Hey Dr. Becky, why would you expect so see MOND effects on such "small" scales like a few AU? Wasn't MOND suggested to only differ from Newtonian on large scales like several thousand or more light years, so only at large differences from the galactic center, where the radial speed curve of stars is flattening out?
And sorry, but the Hernandez data from 2012 can by no means be fitted with any of the presented curves, no matter if Newtonian or MOND. The way the dots are scattered, suggests some very different correlation, if any, too me (with basic statiscal knowledge). How would it look in a non-logarithmic diagram? 🤔 How comes they claim such regression or fitting curves?
There are some other theories I frequently ask scientists on RUclips to look at, with little effect 😢 I think about "Retarded Gravity" by Asher Yahalom. Is it worth talking about? Some other propose to regard some relativistic effects on galaxy scales. Unfortunately, I can't figure out the math myself 😢
What if mass has a greater influence on "Stretched" Space-Time? Has that been considered as a solution to negate the need for 'Dark Matter'? Dark matter seems like such a band aid type of solution.
Credit to good science here. A proponent of a theory who, when new data (and methods in this case) became available, used that data to rigorously test that theory - and then finds the complete polar opposite - is science approaching its ideals and nice to see. Big credit also to the GAIA mission for collecting these observations!
Praise to good analytical science, while no praise to sloppy, even bad science, even if it was an alternate idea to a tough question. His paper should have never left the arXiv Pre-print server if he used it. A 10 sigma significance for breakdown of standard gravity at weak acceleration is direct evidence, says Chae, it is, until it really wasn't. Thank God for somebody questioning the sloppy math analysis and engaged in true peer review. Too much bad and contradictory science is being published by professors into certain media the last few decades. It's like "we just have to get it (my name) published....excellent science be damned." To be fair, I can't always blame the authors (professors), the schools themselves drive this issue and of course the schools might point to the expectations of donors...etc..
A scientist is someone who has a guess at the world and then tries to prove themselves wrong. A pseudo scientist is someone who has a guess at the world and then tries to prove themselves right.
@DrBecky . I have just finished reading, (well listening to) your book. I enjoyed it from beginning to end and would recommend it to anyone with an interest in astronomy. Like Stephen Hawking's books, the book explains things in terms that people who are not scientists can understand, and therefore making the more complex books readable. A thumping good read. hats off you and there's a pangalacticgargleblaster waiting behind the bar for you at Milliways.
Getting closer to the end of the video.. presented well, interesting thanks... that said this feels like arguing over things they dont have enough(or accurate)information to be doing.
That first chart with the negative slope is utter nonsense. The slope, is by simple visual analysis, zero or very close to it. Check the relationship with the Theil-Sen Median Slope method. You seem to be searching for oak trees in a pine forest.
that graphic showing the bending of space time @3:38 has a maximal dimple displacement at the middle of an object, but doesn't force tend to zero at the center of the earth? Is it the integrated space time path from center to flat space still being greater being represented even though force goes to zero?
Yeah in general relativity what you can do with gravitational fields you can break it down to smallest thing to the largest thing and the volume of gravitational fields along with the volume of the universe. You also have to bring into this space in the universe with no gravitational fields this where it becomes mind boggling.
Did they do their analytical method to the worse data? That would help eliminate the possibility of a systematic bias in the elimination criteria that may be picking out only systems that collapse into Newtonian Dynamics and not ones that exhibit the difference for some unknown reason. If the worse data shows a weaker conclusion in favor of Newtonian gravity or maybe no conclusion at all, that would support their work. If it shows MOND that is an interesting result that would require a closer look at the elimination criteria.
The elimination criterion was the uncertainty in the velocity. If your effect only shows up in the least accurate data I think that requires questioning the effect, rather than why you rely on better data.
@@garethdean6382 Unless there is some systematic reason for the uncertainty. I'm not defending MOND here, I HIGHLY doubt it is accurate. But I'd like to know why a little more about why it looked so accurate in the first study. I don't think they were thorough enough on that exploration, at least from what was presented here. Reading the original paper is beyond my capabilities as someone who wasn't good enough at math to complete the physics major.
I see. I don't know how much can be gleaned from the original paper. Looking over the second paper's discussion they seem to suggest there's a bias in the least accurate data; that is, any given measurement isn't equally likely to be inaccurate but instead the lowest speeds (and widest binary separations) have the largest errors and tend to inflate the measured velocities. (Which makes sense, an accuracy of +-1 will mean more to a measurement of 3 than one of 30.) So the widest pairs, where the effect of distance would be greatest, would show both the largest errors and an unexpectedly higher than expected speed.You'd see MOND seemingly arising right where it was expected, due to the way GAIA worked. If true then narrowing the error bars on these data points should also diminish the observed effect.
I am so impressed by the rigor of a scientist who did not hesitate to show where his earlier paper missed the mark. Just as a lay person, I really respect a scientist who can say, "we were wrong about that explanation." And can then go on to say "we still haven't ruled out using these other approaches." That's a really open mind. Yay, Banik, and yay, Dr. Becky!
He's a big of a publicity seeker, but could see where the big news would be. Quite clever to push a clearly incorrect wackjob theory, then use actual data when it's clear which was the tide was turning. It's a technique used all the time by charlatans and religious leaders, "once I was lost", "I was blind but now I see". See also those "reformed criminals" who help law enforcement, or even talking to an ex-smoker. I wonder if the Chinese guy was in on it or just a clueless fame seeker himself?
The “dueling sigmas” aspect is a nice reminder that these sigma values _aren’t_ measuring the certainty of a given theory being right, but merely of a certain result being due to chance given the data used and assumptions made. Our certainty that the entire Gaia dataset wasn’t replaced wholesale by alien hackers is surely way below even 10-sigma.
Yeah, p < α% means that _if the null hypothesis is true_ and _if the methodology and data are correct,_ then if you repeat this experiment many times with different samples, only α% of them will have results this extreme in the long run. It doesn't tell you the probability in the other direction, i.e. it doesn't say that there is an α% chance that either the null hypothesis is false or there was something wrong with the experiment. There may be other reasons why the null hypothesis being false is even less likely than α%. For instance, suppose I pick up a die that I have no particular reason to suspect is loaded to roll 6s, but I decide to test that hypothesis anyway by rolling it three times. If I roll three 6s in a row, that gives me p = < 0.005, which in most fields (not astronomy) would be considered highly significant. But that doesn't mean there's a 99.5% chance that the die is loaded to roll 6. After all, most dice are not loaded, and most loaded dice are not loaded in that way. Even though the p value is low, the alternative hypothesis is probably still wrong. It's more plausible that I just got lucky than that this die I selected arbitrarily happens to be loaded. And then there is the issue that there is normally not a single alternative hypothesis. Even if I conclude that the data favors one hypothesis over another, that doesn't mean either hypothesis is correct. Maybe the die rolls failed to be uniformly distributed, but not because the die was loaded. Maybe it is actually misspotted with six 6s, or maybe it sat out in the sun one day and got warped, or maybe there's some sticky stuff on the 1 spot, or maybe lots of other things. That's a real problem in experiments like this one where zillions of simplifications and assumptions are made regarding things like data selection, binning, quantization, and even selection of statistical test, not to mention the many physical assumptions that go into the model. Even if each of these is reasonable on its own, combining many mediocre parameters can give a garbage result. SO what looked like 10σ can turn into -16σ in the blink of an eye.
@@SloverOfTeuth Of course I just made it up. My point is that 10-sigma and 16-sigma both represent odds so bogglingly close to zero that _any_ possible external source of error, however outlandish, is bound to swamp them. I chose an example that was silly but clearly not _impossible_.
Great follow-up on the first video on MOND; thanks for getting it out so quickly after the paper came out. So good to see the scientific method being so clearly shown in operation using the same data set but applying greater rigour to the body of data used.
Wow! Dr Becky...your explanative clarity of this complex material is wonderful!..you really do bring an appreciation of the science papers to even a lay science curious individual like me!
Things I love. 1) the way you can summarize the work so far, explain key differences and the consequent outcomes. 2) Fights between scientists. They bring all the data and destroy papers so completely. I wait to see Chae's response. 3)un-cen-turtle-tees.
Statistics was probably my least favourite subject in uni, narrowly edging out the dreaded “partial differential equations” for that dubious honour. I’m so glad that there are smart people in the world (like Dr Becky and the other authors mentioned in this vid) who can not only understand the math, but actually apply it to solve the great mysteries of our age :)
I haven't set foot in a university in more than a decade, and I had a nightmare about walking into my PDE course to find I'd forgotten that we had a midterm that day.
