I'm glad that Sabine correctly describes Wolfram, Weinstein, and Lisi's theories as what they really are.
I'm disappointed that she trots out the tired old "zero sum game" rhetoric. I work on small-scale experiments, a field that Sabine portrays as starved of funding because of string theory. But that's not at all how funding actually works. Generally, all of our boats rise or fall together.
String theory is so distant from my field that defunding it and expecting funding for me to rise is like defunding pharmacology and expecting an immediate revolution in palaeontology. Removing funding from one subfield doesn't move it to your favorite field, it just removes it from science in general.
> I work on small-scale experiments, a field that Sabine portrays as starved of funding because of string theory. But that's not at all how funding actually works. Generally, all of our boats rise or fall together.
Please could you quantify this a bit? The competition for grants seems to be fierce. It's hard to see how such a competitive game is not close to zero-sum.
The question is: As scientists and civilians,
1) how to we model/quantify the opportunity cost of 50 years of String Theory?
2) What was the expected value of String Theory (what promises were made?). What has been the ROI?
Hypotheses: What string theory did, among other things, might have been:
1) crowd out the market for ideas.
The perception was created by the media and within academia,that String Theory was/is the hottest idea available. Arguably humanity's most brilliant theoretical physicists and mathematicians seem to have been seduced/compelled into spending their lives working on it. Perhaps one might quantify how many graduate students ended up not exploring other ideas because of the primacy of string theory.
2) caused a perception that physics research is cheaper than it is, effectively reducing the cost expectation in the market.
If string theory is the best idea, and it is purely mathematical, then a university or institution could easily claim to have a cutting-edge physics department, not by funding infrastructure for expensive experiments, but by hiring a string theorist. This in turn might lead to graduate students who can only be trained in string theory. Other institutions might be induced to reduce their costs to the level of string theory institutions/departments. If String Theory is the best idea, how do I justify spending 100x more money than the String Theory department in order to run experiments or train graduate students to do experiments or even work on their crazy new ideas?
3) Lowered the standards to which Physics is held.
If the hottest field in Physics is not required to have any relation to reality, and requires no physical evidence by which ideas are tested for almost half a century, then one might convince themselves that physics can be done purely abstractly, rather than as physical science. For the entire field of Physics to be led in this way, is terrifying.
You seem to assume that the people who control physics funding are a bunch of schmucks who're just shoveling money at string theorists. This is very far from the truth.
* FY 2021 request: $818M
* amount going to all theory /computation: $122M
* amount going to 'Theory': $30M
That theory funding is going to be split among people working on a wide variety of topics, from phenomenology to cosmology to strings. (In fact, the word 'string' doesn't even appear in the document.) In budgetary terms, string theory is a complete sideshow. It gets a lot of media attention, but most string theorists are earning their salary by teaching, rather than from grants. Likewise, very very few string theory grad students are being paid from grant money.
I think your worries about string theorists leading the physics community astray are rather overblown. If it was going to happen, it would have happened twenty years ago, when string theory was at the apex of its prestige. What you generally saw was that places like Rutgers would go from having half a dozen theorists out of a department of 60 physicists to having a dozen theorists out of a department of 66. But the real physicists weren't corrupted or crowded out. They just went right on doing science.
I don't think that's Sabine's point or the parent comment's point, which is mainly about human capital opportunity cost, not funding. Regarding funding, Sabine IS against building a bigger particle collider, which WOULD cost a massive amount of money. The LHC cost ~$5000M.
The parent comment is explicitly discussing funding.
As for the human capital question, it seems absurd to me to call string theorists the 'most brilliant' and then try to dictate what they should work on.
> The parent comment is explicitly discussing funding.
Funding is relevant. Tenure, graduate students, postdocs etc, all cost money. That money has to come from somewhere. Money is intimately tied to the human labour/capital question because there's a bijection between time and money. One of my main questions was: how do we measure this cost and reason about it in order to understand the past 50 years of dismal results and make reasonable decisions about the next 50 years?
> As for the human capital question, it seems absurd to me to call string theorists the 'most brilliant' and then try to dictate what they should work on
What is absurd is to claim that questioning the results of brilliant men and women is absurd. They are not gods to be worshipped blindly. In a startup, would it be absurd for me to question my co-founder and his team's results even though his IQ is 2 standard deviations beyond mine? Let's look further back for a sloppy analogy: is it absurd to criticise Newton's study of alchemy? Was it absurd for his contemporaries, rarely his intellectual equals, to question the results of such studies?
