I’d say a tentative no. While in a very loose principle there are experiments that could identify some aspects of reality that are consistent with it these are (a) not currently possible and (b) likely never will be.
Even beyond those, the string theory landscape is a massive challenge for the concept of falsifiability within string theory. First, we’ll probably never find the “right” solution, and secondly - even if we did - wha does it mean when you can have basically any solution so it’s almost inevitable one will fit observational data?
That doesn’t mean it’s wrong, per se, as it would then be consistent with a mechanism like Eternal Inflation and we could make some sort of anthropic argument as to why we’re in this vacua. But I think we’re beyond the scope of what Popper meant by falsifiability now.
I’m not sure that’s necessarily a bad thing, anyway. It’s inevitable the more and more you talk about the most fundamental particles, the origin of the universe etc etc, that you will come up against ideas that aren’t falsifiable in the strict Popperian sense. At some point we’re likely to get a situation “this is just the way it is but this is a theory that’s consistent with everything we do/can observe”.
But that doesn’t mean Popper’s ideas aren’t a good guiding principle that we should always be striving for falsifiability, it’s just that the entire concept of falsifiability has a limit when you talk about the most fundamental questions. We just never know where the limit is, so what we think is not falsifiable today we may come up with an idea/experiment that makes it falsifiable in the future - so we should always be trying to think of them.
> It’s inevitable the more and more you talk about the most fundamental particles, the origin of the universe etc etc, that you will come up against ideas that aren’t falsifiable in the strict Popperian sense. At some point we’re likely to get a situation “this is just the way it is but this is a theory that’s consistent with everything we do/can observe”.
What? Either verifiability is a principle that all science has to abide by, or none of science has to abide by. You can't say "this field gets to opt-out, while the rest have to follow the falsifiable principle." What other fields of science get to opt-out? Do astrologers get to opt out of falsifyability too? Why or why not?
you might be sad to find out that “verifiability” is already not abided by by all science. Many field scientists don’t care about it. For example the theory about the meteorite causing the end cretaceous mass extinction is certainly a scientific theory but probably isn’t falsifiable
I agree. I think what happens is that scientists formulate hypotheses that they can't test (such as the hypothesis that an asteroid wiped out the dinosaurs), but when the general population hears that a untestable hypothesis has widespread support amongst scientists, they all assume that the widespread support must have occurred because of hard evidence. So over time untested scientific hypotheses become scientific fact. This an effect that some have coined "consensism", and it's something that I don't agree with, but unfortunately is very widespread.
well nothing is ever nor can ever be proven in science, so in that view everything is just a hypothesis that has widespread support amongst scientists. And the idea that a meteorite wiped out the dinosaurs is supported with hard evidence, like the global iridium layer and the huge crater in mexico. The point is not all science has to be “testable” for it to be science. The meteorite theory is supported by hard evidence yet since it happened in the past is fundamentally not testable.
> well nothing is ever nor can ever be proven in science,
I disagree. I believe there are some things that can "proven" to a degree that is very close to absolute. For instance, pouring vinegar and baking soda together gives the fizzy reaction. Its easy to observe, replicate and wrap your brain around whats happening.
That’s one of the things I’m referring to (amongst other potential experiments) in my first paragraph. How do you rule out all versions of SUSY consistent with all versions of string theory when SUSY can basically exist at arbitrarily high energy scales? Certainly ones well beyond what we’re going to practically test in any reasonable timescale.
Have you read about the string theory “landscape” problem? In many string compactifications there appears to be a staggering number of distinct metastable vacuum solutions, often quoted as around 10^500.
Many critics take this to mean string theory is effectively non-falsifiable and lacks predictive power: if one vacuum’s low-energy predictions conflict with observation, another vacuum can be chosen to fit instead.
(String theory supporters hope, but have not yet demonstrated, that additional consistency or vacuum-selection principles could drastically narrow the option.)
On this topic, where the rubber meets the road is the question “Should we hire string theorists in physics departments?” Or is this really a branch of mathematics, since there are no experiments to be done to test the validity of string theory.
I tend to think that theoretical physicists should be guided by data, or at least make some falsifiable predictions as part of their research output.