@@eazegpi I was fortunate to have a really excellent supervisor for my first year probability course at Cambridge as part of the Maths Tripos. The lecturer was OK, but the small group tutorials or supervisions with just two undergraduates were really very critical. My supervisor Bryn Garrod probably made a very substantial difference to my understanding of statistics. I knew at the time that I wanted to do astrophysics later (I moved over to the Natural Sciences Tripos for my second year), but statistics was not something I understood well. I automatically skipped all the statistics questions on the special admissions test for the Maths Tripos at Cambridge known as STEP, which limited the choice of questions (I never even looked at that part either in practice papers or in the real exam because I just knew I would not be able to answer it, let alone do so in a timely manner). Bryn made a vast difference by really explaining the concepts very carefully. He was also a really great friend in other ways: we were both on the cricket team and he helped me to settle into Cambridge. It was quite a sad day when at the end of my first year it was his time to leave Cambridge. I can hardly believe that ten years after automatically skipping all the statistics questions on STEP papers, I managed to do such a careful statistical analysis that will forever be remembered as basically extending the precise Solar System constraints on modified gravity theories outwards by an extra factor of 100, so from about 100 AU to about 10,000 AU or 10 kAU, albeit at lower precision. Obviously I did have to work hard, not only in my first year at Cambridge but later on, doing a special statistics course at the start of my PhD to understand MCMC better. But I am sure the foundational concepts of statistics and probability theory are something I learned from Bryn. I was not scared of statistics afterwards, and gradually started being the statistics expert on projects I was involved with.
Thanks for a great video. I'm starting to become skeptical that Bayesian statistics is being applied correctly in the majority of these papers. Between these two papers something is wrong, because the statistics should account for possible error in the data. Getting that level of confidence for either paper seems fishy.
Beyond the result of dismissing MOND, I’m really impressed with the methodology and rigour of the analysis. This is science at its best. But, I’m with you. I’d rather be lying on a beach at night, sipping a margarita, and staring up at the stars. Especially with a 40 inch telescope available when I feel like getting up and getting “serious”.
I actually love disseminating the draft to trusted science communicators to prepare before the wider press gets hold of it too. I don’t know if this was the aim but I can see this strategy working extremely well to preempt the wild speculation that often happens when mainstream media scrape a pre published paper.
Hi @DrBecky ! Aren't they making a prior assumption on the distribution of eccentricities? Do they have a way of supporting their chosen distribution which was then used to model data from the different gravity models?
This is explained in some detail in the paper. The eccentricity distribution is a free parameter, though it needs to be of the power-law form. There are good reasons why the eccentricity distribution has very little effect on the inferred gravity law.
@@indranilbanik3424 Kind thanks for the answer Dr. Banik! I haven't read the paper and was just curious whether any of the assumptions made on these unknowns had an impact on the conclusions.
@@FredPlanatiaThis is addressed through multiple variations to the nominal analysis in which different choices are made for assumptions we need to make but are unsure of and could plausibly have done a bit differently. That is in the discussion section, with a nice summary table showing how much the inferred gravity law parameter shifts in each case. All analysis variants overwhelmingly prefer Newtonian gravity over MOND.
@@indranilbanik3424 Thankyou. I will have a look. Its very interesting and encouraging to see this type of collaboration in a research paper. It would be very interesting to hear how the collaboration came about and how the analysis and various tests were agreed upon by the different parties to arrive at a conclusion everyone felt was sound and unbiased. I don't mean in a YT comment reply to me, but rather in an interview format with @DrBecky. I think that would be an extra cool demonstration of science at work!
I admire Banik's rigorousness for a definitive scientific conclusion, unlike many researchers who throw their research (worse yet, with statistical sugar and hope that people buy it). I need to look back on "my" research mentality, too. On the other thought, I always wondered who was genuinely convinced of MOND except for a few astrophysicists studying it.
It was already pretty dead with its failure to deal with baryon acoustic oscillations in the CMBR, to be fair. MOND has always been fringe with a strong taste of crank.
Dr. Becky, I feel lucky to be able to learn from your scientifically trained assessments! Up until now I had not heard of Bayesian Statistics! Your unique viewpoint in itself helps me to understand more than I would have otherwise bend able to ascertain on my own!
That leaves the small problem that in a century of looking, we have not found dark matter. There are other studies that claim to have falsified dark matter to a high sigma. Maybe, as Dr. B suggests, we need a new theory of gravity. I would also like to know how the authors of the first paper respond to,the second.
we have found dark matter. We've found it in places where it shouldn't be based on the amount of normal matter that's present. And we've found places where there isn't as much dark matter as there should be based on the amount of normal matter. If gravity just needed to be modified, we'd always find the exact same correlation between matter and dark matter but that isn't what we see. We just don't know what it is, but it's some kind of stuff that can be or not be in a particular place, not just some mathematical error.
We haven't found a particle. We have observations that strongly support its existence. The lensing observations of colliding clusters being the most persuasive. Last time I looked, MOND needed 'some' dark matter to explain them.
@@ianw7898 Dark matter is a theory that has worked in some contexts, not in others. Initially, they imagined something uniformly spread throughout the universe. That did not work out. They kept having too make the math more complex. They have found galaxies with little dark matter, assuming it exists. The same people Dr. B talked about emphasized that their conclusions should not be seen as support for dark matter. The bottom line is we don’t fully understand gravity. The general theory of relativity is the best we have done, but in some contexts it does not work. My own guess is that if we ever figure it out, it won’t be dark matter. Most likely, we would have found it by now, if it existed, and we have looked hard for a century. People, including astrophysicists, have a hard time not getting attached to inadequately proven ideas. We need to learn to say we just don’t know.
I am 9 years old and i love ❤ space and i want to be a astrophysicists when i grow up. i love your videos i also have a question for you ,you know how before the big bang their was nothing i don't get how there was nothing no space no nothing it just seems really confusing and i'm hoping you can answer this question bye😊❤🎉😮
I would have like a bit more discussion on the data screening processes. What possible biases are introduced by the filtering? Would it be a logically expected result to see something like MOND come out if you fed in random errors to velocities?
I discuss this in the section of my paper related to the comparison with prior results. Yes, velocity errors can definitely make an underlying Newtonian population look Milgromian. That is what happened in the analysis of Kyu. I have explained in some detail why that is. I also explicitly demonstrate this by finding a MOND signal in his data, but later applying a stricter quality cut, which removes the signal.
Orbital velocity depends on two factors, as I understand it: mass of the objects and distance between the two objects. So we can "measure" (calculate) the combined mass of the two objects from Doppler velocity if we can measure the distance between the objects, or we can measure (calculate) the distance if we know the mass. We deduce the mass (I am assuming) from spectral luminosity which provides us with estimations according to type, age, etc. But, what if our basic assumptions about this are incorrect in some cases? Perhaps a star with sn enormous iron core will provide more mass than we conclude because our only model that fits the spectra doesn't include a massive higher element core, but, rather a carbon or oxygen core? By narrowing the data filters we are effectively reducing the input error percentages?
What I still don’t understand is, what is MOND description of gravity? General relativity has spacetime, mass curves spacetime, how spacetime is curved tells you how mass will move. So if GR is wrong and there is no spacetime. What or how is gravity described in MOND?
It isn't, but there have been attempts at making a MOND-like version of general relativity (eg. TeVeS, which includes tensor (gravity), vector and scalar fields instead) but they haven't gotten very far.
I just love that the lead author was really hoping for a different outcome (I mean probably). I mean that had to be hard for him too, but this is how science wins.
I love your voice! The accent, rhythm, enunciation, variation in speed and emphasis - all make the content not just lovely to listen to, but easier to understand. Thanks, Dr. Becky!
Eric Laithwaite defeated gravity decades ago.. Using a very large weight lifting weight spinning fast on the end of a 6 foot bar.. Scale it up a bit... To Star or Galaxy size.. Bearing in mind for instance the stars are spinning etc as well.. Add in the equivalent of super conductors, scaled up and heat cold gradient etc.. A hole heap of things can happen.
The reason why nothing makes sense is because Neuton took Kepler’s equation for charged particles and substituted mass. Where does it say charge= mass? Then why are they used as the same terms. This is why they think of stupendous densities and energies. There is one fundamental force in the cosmos and everything is a manifestation of the material transforming Ætherial radiation resulting in magnetic fields and all we perceive is the magnetic field. Electricity minus magnetism is dielectric, The photo electric effect of light this is actually Einsteins only contribution to science. Then he proceeded to conceal this fact, hence E=mC2 and why it’s never used in anything. Everything is electrical because everything involves the electron which is the Interface between the material and the Ætherial.
I looove your nails! I'll do a blue with green chrome powder next, thanks for the inspiration! And obviously, thanks for communicating this result. The evidence seems crushing for MOND.
Same! They're so eye-catching, they really accentuate the gesticulating (in a good way!). Also am here for the science obviously, one of my favourite science communicators, but yeah you go girl!
@@indranilbanik3424 Often on YT, including a link to anything other than YT means the comment doesn't appear. Depends on the setting that the channel owner has. Otherwise you get links to all sorts of pseudoscientific nonsense that is not even close to the peer-reviewed literature! Probably best to just name the paper.
