On December 26, 2016, Vera Rubin died at the age of 88. She is the scientist whose team made observations of anomalous stellar velocities in the spiral arms of galaxies that provided a great impetus to the dark matter hypothesis.
The idea of dark matter had been around from around the first quarter of the 20th century but received a major boost in the late 1970s with more careful measurements by Rubin and co-workers using new technology of the speeds of stars in the spiral arms of galaxies. These more precise data could not be explained using existing, well-established theories of gravity based on the observed amount of familiar matter that we knew about but could be explained by postulating that galaxies were immersed in a halo of matter that we had not detected as yet that extends well beyond the visible edge of the galaxies. The mass of this dark matter creates the gravitational force that causes the anomalous velocities. Subsequent measures of stellar phenomena have bolstered the idea of such matter existing.
The problem is that three decades of searching have not been able to produce any direct detection of dark matter particles. If this matter is so abundant and all around us, why have we not observed it? This is indeed a pressing question in the field of astrophysics and cosmology and a multiplicity of theories have emerged that seek to explain the amount of dark matter, why we have not observed it so far, and what possible experiments might be performed to directly detect these dark matter particles. Clearly, if they exist, their interactions with ordinary matter must be extremely weak since we are immersed in it and yet do not feel it, and the major searches currently underway involve setting up heavily shielded, highly sensitive detectors in deep underground mines to minimize the possibility of other cosmic particles falsely triggering the detectors.
The problem is that as the detectors get more and more sensitive, they will start to pick up other faint signals that could mask any dark matter signals that might be produced. Some astrophysicists are suggesting that it is close to the time to reject dark matter as a failed hypothesis.
However, many scientists believe time is running out for the hunt, which has lasted 30 years, cost millions of pounds and produced no positive results. The LZ project – which is halfway through construction – should be science’s last throw of the dice, they say. “This generation of detectors should be the last,” said astronomer Stacy McGaugh at Case Western Reserve University in Cleveland, Ohio. “If we don’t find anything we should accept we are stuck and need to find a different explanation, perhaps by modifying our theories of gravity, to explain the phenomena we attribute to dark matter.”
Other researchers reject this view: “Theory indicates we have a really good chance of finding dark matter particles,” said Chamkaur Ghag, chair of the Dark Matter UK consortium. “This is certainly not the time to talk of giving up.”
This is an excellent example of an interesting issue that I go into in my forthcoming book The Great Paradox of Science: Its surprising success against all odds (the title is a work in progress too and keeps evolving!) about how we decide in science that a search for something that we believe should exist but that has come up negative should be abandoned. In other words, when should the absence of evidence be taken as evidence of absence? It is a fascinating question and I go into it in some depth in my book, and I alluded to it in an earlier post where I provided an except of the introductory chapter that introduces the problem of proving a negative, that something does not exist.
I’m starting to think we may — in some branches at least — be approaching the end of science. Maybe John Horgan was right.
@polishsalami, we’ll still be learning the basics of biology 1000 years from now, at present funding rates.
Its possible, I suppose that dark matter goes the way of phlogiston theory. It will certainly be interesting to see how that affects other hypothesis and theories in physics and cosmology.
I’m surprised that there aren’t religious buffoons lining up to proclaim that it’s their god/s.
Where is the dark matter?
Why its in god’s heaven of course.
“Clearly, if they exist, their interactions with ordinary matter must be extremely weak since we are immersed in it and yet do not feel it”
Or why have I never heard any issue with the orbits of planets in our own solar system. This has been much of my problem with what has been claimed about dark matter. From what I understand (but maybe it’s just my ignorance), we don’t really have any unsolved mysteries with the orbital mechanics in our own solar system. (Well, other than perhaps a 9th planet.) But, if there is supposed to be all this dark matter, shouldn’t this have been a problem that came up a few centuries ago?
I have, for a while now, taken dark matter (and, likewise, dark energy) to be little more than code for, “We don’t know what’s going on,” and anything said beyond that is wild speculation. It is a shame, though, that some people speak about dark matter as if it is more well understood.
