Dark Matter Experiment Finds Unexplained Signal | Hacker News Reader

Dark Matter Experiment Finds Unexplained Signal

quantamagazine.org

106 pointspseudolus50 comments

"Researchers say there are three possible explanations for the anomalous data. One is mundane. Two would revolutionize physics."

The "mundane explanation" is contamination from inside the experiment. Hopefully it's not that!

When you're in this game long enough, you realize that the thing you want most is not to be wrong. Experimental physics (and every experimental/field branch of science) is a search for what is true.

If the dark matter is axions, that's great. If it isn't, that is okay, too. The key problem is that we don't know what gives rise to the phenomena that have led us to believe that dark matter exists.

The thing we should hope for and aspire to, from every experimental endeavor, is a complete and accurate assessment of uncertainties. A complete assessment is only ever a goal, as the "unknown unknowns" are intrinsically difficult to discover; barring that, one can always list and accurately assess all of the known uncertainties.

It is exciting every time a new or unexpected signal appears in a precision experiment -- perhaps today is the day a great truth is exposed -- but the thing we should always hope for is not an exciting result but, rather, a reliable one.

> When you're in this game long enough, you realize that the thing you want most is not to be wrong.

I suppose most of us are commenting from the outside perspective, where the thing you want most is a warp drive - so any revolution in understanding of the universe gives hope that we'll be able to do things we now consider impossible.

I think you've nailed it. All we really want is to find out we can do warp drive or build a Stargate. I'm only partly joking.

Absolutely. Extraordinary claims require extraordinary evidence.

But realistically it's probably the mundane explanation, just like the faster than light travel a few years ago.

Yep. You almost don't even have to look at the details. Any time you see a headline that says some experimental result "could revolutionize physics" the odds will be very much with you if bet against.

Quasars, pulsars, and the Cosmic Microwave Background all started as unexplainable experimental data: I know the most likely explaination is the mundane one, but what if it is not?

All those three were clear, sharp signals whose origin was not established. This is totally the opposite, a signal who dances between noise/contamination and an actual result(between 1-2 sigma and a actual "discovery" of 5+ sigma).99% of those cases are bogus.

Yes, that's true. The problem is that the more our knowledge converges on the actual truth, the less likely new discoveries become. It has been 50 years or so since the last really new thing was discovered in particle physics despite heroic efforts. The more time goes by, the more likely it is that this is because the next breakthrough is simply beyond our reach.

Um, Higgs Boson?

Now, true, it was expected rather than a surprise. But it was still a discovery.

Visit https://en.wikipedia.org/wiki/Timeline_of_particle_discoveri... and, since we are talking about subatomic particles, remove 1995's antihydrogen and 2011's antihelium-4. Note that the criteria listed on the page are a little generous. What's that graph look like, plotted on time?

1930s: Three subatomic particles

1940s: Two subatomic particles

1950s: Three subatomic particles

1960s: Three subatomic particles

1970s: Four subatomic particles, that's going up but then ...

1980s: only one subatomic particle

1990s: One subatomic particle

2000s: Two subatomic particles, and that is if you are being generous about including the quark-gluon fireball, which I wouldn't be

2010s: One subatomic particle

There are only so many to discover, if these are fundamental, and so the timeline of discoveries per decade must trend down.

IMO, the situation is much worse than this. There haven't been any new particle predictions since the 1970's. The observations since then have all been confirmations of earlier predictions.

But in the Standard Model each “generation” of particles is so much heavier than the previous one, and in practice there is a limit on the power of the accelerator we can build.

Even then there is no reason to believe that there is infinite number of subatomic particles.

But what happens if you add more energy? Do we know that past some point, new particles will stop appearing?

"Know" in the sense that we can be certain, one hundred percent, never doubt? No.

In the sense that you would have to convert the Milky Way's entire mass into a particle accelerator large enough, over the course of millions of years, to get that much energy to probe even further, ain't nobody gonna do that. You could go out even more, why not convert the Local Group? At some point you must accept that we're not going to undertake a billion year engineering project to do that and so it is irrelevant.

"More" will run out as a useful strategy eventually.

However, there are other clues, which largely live in symmetry. How do we "know" that there are three families of leptons and quarks? Even in the 1990s, there were certain cosmological-level hints that the universe would look very different if there were five; four was just barely allowable if you squinted very hard.

Don't get me wrong, you can slam together a lot of energy in one place and get funny things, but they won't be funny things that are new fundamental particles, which are governed by different symmetries, which you will see if you begin browsing the, ah, "zoo" of subatomic particles. For a particle, an anti-particle (even if it is itself), as a symmetry example.

And so you would need new symmetries to explore, which brings us back to the possible axion of the article, which was suggested by a symmetry. In general, new particles are suggested either by symmetry or in some cases, simple conservation, like the humble neutrino, which was originally a placeholder particle to say, "Where is the spin in that reaction going?"