According to Marcel Pawlowski's thread on X/Tweeter things are not so clear yet. Chea's sample contains more tight binaries which are used to calibrate the analysis (no MOND effect expected), he finds also less triple systems than Banik et al. In case Banik et al mistakenly included triple systems in their sample, this tends to favor Newton's behaviour. Incidentally Figure 12 of Banik et al preprint shows a better fit with MOND than Newton for the wider binaries of their sample (r_sky > 5 kUA)
It would be interesting if we found out that General Relativity was also out by a small fraction with very high confidence, implying that general relativity is also just a set of special cases of an even more overarching theory that reduces to special relativity in most common cases.
Hi-Dr Becky I am a big fan of your content and a massive science nerd unfortunately I don’t have a PHD.. Only a couple of diplomas in engineering and computer science.. The reason I have commented is are you still taken suggestions for video reactions.. The series I would love to see you react to if you have time in your busy schedule is a series called FRINGE which is about a team that investigate fringe science events it is a very very good show.. I will try to send you a small clip.. Thank you From a big fan..
Ah I was trying to beat my record which was giving the first like to ur videos but today i was just sooo amazed by a book about universe I missed it but i promise I’ll beat it next time I’ll be here in nanoseconds 😂😊
There are galaxies with very little or no dark matter effect, and galaxies with large dark matter effect but very little visible matter,. MOND enthusiasts have not yet even provided a viable theory for it that can explain all the things general relativity explains. Yet people still argue about whether MOND could be right. What happened to the days of Einstein, Dirac, Bohr, Fermi, Feynman, Heisenberg when physicists were smart?
When I first heard of MOND, it was with the voyager anomaly, which has subsequently been explained without using MOND. Frankly at that point I lost interest and the data at that time was really mixed. Thanks for the clear explanation - and congrats to the authors of this stellar analysis.
As dark matter is so elusive, I had high hopes for MOND. But if the GAIA data is casting serious doubts, I expect EUCLID will provide the nails for its coffin . .
As someone that always looks for answers, I hate that there are areas we still don't have answers to! Our current understanding of gravity only works if we add 3 time the matter and say it behaves similar to but not exactly like matter. Oh, and also doesn't work on small scales so we also need "special" in addition to "general". All this says nothing about our ability to actually calculate what we want on the scale we want either. More mass objects just cause an exponential increase in calculations needed and when we got 100-400 billion stars and more planets ... sheesh!
My take on this is simple: If you need to "make up" dark matter to make your calculations/ equations work, your theory is probably totally wrong and you're "dancing in the dark" (pun intended) 😉
my theory is pretty simple. If you just needed to tweak the equation used to calculate gravity there would always be the same amount of dark matter everywhere. But that's not what the universe looks like. But there's galaxies where there is too much dark matter for the amount of normal matter, and galaxies where there is too little dark matter for the amount of normal matter. That proves that it is stuff, even if we don't know what that stuff is.
Yeah, this is what many laymen think. Thinking this way is too simple. The idea of dark matter isn’t an arbitrary I.e. general purpose, fudge factor. Rather, there are dynamics for how dark matter is believed to behave, and how it would behave seems to match up with observations. Gluons also, aiui, do not directly interact with photons. Yet, we conclude their existence, and QCD, based on what we do observe. Do you have an idea for why you think that dark matter is “dancing in the dark”, but QCD isn’t?
Galaxy rotation speeds are not the only evidence for dark matter, but it just so happens that to understand the better evidence you need to understand fourier analysis first, so its inaccessable to most people.
This illustrates really well one thing I like so much in science: admitting when one is wrong. Banik disproves the theory he was such a big proponent of himself (with his team). Meanwhile in France we have Raoult still trying to prove hydroxychloroquine cures everything...
yes, i think that is one of the things science can teach. This humility regarding your own ideas, being able to bow to the facts and data. But due to human nature and hubris and now the financial factor has been added, it becomes difficult to recognize that you were wrong.
Are 10 sigma or 16 sigma levels of confidence actually meaningful? Those kind of numbers seem preposterous! I used to think particle physics was pretty demanding wanting 5 sigma results before calling a discovery, but here we have a supposed 10 sigma result being overturned by a 16 sigma result? I can't help but think that kind of numerical confidence level can't be realistic at all and could even be a warning sign that something is up somewhere. Can we meaningfully test anything and claim that level of confidence in the result?
Hey I remember back in the late 1990's and early 2000's there was another theory about the speed of light slowing over the age of the universe. Effectively saying that the universe isn't accelerating, just our light based meter stick is getting shorter so the distances look like it's growing. Anyway I was wondering if you have ever heard of this and what was it that disproved it as it never really caught on. And if it was just thrown out of hand, wouldn't now be a good time to revisit some of the more "crackpot" theories to see if any of them match the data better? (I don't remember anyone using Bayesian analysis back then.)
As a Scot, I like to imagine the Banik paper as an insult to the original claim is a Banik burn... The Battle of Bannockburn... I'll... the... door... here... bye.
Dark matter still seems to be "justified" by thinking of galaxies as scaled up versions of solar systems - which they aren't. Black holes actually make up a relatively small portion of the total mass of their constituent galaxies unlike stars in solar systems. Most of the mass of our galaxy is made up of red Dwarfs which aren't visible to the naked eye, so how many would you expect to see on M31 etc? Is there a paper which actually calculates the velocities of stars as a function of distance from the central black hole. Does it take account of all the gravitational contributions of all the stars that lie in between the black hole and any given star? I have never seen such a paper but would be interested to read it if anyone knows of a public domain copy.
Watched for years, but I'm writing something SciFi ish, and you're inadvertently changing my plot with every new episode! Thank you for a truly mind blowing Channel. Huge hugs.
I think you should make a hoodie or t-shirt with a word cloud of all your mispronounced words, maps by the number of occurrences. Just for fun... your bloopers are a joy...
All of which goes some way to explain why we call the first day after the weekend, MONDay. It's just full of stress, ambiguity and uncertainty, and it defies simple explanation. It does seem like a convenient answer when you first look, but when you get down into the details ... One does get down to questions like Why does my reality circle clockwise or anti-clockwise while it's going down the S-bend ....
The one I wish I had the resources to investigate is one I'd term EAGR, enhanced application of general relativity. Basically it's the idea that there is ambiguity in what inputs we should give the the GR tensor and we might get some miles out of tweaking, actually increasing, our inputs. A galaxy itself is a very high energy system besides the mass that is in it. Massive things are moving at relativistic speeds vs one another. Huge masses are separated at distance. Rotational flow of mass at relativistic speeds drags space-time. There is time dilation. If we plug in some more of these values we can get more mass than observed mass especially in the middle and edges. And if more energy->more mass->more force->more energy->more mass->more force. The whole thing could compound on itself. Hopefully it would converge, but maybe you really do get 6x more mass by the time it converges.
So not an argument for dark matter but another nail in MOND’s coffin especially as Banik is the lead author - really thorough and impressive paper. Thanks for explaining it so well Becky. The weirdest space news I saw this week was the claim that a satellite galaxy of the Milky Way has been found comprising almost solely of dark matter - I can’t find any real details about it though.
Dark matter can not be tested and proved or refuted. Of any observation does not fit the dark matter predictions, one Just change the massa and distribution of dark matter tô explain the results. Its the ultimate dragon in my garage theory
@@geok1ng Ehh, if someone said there was an invisible dragon in their garage, I'd say "weight it to prove it exists". Same is true of dark matter. It isn't crazy to theorize that there are particles that exist that don't interact via the electromagnetic field. Just makes it harder to measure them. If anything, it almost feels more crazy to me, that more particles wouldn't exist that don't interact via some of the fundamental forces. The idea that everything that exists must be able to be "seen" is as crazy to me as saying all humans can see and hear is all there is to be seen and heard in the universe.
@@geok1ngyou say it can't be tested, and that they just change the mass and distribution to explain the results. But do you know what happens next? They test to see if that's accurate. There is testing all the time to see if various models work.
Wow you explained this so well i feel like i understand it, an really that's impressive if you can take the average person an start explaining gravitational physics to them an make them feel like they know whats going on even though they haven't been in school in 15+ years lmao. Kudos to you Dr. Becky.
Talking about GAIA and dark matter. Will you cover on your channel this paper by Yongjun Jiao et al. (2023), "Detection of the Keplerian decline in the Milky Way rotation curve"? I'm really interested what your thoughts about this.
With respect to Banik, I do not think he is dumping his support for MOND. He is just saying that Chae et al's work does not support it. Personally, I gave up on MOND being relevant a long time ago.
I think it’s conclusion is that MOND does not apply to these type of gravitational interactions between two stars. It hasn’t ruled out for Galactic scales yet.
This is the kind of science that makes me happy. Someone who proposed something, continued to test and validate, is presented with new info that nearly completely refutes their claim, then accepts this. Props to that team for not having blinders or getting railroaded on proving their theory was correct.
