. So far in O3 they haven't been successful (8 retractions and 10 without counterpart), but that may still change. After all, it's not just about functionality, but also reliability.

      Delete
    9. A constraint is not a prediction.

      Delete
    10. What does "a constraint is not a prediction" mean here? If not pointing to a handful of sources is enough how does matter that fermi hot the alert out first matter?

      @guest The astronomers have pushing LIGO to publish before the are certain to enable followup. Therefore there is a lot of retractions once the more thorough analyses are done. The other events were not as close which makes them harder to observe.

      @JeanTate A coincident event is extremely unlikely but can never be ruled out. That is why scientist report discoveries based on a probability of the same observation given there is nothing to observe.

      Delete
    11. @Matti: I agree. As I said in an earlier comment, these false-alerts aren't falsifications (LIGO is almost impossible to falsify), but false positives.

      The fact remains that contrary to initial expectations, we haven't yet seen a single successful real-time counterpart prediction. There is really no reason whatsoever to be against this basic scientific requirement.

      170817 was at best a delayed prediction: both the trigger and the first skymap were delayed. Only because of this critics can ask who knew what and when, whether or not it was reported on the GCN. As you know, LIGO has never published its 170817 waveform template, despite several groups asking for it.

      In addition, the transient continues to remain ambiguous from a strictly EM perspective. Its distance could never be determined directly, and those claiming r-process elements could in fact not determine even a single element. Hence the Italian group proposing their white dwarf counter-model.

      Just last week LIGO reported another event that turned from 87% terrestrial to 95% black hole candidate. It looks like in the case of LIGO, Heaven and Earth are rather close together :-)

      Delete
    12. @Matti: As the saying goes, been there, done that, got the T-shirt. "A coincident event is extremely unlikely but can never be ruled out. That is why scientist report discoveries based on a probability of the same observation given there is nothing to observe."

      One of the fascinating things about astronomy is that it studies objects vastly more complicated than elementary particles. Yes, Statistics 101 is very helpful. But are all distributions Gaussian? Are there no black swans?

      For the 10,000-th 1a SNe, we're good; for the first (and so far only?) kilonova+GWR transient? Not so much ...

      Delete
    13. @Guest: "who knew what and when" As noone knew about the kilonova LIGO must have found anything they knew in their own data. Unless the critics believe LIGO/Virgo has an secret much better experiment and are just pretending to see things with GWs.

      @JeanTate: We look at the sky a lot so we can calculate the probability of a coincident independently of the underlying distributions, which should be poisson. But what p value is good enough is kind of unrelated to my argument and up to each person.

      What I have a problem with is the statement "The problem with this event is that the announcement from the collaboration about their detection came after the announcement of the incoming gamma ray. Therefore, the LIGO detection does not count as a confirmed prediction, because it was not a prediction in the first place – it was a postdiction. " in the context of "Have we really measured gravitational waves?". What have LIGO/Virgo measured if not gravitational waves when they managed to improve on the prediction by a factor 100 million? How is this not evidence that LIGO/Virgo is working?

      Delete
    14. @Matti: Unfortunately, from a skeptical perspective, one cannot exclude LIGO had access to other sources of information, GRB or otherwise. Powerful telescopes can locate such sources within seconds, faster than LIGO even noticed their (delayed) trigger.

      It took LIGO five more hours to publish their first skymap, and they have never published their waveform template. BICEP2 and other experiments have already shown that using unpublished third-party information isn't impossible.

      One still cannot exclude the optical transient is a foreground or background event (no spectral distance measurement), or an event different from the GRB detected 12 hours earlier, or a WD-WD merger looking like a BNS kilonova.

      If you want to exclude such unlikely scenarios, LIGO has to achieve, after more than four years, a first successful real-time counterpart prediction.

      If you believe in LIGO, as I do, you should be hopeful LIGO will soon achieve such a prediction, as 170817 clearly didn't meet this basic scientific requirement.

      Delete
    15. @Matti, re "We look at the sky a lot so we can calculate the probability of a coincident independently of the underlying distributions, which should be poisson. But what p value is good enough is kind of unrelated to my argument and up to each person."

      I think we are talking past each other, as the saying goes.