A. the deeper you go into the particle physics, the word theory becomes really vague compared to other science. They are more like models or suggestions.
B. the art of Theoretical Physics rely on simplicities. its not like the ways on popperian sense, but way more beautiful. you can make bunch of complicated 'real' theories based on real simple theories in all sciences, but non will stand out.
A lot of people here are confidently claiming string theory is unfalsifiable. Which in some respects is true both in literal terms because it's testable resolutions are very far beyond our present reach, and the breadth of alternative interpretations that are possible within the framework is vast.
However, there was an expectation that the LHC would access data at energy scales and at sizes relevant to many important contentions of string theory, and nothing in the way of uniquely supporting data was found, and as a consequence the research program doesn't enjoy the same status as a preferred theory of everything candidate that it might have 10 or 20 years ago.
Meanwhile, holography has received a lot of momentum and a lot of its momentum was because of Anti-DeSitter space interpretations of black holes that made interpretation of information more coherent than it had been. That is the cornerstone of holography research and is a product of string theory. It's already making testable and workable predictions of behavior of nuclear physics and plasma physics.
If string theory as a program never gets an inch more data in its support it will still be vindicated as the program that brought us holography. So in that sense it's been importantly falsifiable. But also the fact that it lost significant momentum due to getting no added ammunition from the LHC experiments was falsifiability in a sense that was bad for the theory.
well, really nothing is falsifiable in the Popperian sense. This seems to be a common misunderstanding of Popper. Popper didn’t just think “a scientific theory has to be falsifiable”. He thought falsifying was all that mattered in science and that a single failed prediction from a theory fully proved that theory wrong.
To popper, everything a scientist does is in attempt to falsify theories. Every experiment they come up with is to falsify theories and new scientists in school are trained to only falsify theories. As far as I know, no one is trying to falsify string theory because noone even knows how yet, so Popper would think scientists don’t (or shouldn’t) care at all about it and that it’s not a real scientific theory.
Maybe if we can start testing it in 100 years then popper would consider a real theory, but even then we probably won’t be testing it with the sole purpose of trying to prove it wrong.
And to popper, once something is falsified by a single observation then it’s proven capital-F False. But that is just not what actually happens in science. When we make failed predictions we don’t throw out the theory, we find out what went wrong with the experiment; we say it was a statistical anomaly; or we add an extra planet or dark matter to the theory to make it work.
Logically, Popper said if a theory H makes a prediction P, then we can write H -> P, and if P fails then we have not P, then by modus tollens we get not H. Hence H is False.
But Quine and Duhem pointed out that really when we do experiments, we’re not testing H -> P, we’re testing (H and H1 and H2 and H3 and …) -> P, where all those other H are extra things we assume are true, like “all our equipment works like we intend”, “all the other theories that this experiment relies on are true”, “there aren’t any hidden planets or dark matter we can’t see”, etc etc and there are infinitely many of those things that we don’t even know about. So getting (not P) once is just about useless for proving not H.
Thats not to say scientists don’t care about something that looks like Popper’s falsifiability. Some scientists (especially lab scientists like physicists and chemists) care that theories can make testable predictions, while other scientists (especially field scientists like geologists) don’t care about that quite as much.
But one thing even physicists do seem to care about at least as much as testability is explanatory power. We want theories to explain things that we already see. For example historians show that Einstein cared a great deal about Relativity being able to correctly explain the precession of the perihelion of mercury, maybe even more than it was able to make testable predictions. And even though string theory can’t be tested right now, lots of physicists still care about it because it can explain a lot.
No, physicists do not care about explanatory power at least as much as testability. Unless by "physicist" you mean "philosopher".
I can "explain a lot" with my theory that a god decided it should be so. In fact I can explain absolutely everything that way. Is that a scientific theory? Or is it necessary that scientists like the explanation because it is decorated with a lot of pretty math, like string theory is? In which case, is science a matter of opinion?
(String theory is not very different from the god theory, in that it "explains a lot" but doesn't predict anything useful.)
