# Stupidity of the pop science consensus about «many worlds»

Luboš Motl, November 23, 2016

https://youtu.be/dzKWfw68M5U?si=ccQCMdAoyMhAx-KM

Most of the sentences in the video are either downright wrong or at least seriously misleading.

The weird rules of the subatomic world are very different from the laws of the large Universe.

When one is careful, this is just a correct sentence. But those people are never careful so it’s obvious that they actually say lots of wrong things in between the lines.

One of them is, as we will see momentarily, that the laws of quantum mechanics do *not* apply in the large Universe. But they do. The laws of quantum mechanics apply everywhere in the Universe. It’s just that in the context of large objects, the laws of classical physics *also* apply – albeit just approximately. But the laws of quantum mechanics never cease to apply. The relationship between classical and quantum theories is fully analogous to the relationship between non-relativistic and relativistic physics: in both cases, the older theory is a limit, either \(1/c\to 0\) or \(\hbar\to 0\) limit, of the newer theory. But the newer theory is always right, even at arbitrarily low speeds or for arbitrarily large objects.

The video – and equivalent films, books, newspaper articles etc. – contain a huge number of places that make it clear that their authors just don’t understand that quantum mechanics applies *everywhere* in the Universe.

Also, the laws of quantum mechanics aren’t really «weird». They’re as non-weird as the world around us because they exactly agree with everything we know about the world around us.

Now:

The question is when and why the weirdness of QM gives way to classical physics.

As I just said, quantum mechanics never «gives way». At most, they *share* the realm of macroscopic processes. But quantum mechanics never gives up its right to govern everything.

Bogus pop science videos and books are brainwashing their listeners and readers all the time. If you look at them carefully, you will see that every third sentence is filled with some absolutely misguided «plan» to get rid of quantum mechanics, suppress it, deny it, disprove it, reinterpret it, avoid it, or make it disappear. Quantum mechanics never disappears or gets distorted or bent and by talking in this way, you only show your complete misunderstanding of modern physics.

An answer?

An answer is that the Universe is so much weirder than we imagined. Or should I say: the multiverse.

The Universe follows completely universal, logically consistent, sensible laws of quantum mechanics. The laws weren’t known up to 1925 but once they were discovered, physicists understood them and they made sense. And no, you shouldn’t say the «multiverse» because the «many worlds» story is just a fairy-tale. Moroever, even if one chose to talk about the «many worlds» fairy-tale in the context of quantum mechanics, it’s wrong to use the term «multiverse» – a concept that refers to a set of actual geometrically separated universes in inflationary cosmology etc.

One of the strangest features of QM is the superposition.

There is nothing «strange» about the superposition. The very concept «superposition» has been used – for waves – for centuries before QM was discovered. Mathematically, superpositions in QM are exactly the same as the superpositions of electromagnetic waves etc. Physically, the waves have a completely different interpretation. But it’s not a «stranger» interpretation. It’s just different.

Here, he says that the wave functions are probability clouds and discusses the double slit experiment. So far so good. Particles probe both slits and all histories. He says that the histories seem to «converge» to the outcome observed at the end.

The Copenhagen school thought that the act of measurement collapses the possibility space into a single reality.

The fact that some entity behaving like waves is needed to predict physics is clear from the experiments, and so is the fact that one observes one particular location of the particle on the photographic plate. These are not artifacts of some «interpretation». These are experimentally proven facts. And indeed, they also have to be pretty much axioms – universal postulates – of quantum mechanics, a framework designed to explain and predict all these phenomena.

The collapse marks the boundary between the classical and quantum realms.

Yes and no. The relationship between the collapse and the boundary is that before the collapse, classical physics is guaranteed to be an inadmissible description of the events. After the collapse, we’re getting some data – results of measurements – that may be talked about in the same way as the objectively real values of observables in classical physics. But the detailed laws that they follow in quantum mechanics are still different than they were in classical physics. In classical physics, they are calculable – or they deterministically evolve – from the values of quantities in the past. In quantum mechanics, they’re uncertain and different outcomes have calculable probabilities. They are calculable from the interference between different intermediate histories.

Schrödinger found it ridiculous and proposed his cat to highlight the absurdity.

He invented the cat to mock quantum mechanics but the important information is that Schrödinger was completely wrong – and known by the top physicists of his time to be wrong – and the laws he wanted to mock are *exactly* how Nature works. Viewers of such videos are being trained as followers of Schrödinger of a sort – they are being deceived about the very fundamental facts concerning quantum mechanics.

The host describes the cat experiment at some superficial level and asks:

But doesn’t it mean that the cat should also be in a superposition? If so, it should be both dead and alive.