      What "coincidence" are we trying to estimate a probability for?

      That two random regions of sky, e.g. a box x" by y" (A) and one a" by b" (B), will overlap?

      Add time: A is within [t1, t2]; B is within [t3, t4]?

      Add distance (or redshift or ...): A is within [d1, d2]; B [d3, d4]?

      But none of this is "random". For example, we have some constraints, based on decades of observation, on where GRBs occur, in terms of sky regions, time intervals, and redshift ranges. How do we incorporate that into any estimate of probabilities?

      Delete
  26. What LIGO has produced with certainty is the world's most sensitive seismograph, which is possibly also affected by extraterrestrial events. Given the ultra violent and nonlinear nature of the earth's core, a must should therefore be an equally sensitive seismograph which is *not* affected by extraterrestrial events.

    ReplyDelete
  27. I've been reading the comments on your YouTube video on this topic. So far, I've reported one piece of porn, two marketing pieces for the Electric Universe (i.e. Velikovsky nonsense), and one promoting some religious cult.

    ReplyDelete
    Replies
    1. Yeah, sorry about that. It's too many comments for me to screen, so I don't really know what to do. But the YouTube spam filter works much better than Blogger's. The vast majority of junk actually does get flagged automatically, so you never get to see it. Thanks for your help with this.

      Delete
    2. "one piece of porn, two marketing pieces for the Electric Universe (i.e. Velikovsky nonsense), and one promoting some religious cult"

      I'm surprised that there wasn't something which combines all three.

      Delete
    3. Update: of the comments I reported, they're all back (except for one that explicitly threatened violence)!

      When you report, you get a message like this: "If we find this content to be in violation of our Community Guidelines we will remove it."

      So I guess linkspam marketing and soft core porn are not in violation of YouTube's so-called Community Guidelines.

      Can you, Sabine, yourself delete YouTube comments, on your own video? Delete them so they don't come back?

      Delete
    4. Update 2: I spent some fruitless, frustrating hours trying to understand how YouTube works, re comments on a video. There are dozens of documents/webpages/etc on YouTube policies, but precious few which give clear, unambiguous guidelines for comments and how to "Report" them appropriately. Yes, promotional material/comments is definitely reportable, but screamingly obvious promotional material for Electric Universe (e.g. "check out this: {YouTube link to a promotional video}") gets a pass from whoever reviews reports.

      I give up.

      Delete
  28. Sabine,

    Searching around the internet for more about this I came across an article titled "Fake-News-from-the-Universe" by Alexander Unzicker hosted at heise.de in July 2019. I don't know if you are aware of it, it provides more specific information on the failings of the collaboration.

    (do links work in comments? Link:) www.heise.de/tp/features/Fake-News-from-the-Universe-4464599.html

    Damn those German inquiring minds!

    I particularly liked the part with the modification of the subject line of the alert mail for GW170817 to make it look that Ligo/Virgo "pointed" first.

    Also do all the subsequent failed predictions (like the one of May 10, 2019) count as a falsification of the predictive ability of the instrument?

    I mean if there are many mispredictions and eventually one correct one, should there then be followed by a meta-analysis of the chance of the prediction beeing just due to chance?

    Because since the early days it was said that 10-20 detections and about 1-2 of simultaneous optical matchings should be produced. So far we have only a single one? Has there really been nothing more since the summer?

    Lastly, are there really simultaneous "glitches" in three laboratories?

    thanks,

    ReplyDelete
    Replies
    1. Since someone complained about the term afro-engineering above, let me say, as a German citizen with an inquiring mind, that I am deeply offended to be thrown into the same category as Unzicker.

      Delete
  29. Hi Technomagos,

    These are not strictly falsifications (it's almost impossible to falsify LIGO), but mostly false positives. LIGO retracted about half of them, the other half simply remained without optical counterpart. We will never know if these were real GWs.

    But at some point, the many false positives, and the absence of real positives, could indeed question LIGO's reliability, or even functionality. All we currently have is the murky case of 170817, which wasn't a real-time prediction.

    Regarding glitches, yes there are often dozens of correlated multi-detector glitches every day. People easily forget how *extremely* sensitive these instruments really are. The question of course is, sensitive to what...