And likewise a can make a falsifiable theory that God will descend on the populace in 27 years, 2 months, and 16 days and send everyone to heaven. That doesn’t make it a good theory that anyone would care about. When a theory explains a lot and isn’t just mumbo jumbo then scientists like it.
>> string theory is no different than the god theory in that it explains a lot but doesn’t predict anything useful
You’ve missed the entire point. String theory is not “no different than the god theory” because it is consistent with other theories, gives us good insight into the universe, and is generally actually meaningful and useful. Or would you like to go to all the phd physicists at top tier universities around the world who study string theory and tell them their theory is as useful as trash?
A theory doesn’t have to make any predictions for scientists to like it. It just has to tell us something interesting about the world. “God made everything happen that way” tells us absolutely nothing interesting.
Here’s another theory that we can’t test at all yet: The universe is 85% dark matter that we can’t see. This doesn’t make any predictions, yet it explains a whole lot to make it consistent with relativity. Is this also just as useful as your god theory?
>...that anyone would care about....then scientists like it......for scientists to like it....has to tell us something interesting about the world....that anyone would care about....then
scientists like it...gives us good insight...
You are consistently confusing science with opinion. Opinion is for philosophers, not scientists.
> Or would you like to go to all the phd physicists at top tier universities around the world who study string theory and tell them their theory is as useful as trash?
Yes, string theory has not proven useful for anything.
You seem to think all scientists are really just philosophers then. Maybe the only true scientists are freshmen undergrads who have yet to form any opinions and adhere strictly to lab manuals to do science
In philosophy, a convincing argument is as good as proof. But in science, only reproducible observations count---scientists have opinions (aka theories), but most scientists (don't know about string theorists) know that theories are not evidence, only reproducible empirical evidence is evidence.
The opinions and theories of scientists have been wrong or not-even-wrong more often than they have been right.
This is all obvious, it is not something that needs to be explained (except perhaps to freshman undergrads).
you seem to have a warped view of what science or philosophy of science is. If we clearly see scientists doing things that your idea of science would say isn’t something science does, then you can’t say those people aren’t scientists, you have to say your idea of science is wrong.
Your idea of science would say that a huge majority of scientists are not actually scientists, and that a huge majority of science theories are not actually science theories. The only conclusion you should come to from that is that your idea of what science is wrong. Not that those people aren’t doing science.
Scientists clearly don’t only take “reproducible empirical evidence” as evidence for theories. So that part of your idea of what science is cannot be correct.
Scientists clearly incorporate opinions into their theories and their work all the time, so your idea that “opinion is for philosophers, not scientists” cannot be correct
Scientists (including physicists) clearly do care about explanatory power not much less than testability, so your idea that they don’t cannot be correct.
String theory clearly has been and is useful to physicists, so your idea that it isn’t because its not testable cannot be correct.
These are not my opinions. These are things that we see scientists do.
No, none of what you said is true, Outside of string theory, you can produce no examples of scientific conclusions being evidence-free or "incorporating opinions".
Certainly there are theories that have not yet been verified or falsified, but will be some day. There's no problem with that. Science progresses by repeated cycles of proposition (theory) and verification.
But perhaps you are talking about political science, christian science, dismal science (economics), etc. I was talking only about physical science.
Mooks79 | a day ago
I’d say a tentative no. While in a very loose principle there are experiments that could identify some aspects of reality that are consistent with it these are (a) not currently possible and (b) likely never will be.
Even beyond those, the string theory landscape is a massive challenge for the concept of falsifiability within string theory. First, we’ll probably never find the “right” solution, and secondly - even if we did - wha does it mean when you can have basically any solution so it’s almost inevitable one will fit observational data?
That doesn’t mean it’s wrong, per se, as it would then be consistent with a mechanism like Eternal Inflation and we could make some sort of anthropic argument as to why we’re in this vacua. But I think we’re beyond the scope of what Popper meant by falsifiability now.
I’m not sure that’s necessarily a bad thing, anyway. It’s inevitable the more and more you talk about the most fundamental particles, the origin of the universe etc etc, that you will come up against ideas that aren’t falsifiable in the strict Popperian sense. At some point we’re likely to get a situation “this is just the way it is but this is a theory that’s consistent with everything we do/can observe”.