Yes, the allowed space of states for the cat is a linear Hilbert space – which is true for cats or any other physical system in Nature – so it is generally found in a superposition. And no, it doesn’t mean that the cat is «both» dead and alive at the same moment. The superposition of two states means that the physical system is *either* in one *or* the other. The superposition doesn’t mean that two objects or objects in two conditions exist *simultaneously*. It means that two (or more) *possibilities* exist simultaneously. The superposition is an addition but when we’re adding wave functions, we’re not adding real objects like apples, oranges, or cats. We’re adding probability amplitudes – which is much more similar to the addition of probability distributions than to the addition of apples, oranges, or cats.

It’s possible for the cat to be dead *and* it’s possible for it to be alive. But in the cat experiment, it is *not* possible that there exist two cats at the same moment. The addition \(+\) in between two wave functions corresponds to the logical addition i.e. the word «OR», not «AND». To get «AND», you need a multiplication. In particular, the wave function describing two independent objects is the tensor product \(\ket\alpha\otimes \ket \beta\) of their individual wave functions.

But why can’t the cat collapse its own wave function?

When the cat is perceiving the world, it surely can. From its viewpoint, it collapses the wave function just like any observer does. But the point is that the precise form of the wave function *depends on the observer*. So if a cat is perceiving things, it should be using different wave functions than the wave functions that an external observer needs to use.

The guy asks whether the observer or the Universe is blurred. Yes, the human beings and anything else in the Universe – and the Universe as a whole – is generally found in a superposition of various states. But the wiggly fuzzy video sequences are demagogic because they show something that looks different from the world as we know it. In contrast with that, the superpositions used by quantum mechanics show the world *exactly* as we know it.

Many adherents to the Copenhagen interpretation now have a more sensible description of what’s going on: the superposition doesn’t extent do macroscopic scales.

This is complete rubbish. Everyone who understands quantum mechanics at least a little bit knows that all physical systems are described by linear Hilbert spaces i.e. superpositions extend *everywhere*.

The superpositions disappear when different quantum histories diverge.

The superpositions don’t disappear in the decoherence. The only thing that happens is that the relative quantum phases become unobservable in practice.

He says that «coherence» means that «the waves sufficiently overlap». But that’s simply not a correct identification. Coherence is absolutely independent from the geometric overlap of the individual terms. The parts of the wave function for a particle behind the two individual slits don’t geometrically overlap but they are able to create interference patterns. On the contrary, wave functions describing different states of our wet and messy brains are defined in the same contiguous space of possibilities but the quantum coherence is largely lost.

Many verbal defenders of the «many worlds» identify «coherence» with the «geometric overlap». By conflating these completely different things, they show how incredibly sloppy they are.

When coherence is lost, the alternative histories cannot interact with each other.

This kind of talk shows another absolutely embarrassing conflation of concepts. He confuses «interactions» and «interference». Interactions are given by terms in the Hamiltonian that prevent us from splitting the Hamiltonian into the sum of the Hamiltonians for non-interacting pieces. The non-interacting

\[ H = H_1+H_2 \] is replaced with \[H = H_1+H_2 + V_{12}, \]

if I am just a little bit schematic (but basically accurate). This form of the Hamiltonian introduces the interaction \(V_{12}\) between two *objects* \(1,2\) that would be independently described by separate Hamiltonians \(H_1,H_2\). On the other hand, the interference of wave function exists despite the fact that the evolution operator transforming the wave functions is *precisely linear*. So the two interfering pieces of a wave function don’t interact at all. When a wave function for an electron looks like the sum of two faraway packets on a collision course, these packets will just go through each other without *any* repulsion or recoil or collision or interaction. This is guaranteed by the linearity which *is* precisely obeyed in quantum mechanics and *has to be* obeyed – basically because it gives rise to the linear laws for probabilities such as

\[ P(U)+P(V) = P(U\text{ or } V) + P(U\text{ and }V). \]

Terms in a wave function which is a superposition don’t interact before decoherence and they don’t interact after decoherence, either. Decoherence doesn’t add any or subtract any *interactions*.

If he wanted to say that the diverging parts of the wave function get separated and cannot «overlap» again in the future, it’s also wrong, as I said previously. The important observables in our brain decohere from each other quickly but they still occupy the same space of possibilities and can reunite in the future.

The Universe chooses a history, more precisely, it chooses a final result.

Right. And he also correctly says that the constructive interference makes some results more likely than others. A rare segment in the video that is actually correct. However, within seconds, we’re back to stuff like:

The Copenhagen interpretation says that the selection occurs in a fundamentally random way.