    LIGO had one other good opportunity to prove its functionality, and that is continuous GWs from fast-spinning pulsars. Almost everybody expected these, but again LIGO found nothing in all 221 cases. Perhaps LIGO is right, perhaps not, nobody knows. There was not a single media report about this non-discovery.

    ReplyDelete
    Replies
    1. Thanks, this clears the picture a lot. I wonder at what stage will the community start to wonder how to cope with the possibility of Ligo failing to detect mutiple events what was promised

      Delete
    2. @Technomagos: "the possibility of Ligo failing to detect mutiple events what was promised" Well, it has delivered GWR transients from quite a few BH-BH mergers, hasn't it?

      For NS-NS mergers: it depends on which astrophysical models of NS-NS binaries you choose ... in some LIGO/Virgo should have, by now, detected GWR from their mergers (driven primarily by orbital decay due to GWR!); in others, perhaps these occur at most once a decade (with the constraint that they are close enough for the full gamut of EMR to be detectable by equipment available today).

      Delete
  30. September 2019: "The LIGO Scientific Collaboration and the Virgo Collaboration have cataloged eleven confidently detected gravitational-wave events during the first two observing runs of the advanced detector era..." (ArXiv 1908.11170). The paper concludes with this: "...encourages the scientific community to analyze its data." A challenging paper (to say the least) at 54 pages.

    ReplyDelete
  31. Philip,

    My sincere apologies. I was just joking, don't assign any meaning to it. Consider it just poor humor, I know as a Greek I have been offended myself when bundled together with certain public figures, politicians mostly, labeled as "the Greeks".

    I don't even know that author, why is he offending?

    ReplyDelete
  32. I'll continue my series of posts on transients in astronomy (upthread) later.

    A first go at a summary:

    New classes of transient in astronomy generally take quite a while to be fully accepted. A key reason for their acceptance is the "I observed one with my own, independent, equipment" kind of independent confirmation. Where what was observed was a different transient, but of the same kind. And this is so even if the transients - as a class - later turn out to be heterogenous (GRBs are a good example).

    Another key reason is very good localization, sufficient for optical telescopes to be able to identify a likely galaxy to which the transient belongs. This is somewhat strange, as localization can sometimes be better in the gamma ray part of the EMR spectrum (no zone of avoidance, for example, and the Sun never blinds the detectors).

    An almost universal feature of a new class of transient is that, at least initially, astrophysical models are, um, hit and miss at best, especially the "predictive" ones ... the universe/Nature seems to love throwing curve balls!

    So GWR transients seem very normal: little in the way, yet, of independent observations; astrophysical models which don't seem to quite fit (if only re the expected frequency of NS-NS mergers, say); few good associations with EMR transients; etc.

    The good news is, I think, that this is a very hot topic, so lots of effort, by top-flight teams (of observers and theoreticians) will be expended, looking for greater clarity.

    ReplyDelete
  33. It is of some interest to allow 'history' to be a guide. Recall that "Penzias and Wilson had measured the intensity of the cosmic radio static at ONLY one wavelength, 7.35 centimeters." (page 66, The First Three Minutes, Weinberg). Of history, read chapter three in that book and continue with his Gravitation and Cosmology (1972, pages 506-528). You will discover that "confirmed predictions" (the terminology used in the essay above) in cosmology are not as linear as all that. Weinberg begins with "let us consider what sort of spectrum we would expect on purely theoretical grounds..." then "in order to pick out the signal..." then "it remained to verify..." and finally "the case for this view is certainly good enough to warrant further examination." It all amounts to wonderful science, which is precisely my view on LIGO: wonderful science.

    ReplyDelete
  34. Isn't is safe to say that mergers of small pathological objects are unbelievably unlikely? I mean, many orders of magnitude more unlikely than supernovae? This experiment seems tied to events that are intrinsically unlikely. Does anyone worry about this? (Example - try to hit the Sun with a probe you can actually control.)

    -drl

    ReplyDelete
    Replies
    1. If one were to encloak the sun with a spherical screen it would still gravitate in the same way. Now suppose within this cloaking device I compressed the sun into a black hole with a radius on only 3km. The gravitation outside the screen would be entirely the same. This highly idealized thought experiment is an instance of Gauss' law.