But that doesn’t mean Popper’s ideas aren’t a good guiding principle that we should always be striving for falsifiability, it’s just that the entire concept of falsifiability has a limit when you talk about the most fundamental questions. We just never know where the limit is, so what we think is not falsifiable today we may come up with an idea/experiment that makes it falsifiable in the future - so we should always be trying to think of them.
NOLA_nosy | a day ago
Excellent. Very helpful analysis. Thank you.
freework | a day ago
> It’s inevitable the more and more you talk about the most fundamental particles, the origin of the universe etc etc, that you will come up against ideas that aren’t falsifiable in the strict Popperian sense. At some point we’re likely to get a situation “this is just the way it is but this is a theory that’s consistent with everything we do/can observe”.
What? Either verifiability is a principle that all science has to abide by, or none of science has to abide by. You can't say "this field gets to opt-out, while the rest have to follow the falsifiable principle." What other fields of science get to opt-out? Do astrologers get to opt out of falsifyability too? Why or why not?
-Wofster | a day ago
you might be sad to find out that “verifiability” is already not abided by by all science. Many field scientists don’t care about it. For example the theory about the meteorite causing the end cretaceous mass extinction is certainly a scientific theory but probably isn’t falsifiable
freework | a day ago
I agree. I think what happens is that scientists formulate hypotheses that they can't test (such as the hypothesis that an asteroid wiped out the dinosaurs), but when the general population hears that a untestable hypothesis has widespread support amongst scientists, they all assume that the widespread support must have occurred because of hard evidence. So over time untested scientific hypotheses become scientific fact. This an effect that some have coined "consensism", and it's something that I don't agree with, but unfortunately is very widespread.
-Wofster | a day ago
well nothing is ever nor can ever be proven in science, so in that view everything is just a hypothesis that has widespread support amongst scientists. And the idea that a meteorite wiped out the dinosaurs is supported with hard evidence, like the global iridium layer and the huge crater in mexico. The point is not all science has to be “testable” for it to be science. The meteorite theory is supported by hard evidence yet since it happened in the past is fundamentally not testable.
freework | 21 hours ago
> well nothing is ever nor can ever be proven in science,
I disagree. I believe there are some things that can "proven" to a degree that is very close to absolute. For instance, pouring vinegar and baking soda together gives the fizzy reaction. Its easy to observe, replicate and wrap your brain around whats happening.
Mooks79 | a day ago
Lucky that’s not what I meant, then. Albeit I used the word theory inappropriately in that quote.
vwibrasivat | a day ago
A plausible future experiment could rule out supersymmetry. Or possibly future physics loses SUSY as a framework.
Without SUSY, string theory goes to the grave. Yes?
Mooks79 | a day ago
That’s one of the things I’m referring to (amongst other potential experiments) in my first paragraph. How do you rule out all versions of SUSY consistent with all versions of string theory when SUSY can basically exist at arbitrarily high energy scales? Certainly ones well beyond what we’re going to practically test in any reasonable timescale.
vwibrasivat | a day ago
> SUSY can exist at arbitrarily high energy scales
citation?
Mooks79 | a day ago
Is that really the best you can do? This is well known, have a google.
ProfAndyCarp | a day ago
Have you read about the string theory “landscape” problem? In many string compactifications there appears to be a staggering number of distinct metastable vacuum solutions, often quoted as around 10^500.
Many critics take this to mean string theory is effectively non-falsifiable and lacks predictive power: if one vacuum’s low-energy predictions conflict with observation, another vacuum can be chosen to fit instead.
(String theory supporters hope, but have not yet demonstrated, that additional consistency or vacuum-selection principles could drastically narrow the option.)
vwibrasivat | a day ago
Future experiments could have an effect on the validity of supersymmetry. Can string theory survive without SUSY? your thoughts?
Expatriated_American | a day ago
On this topic, where the rubber meets the road is the question “Should we hire string theorists in physics departments?” Or is this really a branch of mathematics, since there are no experiments to be done to test the validity of string theory.