This is science at its best - objectively answering the question "what model fits this data best" - it takes a lot of humility and scientific honesty to be first author on a paper saying that in this scenario there is no evidence for the theory you are basing your career on. I can imagine him looking at that first pass data saying "that looks messy, we can clean this up and do better" followed by many rounds of research group discussions of "what have we missed, why is this not showing MOND", eventually leading to the high quality analysis we see in this paper.
It wasn't the same team doing this study. The first study was done by proponents of MOND. The second one was done by detractors of MOND. Both saw what they wanted to see, in the same data.
There is an explanation on Redditt on how Banik also made an error. Due to a modelling error. This is explained well by Hernandez and Chae: I don't understand it at all but this still shows that the explanations are not decided at all.
Love your channel. Outer space has always been an interest, being a big sci-fi fan, and you would make a great Professor. Could you please answer a question for me? I heard that Apophis was supposed to come about 38k miles away from Earth in 2029. But since then I heard that Earth has started shifting on it's axis. Could this shift be enough to put the Earth in the path of Apophis now? Could you please look into this & get back to me or make a video about it? Thank you
I'd like to know if that change in the tilt is enough to make a difference of 38k miles. It doesn't seem out of the realm of reality that it could. I thought I'd ask someone that might be able to find out. I highly doubt that I'll still be around by then to find out, so thought I'd ask before it's to late.@@XellithUS
@@piratelordgrumpy9659 Its just a little bit of axial precession. Like how spinning tops wobble a bit (actually exactly how spinning tops wobble a bit). This isn't the earth being physical moved up or down on its axis of rotation, its just a small change in the direction that axis points. It doesn't change the orbital path.
@@piratelordgrumpy9659 The Earth wont change its orbital path unless its acted upon by a force. A tilt in its Axis will not push or pull it, therefore it wont change its orbit. Newtons 1st law. This is an oversimplification, but if you have a spaceman going around the planet in his space suit, they cant change orbit unless they have a jetpack, no matter how much they spin or tilt on the spot.
Euclid made a guest appearance in that animated image shot of the JWST & GAIA orbits. They just released the first science images and it looked so awesome; what a fabulous telescope.
Great video. Mond have never been able to explain both rotation of galaxies and gravitational lensing of galaxy clusters so I have always thought that was a bad conjecture.
Wormholes linking areas of gravity. Closing in less then a nanosecond but Long enough to chang gravitational paths would explain everything in datk matter
So your idea relies on a theoretical phenomena to explain a thing we cannot observe. Im gonna need more than your word salad before i climb aboard that train.
invisible imaginary stuff needed to make old theories more correct is more like it. Wait until you find out how wrong the redshift is and all the distance calcs are incorrect.
THIS is why I love science. It would be heartbreaking for your article to be refuted, but we are now one (small) step closer to the truth, and it was done through reason and rigour. I have a lot of respect for Chae et al. They thought they had something new and took a chance, which is how we innovate. Banik et al. gave it the hard look that it needed. They went the extra mile in their very thorough paper. This is how we learn and create a better world.
Hey Dr. Becky thanks for the video and keep them coming. I'm starting to think that we are in a new epicycle version of cosmology. MOND, Dark Matter, Dark Energy all require us to jump through invisible hoops. Hopefully I will live long enough for someone to figure out why the universe doesn't care for our current theories of gravity.
Had a watch. Humm, its all about the manipulation of horribly poor data and the outcome between the two being a small deviation. Dumping 2/3 of this data because it is too far away from either result I'm not sure is the right thing to do too.
You always make these videos so interesting, thank you for making high-level science accessible to anyone. I love the idea of models having built-in sensitivity tests that compare and contrast how different parameters will affect the final result in the model. As you demonstrated, small assumptions, whether systemic or man-made, can make a huge difference in the type of data you'll collect, how you analyze the data, and ultimately what you conclude from the data. Recently, I read something online about measuring a gravitational wave that had a wavelength of over a light-year long, and that the measurements require the technology behind it to be incredibly sensitive. Similarly, on the opposite end of the energy spectrum, there are limits to what we've been able to detect. So I guess my question is this: How do you deal with all the uncertainty in cases like the one mentioned above where you have only a handful of detections ever recorded and brand new technologies to measure them with? How do you tell when data points should be rejected? Is there a certain threshold of detections you need to hit before you can claim statistical confidence in the data? I ask because my last job was calibrating radiation survey detectors, and when we hit low enough exposure rates, the manufacturers suggested taking an integrated rate over a certain period of time and calculating things that way, since there weren't enough individual detection events. Do you do something analogous in your field? In my particular example, that method wouldn't be very effective since the wave itself takes a year to pass by. If you're still reading, I appreciate it! I've always wanted to know how you're able to get good such certainty in your results with all the error bars and uncertainties flying around in the measurements and models. I'm assuming some kind of statistical tests occur during the analysis process, including the one that results in determining sigma, maybe? Thanks!
can i ask why this stuff is always comparing GR to MOND rather than some modern modified gravity theory? i work kind of tangentially to modified gravity (mostly on a particular theory) at arguably one of the best universities for gravitational physics and i'm not aware of anyone seriously working on MOND. is it just because to the public MOND = modified gravity? or are people just not aware of modern relativistic modified gravity theories?
@@tonywells6990 just as an example, phillip mannheim claims that conformal gravity predicts flat rotation curves and can do away with dark energy. i haven't worked on this theory personally but i don't see how it's any different
Just looking at the raw plots of points screams to me that data refinement is needed here. The model curves for both MOND and Newtonian gravity are basically linear, but the data points are scattered over a fairly two-dimensional part of the graph, which at the surface indicates that more is going on than just either model. As the Banik paper shows, the "more going on" is errors in the data, due to the factors you indicated.
Does MOND predict any of the observations usually attributed to GR effects: Anomalous precession of Mercury’s orbit, etc.? If not? How do MOND proponents account for those observations?
Go to sponsr.is/cs_drbecky and use code DRBECKY to save 25% off on subscription today. Thanks to Curiosity Stream for sponsoring today’s video.
Hey Dr. Becky, why would you expect so see MOND effects on such "small" scales like a few AU?
Wasn't MOND suggested to only differ from Newtonian on large scales like several thousand or more light years, so only at large differences from the galactic center, where the radial speed curve of stars is flattening out?
And sorry, but the Hernandez data from 2012 can by no means be fitted with any of the presented curves, no matter if Newtonian or MOND.
The way the dots are scattered, suggests some very different correlation, if any, too me (with basic statiscal knowledge). How would it look in a non-logarithmic diagram? 🤔
How comes they claim such regression or fitting curves?
There are some other theories I frequently ask scientists on RUclips to look at, with little effect 😢
I think about "Retarded Gravity" by Asher Yahalom.
Is it worth talking about?
Some other propose to regard some relativistic effects on galaxy scales.
Unfortunately, I can't figure out the math myself 😢
What if mass has a greater influence on "Stretched" Space-Time?
Has that been considered as a solution to negate the need for 'Dark Matter'?
Dark matter seems like such a band aid type of solution.
Quantized Inertia elegantly combines Rindler Horizons, Unruh Radiation, and the Casimir Force to explain wide binaries precisely!
Credit to good science here. A proponent of a theory who, when new data (and methods in this case) became available, used that data to rigorously test that theory - and then finds the complete polar opposite - is science approaching its ideals and nice to see. Big credit also to the GAIA mission for collecting these observations!
Yeah, it doesn't get more sciencey than that.
It's not new data. It's the same data but screened via different statistical methods.
@@Yutani_Crayven gotcha. same praise applies
Praise to good analytical science, while no praise to sloppy, even bad science, even if it was an alternate idea to a tough question. His paper should have never left the arXiv Pre-print server if he used it.
A 10 sigma significance for breakdown of standard gravity at weak acceleration is direct evidence, says Chae, it is, until it really wasn't.
Thank God for somebody questioning the sloppy math analysis and engaged in true peer review.
Too much bad and contradictory science is being published by professors into certain media the last few decades. It's like "we just have to get it (my name) published....excellent science be damned."
To be fair, I can't always blame the authors (professors), the schools themselves drive this issue and of course the schools might point to the expectations of donors...etc..
A scientist is someone who has a guess at the world and then tries to prove themselves wrong.
A pseudo scientist is someone who has a guess at the world and then tries to prove themselves right.
Space and physics is ballin’. Boy do I regret taking ag science in hs.
@DrBecky . I have just finished reading, (well listening to) your book. I enjoyed it from beginning to end and would recommend it to anyone with an interest in astronomy. Like Stephen Hawking's books, the book explains things in terms that people who are not scientists can understand, and therefore making the more complex books readable. A thumping good read.
hats off you and there's a pangalacticgargleblaster waiting behind the bar for you at Milliways.
Great video and explanation. Thank you for making it.
Getting closer to the end of the video.. presented well, interesting thanks... that said this feels like arguing over things they dont have enough(or accurate)information to be doing.