      While a black hole may be very small according to the scale of the horizon, the gravitational field of the black hole extends as far as it would if the mass were in a more ordinary star. Many of these black holes observed in mergers are 50 solar masses, and this is about the mass of some supergiant stars or high mass O-class stars. The gravitational field beyond the surface of these stars is entirely the same, in a static situation with the mass distribution in the star, as if the star were replaced by a black hole of the same mass.

      This then means the effective cross section of a black hole for scattering is much larger than the horizon radius. They can indeed collide, and if two are gravitationally bound in an orbit they inevitably will coalesce.

      Delete
    2. Thanks, but the average distance between stars is something like 4 light years. Interaction of any sort is unlikely, much less a merger. Right?

      -drl

      Delete
    3. @drl: "I mean, many orders of magnitude more unlikely than supernovae?" You may be interested to know that there's a great many papers on the progenitor(s) of type 1a supernovae ... SD (single degenerate, i.e. a white dwarf) or DD (double degenerate (i.e. merger of two white dwarfs). Here's a set of slides on the topic (pretty random, there are hundreds of these): http://www.eso.org/sci/meetings/2013/DSLG/Presentations/S_II-Ruiter.pdf

      So how does this come about? First, recognize that binary stars (two stars in mutual orbit around the CoG) are common, and that our solar system (not a binary) may be somewhat unusual. Second, each star in a binary evolves, and in that evolution one star may greatly influence the other (like our Sun will go red giant and swallow Mercury, for example, only much more extreme). Third, compact binaries will lose orbital energy due to GWR. Fourth, astrophysical models tend to show that DD SNe 1as are quite likely ... they seem to happen in about the right timeframe after star-bursts.

      Oh, and have you heard of "blue stragglers"? They're found in globular clusters, and our best models for their formation is merging stars.

      Finally, the binary pulsar which gave Hulse and Taylor their Nobel will end up as a merger. There are several known neutron star-neutron star (NS-NS) binaries; they too will one day end up merging.

      Delete
    4. Also about a third or so of star systems are double star systems. If these evolve into neutron stars or black holes the gravity tango is assured.

      Delete
    5. The average separation of binary systems is 1000 AU. What tango? Over what time scale? I get the strong feeling that this is wishful thinking, not backed up by an actual calculation of how probable a merger of even two stars of some diameter.

      -drl

      Delete
    6. There are plenty that are in tight orbits. The classic case is Algol that is close enough to have accretion between the two stars. It only takes a few of these to set up mutually orbiting black holes.

      Delete
    7. @drl: the study of binary stars is old indeed. There are textbooks written on the topic. An average separation is useless for making intuitive conclusions, you need the distribution. Check out this for a brief overview on interacting binaries:
      https://jila.colorado.edu/~pja/astr3730/lecture32.pdf

      Delete
  35. You may find information that would support the claim of gravitational waves having been discovered by LIGO

    "In a special public lecture webcast at Perimeter Institute on October 23, 2019, Gabriela González will provide a first-hand account of LIGO’s century-in-the-making breakthrough, and explain observations made as recently as this year. " https://youtu.be/ll9OwIWe01w

    ReplyDelete
  36. "... is that no one seems to actually know where the curve in the famous LIGO plot came from. ... if we don’t know how the predictions were fitted to the data."

    This is so symptomatic for the actual “science business” and it was one of the reasons for me to quit doing “professional science”. I published papers where I had to struggle with reviewers and editors who had the opinion that details of the data analysis are not of much interest for the reader, they should be published in an appendix or elsewhere. Most of interest was a sound, colorful and handy presentation of the results in order to increase the number of readers and, of course, increase the impact factor of the journal.

    Today, science is a billion-dollar business. So, get used to it or leave.

    ReplyDelete
  37. A sad episode recounted: Virginia Trimble, astronomer and Joseph Weber’s second wife, said it was only after Weber’s death that his contributions began to be publicly recognized. “They were nasty until he died because of competition for funding,” and “Once he died and wasn’t competing for money, slowly but surely all the people who had gotten the money said yes, he founded their field.” (10/20/2107, Chicago Tribune).
    We read: "Experiments using billion-dollar machines had to be approved by committees, which based their decisions not only on the objective merits of the proposals but on subjective judgments of the applicants’ reputations and standing in their fields."
    (2015, Hiltzik, The Origins of Big Science).