One of the problems in science is that people sometimes work as hard at dismissing competing theories as supporting their preferred one. Many times in science, unrelated and competing ideas have turned out to be different forms of the same answer (e.g. Wiles’ work on Fermat’s theorem, ceramics and superconductors, among others). Would we have more answers or even a UFT if people were more cooperative than competitive, willing to try and prove each other’s work?
https://www.quantamagazine.org/20161129-verlinde-gravity-dark-matter/
Mano
You didn’t even address the strongest piece of evidence for dark matter! It’s not just the rotation rates of galaxies. It’s also gravitational lensing in the bullet cluster. It’s also particular characteristics in the cosmic microwave background.
IIRC: The short version: A while ago, using math and observations, they predicted another large planet beyond Neptune. However, it’s likely that the observations of one historical observatory were bad. The logs of that observatory mentioned a cleaning at a certain date, and I forget the exact context, but the context made it suspicious in hindsight. Once you remove the data from that observatory, then there is no missing mass in the solar system, and no evidence for a missing planet #9.
PS:
To everyone:
> Dark Matter vs Modified Gravity | Sean Carroll
While dark matter was first suggested by the rotation curves of galaxies, it has made a lot of testable predictions since then. A few (and commenters should listen to Sean Carroll’s talk):
Lensing of the bullet cluster, as a commenter mentioned
Growth of structure (from tiny fluctuations 300,000 years after the Big Bang to galaxies today)
Masses of galaxies as inferred from gravitational lensing compared to their mass in stars and gas
The fact that MOdified Newtonian Dynamics (MOND) matches galaxy rotation curves isn’t a very strong argument in its favor: that’s exactly what it was made up to do, and it’s not clear that it can get all of the other things right. Sure, our theory of gravity could be wrong (and people are constantly making up new ones!), but we are doing awfully well with general relativity plus dark matter. Dark matter could basically be anything, and it need not interact at all with ordinary matter. That would indeed be pretty depressing. We do have ways of detecting different types of dark matter: black holes or MACHOs would interact with stars and change their distributions; they would also cause gravitational lensing. We haven’t seen this at the required level, placing limits on MACHO dark matter. Axionic dark matter could be detected by interactions transforming them into photons within a highly magnetized resonant chamber, while very low dark matter particle masses produce different properties for dwarf galaxies. Again, these are ongoing topics of research in addition to the WIMP searches. And of course, we’re always hoping that our colliders find something unexpected.
For the orbital dynamics, note that the dark matter density in the Solar neighborhood is expected to be about 0.01 Solar masses per cubic parsec. In other units, within Neptune’s orbit, it’s equivalent to a single ~50 km-diameter asteroid, more than a trillion times smaller than the mass of the Sun.
Hi Mano,
I’m sure there are wholes in the logic, and it’s more apposite to the linked article, but:
When we say proving something exists is straightforward, but proving a negative is logically impossible, we are agreeing to special pleading in my opinion.
There is no difference, logically, between proving a positive and proving a negative. The logical apparatus is the same, say 1st order quantitative logic. We can prove a negative with axioms and rules of derivation. Mathematics uses these types of proofs all the time.
So, logically, we can prove a negative.
When we get to the existence of something, or non-existence, then it’s a matter of how true the premises are. If we hold the premises true for a logically valid proof of a positive, say a conditional derivation, then we hold that something exists. If we hold that the premises are true for a logically valid proof of a negative, say an indirect derivation, then we ought to hold that something doesn’t exist. Otherwise, it’s special pleading for the positive, but not the negative.
When someone says, absence of evidence is not evidence of absence, why, but for special pleading, do they grant the presence of evidence, is evidence of presence? I mean, you could always be misinterpreting the evidence, as all theories are underdetermined by evidence…
whole holes even….
Forget about the types of deriviation. All that matters is if the argument is sound. I really should think my way through things before posting. In any case. If we hold the premises true, and the premises are negative, then if the conclusion of a logically valid argument using these premises must be is negative, we’ve proven a negative. If we grant that we can prove a positive, unless we’re special pleading, we grant we can prove a negative.
Someone might say, ‘yeah, but, all swans were thought to be white, by your argument there was no such thing as a
black swan, yet I see them in Australia’, which means that argument wasn’t sound (false premise), not that it wasn’t logical. The same is true for proving the existence of black swans, that we can see black swans doesn’t prove that we haven’t mistaken the evidence somehow, and they really aren’t black swans, but something that looks just like a black swan. The level of confidence is probably approaching 100%, but it can never be certain. Switch it around, and it holds for a negative, the level of confidence for something might be approaching 100%, but it can never be certain.