I tend to think that theoretical physicists should be guided by data, or at least make some falsifiable predictions as part of their research output.
rogerbonus | a day ago
At some level we inevitably are going to have to rely on inference to the best explanation.
terAc5 | 20 hours ago
A. the deeper you go into the particle physics, the word theory becomes really vague compared to other science. They are more like models or suggestions.
B. the art of Theoretical Physics rely on simplicities. its not like the ways on popperian sense, but way more beautiful. you can make bunch of complicated 'real' theories based on real simple theories in all sciences, but non will stand out.
josefjohann | 5 hours ago
A lot of people here are confidently claiming string theory is unfalsifiable. Which in some respects is true both in literal terms because it's testable resolutions are very far beyond our present reach, and the breadth of alternative interpretations that are possible within the framework is vast.
However, there was an expectation that the LHC would access data at energy scales and at sizes relevant to many important contentions of string theory, and nothing in the way of uniquely supporting data was found, and as a consequence the research program doesn't enjoy the same status as a preferred theory of everything candidate that it might have 10 or 20 years ago.
Meanwhile, holography has received a lot of momentum and a lot of its momentum was because of Anti-DeSitter space interpretations of black holes that made interpretation of information more coherent than it had been. That is the cornerstone of holography research and is a product of string theory. It's already making testable and workable predictions of behavior of nuclear physics and plasma physics.
If string theory as a program never gets an inch more data in its support it will still be vindicated as the program that brought us holography. So in that sense it's been importantly falsifiable. But also the fact that it lost significant momentum due to getting no added ammunition from the LHC experiments was falsifiability in a sense that was bad for the theory.
rb-j | a day ago
This question deserves more upvotes.
EdCasaubon | a day ago
It does. So does your comment.
rb-j | 4 hours ago
Well, at least I recovered from -4. It may have gone lower, I dunno.
EdCasaubon | a day ago
No. Which is a damning indictment in my book.
-Wofster | a day ago
well, really nothing is falsifiable in the Popperian sense. This seems to be a common misunderstanding of Popper. Popper didn’t just think “a scientific theory has to be falsifiable”. He thought falsifying was all that mattered in science and that a single failed prediction from a theory fully proved that theory wrong.
To popper, everything a scientist does is in attempt to falsify theories. Every experiment they come up with is to falsify theories and new scientists in school are trained to only falsify theories. As far as I know, no one is trying to falsify string theory because noone even knows how yet, so Popper would think scientists don’t (or shouldn’t) care at all about it and that it’s not a real scientific theory.
Maybe if we can start testing it in 100 years then popper would consider a real theory, but even then we probably won’t be testing it with the sole purpose of trying to prove it wrong.
And to popper, once something is falsified by a single observation then it’s proven capital-F False. But that is just not what actually happens in science. When we make failed predictions we don’t throw out the theory, we find out what went wrong with the experiment; we say it was a statistical anomaly; or we add an extra planet or dark matter to the theory to make it work.
Logically, Popper said if a theory H makes a prediction P, then we can write H -> P, and if P fails then we have not P, then by modus tollens we get not H. Hence H is False.
But Quine and Duhem pointed out that really when we do experiments, we’re not testing H -> P, we’re testing (H and H1 and H2 and H3 and …) -> P, where all those other H are extra things we assume are true, like “all our equipment works like we intend”, “all the other theories that this experiment relies on are true”, “there aren’t any hidden planets or dark matter we can’t see”, etc etc and there are infinitely many of those things that we don’t even know about. So getting (not P) once is just about useless for proving not H.
Thats not to say scientists don’t care about something that looks like Popper’s falsifiability. Some scientists (especially lab scientists like physicists and chemists) care that theories can make testable predictions, while other scientists (especially field scientists like geologists) don’t care about that quite as much.
But one thing even physicists do seem to care about at least as much as testability is explanatory power. We want theories to explain things that we already see. For example historians show that Einstein cared a great deal about Relativity being able to correctly explain the precession of the perihelion of mercury, maybe even more than it was able to make testable predictions. And even though string theory can’t be tested right now, lots of physicists still care about it because it can explain a lot.
SeeBuyFly3 | a day ago
No, physicists do not care about explanatory power at least as much as testability. Unless by "physicist" you mean "philosopher".