What I want from this debate, is a rigorous way to account for the past 50 years of string theory in terms of human intellectual labour cost, general opportunity cost, and return on investment.
You seem strangely hostile and defensive. But what I'm looking for is merely for the math to speak for itself so that the debate can be settled once and for all.
I've been a bit short with you because you've written long speculative comments without making any attempt to look up the numbers yourself. That's something to be embarrassed about.
I don't have anything at stake here, personally. I am not a string theorist, nor am I dependent on grant funding. But I've worked in and out of academia as a scientist, and it offends me to see you play-acting as a critic of science without even understanding the lay of the land.
> I've been a bit short with you because you've written long speculative comments without making any attempt to look up the numbers yourself. That's something to be embarrassed about.
That's an assumption on your part, one that is both incorrect and unfortunate. (as far as theoretical physics is concerned.)
Embarrassment, however, is largely irrelevant when trying to understand anything in a systematic way.
> and it offends me to see you play-acting as a critic of science without even understanding the lay of the land.
Offendedness. Well, we simply have different experiences and perspectives within academia. We see the land differently. If that's your threshold for offence then life must be difficult indeed. I don't envy your position. If it will somewhat assuage your being offended I should tell you that, as a mathematician, I champion science above any other human activity. But who I am and what I do should be irrelevant when testing hypotheses.
Yeah, in grad school the theory grad students were the ones who had to TA a bunch to cover gaps in their funding.
When you have a couple million tied up in lab equipment it's easy to divert some to grad students to keep it running but us theory folk were switching projects every couple years just to keep the lights on as it were
Well, maybe. Or maybe concentrating funding in less promising fields and approaches leads to disappointing results, which in turn leads to lower funding in the long term.
I did enjoy watching the spacetime videos although I also could not understand most of it. But I feel like the content and topics jumped around a lot and they would present string theory more or less as fact. It was a grabbag of cool ideas but it would be more fruitful I think to build upon layers of knowledge
I'm right there with you - I've watched their videos for better part of 2 years. Each video I've seen most likely twice while some of them I need to watch 4-5 times to understand.
At times, I'll think I understand it. Come back 6 months later having a bit more understanding of related topics only to realize I understood absolutely none of the original video and am just beginning to get it =/. Seemingly innocuous sentences carry insane amounts of information (i.e. Do anything to a blackhole and the surface area never decreases except in Hawking Radiation)
The topics do appear jump around a lot, but there's method to the madness. Topics such as Blackhole/Thermodynamic entropy require foundations that appear unrelated at first.
Sure, they have video lists[0] to help but I've rewatched this one so many times.
> they would present string theory more or less as fact
To be fair, they have a why ST is right[1] and wrong[2].
I've been watching PBS Space Time since they started years ago. A lot of their videos build on previous videos, so to fully understand a concept you might need to go down some rabbit-holes. But it is worth it.
I'd much rather have channels like this that welcome actually getting into the details a bit, rather than the hundreds of channels that have broad, surface-level facts of a subject lazily combined with some stock-footage.
I disagree that we don't need to unify gravity and the other 3 forces under one model.
We need to unify them because there is obviously a missing link between the two theories (QM and GR), since spacetime is one and the same and there are properties defined in one theory that clearly affect the other one (but we don't know them yet).
I believe you are being imprecise with your terms. Understanding how gravity applies on a quantum level is a theory of "quantum gravity". That is a very different concept than a "unified" theory. Your post uses appears to use the concepts interchangeably, although OP took care to show they are different.
Another thing to note here is that the quantum theories are completely compatible with Special Relativity. That is, in non-curved spacetime Quantum Field Theory makes all the same predictions as special relativity does, plus all the quantum weirdness.
When scientists talk about unification, they're talking about the idea that all the forces are fundamentally the _same force_ at high enough energy (or early enough times after the big bang).. They've only thus far been successful at unifying the electromagnetic and weak forces.
The author describes the unification process of the non-gravity forces under various symmetries and that there are desires to find a way to unify these under a GUT, and also describes there is a need to modernize gravity so it fits logically with the QFTs.