Long story short, we can prove a negative just as well as we can prove a positive.
I agree with many of the commenters that there is other circumstantial evidence for dark matter other than spiral galaxy velocities. This is why scientists have granted it provisional existence status. But at some point there will have to be some positive evidence for it to be conclusive.
It’s kind of like the Higgs particle recently (and the neutrino earlier). It was believed to exist because of strong circumstantial evidence. But the positive detection was the clincher. If efforts to detect it had continued to be thwarted, at some point its existence would have been called into question, like with phlogiston a few centuries ago.
Brian E.,
One can use axioms and mathematics to prove the negative of a proposition, though even there we run into the Godel problem that no non-trivial axiomatic system can be guaranteed to be either consistent or complete and so the truth of any theorem that is proven is suspect. But how does one do that for the nonexistence of a physical entity?
As for proving the existence of an entity, the only way to do so logically is if assuming the non-existence of the entity leads to a logical contradiction, something that David Hume pointed out ages ago when he said that “there is an evident absurdity in pretending to demonstrate a matter of fact, or to prove it by arguments a priori. Nothing is demonstrable, unless the contrary is a contradiction. Nothing, that is directly conceivable, implies a contradiction. Whatever we conceive as existent, we can also conceive as non-existent.” (Dialogues Concerning Natural Religion (1779; published posthumously and quoted at D, 9.5/189; cp, EU,12.28–34/ 164–5)
I don’t understand the distinction, except as an imprecise measure of “strengths” of evidence which exist on a continuum, from very strong to very weak, aka as a description of the evidence and how convincing the evidence is.
Hi Mano,
If one can prove the negative of a proposition, one has proven a negative. That was my first point. Godel’s incompleteness theorems are not picky, apply to certain mathematic systems, and, if they apply to a negative, they apply to a positive. I don’t think they’re relevant only to negatives.
I love me some Hume, and do not doubt the Canny Scot for one second. Now, then there seems to no obvious contradiction to saying that there are no black swans, we are just mistaken in calling those black swan like things we see in Australia black swans, they are actually…..(insert arguments). As I said, all theories are underdetermined by evidence, so no amount of evidence can actually prove logically that black swans exist. We can only get to a point where we say, ‘well if that’s not a black swan, you’re a tool, stop wasting my time!’ The same goes for negatives. We can show that we’ve never seen a unicorn, but someone can offer ad-hoc reasons to why we’ve never found one.
A bit of terminology. To prove something logically, I mean that it is 100% certain. This only occurs in axiomatic systems of logic and mathematics. If we’re talking about the existence of physical things, based on evidence, that is never possible. We can only approach 100%. This is my point, we can never show that something physical does or doesn’t exist. Only that our confidence, or the evidence leads to near 100% certainty.
If we can’t prove a negative, we can’t prove a positive.
Better phrased as ‘If we can prove that something exists (to our satisfaction, then we can prove that something doesn’t exists (to our satisfaction)’ and to quibble with the latter while not quibbling with the former is special pleading.
The whole can’t prove a negative is false, a red herring, and had hoists the purveyor on his/her own petard. What goes for the negative, goes for the positive.
OK, I’m heading into rant territory! Thanks for your indulgence Mano.
Dark matter (if it exists) is extremely diffuse, so the mass of it within the solar system is negligible (only that of a medium-sized asteroid). Ethan Siegel address the question in a post here: Does Dark Matter affect the motion of the Solar System?.
EnlightenmentLiberal @16:
It’s not a complicated distinction.
Something can be proven to exist by pointing to it, but the converse is not true — that you can’t point to something doesn’t mean it doesn’t exist.
—
Brian E @17:
But that’s a triviality; yes, every proposition can be expressed as the negation of its negation.
The distinction at hand is whether something does or does not exist, not how that proposition is expressed.
You’re clumsily alluding to the distinction between analytic and synthetic claims, but failing to note Mano’s distinction between propositions and observations (brute facts).
Propositions are abstractions, but brick walls are brick walls.
—
If you want to use logical terms, then dark matter is (at least currently) an abductive inference, not an observation.