It’s the theory known as quantum mechanics, and not just some «interpretation», that postulates that the outcomes are intrinsically probabilistic. Also, this fact may be pretty much safely verified experimentally. The host would call it a «nondeterministic interpretation». Well, it’s just a wording making it harder for him to learn physics because this wording encourages its user to remain doubtful and confused about the unquestionable fact that the fundamental laws of Nature are intrinsically probabilistic in character.

However, there’s another interpretation, one in which nothing collapses.

It’s not an interpretation, it’s a wishful thinking dreaming about the existence of a *completely different* theory that could describe Nature. However, it’s easy to see that no such alternative theory can be viable. The bloke talks about everything existing in copies and he says

It sounds outrageous but it is a very serious interpretation of quantum mechanics.

It doesn’t sound outrageous and it’s not serious. It’s just a stupid crackpot fairy-tale that has nothing to do with physics. We also hear that Hugh Everett proposed this idea in his PhD thesis. Well, the thesis didn’t actually contain claims about the many worlds – even though a reason could have been adviser John Wheeler’s discouragement of his student. Different pieces of this silliness were added (or perhaps reinstated) later by others, especially by DeWitt.

«Many worlds» say that all the worlds in which the electron lands at any place \(\vec X\) of the photographic plate after a double slit experiment simultaneously exist.

There’s no canonical way to split a wave function \(\ket\psi\) into a sum of pieces that could be interpreted classically. Only the sum, the total vector \(\ket\psi\), contains the physical information. To write this vector as a sum may be useful but it may be done in infinitely many equally valid ways and which sum is more useful than others depends on the future observations we want to perform on the system (the decomposition into the sum of eigenstates of an operator that is going to be measured is typically more useful).

A particular preferred splitting of the wave function cannot in any way «exist» before the future observations.

All the worlds with different locations \(\vec X\) of the electron on the photographic plate are equally likely.

That’s too bad because literally in every experiment in the world (except for a measure-zero fine-tuned subset) we may see that the different outcomes are *not* equally likely. So this «theory» or «alternative interpretation» is just dead. Could you please stop talking about this rubbish then? Needless to say, No. These folks don’t give a damn whether their fairy-tale has been falsified.

He takes the attitude that the more likely outcomes should correspond to «many copies» of the same Universe.

Too bad that nothing of the sort may be derived from any conceivable theory. Moreover, probabilities are continuous (as we know from the experiments and as we can calculate from proper quantum mechanics) while the «fraction of the Universes with a certain property» has to be a rational number, to say the least. In quantum mechanics, the probabilities are continuous and fundamental – they’re the most elementary quantities that can be calculated from the equations of Nature. Their being irrational (not a ratio of integers) is a simple way to see that no framework in which they would be derived as a «fraction of some Universes» may ever be found.

While talking about the «many worlds» fantasies, the people reveal their wishful thinking and formulate «demands» how they want to Nature to behave. If their brains were functional, they would also be able to figure out that no theory obeying the demands may exist.

It seems as crazy to build uncountably many Universes to avoid the probabilities in QM – as to build a new house to avoid washing dishes.

Well, right, but this metaphor is surely not the biggest problem of the «many worlds» religion.

The Copenhagen interpretation itself proposes many worlds in the superposition.

No, the Copenhagen interpretation includes (primarily) exactly the opposite statement that the superposition describes one copy of a physical system whose future measured properties aren’t uniquely known and can only be predicted probabilistically.

The interpretations are basically the same except that the Copenhagen interpretation merges the histories into one timeline.

The Copenhagen interpretation – again, he means quantum mechanics – doesn’t merge any histories. The other histories just never existed. When I throw a die and get 6, even though all six outcomes were *a priori* possible, I haven’t created any Universe(s) where the die would show 4. It was a possibility but once the outcome 6 is measured, the other possibilities are known not to be real in any sense. In this example, it is enough to imagine the classical probabilities. But the uncertainty and probabilities in quantum mechanics have the *same* interpretation.

After all, even the uncertainty about an ordinary real-world die may be described by quantum mechanics. The more we know about the atoms, the more capable of predicting the outcome we may become. So the uncertainty of the outcome depends on the precision of our knowledge. The same statement is true in quantum mechanics as well. The only new aspect of quantum mechanics is that it implies that the precision of anyone’s knowledge of any initial state is bounded by the uncertainty principle so some uncertainty about the outcomes is unavoidable, too.

Many worlds may be the purer interpretation of the mathematics of quantum mechanics.