    ReplyDelete
  38. In these cases, there should also be electromagnetic radiation emitted which telescopes can see. For black hole mergers, one does not expect this to be the case.

    Are you saying that black hole mergers do not emit observable EM radiation? Why would this be the case and do you have any references I can read?

    ReplyDelete
    Replies
    1. Alex,

      That's right because black holes are not made of matter that could emit such radiation. They may be surrounded by matter that can emit radiation but this emission would be too small to be observable. I don't have a reference at hand, sorry.

      Delete
    2. A search using "progenitors of binary black hole mergers" produces several hits for different papers. Unfortunately, arXiv is unreachable for me, just now, so I can't check them out. But when up, try 1702.08056, 1801.05433, and 1605.03839.

      The LIGO BH-BH merger GWR transients (i.e. before Virgo joined the party) seem odd: how do such quite massive BH binaries form? Lack of good astrophysical models for their evolution means we can say essentially nothing about the incidence of accretion disks, say, or anything that may create an EMR transient. However their estimated distances (from GWR merger models) suggest that any associated EMR transients we could see may be very faint in any waveband/region. The same may be true for any neutrino transients.

      Delete
    3. The only way two vacuum black holes could emit electromagnetic radiation in a coalescence is if they are charged. Two charged black holes in a merger will act as an evolving electric dipole.

      These black holes would require a substantial electric charge for this EM radiation to be appreciable. We do not expect that to occur ordinarily. The more charged a black hole becomes the smaller is the black hole cross section for another incoming particle of the same sign of charge. So it becomes difficult to stuff more of the same charge into a black hole. We also do not expect there to be highly charged matter that implodes into black holes to begin with.

      Delete
  39. Before leaving the discussion, take a look at LIGO's latest event, which turned from 87% terrestrial (i.e. noise) to 95% black hole candidate (not a glitch, but perhaps different triggers by different search pipelines). I hope you realize the challenges we're facing, whether or not 170817 was genuine. https://gracedb.ligo.org/superevents/S191105e/view/

    Cheerio!

    ReplyDelete
  40. You write "Hulse and Taylor did this by closely monitoring the orbiting frequency of a pulsar binary. If the system loses energy, the two stars get closer and they orbit faster around each other. The predictions for the emission of gravitational waves fit exactly on the observations. Hulse and Taylor got a Nobel prize for that in 1993."

    This is a frequent misconception but still wrong. Hulse and Taylor got the Nobel Prize for the *discovery* of the NS system in question. The measurements of the orbital decay in the years after the discovery were made after Hulse had moved on to different fields of science, instead those studies that then led to an indirect observation of GWs were done by Joel Weisberg, Taylor, and others. E.g. the iconic plot you show in your video near 2:45 is from Weisberg & Taylor: "The Relativistic Binary Pulsar B1913+16: Thirty Years of Observations and Analysis" https://ui.adsabs.harvard.edu/abs/2005ASPC..328...25W/abstract , see references in that paper for earlier studies by Weisberg, Taylor et al.

    A longer description of this whole story and J. Weisberg's often neglected role in it (and the fact that Hulse was no longer involved in it) can be found here: "The binary pulsar and the quadrupole formula controversy" by Daniel Kennefick : Eur. Phys. J. H42, 293–310 (2017) , online here : https://link.springer.com/content/pdf/10.1140%2Fepjh%2Fe2016-70059-2.pdf

    ReplyDelete
  41. Apropos, when guessing whether the L/V data are real GWs, it may be useful to consider strangeness (if any) in the supposed GWs, like e.g.:
    https://astrobites.org/2019/10/31/77569/
    https://arxiv.org/abs/1910.03601

    ReplyDelete

PLEASE READ THE COMMENT RULES BEFORE COMMENTING.

Comment moderation on this blog is turned on.
Submitted comments will only appear after manual approval, which can take up to 24 hours.
Comments posted as "Unknown" go straight to junk. You may have to click on the orange-white blogger icon next to your name to change to a different account.