That first chart with the negative slope is utter nonsense. The slope, is by simple visual analysis, zero or very close to it. Check the relationship with the Theil-Sen Median Slope method. You seem to be searching for oak trees in a pine forest.
that graphic showing the bending of space time @3:38 has a maximal dimple displacement at the middle of an object, but doesn't force tend to zero at the center of the earth? Is it the integrated space time path from center to flat space still being greater being represented even though force goes to zero?
More data would be nice, and more, larger telescopes etc
Yeah in general relativity what you can do with gravitational fields you can break it down to smallest thing to the largest thing and the volume of gravitational fields along with the volume of the universe. You also have to bring into this space in the universe with no gravitational fields this where it becomes mind boggling.
Did they do their analytical method to the worse data? That would help eliminate the possibility of a systematic bias in the elimination criteria that may be picking out only systems that collapse into Newtonian Dynamics and not ones that exhibit the difference for some unknown reason. If the worse data shows a weaker conclusion in favor of Newtonian gravity or maybe no conclusion at all, that would support their work. If it shows MOND that is an interesting result that would require a closer look at the elimination criteria.
The elimination criterion was the uncertainty in the velocity. If your effect only shows up in the least accurate data I think that requires questioning the effect, rather than why you rely on better data.
@@garethdean6382 Unless there is some systematic reason for the uncertainty. I'm not defending MOND here, I HIGHLY doubt it is accurate. But I'd like to know why a little more about why it looked so accurate in the first study. I don't think they were thorough enough on that exploration, at least from what was presented here. Reading the original paper is beyond my capabilities as someone who wasn't good enough at math to complete the physics major.
I see. I don't know how much can be gleaned from the original paper. Looking over the second paper's discussion they seem to suggest there's a bias in the least accurate data; that is, any given measurement isn't equally likely to be inaccurate but instead the lowest speeds (and widest binary separations) have the largest errors and tend to inflate the measured velocities. (Which makes sense, an accuracy of +-1 will mean more to a measurement of 3 than one of 30.)
So the widest pairs, where the effect of distance would be greatest, would show both the largest errors and an unexpectedly higher than expected speed.You'd see MOND seemingly arising right where it was expected, due to the way GAIA worked. If true then narrowing the error bars on these data points should also diminish the observed effect.
@@garethdean6382 Thank you. That is very helpful. Makes sense now.
Gravity is just time running at a different rate between two points in space. Mass can cause that but so do other things we can't 'see.'
This is completely irrelevant to the science, but I have to look into how young Banik is. That picture makes him look like he's a teenager!
I am so impressed by the rigor of a scientist who did not hesitate to show where his earlier paper missed the mark. Just as a lay person, I really respect a scientist who can say, "we were wrong about that explanation." And can then go on to say "we still haven't ruled out using these other approaches." That's a really open mind. Yay, Banik, and yay, Dr. Becky!
A good scientist that all scientists should aspire to emulate
Thanks very much for your comments about my paper!
It's a rare breed of scientists. Unfortunately, it will haunt their career. All the incentives are lined up for bad science.
otherwise it would be called religion, not science.
He's a big of a publicity seeker, but could see where the big news would be.
Quite clever to push a clearly incorrect wackjob theory, then use actual data when it's clear which was the tide was turning.
It's a technique used all the time by charlatans and religious leaders, "once I was lost", "I was blind but now I see". See also those "reformed criminals" who help law enforcement, or even talking to an ex-smoker.
I wonder if the Chinese guy was in on it or just a clueless fame seeker himself?
The “dueling sigmas” aspect is a nice reminder that these sigma values _aren’t_ measuring the certainty of a given theory being right, but merely of a certain result being due to chance given the data used and assumptions made. Our certainty that the entire Gaia dataset wasn’t replaced wholesale by alien hackers is surely way below even 10-sigma.
Never underestimate those trisolaran sophons!
Yeah, p < α% means that _if the null hypothesis is true_ and _if the methodology and data are correct,_ then if you repeat this experiment many times with different samples, only α% of them will have results this extreme in the long run. It doesn't tell you the probability in the other direction, i.e. it doesn't say that there is an α% chance that either the null hypothesis is false or there was something wrong with the experiment. There may be other reasons why the null hypothesis being false is even less likely than α%.
For instance, suppose I pick up a die that I have no particular reason to suspect is loaded to roll 6s, but I decide to test that hypothesis anyway by rolling it three times. If I roll three 6s in a row, that gives me p = < 0.005, which in most fields (not astronomy) would be considered highly significant. But that doesn't mean there's a 99.5% chance that the die is loaded to roll 6. After all, most dice are not loaded, and most loaded dice are not loaded in that way. Even though the p value is low, the alternative hypothesis is probably still wrong. It's more plausible that I just got lucky than that this die I selected arbitrarily happens to be loaded.
And then there is the issue that there is normally not a single alternative hypothesis. Even if I conclude that the data favors one hypothesis over another, that doesn't mean either hypothesis is correct. Maybe the die rolls failed to be uniformly distributed, but not because the die was loaded. Maybe it is actually misspotted with six 6s, or maybe it sat out in the sun one day and got warped, or maybe there's some sticky stuff on the 1 spot, or maybe lots of other things. That's a real problem in experiments like this one where zillions of simplifications and assumptions are made regarding things like data selection, binning, quantization, and even selection of statistical test, not to mention the many physical assumptions that go into the model. Even if each of these is reasonable on its own, combining many mediocre parameters can give a garbage result. SO what looked like 10σ can turn into -16σ in the blink of an eye.
Run me through how you estimated that probability. Or, let me guess, you just made it up.
@@SloverOfTeuth The probability of rolling three consecutive sixes on a fair six-sided die is (1/6)³ = 1/216 < 1/200 = 0.005.
@@SloverOfTeuth Of course I just made it up. My point is that 10-sigma and 16-sigma both represent odds so bogglingly close to zero that _any_ possible external source of error, however outlandish, is bound to swamp them. I chose an example that was silly but clearly not _impossible_.
Great follow-up on the first video on MOND; thanks for getting it out so quickly after the paper came out. So good to see the scientific method being so clearly shown in operation using the same data set but applying greater rigour to the body of data used.
This was a very in-depth and well explained video that really helps us understand the gravity of the situation in the field.
Har har.
Get out 😂
Wow! Dr Becky...your explanative clarity of this complex material is wonderful!..you really do bring an appreciation of the science papers to even a lay science curious individual like me!
Things I love.
1) the way you can summarize the work so far, explain key differences and the consequent outcomes.
2) Fights between scientists. They bring all the data and destroy papers so completely. I wait to see Chae's response.
3)un-cen-turtle-tees.
Thank you so much for this video. Your talent in communicating such a complex subject is the envy of us all.
Statistics was probably my least favourite subject in uni, narrowly edging out the dreaded “partial differential equations” for that dubious honour. I’m so glad that there are smart people in the world (like Dr Becky and the other authors mentioned in this vid) who can not only understand the math, but actually apply it to solve the great mysteries of our age :)
I haven't set foot in a university in more than a decade, and I had a nightmare about walking into my PDE course to find I'd forgotten that we had a midterm that day.
I'm with you on the statistics part (I had a separate course on probablities, also hated it). But I did love my pde course.
A lot of teachers are plain boring and have no intuition. You just weren’t lucky it’s interesting
@@eazegpi I was fortunate to have a really excellent supervisor for my first year probability course at Cambridge as part of the Maths Tripos. The lecturer was OK, but the small group tutorials or supervisions with just two undergraduates were really very critical. My supervisor Bryn Garrod probably made a very substantial difference to my understanding of statistics. I knew at the time that I wanted to do astrophysics later (I moved over to the Natural Sciences Tripos for my second year), but statistics was not something I understood well. I automatically skipped all the statistics questions on the special admissions test for the Maths Tripos at Cambridge known as STEP, which limited the choice of questions (I never even looked at that part either in practice papers or in the real exam because I just knew I would not be able to answer it, let alone do so in a timely manner). Bryn made a vast difference by really explaining the concepts very carefully. He was also a really great friend in other ways: we were both on the cricket team and he helped me to settle into Cambridge. It was quite a sad day when at the end of my first year it was his time to leave Cambridge. I can hardly believe that ten years after automatically skipping all the statistics questions on STEP papers, I managed to do such a careful statistical analysis that will forever be remembered as basically extending the precise Solar System constraints on modified gravity theories outwards by an extra factor of 100, so from about 100 AU to about 10,000 AU or 10 kAU, albeit at lower precision. Obviously I did have to work hard, not only in my first year at Cambridge but later on, doing a special statistics course at the start of my PhD to understand MCMC better. But I am sure the foundational concepts of statistics and probability theory are something I learned from Bryn. I was not scared of statistics afterwards, and gradually started being the statistics expert on projects I was involved with.
This video should've been posted on MONDay.