But then the author seems to make a point of that a ToE has to be a GUT with gravity in it and that might not be how nature works?
Which I agree with but I would not have used that semantics of ToE. I think most physicists at this point would settle for any logical merging of gravity with the QFTs that is consistent even if it postulates more free parameters and not less :)
Anyway, just a nitpick on the terminology.. just seemed a bit click-baity (but I'm sure that was not the intention).
> There is no reason that nature should actually be described by a theory of everything. While we *do need a theory of quantum gravity to avoid logical inconsistency in the laws of nature, the forces in the standard model do not have to be unified, and they do not have to be unified with gravity. It would be pretty, yes, but it’s unnecessary. The standard model works just fine without unification.
We don't need the theories to be unified for them to be consistent. Putting all our efforts into that unification could actually be leading us astray and postponing the day when we've figured out how they're consistent.
David Deutsch is the only person who has a theory of everything, not just for mechanics. He connects quantum theory (he wrote the first quantum algorithm), evolution (of genes and memes), and computation into the only true theory of everything. It's elegant, meaning the explanation of the theory is beautiful and highly structured. He presents it in The Fabric of Reality (1997) and has since been building it into constructor theory http://constructortheory.org/
I don't think he has attempted to fold epistemology into physics or has a deep understanding of evolutionary epistemology and how it connects with computation or mechanics.
> Having said that, what do you think I think about Lisi’s and Weinstein’s and Wolfram’s attempts at a theory of everything? Well, scientific history teaches us that their method of guessing some pretty piece of math and hoping it’s useful for something is extremely unpromising
Is this really true? Perhaps we can make this critique of string theory but what about other cases?
Einstein had the physics idea for general relativity first, and adopted the mathematics of differential geometry to try to implement it. People had the idea for quantum mechanics first, and then discovered that there already existed mathematics for it. For quantum field theory, there was no existing mathematics for it, so they invented their own. For gauge theories (such as the Standard Model), I'm not quite sure the initial motivation.
On the other hand, I think you could argue that Lagrangian and Hamiltonian mechanics were developed for aesthetic reasons, and these both proved influential for twentieth-century physics.
Lagrangian and Hamiltonian mechanics were useful early on, though. They make perturbative calculations much easier in classical mechanics.
If you want to analyze the effect of one planet on the time evolution of another planet's orbital parameters, Hamiltonian mechanics is the formalism to work in.
Yes, we do need a "theory of everything," or at a minimum we need a model of quantum gravity. Consider the color of gold, as a famous example; gold has its unique and curious color because of a combination of relativistic and quantum effects at the subatomic level [0].
Edit: Wow, fast downvotes! Sorry, I'll add another paragraph: The reason why we cannot use quantum mechanics (QM) or general relativity (GR) alone to explain the color of gold is because we know that the color comes from photons being emitted by electrons, as in QM, but the electron orbitals are distorted as if the electrons were moving at a serious fraction of the speed of light, as in GR. I think that this flatly contradicts her when she says:
> So far, this is a purely theoretical problem because with the experiments that we can currently do, we do not need to use quantum gravity. In all presently possible experiments, we either measure quantum effects, but then the particle masses are so small that we cannot measure their gravitational pull. Or we can observe the gravitational pull of some objects, but then they do not have quantum behavior. So, at the moment we do not need quantum gravity to actually describe any observation.
The color of gold, the liquidity of mercury, and even the ability of lead-acid batteries to turn over car starters are all everyday observations which need both QM and GR to explain. But, as she correctly notes, we can't put QM and GR together yet! So we do need some sort of quantum gravity.
I’m a physicist and I downvoted. Relativistic quantum field theory (which explains gold’s color) and quantum gravity are two entirely different things. We already have a relativistic quantum theories which combine quantum mechanics with Einstein’s special relativity, and explains everything we know about chemistry. Quantum gravity would combine the standard model with general relativity, which is only really needed for non-human scale things like black holes and the origin of the universe.
Special relativity and general relativity are two very different things. Don’t conflate them.
Downvotes might be because the article distinguishes between a theory of quantum gravity and a theory of everything:
> Physicists would much rather have all these forces unified to one, which means that they would all come from the same mathematical structure. [...] At the energies that we have tested it so far, the symmetry would have to be broken, which gives rise to the standard model. This unification of the forces of the standard model is called a “grand unification” or a “grand unified theory,” GUT for short.