«Many worlds» don’t have anything to do with the mathematics of quantum mechanics. The whole purpose of *any* calculation in quantum mechanics is to make predictions of physical outcomes. And these predictions have to be fundamentally probabilistic. Quantum mechanics predicts *probabilities*. Because there are no probabilities in a «many worlds interpretation», such an interpretation has *zero* in common with mathematics of quantum mechanics which is all about the calculation of the values of probabilities.

«Many worlds» is just an independent fairy-tale that some people try to add on top of quantum mechanics, just like some people want to add Noah’s Ark to Darwin’s theory. But the pieces don’t have anything to do with each other and when one looks just a little bit carefully, they are absolutely incompatible with each other.

Many worlds is more economical in the number of concepts it adds to quantum mechanics.

The price we pay is that it can’t agree with *anything* that we know about Nature. But this «economy» is deceitful for so many reasons. First of all, the Copenhagen rules *are* the rules of quantum mechanics. They don’t have to be added and they can’t be added. They’re already there and they’re the only rules that connect the mathematical symbols and structures with our real-world perception in *any way*. It you removed them from quantum mechanics, you would get a theory saying *nothing at all* about the physical world. What you learn in a course on linear algebra or any other mathematical course may be *useful* for your doing physics but it contains *no physics* by itself. One needs to add *some physical rules or postulates about Nature and observations* and the «axioms of the Copenhagen interpretation» are really the *physical beef* of quantum mechanics.

Second, even if one imagined that some «many worlds»-based theory would be possible, it’s spectacularly obvious that it couldn’t be economical in this sense at all. It would have to contain completely new mechanisms that split the world and that decide the «lines» along which the worlds are split, precise timing when they are split, how many of them are the result of the splitting, and some special «bonus gadgets» that allow the fraction of the post-splitting worlds to be irrational, among many other things.

No one can write such a theory and it’s pretty clear that one can’t exist but if it existed, it would be extremely far from an economic theory, let alone a more economic theory than conventional quantum mechanics. What is economic is just the «fairy-tale» – and its economic because the believers in this pseudoscientific religious cult have *zero interest* in producing and verifying actual scientific predictions. Their fairy-tale hasn’t ever achieved and cannot ever achieve anything of the sort but these people just don’t care a tiny bit. They’re economic – because they have thrown out the baby with the bath water.

Everett’s idea wasn’t taken too seriously when it was first proposed.

It’s still not taken seriously by physicists who are up to their job. It’s just a favorite fairy-tale among pop-science writers. Yes, some of these writers are also physicists during the workdays. But their «many worlds» hobby has nothing to do with their careers. No serious physicists are using «many worlds» in their actual research in any way.

It wasn’t famous partly because he was a graduate student who disappeared into the Pentagon.

He disappeared into the Pentagon *because*, unlike some other graduate students, he couldn’t find a job. He couldn’t find a job because he hasn’t demonstrated his ability to do physics research. Well, more disturbingly, he hasn’t demonstrated that he really understood the foundations of quantum mechanics well – even though he claimed those to be the focus of his research. He was a subpar graduate student and *therefore* he didn’t continue as a researcher. The history as presented by the bloke is completely upside down. He suggests that more famous physicists start as famous ones and aren’t constrained by the stigma of the graduate school, and therefore their ideas are more influential. This is silly, of course.

I am sorry but in the modern era, almost every physicist is a graduate student at an early stage of his career. But some other physicists do something insightful enough so that it impresses their colleagues and these ex-students are offered postdoc jobs and other jobs. Hugh Everett didn’t because the stuff he wrote was mostly interesting for pop science writers (especially a few decades after he left research), not scientists.

But another reason of resistance must be the near-existential-crisis caused by the idea that there are many versions of ourselves.

There is nothing «existential» about that hypothesis. It’s a hypothesis just like any other. When one scientifically tests the idea, he finds out that it disagrees with the observations – even the most elementary features of any observations.

Many worlds is somewhat mainstream and supported by mathematics of quantum mechanics.

I don’t want to repeat that all the believers in this stuff are clueless. But less personally, it’s just a plain lie that the «many worlds» are supported by mathematics of quantum mechanics. They have no overlap and, when looked carefully, sharply contradict one another. Despite any wishful thinking about the «details», «many worlds» can’t even agree with such an elementary fact about the mathematics of quantum mechanics as the continuity of the probability amplitudes (and, therefore, the resulting probabilities).

Unlike Copenhagen, «many worlds» are deterministic.