It's the time change in the US - by the time I saw it, it was extremely red-shifted.
Thanks for a great video. I'm starting to become skeptical that Bayesian statistics is being applied correctly in the majority of these papers. Between these two papers something is wrong, because the statistics should account for possible error in the data. Getting that level of confidence for either paper seems fishy.
The second paper does have an accounting of error in it. There is no real issue with the methodology here.
Beyond the result of dismissing MOND, I’m really impressed with the methodology and rigour of the analysis. This is science at its best.
But, I’m with you. I’d rather be lying on a beach at night, sipping a margarita, and staring up at the stars. Especially with a 40 inch telescope available when I feel like getting up and getting “serious”.
I actually love disseminating the draft to trusted science communicators to prepare before the wider press gets hold of it too.
I don’t know if this was the aim but I can see this strategy working extremely well to preempt the wild speculation that often happens when mainstream media scrape a pre published paper.
Hi @DrBecky ! Aren't they making a prior assumption on the distribution of eccentricities? Do they have a way of supporting their chosen distribution which was then used to model data from the different gravity models?
There’s a link to the research paper in the description.
This is explained in some detail in the paper. The eccentricity distribution is a free parameter, though it needs to be of the power-law form. There are good reasons why the eccentricity distribution has very little effect on the inferred gravity law.
@@indranilbanik3424 Kind thanks for the answer Dr. Banik! I haven't read the paper and was just curious whether any of the assumptions made on these unknowns had an impact on the conclusions.
@@FredPlanatiaThis is addressed through multiple variations to the nominal analysis in which different choices are made for assumptions we need to make but are unsure of and could plausibly have done a bit differently. That is in the discussion section, with a nice summary table showing how much the inferred gravity law parameter shifts in each case. All analysis variants overwhelmingly prefer Newtonian gravity over MOND.
@@indranilbanik3424 Thankyou. I will have a look. Its very interesting and encouraging to see this type of collaboration in a research paper. It would be very interesting to hear how the collaboration came about and how the analysis and various tests were agreed upon by the different parties to arrive at a conclusion everyone felt was sound and unbiased. I don't mean in a YT comment reply to me, but rather in an interview format with @DrBecky. I think that would be an extra cool demonstration of science at work!
I admire Banik's rigorousness for a definitive scientific conclusion, unlike many researchers who throw their research (worse yet, with statistical sugar and hope that people buy it). I need to look back on "my" research mentality, too. On the other thought, I always wondered who was genuinely convinced of MOND except for a few astrophysicists studying it.
It was already pretty dead with its failure to deal with baryon acoustic oscillations in the CMBR, to be fair. MOND has always been fringe with a strong taste of crank.
Also true.
Dr. Becky, I feel lucky to be able to learn from your scientifically trained assessments! Up until now I had not heard of Bayesian Statistics! Your unique viewpoint in itself helps me to understand more than I would have otherwise bend able to ascertain on my own!
That leaves the small problem that in a century of looking, we have not found dark matter. There are other studies that claim to have falsified dark matter to a high sigma. Maybe, as Dr. B suggests, we need a new theory of gravity. I would also like to know how the authors of the first paper respond to,the second.
we have found dark matter. We've found it in places where it shouldn't be based on the amount of normal matter that's present. And we've found places where there isn't as much dark matter as there should be based on the amount of normal matter. If gravity just needed to be modified, we'd always find the exact same correlation between matter and dark matter but that isn't what we see. We just don't know what it is, but it's some kind of stuff that can be or not be in a particular place, not just some mathematical error.
We haven't found a particle. We have observations that strongly support its existence. The lensing observations of colliding clusters being the most persuasive. Last time I looked, MOND needed 'some' dark matter to explain them.
@@ianw7898 Dark matter is a theory that has worked in some contexts, not in others. Initially, they imagined something uniformly spread throughout the universe. That did not work out. They kept having too make the math more complex. They have found galaxies with little dark matter, assuming it exists. The same people Dr. B talked about emphasized that their conclusions should not be seen as support for dark matter. The bottom line is we don’t fully understand gravity. The general theory of relativity is the best we have done, but in some contexts it does not work. My own guess is that if we ever figure it out, it won’t be dark matter. Most likely, we would have found it by now, if it existed, and we have looked hard for a century. People, including astrophysicists, have a hard time not getting attached to inadequately proven ideas. We need to learn to say we just don’t know.
@@Walter-uy4or The evidence supports dark matter. It is as simple as that.
@@ianw7898 Sorry, but that is simply not true. Some evidence does, some does not.
I am 9 years old and i love ❤ space and i want to be a astrophysicists when i grow up. i love your videos i also have a question for you ,you know how before the big bang their was nothing i don't get how there was nothing no space no nothing it just seems really confusing and i'm hoping you can answer this question bye😊❤🎉😮
I would have like a bit more discussion on the data screening processes. What possible biases are introduced by the filtering? Would it be a logically expected result to see something like MOND come out if you fed in random errors to velocities?
This was my question too. How was the quality cut done in a way that ensures the data wasn’t skewed?
Agreed. Though the errors may not be random, they may be systematic.
There have been previous studies that claim GR explains wide binaries to as much as 16-sigma. Getting accurate data is the most difficult part.
I discuss this in the section of my paper related to the comparison with prior results. Yes, velocity errors can definitely make an underlying Newtonian population look Milgromian. That is what happened in the analysis of Kyu. I have explained in some detail why that is. I also explicitly demonstrate this by finding a MOND signal in his data, but later applying a stricter quality cut, which removes the signal.
Orbital velocity depends on two factors, as I understand it: mass of the objects and distance between the two objects. So we can "measure" (calculate) the combined mass of the two objects from Doppler velocity if we can measure the distance between the objects, or we can measure (calculate) the distance if we know the mass.
We deduce the mass (I am assuming) from spectral luminosity which provides us with estimations according to type, age, etc. But, what if our basic assumptions about this are incorrect in some cases? Perhaps a star with sn enormous iron core will provide more mass than we conclude because our only model that fits the spectra doesn't include a massive higher element core, but, rather a carbon or oxygen core?
By narrowing the data filters we are effectively reducing the input error percentages?
What I still don’t understand is, what is MOND description of gravity? General relativity has spacetime, mass curves spacetime, how spacetime is curved tells you how mass will move. So if GR is wrong and there is no spacetime. What or how is gravity described in MOND?
It isn't, but there have been attempts at making a MOND-like version of general relativity (eg. TeVeS, which includes tensor (gravity), vector and scalar fields instead) but they haven't gotten very far.
MOND was invented to do away with relativity and dark matter. It now requires both in a vain attempt to remain relevant.
This is what makes science great. The complete willingness to rip each others' (and their own) research and conclusions to shreds.
I just love that the lead author was really hoping for a different outcome (I mean probably). I mean that had to be hard for him too, but this is how science wins.
There really are some incredibly smart people out there and though I do not understand their work very much at all, I have a huge respect for them.
I love your voice! The accent, rhythm, enunciation, variation in speed and emphasis - all make the content not just lovely to listen to, but easier to understand. Thanks, Dr. Becky!
Muito interessante esta alternativa à Gravidade. Parabéns pela explicação Dr. Beck ❤
Eric Laithwaite defeated gravity decades ago..
Using a very large weight lifting weight spinning fast on the end of a 6 foot bar..
Scale it up a bit...
To Star or Galaxy size..
Bearing in mind for instance the stars are spinning etc as well..
Add in the equivalent of super conductors, scaled up and heat cold gradient etc..
A hole heap of things can happen.
Eric failed gyroscope class.
Laithwaite was a lunatic.
@@tonywells6990
?
Your videos are pure bliss, thank you for putting in the work on these.
The reason why nothing makes sense is because Neuton took Kepler’s equation for charged particles and substituted mass. Where does it say charge= mass? Then why are they used as the same terms. This is why they think of stupendous densities and energies. There is one fundamental force in the cosmos and everything is a manifestation of the material transforming Ætherial radiation resulting in magnetic fields and all we perceive is the magnetic field. Electricity minus magnetism is dielectric, The photo electric effect of light this is actually Einsteins only contribution to science. Then he proceeded to conceal this fact, hence E=mC2 and why it’s never used in anything. Everything is electrical because everything involves the electron which is the Interface between the material and the Ætherial.
Can you do a video explaining the theory that dark energy or dark matter might be a fifth force or why that’s false?
Love videos like this that get somewhat into the nitty gritty (while still being accessible to a layperson)! Thank you!
I looove your nails! I'll do a blue with green chrome powder next, thanks for the inspiration!
And obviously, thanks for communicating this result. The evidence seems crushing for MOND.
Same! They're so eye-catching, they really accentuate the gesticulating (in a good way!). Also am here for the science obviously, one of my favourite science communicators, but yeah you go girl!
My research team and I published something very similar on MOND. Its up on Arxiv published by Dr. Sophia Natalie Cisernos
Perhaps add an Arxiv link in your comment or in a reply to it.