> What physicists mean by a theory of everything is then a theory from which all the four fundamental interactions derive. This means it is both a grand unified theory and a theory of quantum gravity.
As she says, that's specifically "grand" unified theories (GUTs). A theory of everything might be imagined as similar to a GUT, in that they should both have all of the same stuff on the blackboard and arranged into roughly the same sorts of equations, but a theory of everything might not be able to express all four forces in terms of just one force. Indeed, what we mean by "force" is two different things entirely: in GR, "force" is the shape of the fabric of space, while in QM, "force" is purses of coins (I don't want to say "packets of data") traveling through that space.
I don't quite agree with Wikipedia [0] on this; they say:
> Unifying gravity with the electronuclear interaction would provide a theory of everything (TOE) rather than a GUT. GUTs are often seen as an intermediate step towards a TOE.
We all have different opinions on this terminology, it seems!
Not a physicist. But perhaps this is referring to combining quantum mechanics with special relativity, not GR. So these examples are not examples of problems where you need a theory of quantum gravity.
Wrong. Special relativity ignores gravity interactions. This article is explicitly mentioning gravity. Describing gravity is why Einstein researched/invented GR.
I think the main questions we try to answer with a theory of everything is about gravity. How does gravity work at quantum level? Is it possible to convert other fundamental forces into gravity?
Yes, and OP is explicitly saying that is incorrectly conflating two separate things. The questions you list would be answered by a theory of quantum gravity. That theory can be consistent and answer all of our questions. A theory of everything is superfluous.
Don't Maxwell's equations all apply to the electromagnetic interaction? Why would unifying Maxwell's equations help unify electromagnetic forces with the weak and strong nuclear forces?
At the moment, we are using two mutually incompatible theories to explain sunlight. We need gravity to explain how the sun formed and we need QFT to understand how it produces radiation.
I don't mean to imply that humans are terrible. I just mean that we think we have a lot of answers when we don't.
I came up with my theory when I read about Norman Borlaug and how he developed GMO semi-dwarf wheat which is purported to have saved a billion lives from starvation.
But people now blame GMO crops for all manner of problems like unhealthy gut, Autism Spectrum Disorders, gluten sensitivity, etc.
In sum, we're just trying to do the best we can, and we can't see all the possible consequences of our actions.
A bit tangential to the article but I just had this thought:
Human knowledge has gaps because humans do not live long enough to effectively pass enough knowledge from generation to generation. Each generation generates knowledge and records it in one form or the other AND also passes it with direct verbal and non-verbal communication. However, since the overlap between two generation's cognitive prime age is narrow, not every thought, intuition, idea is passed along well enough. This creates a continuous rot in knowledge among generations and hence the humanity causing perpetual gaps in knowledge.
This is the main reason why humanity at large does not have perfect knowledge of things such as universal forces, exact nature of universe, ability to predict random events and so on.
the overlap between two generation's cognitive prime age is narrow
That argument fails on the above point - you picture generations as discontinuous (or nearly so), whereas in the normal, non-technical sense - as here - they are simply a useful picture of an aggregation over a continuum of ages in the population as a whole.
Working and academic environments - beyond a certain level of competence - are populated by individuals of varying ages and experience which are more likely to be demarked by managerial levels and career choices than individual knowledge and aptitude.
Not sure I buy the "narrow overlap of generations" as the main reason for imperfect knowledge. Yes humans don't live long enough. But generations and population ages vary smoothly in a cohort. May be centuries ago, when the effective number of people you interact with is small and finite due to slow transport of information, people and goods. But I would argue since the advent of books, organized schools and universities this is less of a problem. Now with the internet this is really a non issue. One has all the lectures that one could possibly want to listen and all the books and papers to spend every waking moment consuming. So I guess fundamentally we are only limited by our finite lifespan, and may be working memory that has a good decay rate.
> Now with the internet this is really a non issue.
A good chunk of valuable information disappears all the time from Internet. Many thoughts expressed in personal blogs usually go away within a few years. See, for example, demise of Geocities and Yahoo Groups. While only a small part of this disappeared information is valuable, some of it incredibly useful and important.