(The observation is that a phenomenon exists which is explainable by the existence of dark matter)
Dunc @19: tbrandt addressed the question in the last paragraph of comment #10.
Brian E:
Yes, you can start with a set of axioms and use deductive logic to prove things. But what you’re proving is, essentially, “if these axioms are valid, then P.” That doesn’t tell you how to select useful axioms. If you’re doing geometry, you can choose axioms that apply to a plane, or axioms that apply to a sphere (among other options). In plane geometry, you can prove that there are no triangles whose angles don’t add up to 180 degrees. But if you take a globe, draw a triangle connecting the North Pole, Hong Kong, and Mumbai, and measure its angles, you won’t get 180°.
In this context, what needs to be proven is that dark matter exists, not that it doesn’t. “Here is a black swan” is an existence proof that black swans exist; we’ve previously defined a black swan as being a blackbird of the category “swan” (so a crow doesn’t count,. and neither does a black coot) With dark matter, we’re at the stage of observing effects. In the swan analogy, we’re sure there are no white swans here, and something has pulled up young reeds from the marsh, and chased the Canada geese from their usual nesting area, and a child was buffeted hard enough to break his arm, but his description of what happened is confused and nobody else was watching. [I used to live next to a salt marsh that was sometimes inhabited by mute swans, hence the ornithological examples.] That doesn’t add up to proof of a black swan, but “no black swans” leaves unexplained events.
Physicists concluded that there is no luminiferous ether because if it had existed, its effects could have been measured, and Michelson and Morley made the measurements and found no effect. There may not be dark matter, but if there isn’t, the effects that are being attributed to it have some other cause(s) that should be looked for.
Rob @21: Ah, so he did. Still, I think the link to Ethan’s piece is worthwhile…
Thanks for the reply.
Mano is the one who says
in relation to science.
So, if it’s a triviality, I don’t disagree, but its repeated by everyone, and I thought I’d point out that if you’re talking about deductive systems, then, it’s a false statement, and why repeat it when talking about science, which isn’t a deductive system?
When talking about existence, we aren’t talking about proving a negative, that’s a red herring as I commented above. But one Mano brought into the discussion. What we’re talking about is proving something exits or doesn’t. If you say you can’t prove something doesn’t exist, then you are hoist on your own petard when claiming something does exist. There will never be enough evidence to show something does or doesn’t exist. We just get to a point where we say the evidence is strong/probable/etc that something does or doesn’t exist.
Instead of “The Great Paradox of Science: Its surprising success against all odds”, which is a bit wordy, wouldn’t “The Paradoxical Success of Science” be better? A subtitle could be “What were the odds? “, but seems unnecessary.
Also, would it be more accurate to say that one cannot establish universal negative existence claims?
IIRC, IMO, by this standard, atoms are also an abductive inference, and not a (direct) observation.
This statement is not right, and it’s not even wrong. It’s too confused to be wrong.
First, I think you would agree with me that there is no such thing as “absolute proof beyond all doubt” in science. I accept your imprecise shorthand, which I often use, but in this context, we must be clearer.
By pointing to a chair within arm’s reach which we can see and touch, that’s really strong evidence for the existence of the chair. Incredibly strong. However, it’s still not beyond all possible doubt. I may have to start invoking scifi to offer alternatives, but it’s not entirely outside the realm of plausibility. If it’s just literally seeing, then that is much less strong. Professional magicians all the time “fool our senses” in similar ways. Context would be very important too.
There are no (full size) T Rexes walking the Earth today. Not a one. I know this along several lines of evidence. It’s not merely that I do not accept their existence today. I positively affirm their non-existence. To reach this belief, the strongest line of evidence is that they’re really big, and if there were any, then someone would have seen one by now. In essence, this is an argument of the form “I cannot point to something, and therefore it does not exist”. Obviously, nuance and context are important, but nuance and context are almost always important.
Carl Sagan was simply wrong when he said “absence of evidence is not evidence of absence”. Put correctly, “absence of expected evidence is evidence of absence”. Again, the evidence is of varying strengths, and context is important.