@@indranilbanik3424 Often on YT, including a link to anything other than YT means the comment doesn't appear. Depends on the setting that the channel owner has. Otherwise you get links to all sorts of pseudoscientific nonsense that is not even close to the peer-reviewed literature!
Probably best to just name the paper.
According to Marcel Pawlowski's thread on X/Tweeter things are not so clear yet. Chea's sample contains more tight binaries which are used to calibrate the analysis (no MOND effect expected), he finds also less triple systems than Banik et al. In case Banik et al mistakenly included triple systems in their sample, this tends to favor Newton's behaviour.
Incidentally Figure 12 of Banik et al preprint shows a better fit with MOND than Newton for the wider binaries of their sample (r_sky > 5 kUA)
It would be interesting if we found out that General Relativity was also out by a small fraction with very high confidence, implying that general relativity is also just a set of special cases of an even more overarching theory that reduces to special relativity in most common cases.
Mond and Newton/SRT both hint at an overarching theory at/near the Planck level. Phenomenology breaks Sigma.
@@808bigisland They do not imply it. What does imply it is semiquantization of the general theory of relativity.
Hi-Dr Becky I am a big fan of your content and a massive science nerd unfortunately I don’t have a PHD..
Only a couple of diplomas in engineering and computer science..
The reason I have commented is are you still taken suggestions for video reactions..
The series I would love to see you react to if you have time in your busy schedule is a series called FRINGE which is about a team that investigate fringe science events it is a very very good show..
I will try to send you a small clip..
Thank you
From a big fan..
Ah I was trying to beat my record which was giving the first like to ur videos but today i was just sooo amazed by a book about universe I missed it but i promise I’ll beat it next time I’ll be here in nanoseconds 😂😊
There are galaxies with very little or no dark matter effect, and galaxies with large dark matter effect but very little visible matter,. MOND enthusiasts have not yet even provided a viable theory for it that can explain all the things general relativity explains. Yet people still argue about whether MOND could be right. What happened to the days of Einstein, Dirac, Bohr, Fermi, Feynman, Heisenberg when physicists were smart?
When I first heard of MOND, it was with the voyager anomaly, which has subsequently been explained without using MOND. Frankly at that point I lost interest and the data at that time was really mixed.
Thanks for the clear explanation - and congrats to the authors of this stellar analysis.
Wasn't it the Pioneer anomaly?
@@ianw7898 Yes, I was remembering incorrectly.
@@tacitus7797 And that was explained some years back, iirc.
As dark matter is so elusive, I had high hopes for MOND. But if the GAIA data is casting serious doubts, I expect EUCLID will provide the nails for its coffin . .
Great explanation of a complex topic. Thanks again to Dr, Becky!
It's still crazy to me that a dude with a telescope way back when got this close to being right about describing the effects of gravity
Perhaps one of your best videos with a simple explanation of a complex subject.
This one was a banger. Thanks as always Becky!
As someone that always looks for answers, I hate that there are areas we still don't have answers to! Our current understanding of gravity only works if we add 3 time the matter and say it behaves similar to but not exactly like matter. Oh, and also doesn't work on small scales so we also need "special" in addition to "general". All this says nothing about our ability to actually calculate what we want on the scale we want either. More mass objects just cause an exponential increase in calculations needed and when we got 100-400 billion stars and more planets ... sheesh!
My take on this is simple: If you need to "make up" dark matter to make your calculations/ equations work, your theory is probably totally wrong and you're "dancing in the dark" (pun intended) 😉
my theory is pretty simple. If you just needed to tweak the equation used to calculate gravity there would always be the same amount of dark matter everywhere. But that's not what the universe looks like. But there's galaxies where there is too much dark matter for the amount of normal matter, and galaxies where there is too little dark matter for the amount of normal matter. That proves that it is stuff, even if we don't know what that stuff is.
Yeah, this is what many laymen think. Thinking this way is too simple.
The idea of dark matter isn’t an arbitrary I.e. general purpose, fudge factor. Rather, there are dynamics for how dark matter is believed to behave, and how it would behave seems to match up with observations.
Gluons also, aiui, do not directly interact with photons. Yet, we conclude their existence, and QCD, based on what we do observe.
Do you have an idea for why you think that dark matter is “dancing in the dark”, but QCD isn’t?
Galaxy rotation speeds are not the only evidence for dark matter, but it just so happens that to understand the better evidence you need to understand fourier analysis first, so its inaccessable to most people.
@@deltalima6703time is slowing down so that can explain everything
MOND being wrong doesn't necessarily make dark matter right.
I think mass isn't just IN space-time, but is OF space-time.
This illustrates really well one thing I like so much in science: admitting when one is wrong. Banik disproves the theory he was such a big proponent of himself (with his team). Meanwhile in France we have Raoult still trying to prove hydroxychloroquine cures everything...
yes, i think that is one of the things science can teach. This humility regarding your own ideas, being able to bow to the facts and data. But due to human nature and hubris and now the financial factor has been added, it becomes difficult to recognize that you were wrong.
Are 10 sigma or 16 sigma levels of confidence actually meaningful? Those kind of numbers seem preposterous! I used to think particle physics was pretty demanding wanting 5 sigma results before calling a discovery, but here we have a supposed 10 sigma result being overturned by a 16 sigma result? I can't help but think that kind of numerical confidence level can't be realistic at all and could even be a warning sign that something is up somewhere. Can we meaningfully test anything and claim that level of confidence in the result?
Hey I remember back in the late 1990's and early 2000's there was another theory about the speed of light slowing over the age of the universe. Effectively saying that the universe isn't accelerating, just our light based meter stick is getting shorter so the distances look like it's growing. Anyway I was wondering if you have ever heard of this and what was it that disproved it as it never really caught on. And if it was just thrown out of hand, wouldn't now be a good time to revisit some of the more "crackpot" theories to see if any of them match the data better? (I don't remember anyone using Bayesian analysis back then.)
I think it’s called “tired light”
As a Scot, I like to imagine the Banik paper as an insult to the original claim is a Banik burn... The Battle of Bannockburn... I'll... the... door... here... bye.
Nice way to explain it! Very much appreciated!
Dark matter still seems to be "justified" by thinking of galaxies as scaled up versions of solar systems - which they aren't. Black holes actually make up a relatively small portion of the total mass of their constituent galaxies unlike stars in solar systems. Most of the mass of our galaxy is made up of red Dwarfs which aren't visible to the naked eye, so how many would you expect to see on M31 etc? Is there a paper which actually calculates the velocities of stars as a function of distance from the central black hole. Does it take account of all the gravitational contributions of all the stars that lie in between the black hole and any given star? I have never seen such a paper but would be interested to read it if anyone knows of a public domain copy.
Watched for years, but I'm writing something SciFi ish, and you're inadvertently changing my plot with every new episode!
Thank you for a truly mind blowing Channel. Huge hugs.
I think you should make a hoodie or t-shirt with a word cloud of all your mispronounced words, maps by the number of occurrences. Just for fun... your bloopers are a joy...
All of which goes some way to explain why we call the first day after the weekend, MONDay. It's just full of stress, ambiguity and uncertainty, and it defies simple explanation. It does seem like a convenient answer when you first look, but when you get down into the details ... One does get down to questions like Why does my reality circle clockwise or anti-clockwise while it's going down the S-bend ....
The one I wish I had the resources to investigate is one I'd term EAGR, enhanced application of general relativity. Basically it's the idea that there is ambiguity in what inputs we should give the the GR tensor and we might get some miles out of tweaking, actually increasing, our inputs. A galaxy itself is a very high energy system besides the mass that is in it. Massive things are moving at relativistic speeds vs one another. Huge masses are separated at distance. Rotational flow of mass at relativistic speeds drags space-time. There is time dilation. If we plug in some more of these values we can get more mass than observed mass especially in the middle and edges. And if more energy->more mass->more force->more energy->more mass->more force. The whole thing could compound on itself. Hopefully it would converge, but maybe you really do get 6x more mass by the time it converges.
So not an argument for dark matter but another nail in MOND’s coffin especially as Banik is the lead author - really thorough and impressive paper. Thanks for explaining it so well Becky.
The weirdest space news I saw this week was the claim that a satellite galaxy of the Milky Way has been found comprising almost solely of dark matter - I can’t find any real details about it though.
Dark matter can not be tested and proved or refuted. Of any observation does not fit the dark matter predictions, one Just change the massa and distribution of dark matter tô explain the results. Its the ultimate dragon in my garage theory
@@geok1ng Ehh, if someone said there was an invisible dragon in their garage, I'd say "weight it to prove it exists". Same is true of dark matter. It isn't crazy to theorize that there are particles that exist that don't interact via the electromagnetic field. Just makes it harder to measure them. If anything, it almost feels more crazy to me, that more particles wouldn't exist that don't interact via some of the fundamental forces. The idea that everything that exists must be able to be "seen" is as crazy to me as saying all humans can see and hear is all there is to be seen and heard in the universe.