You're completely ignoring the argument. The purpose of the internet is not recording information for all eternity. It's to transmit information from person to person. If you want to preserve information then write a book and publish it. It will be archived in organizations that specialize in archival. They allow the internet as a communication method but in theory nothing prevents you from just driving to the location and reading the book on site without the internet.
Either the information was important and was probably backed up somewhere or replicated on other websites while geocities was still alive and other platforms where coming up (the overlap of generations) or the information was not noteworthy in which case it probably wouldn't have been found even if geocities existed today.
I like this idea. But I think there are some advantages from information transfer being incomplete.
Because new humans have to learn what the prior generation learned from scratch, it gives them a fresh canvas to eliminate flaws and purify knowledge from the prior generation.
It's like taking a wrong turn on a roadtrip. You've driven so far and convinced yourself that you're on the right path. But in reality you need to step back and check all the assumptions and choices you've made to get on this path.
People coming in and out of existence facilitates sanity checks on current acquired knowledge.
I've had similar thoughts lately. Here's a simple mathematical model of the process that I felt inspired to create by your comment:
k[i] = (1 - c1) * k[i - 1] + c2
where i is the generation number, k is the amount of knowledge we have, c1 is the loss factor (amount of knowledge lost from one generation to another, 0 ≤ c1 ≤ 1), and c2 is the (constant) amount of new knowledge created by generation i.
In this model, it's easy to show that after long periods of time the amount of knowledge will plateau. In fact, the plateau is
k[∞] = c2 / c1
The basic point of this is to highlight that both knowledge creation and loss are factors, and the loss means that knowledge may not keep increasing indefinitely. Of course, any number of the approximations and assumptions involved in this derivation may not be valid, but I think the possibility that knowledge loss can cause problems is worth considering.
Treating c1 and c2 as constants seems problematic. Different generations have different populations and access to resources such as education, technology, etc. The unknown knowledge to be discovered varies. The methods of preserving and transmitting knowledge vary.
There were certainly cavemen born with Einstein-level intelligence from time to time, but without a foundation to build on, peers to collaborate with, or written language to preserve their knowledge, they would both create and pass on significantly less knowledge than someone in modern times. One generation might lose a large portion of all the knowledge built up for millenia prior during a civilization's collapse; while another generation may stumble upon a discovery which ushers in a short but extremely productive era of knowledge production. If nothing else, if you have 4 billion people in one generation and 8 billion in another then all else being equal the latter generation has twice as many brains with which to store and create knowledge.
Fair points. I've made a few refinements already on paper, switching to an ODE model rather than a difference equation and changing the knowledge production rate to vary. If there are diminishing returns then we would expect c2 to decrease as k increases at large k. As you're aware, perhaps this is offset by increases in population.
As you've indicated, heterogeneity in various parts of the problem make this sort of modeling questionable. It's a crude model not meant to make precise predictions, just show that knowledge loss can be a major factor at long times.
But there's also that saying about how science progresses one funeral at a time. I think incremental progress might be more likely to occur with long lived humans, but breakthroughs may come only with new blood.
Though, if a very long lived human were to become an expert in many different fields, that could also be conducive to innovation.
>However, since the overlap between two generation's cognitive prime age is narrow, not every thought, intuition, idea is passed along well enough.
The overlap is perfect and zero at the same time as "generations" are a descriptive simplification rather than a reflection of reality. Every year people are born whose overlap with the people the year before is off by just one year (more or less). A new PhD student learns from his peers a year further along as well as from the post-doc 5 years ahead and the professor 10 years ahead.
You're saying that if we simply lived long enough we'd be able to predict random events?
I have strong objections to this theory/thought.
You'll have to describe to me how this mechanism works in greater detail because even within my personal lifetime, I do not have perfect anything for anything and I have no clue on how living longer helps that particular problem.
Let's say I throw a rock in a parabolic curve in a plain field. Humans can predict exactly where it will fall. This is because humans know everything about gravity, mass, etc.
However, we cannot predict when and where exactly will a random meteor fall. This is because we do not know exactly everything about the universe and the various forces of universe.
My thought is that a lot of humans might have understood more about those forces in the past but were unable to effectively pass that information for the next generation to build upon. This could be simply because the cognitive prime of the two generation overlaps for a relatively short period of time.