I can check one room of a house for occupants, and if I don’t find any, absent any other particular information, this lowers my odds-estimation of the truth of the proposition “there are occupants in the house”. As I check more rooms, my odds-estimation lowers, until eventually I check all of the rooms, and then my odds-estimation becomes close to 0% (but not 0%, i.e. “maybe they’re hiding”). The T Rex example is exactly the same form of reasoning: replace “house” with “Earth’s land surface”, “occupant” with “T Rex”, and “room” with “tract of land”. Exact same form of reasoning.
jaxkayaker @#25,
That is a good idea for a title!
Also, you are right that one cannot prove, for example, that no magnetic monopoles exist because you would be making a universal negative existence claim.
Which is what I meant by saying that accepting ‘absence of evidence is not evidence of absence’ whilst not also accepting ‘presence of evidence is not evidence of presence’ is special pleading.
Anyway, this digression was really only because I think Mano was going to publish that in a book, the mixing of statements more about deductive systems juxtaposed with statements about science and how we can show logically that something exists (if by logically, we mean a tautology, we can’t in evidence based reasoning (science), and if we mean using a logical argument with evidence, we can, but not to a 100% probability), versus can’t prove a negative (we can in deductive systems, but can’t in science, but that’s no biggie, as it’s the mirror image of proving something exists) might be confusing.
To Mano in post 28
Also, one cannot “prove” anything, where “prove” means “show that something is true beyond all doubt” e.g. “show that something is true to 100% confidence”. This is a simple reflection that the only way that we learn about the world is through our senses and evidence, which when properly done obeys something like Bayes’ (probability) equation. The only things that we can do is convince ourselves, with good reason, on the basis of evidence, that some propositions are more likely to be true, and some propositions are less likely to be true, and we can also provide ballpark numbers for these propositions.
In other words, a good definition of a rational actor is someone who takes actions in order to maximize their expected utility according to their own utility function, and according to their predictions about the possible futures and their relative likelihoods. This definition is inextricably linked with the language of a gambler, because we are all gamblers. For example, every time that we take a car ride, we take the gamble that the utility of the extremely high likelihood of reaching our destination with the car in a short amount of time outweighs the expected extremely unlikely cost of dying in a car crash (etc.). In order to make these kinds of bets that is every single action that we take, we need to develop approximate odds for all relevant propositions (and we need to develop heuristics for developing approximations of these odds, and we need to develop heuristics for determining the appropriate level of analysis and heuristics for particular scenarios, etc.).
We never have absolute certainty, and I say that with a merely high but not absolute level of certainty.
As another example, one might say that we can be sure about the results of purely logical or mathematical proofs, but that is also mistaken. Imagine any tolerably lengthy mathematical proof. There might be a mistake in there. When you or I do a proof, we often make mistakes. Because it survives peer review, we gain additional confidence that there are not mistakes, but never can this sort of empirical process ever raise our confidence to an absolutely certain, 100% level of confidence. Again, we never deal with absolute certainty. This sort of reasoning applies to every pure-logic aka analytic claim, and therefore it applies to all claims, whether analytic or synthetic (or some mix of both).
Brian @#29,
You are quite right, I explore all these things in some detail in my book. These are complex, nuanced questions on the nature of knowledge and I cannot do my arguments justice in the snippets that are possible in the blog and comments. So my posts should be taken as merely general indicators of my thinking and as a hope that they trigger interest in people reading the book when it comes out, which I hope will be sometime in 2017.
EL @#30,
You are right that in science we never ‘prove’ anything in the sense of having mathematical certainty. What we have are reasoned judgments that we feel confident enough in to treat as if they are certain. The word ‘proof’ as used in science means just that.
@27: I always understood this in a general, not specific case for evidence.
The idea is: Absence of “any” evidence does not mean evidence for absence.
In your scenario there is a lot of evidence for absence which you produce by checking the rooms. So you do not have a lack of any evidence, just evidence for something.
The correct metaphor would then be:
You come to a house. You do not enter the house or check any room in any way. (Lack or absence of evidence). You can not conclude that the house is empty (evidence of absence).
When you produce the evidence for absence (by checking the rooms) you can of course say that the house is empty. But you have to check first.
You can of course say that in this more general case the premise is trivial true and you would be right. But considering how often argument from ignorance is used, it is still an important lesson.
To Mano
I suspect you missed my point in my earlier post. Your earlier post included:
Ok. I believe that your intended communication in this previous post is wrong. For the purposes of what I believe that you intended to communicate, I can “prove” that there are no magnetic monopoles.