I can "explain a lot" with my theory that a god decided it should be so. In fact I can explain absolutely everything that way. Is that a scientific theory? Or is it necessary that scientists like the explanation because it is decorated with a lot of pretty math, like string theory is? In which case, is science a matter of opinion?
(String theory is not very different from the god theory, in that it "explains a lot" but doesn't predict anything useful.)
-Wofster | a day ago
And likewise a can make a falsifiable theory that God will descend on the populace in 27 years, 2 months, and 16 days and send everyone to heaven. That doesn’t make it a good theory that anyone would care about. When a theory explains a lot and isn’t just mumbo jumbo then scientists like it.
>> string theory is no different than the god theory in that it explains a lot but doesn’t predict anything useful
You’ve missed the entire point. String theory is not “no different than the god theory” because it is consistent with other theories, gives us good insight into the universe, and is generally actually meaningful and useful. Or would you like to go to all the phd physicists at top tier universities around the world who study string theory and tell them their theory is as useful as trash?
A theory doesn’t have to make any predictions for scientists to like it. It just has to tell us something interesting about the world. “God made everything happen that way” tells us absolutely nothing interesting.
Here’s another theory that we can’t test at all yet: The universe is 85% dark matter that we can’t see. This doesn’t make any predictions, yet it explains a whole lot to make it consistent with relativity. Is this also just as useful as your god theory?
SeeBuyFly3 | a day ago
>...that anyone would care about....then scientists like it......for scientists to like it....has to tell us something interesting about the world....that anyone would care about....then
scientists like it...gives us good insight...
You are consistently confusing science with opinion. Opinion is for philosophers, not scientists.
> Or would you like to go to all the phd physicists at top tier universities around the world who study string theory and tell them their theory is as useful as trash?
Yes, string theory has not proven useful for anything.
-Wofster | a day ago
You seem to think all scientists are really just philosophers then. Maybe the only true scientists are freshmen undergrads who have yet to form any opinions and adhere strictly to lab manuals to do science
SeeBuyFly3 | a day ago
In philosophy, a convincing argument is as good as proof. But in science, only reproducible observations count---scientists have opinions (aka theories), but most scientists (don't know about string theorists) know that theories are not evidence, only reproducible empirical evidence is evidence.
The opinions and theories of scientists have been wrong or not-even-wrong more often than they have been right.
This is all obvious, it is not something that needs to be explained (except perhaps to freshman undergrads).
-Wofster | a day ago
you seem to have a warped view of what science or philosophy of science is. If we clearly see scientists doing things that your idea of science would say isn’t something science does, then you can’t say those people aren’t scientists, you have to say your idea of science is wrong.
Your idea of science would say that a huge majority of scientists are not actually scientists, and that a huge majority of science theories are not actually science theories. The only conclusion you should come to from that is that your idea of what science is wrong. Not that those people aren’t doing science.
Scientists clearly don’t only take “reproducible empirical evidence” as evidence for theories. So that part of your idea of what science is cannot be correct.
Scientists clearly incorporate opinions into their theories and their work all the time, so your idea that “opinion is for philosophers, not scientists” cannot be correct
Scientists (including physicists) clearly do care about explanatory power not much less than testability, so your idea that they don’t cannot be correct.
String theory clearly has been and is useful to physicists, so your idea that it isn’t because its not testable cannot be correct.
These are not my opinions. These are things that we see scientists do.
SeeBuyFly3 | a day ago
No, none of what you said is true, Outside of string theory, you can produce no examples of scientific conclusions being evidence-free or "incorporating opinions".
Certainly there are theories that have not yet been verified or falsified, but will be some day. There's no problem with that. Science progresses by repeated cycles of proposition (theory) and verification.
But perhaps you are talking about political science, christian science, dismal science (economics), etc. I was talking only about physical science.
Key-Beginning-2201 | a day ago
This is why it's a theory.
rb-j | 4 hours ago
By "theory", perhaps you mean "hypothesis"?
General Relativity is a theory. And has withstood falsification in the Popperian sense.
The Standard Model (including Quantum Mechanics) is a theory. And has withstood falsification in the Popperian sense.