> Even in the 1990s, there were certain cosmological-level hints that the universe would look very different if there were five; four was just barely allowable if you squinted very hard.

Could you (or someone) be more specific here? Or give a pointer to where to learn more?

Short of me digging through boxes of old college textbooks and such, I must wave my hands about as I struggle with memory. Here goes:

essentially the number of particle "generations" (with the first generations of leptons being the electron, the positron, the electron neutrino, and the electron anti-neutrino) has an effect on the abundance of the "light elements" like Helium-4 during the early phase of the universe. This was pointed out as a dependence in roughly the 1960s by Hoyle (yes, that Hoyle) and Tayler. Each generation would increase the density of the early universe, causing it to expand faster, and in turn the earlier the transmutations of light elements would have stopped, leaving excess neutrons to go around. This means more Helium-4. So, measure the Helium-4 abundance and you can set some error bars on the number of generations.

So the tau particle, in generation three, was predicted around 1971 and finally detected about six years later. Of course the Nobel for this discovery wasn't awarded until much later but everyone was on tenterhooks: could we find a fourth generation? Could it exist? But we didn't have the measurements of the early universe nailed down at the time.

Measurements of the abundance of these elements around the late eighties or early nineties is in line with the number of generations being three, but it could be four at the "couple of sigma" level of confidence. Indeed, originally it was thought that four was an upper limit in the 1980s.

And so here we are. For a fourth generation, the equivalent of the electron mass would be >> 45 GeV/c². Simply enormous. My guess is that further observations have pushed the confidence of "three" much much higher since then, but I haven't been paying attention.

Of course, the generations aren't confined to leptons (the "lightest" particles), but also to the quarks. So the first generation has its mirror in the up quark, the anti-up quark, the down quark, and the anti-down quark. Again, one of those symmetries I brought up.

Add allll the energy? Big bang. One particle. ducks

Those famous cases are the exception that proves the rule.

In what way do they prove the rule? They fact that these exceptions exist prove the rule is no rule, but just a guideline.

The "rule" being that the vast majority of unexplained results are due to experimental error or noise. The existence of a few famous examples that were indeed new science, narrows the statement from "all unexplained results" to just "the vast majority".

To be fair, the line works much better than "is probably noise".

Yeah, if it was almost any other branch of science, I'd half expect new revolutionary findings.

But in physics, mundanity wins.

There's a reason for this, and it reveals a very deep truth about our universe: at root, the laws of nature are simple. All of the interesting complexity is emergent. That's the reason that, in terms of the potential for producing an endless stream of new discoveries, physics is a drier well than other fields of science.

We don’t know that. It might as well be that the laws we know just are good simple approximations to something much more complex.

If it is complex but can be described simply, Ockham's Razor says it is simple.

No, that is a misunderstanding of Ockham’s razor. If a simpler explanation explains all known phenomena, then of course, choose the simpler explanation. But that doesn’t mean that there aren’t any unknown phenomena that the simpler explanation can’t explain that forces us to choose a more complicated explanation later.

Ockham’s razor doesn’t say anything about reality. It only says anything about which model of reality we should choose given a set of known phenomena.

One way to think about it is that elementary particles are so extremely small that they can't have that much complexity.

Entirely unprovable intuitive reasoning, I know. But I think there is something to it.

Or even the Digamma [1] signal that appeared independently at the same time in two experiments at CERN.

[1] https://en.m.wikipedia.org/wiki/750_GeV_diphoton_excess

And the faster than light travel was "discovered" at the same lab, Grand Sasso National Laboratory.

To be fair, they didn't actually think they'd discovered faster than light travel. They basically went "uhhhh, we double-checked everything and our numbers work out but we're pretty sure this isn't faster than light; here's our paper, can someone please point out wtf we screwed up?

And the fun part of that one is that all the math, data, etc. all lined up to say that. Their analysis of the data was correct, and it turned out to be an equipment issue. A single incorrectly installed fiber optic cable, and an out of spec clock. It wasn't even a bad cable, just not inserted completely correctly.

https://en.wikipedia.org/wiki/Faster-than-light_neutrino_ano...

EDIT: A much better detailed analysis of the failure including pictures of the cable, https://profmattstrassler.com/articles-and-posts/particle-ph...

This is a bit of revisionist history. The way the "discovery" was announced was slightly subdued, they called it an "anomaly", and added the caveats that it could be due to some errors, but they also stated that the "deviation of the neutrino velocity from c would be a striking result pointing to new physics in the neutrino sector". So pretty much, we are not 100% sure, but if this is correct, this is revolutionary.

That's not exactly the tone of someone who asks the larger community to assist with a debugging exercise.