Now, I need to be a little more careful. AFAICT, magnetic monopoles are consistent with some plausible extensions of the Standard Model of physics. So, it’s not the best example.
However, I believe that you intended to communicate an idea that is equivalent to the claim “one cannot prove that there are no immaterial human souls”, or “one cannot prove that there are no unicorns”. Was that your intention? Did I properly decipher your intention from your communication?
Under the meaning of “prove” which is “demonstrate the truth of the claim beyond all reasonable doubt, but not all possible doubt”: I can prove that there are no unicorns. I can prove that there are no immaterial human souls. I can prove universal negative existence claims. They’re really quite easy to prove. I don’t know by what method or standard or process you arrive at this rather silly conclusion.
In particular, can you agree with me that it has been proved that there are no T Rexes walking on our planet today? Now, what’s the difference between that and Unicorns? Are you going to ad-hoc “but Unicorns might be invisible” ?
Again, absence of expected evidence is evidence of absence, and this kind of evidence of absence can rise to the point of “scientific proof”. In the particular cases of immaterial human souls, angels, gods, and all supernatural claims, we do have scientific proof that they do not exist. The evidence that we have in favor of the Standard Model is overwhelming, and the Standard Model leaves no room for such things. Sure, there might be a deist angel or god that doesn’t actually do anything, but that sort of usage without qualification is a disingenuous misuse of the words “angel” and “god”. For angels and gods that actually do something, they cannot exist, because by definition they are something above materialism, and the Standard Model and the evidence for it leave no room for such things. If some thing exists and obeys the Standard Model, then it’s probably not reasonable to use the word “supernatural” to describe it, and for postulates that there exists something that has effect on our material world in a way that does not conform to the Standard Model, we can be sure that such things do not exist because of the overwhelming evidence in favor of the near universal applicability of the Standard Model. In order to entertain the epistemic possibility that supernatural stuff exists, you also have to entertain the epistemic possibility that the Standard Model is very, very wrong, and that is not a reasonable thing to do in light of the incredible evidence that we have in favor of the Standard Model. In other words: Because the evidence for the Standard Model is so good, and because the Standard Model is inconsistent with the existence of the supernatural, therefore we can be very confident that the supernatural does not exist.
Further, this kind of reasoning should have been obvious once Newtonian mechanics was discovered (circa 1687). It was obvious to some. It was obvious to Baron d’Holbach, who is one of the first famous modern western atheists. His writings (circa 1770) are clear, and quite insightful. Some of his particular scientific ideas were wrong, but he was right in his general position that the evidence strongly supports an impersonal, material world that evolves according to simple, impersonal, material, mathematical laws.
To put this in perspective: The Standard Model is now a complete and accurate theory of every single experiment that has ever been done on Earth. We know that there are no “soul particles” or “spirit forces”, because the math of Quantum Field Theory says that if any such thing existed in a way that had any remotely significant effect on our life and physics on Earth, then we would have created it already in a particle accelerator. That’s what the math says. Either you think the math and theory are wrong, or you have to conclude that the supernatural has effectively zero impact on life and physics on Earth.
Note: There is future physics left to be discovered, and our theories are incomplete, but we know this only because of observations of things outside of Earth. There is new physics to be discovered, but none of that physics is happening on Earth, and we know this because the math of Quantum Field Theory says so and because of our experiments with particle accelerators.
EnlightenmentLiberal:
OK, care to do so?
To John Morales
I just did.
EL, you haven’t even defined what you mean by an “immaterial human soul” — presumably you think it should have detectable “soul particles” or “spirit forces”. If so, in what sense is it supposedly immaterial?
EL @34:
This is just gibberish. Please demonstrate how “that’s what the math says”.
EL @#34,
I started writing aa reply here but it got so long that I made it into a separate post here to reach a broader audience.
EnlightenmentLiberal, I saw this and thought of you: Has the Large Hadron Collider Disproved the Existence of Ghosts?.
(You’re not alone!)
It is more than clear that dark matter does not exist at all. I have proven it in the article Dark Matter Does not Exist at all (http://article.sapub.org/10.5923.j.fs.20170701.01.html) And I have also recently shown that the Universe is closed (http://article.sapub.org/10.5923.j.fs.20170701.02.html.