This is another bizarre statement. The outcomes of quantum experiments are *observed* to be random and can’t be deterministically predicted. So any deterministic theory is just wrong. You are free to say that all the wrong outcomes «also exist» but if you want to construct anything that has at least a *tiny chance* to become a scientific theory, you must say something about *what we actually observe*. And we just observe the random outcomes of the quantum experiments whose probabilities seem to be predictable. So *any* viable theory must say something about the prediction of these probabilities.

If there are many copies of yours in «many worlds», you would still need to say something about the probability that you find yourself to be one copy or another. At the end, you need to add the whole quantum mechanics – the whole «Copenhagen interpretation» – on top of your «many worlds» fairy-tale, anyway, because your fairy-tale has contributed *zero* to the actual purpose of any physical theory which is to predict the probabilities.

«Many worlds» observe the apparent randomness of quantum mechanics by an observer bias.

Holy cow. This statement is absolutely vacuous. «A bias» is a negative concept. In this case, the word «bias» is used as a fog emitted in order to obscure a strikingly obvious contradiction between the observations and the «many worlds» fairy-tale. The observations show that the *a priori* equally possible outcomes don’t end up equally real – we observe one outcome but not the others – while the «many worlds» fairy-tale implies that all of them are equally real.

It’s a contradiction and for the cultists to overlook this contradiction, they introduce «a bias». But «a bias» is just a way to say «do not trust the prediction of the many worlds that all outcomes are equally real and equally likely». You *should* trust it because it’s clearly a (wrong) prediction of the «many worlds». But even if you fool yourself into thinking that you may «undo» this (wrong) prediction of the «many worlds», you’re still extremely far from having a physical theory. A viable physical theory should actually predict the probabilities. It should make lots of *positive* and *specific* statements.

In your language, it means that it should precisely quantify the «observer bias». None of the «many worlds» cultists has anything to say about the bias, its origin, or its exact direction or magnitude. The whole «fairy tale» has done *zero* to make predictions possible. The «many worlds» fairy-tale has just postponed the construction of a theory – 100% of the problem has been shifted to a description of the «observer bias», a task that hasn’t been solved at all.

You may (and you really have to) solve this task by the laws of the proper quantum mechanics – dismissed by these cultists as the «Copenhagen interpretation». You really need it to quantify the «observer bias». Assuming that you want to talk about serious physical problems and their description at all, the word «observer bias» becomes just another silly pop-science synonym of «probabilities of observable outcomes». Unless your theory is self-evidently hopeless, you will have changed *nothing at all* about the beef of the «Copenhagen interpretation». You still need to add a theory predicting those, the only viable theory doing so is the «Copenhagen interpretation», and when you think at least a little bit rationally, you will see that all the layers of the «many worlds» are just silly pseudoscientific fairy-tales that contribute *nothing* to your ability to predict, explain, or understand any phenomena in Nature, and this stupid layer should be cut away by Occam’s razor.

The bloke also says a couple of confused sentences about the «free will». In a deterministic Universe, there would clearly be no free will in principle. You can’t «choose your own adventure» in the «many worlds» set if the deterministic Universe is everything that exists. If you add a «soul» on top of the histories with «many worlds», the soul may perhaps «choose» where it wants to enjoy its «adventures» (real «souls» don’t really choose the outcomes, they only choose the questions; Nature chooses the outcomes for them). But once you admit that there is one history preferred and perceived by the «soul», you just admit that this history is more physical than all the others. It’s really the only one you need in physics. The whole system with the alternative «many worlds» history is just a useless addition that contributes nothing to physics or science. And it really contributes negatively because if you look carefully, you will realize that there’s no consistent way to define what the «other worlds» should look like in general.

Matt talks to a Dianna from a Physics Girl show and gets a homework exercise to experimentally prove the roundness of the Earth. She was asked to explain five heavily jargony words in basic English (1,000 top words).

Again, the fight against these pseudoscientific delusions about quantum mechanics seems to be a last cause. Most people *are* incredibly stupid and this kind of idiocy – idiocy of the people believing that the words «quantum mechanics» actually mean «different crackpots’ strategies to distort and deny quantum mechanics» – has been spreading exponentially and only a small set of thousands of physics experts in the world understands what quantum mechanics actually says and why it cannot say just the opposite. The additional millions of readers of popular books etc. are being completely brainwashed, even when it comes to the most fundamental questions.

Given the influence of the public over the funding, I wonder whether the people who want to understand and do understand the *actual* laws of quantum mechanics – or Nature – will be «allowed to exist» in 10 or 20 years from now or whether the physics community will be degraded into an exhibition by the people who don’t really know what they’re talking about and who have to constantly pay lip service to the pseudoscientific beliefs of the majority.