And the recent announcement repeats the same pattern. "This could be due to some equipment malfunction, but if not, then this is big".

The XENON experiment is located at Grand Sasso but it is a collaboration between multiple laboratories

It wasn't a single group looking at those results

Exactly what I thought when I read that it was in Italy.

I mentally rolled my eyes and thought "here we go again".

(My partner's from Italy and I spend at least a month there each year, so I hope I don't come across as being entirely needlessly snarky saying the above! :) )

Researchers say there are three possible explanations for the anomalous data. One is mundane, 99.98%. Two would revolutionize physics, 0.01% and 0.01%.

Experimental physicists journey is similar to startup founder

They have to figure out which "idea" they want to commit for a very long time. They have to convince a rich person/org to fund their expensive experiment. Work hard to get things right. Have patience until you get it right.

Success will give you recognition and money and in some cases probably Nobel (being a unicorn). Failure will get you lot of learnings and probably some recognition too ( for proving something wrong). Unfortunately, there is one more state, being in limbo and not going anywhere.

As long as you are having fun in what you do, it is worth it.

Note: I am not an experimental physicist.

I was an experimental physicist, and I'd say that's a reasonable comparison to make.

The time scales are much longer, unfortunately. Instead of taking one or two years to get a company off the ground, it can be more like five for an experiment[1]. I'd attribute this to slower iteration. You end up throwing away or rebuilding a lot of physical things, many of them bespoke, on the way to a working apparatus.

Also, the compensation is much worse for experimental physics. For he first decade of your career, you work as a grad student, then a post-doc, for slightly above poverty wages. This does have the advantage of meaning you work with very motivated people, but you're giving up a lot for that.

[1] The XENON experiments from the article (10, 100, and now 1 Tonne) have been developed over about 15 years, for example

Am I the only one left with a befuddled feeling of missing context when it comes to particle physics? Like what does it mean for dark matter to be axions? "What's a quark?", Etc.

I feel like I have a context for understanding electrons and protons. I have read A Brief History of Time. What do I read next in order to have this article not sound like mumbo jumbo?

The first four paragraphs of the Wikipedia page on quarks seem decent: https://en.wikipedia.org/wiki/Quark

I hadn't heard of axions before, but https://en.wikipedia.org/wiki/Axion and https://en.wikipedia.org/wiki/Peccei%E2%80%93Quinn_theory seem to be a starting point.

I can’t recommend Feynman’s “QED: The Strange Theory of Light and Matter” highly enough. It gives description of Quantum Mechanics in simple terms (while math is oversimplified for my taste, still better than other books, where math is too hard).

https://www.amazon.com/QED-Strange-Princeton-Science-Library...

Sean Carroll's Great Course on Particle Physics (and his other ones) is a really great introduction.

edit: Higgs Boson, Dark Matter/Energy, Time

https://www.thegreatcourses.com/professors/sean-carroll/

With tritium's short half-life, it seems like it could be ruled out by simply waiting.

Tritium is actually being continuously produced by cosmic rays smashing up atoms in the detector. That's why these experiments are buried deep in mines, but even the small amount that get through the earth can cause problems in a sensitive detector.

https://www.sciencedirect.com/science/article/abs/pii/S09276...

Tritium's half-life is 12.5 yr. While I haven't looked up the statistical significance of the 53 excess events, it would take at most six half-lives to hammer that to below one.

Sometimes waiting works (in the absence of any cosmogenics or radionuclide support for the background), but a good assay and background model is generally much faster.

Very cool.

Another nothing burger anomaly being reported on by a journalist who desperately wants to push the narrative that it will change the entire field so you will click on the article.

I've gotta set up filters for "new particle", "unexplained signal," and "potential breakthrough."

Wolchover is a great scientific journalist and I can't think of a less clickbait-y headline than the one provided.

Please set up the filters and take your ignorant bitterness elsewhere.

Great X Journalist is an oxymoron.

Journalism has no art to it. It's a simple algorithm

> find a story you can contort into something interesting

> interview an "expert" for two hours so you can quote mine one sentence about how remarkable the story is

> Publish bullshit article, wait for ad revenue

Maybe this story isn't bullshit, but when your only quote is from a guy who -- by your own admission -- was uninvolved with the research, and all he has to say is "this could maybe be significant," you don't really have a story, do you?

It's not Wolchover that I object to, it's the incentives that are set up to make her subtly overstate the importance of anything she has to cover (disclaimer: I don't know any of her other work).

95% of journalism should be book-keeping. Dryly reporting the quantitative results of research. The fact that she has a quote like that suggests to me that she was digging to make it something more. I don't think that's her fault; it's just her job. But I have to call it out when I perceive it.

you do know that quanta is run by a non-profit. if you want book-keeping go to arXiv.

pretty pathetic that you're this upset

Reply on news.ycombinator.com