Now had the humans lived long enough to cross some threshold of effective communication and longer overlap of cognitive peak between generations, they could have better able to pass their intuition, experience, thoughts, ideas to the next generation. In other words, there would be less loss of information/knowledge which would be just enough to fill the critical gaps to know everything about the universe.
Knowing everything about the universe would enable them to predict everything precisely just as in the example of a parabolic throw of a rock on earth's surface.
> I throw a rock in a parabolic curve in a plain field. Humans can predict exactly where it will fall
No, we can predict this within a margin of error. That margin of error might be small for simple cases but if you add some gusty wind, some rain etc. it can increase quite rapidly.
> However, we cannot predict when and where exactly will a random meteor fall
We can also predict it within a margin of error just like with the rock in the plain field. But you are comparing an event that last a few seconds (from the throw to the landing) to an event that may last years or more depending on what kind of meteor you are talking about. So of course over the span of years, even a very small imprecision turns out to have very big consequences.
If you could throw the rock hard enough so that it orbits earth for some years before coming back, it would be as hard to predict where it would land.
> Knowing everything about the universe
How could you possibly know everything about the universe when some events are purely random?
> In other words, there would be less loss of information/knowledge which would be just enough to fill the critical gaps to know everything about the universe.
So you are saying we happen to live at the exact time, where, had we not this knowledge transfer issue, we would know everything? This is quite self-centered.
Which also simply mean that in a relative close we will now everything, which seems quite ludicrous.
Also take into account quantum theory. There IS no precise location and velocity at the same time. There are only probabilities. At least according to the Copenhagen interpretation
> My thought is that a lot of humans might have understood more about those forces in the past but were unable to effectively pass that information for the next generation to build upon.
This sounds incredibly unlikely. When do you imagine that these people lived? What data were they using to derive their scientific theories? What technology were they using to generate that data?
You can't hope to explain the universe without observing it first, and humans are quite bad at observation without the aide of technology.
Another thing is that the way we store information in our brains is different than the way we store information on paper or in computers. It's not possible to communicate ideas directly between brains so we're always going to be losing information in that transfer.
There are no "generations". It is a simplification when we talk about "boomers" and "generation-X" and "millennials". Every year about the same number of people get born and die. You could say that every year is a new "generation".
Talking about "generations" is a bit like talking about signs of Zodiac. Are you a millennial? Or are you Capricorn?
Don’t we just end up running into Gödel’s incompleteness theorem?
Human ability to close this “linked chain” of generations will always have a “missing link”?
An asteroid or... great filter due to environmental collapse!
To me it always felt intuitively like being able to have perfect knowledge of all wave function collapses, all the states of every particle ever scale of a problem.
There’s entropy too; no more information to pull from the universe sounds a lot like heat death to me.
Are we chasing a biochemical feedback loop? There’s a lot of “unproven” math, to be casual Sunday night from a legal weed state about it ;)
How much potential is there all of this boils down to our ability to so thoroughly convince ourselves of something’s correctness and it turns out there’s no Gods on top of that mountain?
That applies to other animals, but not humans. It's almost bizarre that you would assert that because the exact opposite is true. Humans are in every way more "advanced" than other animals on the planet precisely because the information they pass down is useful, comprehensive, and widespread. The last few thousand years has seen the information that gets "passed down" become systemic, and prolific. "Overlap" has nothing to do with anything, as anyone at any age can access a large percentage of important human knowledge going back hundreds to thousands of years, and it's really pretty easy to do so.
Edit in response to downvote: This is an effort to model the whole universe using graph theory. I found it amazing not because how possible it might be the true answer, but because it provides an insight that clicks: a computational world, where any limited space can store only a limited amount of information. At least it is a beautiful mental model.
"Do we need a ToE" is indeed an important question. But if some theory has been proposed and it "looks like" a good candidate of Theory of Everything, then many of us can just skip such questions.
I'm disappointed that she trots out the tired old "zero sum game" rhetoric. I work on small-scale experiments, a field that Sabine portrays as starved of funding because of string theory. But that's not at all how funding actually works. Generally, all of our boats rise or fall together.
String theory is so distant from my field that defunding it and expecting funding for me to rise is like defunding pharmacology and expecting an immediate revolution in palaeontology. Removing funding from one subfield doesn't move it to your favorite field, it just removes it from science in general.