@@geok1ngyou say it can't be tested, and that they just change the mass and distribution to explain the results. But do you know what happens next? They test to see if that's accurate. There is testing all the time to see if various models work.
So Banik cherry picked (quality gated) the data to remove the values supporting MOND 😏
Wow you explained this so well i feel like i understand it, an really that's impressive if you can take the average person an start explaining gravitational physics to them an make them feel like they know whats going on even though they haven't been in school in 15+ years lmao. Kudos to you Dr. Becky.
Talking about GAIA and dark matter. Will you cover on your channel this paper by Yongjun Jiao et al. (2023), "Detection of the Keplerian decline in the Milky Way rotation curve"? I'm really interested what your thoughts about this.
MOND was always an ad hoc theory - designed to fix a known problem, but with no a priori motivation. Very nice work here.
With respect to Banik, I do not think he is dumping his support for MOND. He is just saying that Chae et al's work does not support it.
Personally, I gave up on MOND being relevant a long time ago.
I think it’s conclusion is that MOND does not apply to these type of gravitational interactions between two stars.
It hasn’t ruled out for Galactic scales yet.
This is the kind of science that makes me happy. Someone who proposed something, continued to test and validate, is presented with new info that nearly completely refutes their claim, then accepts this. Props to that team for not having blinders or getting railroaded on proving their theory was correct.
I don't understand the Science, & don't care. However, you seem like a nice person, & so I will try my best to be excited by this new development.
This is science at its best - objectively answering the question "what model fits this data best" - it takes a lot of humility and scientific honesty to be first author on a paper saying that in this scenario there is no evidence for the theory you are basing your career on. I can imagine him looking at that first pass data saying "that looks messy, we can clean this up and do better" followed by many rounds of research group discussions of "what have we missed, why is this not showing MOND", eventually leading to the high quality analysis we see in this paper.
It wasn't the same team doing this study. The first study was done by proponents of MOND. The second one was done by detractors of MOND. Both saw what they wanted to see, in the same data.
There is an explanation on Redditt on how Banik also made an error. Due to a modelling error. This is explained well by Hernandez and Chae: I don't understand it at all but this still shows that the explanations are not decided at all.
Love your channel. Outer space has always been an interest, being a big sci-fi fan, and you would make a great Professor. Could you please answer a question for me? I heard that Apophis was supposed to come about 38k miles away from Earth in 2029. But since then I heard that Earth has started shifting on it's axis. Could this shift be enough to put the Earth in the path of Apophis now? Could you please look into this & get back to me or make a video about it? Thank you
A shift in axis is not a change in orbit.
I'd like to know if that change in the tilt is enough to make a difference of 38k miles. It doesn't seem out of the realm of reality that it could. I thought I'd ask someone that might be able to find out. I highly doubt that I'll still be around by then to find out, so thought I'd ask before it's to late.@@XellithUS
@@piratelordgrumpy9659 Its just a little bit of axial precession. Like how spinning tops wobble a bit (actually exactly how spinning tops wobble a bit). This isn't the earth being physical moved up or down on its axis of rotation, its just a small change in the direction that axis points. It doesn't change the orbital path.
@@piratelordgrumpy9659 The Earth wont change its orbital path unless its acted upon by a force. A tilt in its Axis will not push or pull it, therefore it wont change its orbit. Newtons 1st law.
This is an oversimplification, but if you have a spaceman going around the planet in his space suit, they cant change orbit unless they have a jetpack, no matter how much they spin or tilt on the spot.
Dr. Becky's enthsusiasm and warm laugh, always make me happy .
Even on these dark days ❤
Grtz from the netherlands Johny geerts
Mond sounds like something out of monsters Inc
This video blew my MOND.
Euclid made a guest appearance in that animated image shot of the JWST & GAIA orbits. They just released the first science images and it looked so awesome; what a fabulous telescope.
Great video. Mond have never been able to explain both rotation of galaxies and gravitational lensing of galaxy clusters so I have always thought that was a bad conjecture.
Yeah, I was somewhat liking the concept, but alas, reality cares not for my feelings.
Such an educative video, as the others. I admire you as an astrophysics master student. ❤
I can't imagine you during the lectures btw😎
Wormholes linking areas of gravity. Closing in less then a nanosecond but Long enough to chang gravitational paths would explain everything in datk matter
No it wouldn't, lmao. It wouldn't explain gravitational lensing, for example. Also you need actual math not just some random collection of words.
So your idea relies on a theoretical phenomena to explain a thing we cannot observe. Im gonna need more than your word salad before i climb aboard that train.
Too much advertising on every RUclips channel. I'm almost ready to quit.
invisible imaginary stuff needed to make old theories more correct is more like it. Wait until you find out how wrong the redshift is and all the distance calcs are incorrect.
Posting a random comment....
FOR THE ALGORITHM!!!! ❤
THIS is why I love science. It would be heartbreaking for your article to be refuted, but we are now one (small) step closer to the truth, and it was done through reason and rigour. I have a lot of respect for Chae et al. They thought they had something new and took a chance, which is how we innovate. Banik et al. gave it the hard look that it needed. They went the extra mile in their very thorough paper. This is how we learn and create a better world.
Every scientifically valid hypothesis that get knocked over at the first hurdle adds to the sum of what we know.
Hey Dr. Becky thanks for the video and keep them coming. I'm starting to think that we are in a new epicycle version of cosmology. MOND, Dark Matter, Dark Energy all require us to jump through invisible hoops. Hopefully I will live long enough for someone to figure out why the universe doesn't care for our current theories of gravity.
This is going above and beyond my head but feels like a physics paper equivalent of a rap battle.
Something about dark matter feels like adding epicycles to keep a geocentric model of the universe.
Had a watch. Humm, its all about the manipulation of horribly poor data and the outcome between the two being a small deviation. Dumping 2/3 of this data because it is too far away from either result I'm not sure is the right thing to do too.
You always make these videos so interesting, thank you for making high-level science accessible to anyone. I love the idea of models having built-in sensitivity tests that compare and contrast how different parameters will affect the final result in the model.
As you demonstrated, small assumptions, whether systemic or man-made, can make a huge difference in the type of data you'll collect, how you analyze the data, and ultimately what you conclude from the data.
Recently, I read something online about measuring a gravitational wave that had a wavelength of over a light-year long, and that the measurements require the technology behind it to be incredibly sensitive. Similarly, on the opposite end of the energy spectrum, there are limits to what we've been able to detect.
So I guess my question is this: How do you deal with all the uncertainty in cases like the one mentioned above where you have only a handful of detections ever recorded and brand new technologies to measure them with? How do you tell when data points should be rejected? Is there a certain threshold of detections you need to hit before you can claim statistical confidence in the data?
I ask because my last job was calibrating radiation survey detectors, and when we hit low enough exposure rates, the manufacturers suggested taking an integrated rate over a certain period of time and calculating things that way, since there weren't enough individual detection events. Do you do something analogous in your field? In my particular example, that method wouldn't be very effective since the wave itself takes a year to pass by.
If you're still reading, I appreciate it! I've always wanted to know how you're able to get good such certainty in your results with all the error bars and uncertainties flying around in the measurements and models. I'm assuming some kind of statistical tests occur during the analysis process, including the one that results in determining sigma, maybe? Thanks!
Impressive. And that from a big MOND-supporter.
can i ask why this stuff is always comparing GR to MOND rather than some modern modified gravity theory? i work kind of tangentially to modified gravity (mostly on a particular theory) at arguably one of the best universities for gravitational physics and i'm not aware of anyone seriously working on MOND. is it just because to the public MOND = modified gravity? or are people just not aware of modern relativistic modified gravity theories?
MOND makes a prediction, do modified gravity theories like TeVeS make any predictions that have been observed?
@@tonywells6990 just as an example, phillip mannheim claims that conformal gravity predicts flat rotation curves and can do away with dark energy. i haven't worked on this theory personally but i don't see how it's any different
@@lindsayweir4931I agree MOND seems like a dead end but I suppose its simplicity is intriguing.
Just looking at the raw plots of points screams to me that data refinement is needed here. The model curves for both MOND and Newtonian gravity are basically linear, but the data points are scattered over a fairly two-dimensional part of the graph, which at the surface indicates that more is going on than just either model. As the Banik paper shows, the "more going on" is errors in the data, due to the factors you indicated.
Does MOND predict any of the observations usually attributed to GR effects: Anomalous precession of Mercury’s orbit, etc.? If not? How do MOND proponents account for those observations?
It would not be noticeable at those scales. It is on galactic scales and greater where MOND should show up. And doesn't.
@@ianw7898 If MOND does not predict phenomena we know exist, what use is it?
@@markholm7050 It is of no use. It is on its death bed.
So we either need a new model to replace the existing models -or- we need a new particle never before seen. I'm not betting on either one yet.