QM says no observer, no existence

[honestrosewater]

What should I say to people who claim that QM supports the view that the universe couldn't exist without humans or some kind of "observer"? Really?? I see where this comes from, but what is the truth? I'm not talking about getting into an ontological argument- just what QM really has to say on the subject. I see this so often and want to ask them how they happen to know what QM says, but I would rather know what you nice people think QM says so I can pass it along.
Many thanks.

 

[vanesch]

What should I say to people who claim that QM supports the view that the universe couldn't exist without humans or some kind of "observer"? Really?? I see where this comes from, but what is the truth?

Nobody knows "what is the truth"

The difference between other (former) physical theories and quantum theory is that in quantum theory, the observer plays a central role, while that was much less the case in others. However, will quantum theory remain with us in more or less its present form or not ? Nobody knows. When to give what ontological status to what in quantum theory ? One can have different preferences.
So I think the honest answer to your question is that there is no "truth" to the above statement, in the sense that 1) maybe QM is not the last word and 2) even if it is, this is not required. It can be a point of view which is up to a point defendable, in the same way as the opposite point is up to a certain point, defendable.
It is my opinion that anyone giving an answer yes/no to the above question is just venting his own opinion, but no *requirement* of quantum theory.

cheers,
Patrick.

 

[honestrosewater]

Okay, but what is an "observer" in the theory? Is it a human? A conscious being? Any old measurement device? An abstract event?
I'm talking about when people use QM to support their (philosophical) views. So I am interested in what QM actually supports, not how good of a model QM is. You're saying there are several possible interpretations of what the theory says?

Maybe an example would help: The universe must have been created by an intelligent being because QM says there must be an observer for anything to happen.

Is that an appropriate interpretation of what QM says?

 

[vanesch]

Okay, but what is an "observer" in the theory? Is it a human? A conscious being? Any old measurement device? An abstract event?
I'm talking about when people use QM to support their (philosophical) views. So I am interested in the truth about what QM actually supports, not how good of a model QM is.

Ok, you want my personal view ? Here it goes. I'm a heretic MWI-er (many world interpretation). Heretic, because I'm convinced that you cannot do QM without the Born rule (which MWI-ers somehow try to extract out of unitary QM). But who says Born rule, says observer, preferred basis and all that.
So there are 2 parts to "the world": one is the ontological, physical world, and its physics is completely described by unitary QM (that's the MWI part). So yes, the universe exists without it being observed.
The other part is the consciousness, the subjective world. As you know from solipsism, it doesn't make sense to talk about "other" consciousnesses; only yours counts. The psycho-physical hypothesis associates subjectiveness to a physical object, and that's exactly what you need to apply the Born rule: the Born rule gives you the transition between the objective world (physical world) and the subjective world of the consciousness, which is associated to a subsystem (say, a brain). It describes the subjective experiences that that consciousness undergoes (as random processes dictated by the Born rule) due to the objective world.
This also implies that your subjective experience corresponds only to ONE of an infinity of objective facets of the world.

cheers,
Patrick.

EDIT: I would like to stress that this is just my personal view (which I don't believe to be true :-). It is not universally accepted. There are other views on QM, but which I personally don't like for reasons which only have to do with my personal preferences

 

[slyboy]

I think the answer is that quantum mechanics, in its weakest possible form, makes no claims one way or the other about this. It is just an algorithm, or an operational theory if you like, for translating statements about experimental setups into probability assignments to outcomes of different measurements. As such, it makes no claims about the structure of any underlying reality, or about the necessity of observervers to that underlying reality.

Now, clearly we could think about other theories in the same terms if we were inclined to (and perhaps the logical positivists would prefer us to do so). For example, we could say that themodynamics was just about relating statements about thermometer readings to readings on pressure guages and volume measurements (ignoring the fact that thermodynaics can also be used to describe other systems like magnets for the moment). We could deny that this derives from an underlying reality in which molecules fly around, interact and collide with each other. However, the latter picture is compelling because we can see statistical fluctuations from thermodynamic equilibrium, it allows us to construct statistical mechanics and to study physics far from equilibrium.

Clearly, it would also be preferable to translate the operational newspeak of the quantum formalism into a more realistic picture of the world. The problem is that we have many ways of doing this and none of them are particularly compelling. There is no evidence, like the statistical fluctuations from thermal equilibrium, that pushes us to take one picutre more seriously than any of the others.

However, some of these ways of constructing this reality do not give observers any special role, e.g. Bohmian mechanics, spontaneous collapse, many-worlds, etc. Therefore, to say that QM necessarily gives a special role to the observer is definitely wrong.

 

[vanesch]

Maybe an example would help: The universe must have been created by an intelligent being because QM says there must be an observer for anything to happen.

Is that an appropriate interpretation of what QM says?

Absolutely not. It is even difficult to defend, because the observer must be associated with a physical structure.

 

[Crosson]

Maybe an example would help: The universe must have been created by an intelligent being because QM says there must be an observer for anything to happen.

Is that an appropriate interpretation of what QM says?

The bold is where the argument totally falls apart. The quantum mechanical "wave function" will continue to evolve through time regardless of observations. That is, "things happen", even when no one is looking.

 

[zefram_c]

To expand on the above comment, QM is more than capable to describe the time evolution of quantum objects, including interactions between objects or with externally imposed potential fields. In fact, it is actually deterministic in this manner: the wave function at any future time can be calculated, in theory, given the appropiate starting conditions. It is only at the transition between QM objects and macroscopic, classical objects that the conceptual difficulties arise: if we could show that the interaction between a quantum object and a macroscopic measuring apparatus causes the wave function of the quantum object to change to a sharply localized gaussian (in whatever parameter space one is 'measuring'), the problem would be solved and we would have an entirely self-consistent, pure quantum description of the measurement process. Unfortunately it is not possible to write down the wave function of the measuring apparatus, since it consists of O(10^23) particles. My personal view is that there is a pure quantum solution to the measurement problem along the above lines, but we may never know it.

 

[Les Sleeth]

However, some of these ways of constructing this reality do not give observers any special role, e.g. Bohmian mechanics, spontaneous collapse, many-worlds, etc. Therefore, to say that QM necessarily gives a special role to the observer is definitely wrong.

Nice post.

I've been sort of the philosophical iconoclast for thinking that wave collapse when consciousness observes is simply due to the intrusion of the machinery we use to observe. The truth is, consciousness alone cannot observe anything at the quantum level, and to me that means what is most suspect for the collapse is the technology between consciousness and the photon, and not anything to do with consciousness itself.

 

[Nereid]

And back to 'the observer' if I may? honestrosewater wrote:

Okay, but what is an "observer" in the theory? Is it a human? A conscious being? Any old measurement device? An abstract event?

In the case of the poor old cat, does it die (or live) only when a human views a videotape of what went on in the box, 10 million years later? Or is the cat 'an observer' (so the dead/alive superposition can never arise in the first place)?

 

[Picklehead]

IMO, the cat must be an observer, or else you tempt the question of who collapses the superposition of 'observer witnessing live cat/observer witnessing dead cat' . . . assuming we haven't misdefined 'observer' and that the whole concept of waveform collapse is correct in the first place.

 

[vanesch]

if we could show that the interaction between a quantum object and a macroscopic measuring apparatus causes the wave function of the quantum object to change to a sharply localized gaussian (in whatever parameter space one is 'measuring'), the problem would be solved and we would have an entirely self-consistent, pure quantum description of the measurement process. Unfortunately it is not possible to write down the wave function of the measuring apparatus, since it consists of O(10^23) particles.

Yes, we can: look: |psi>

There are 2 fundamental problems with the view that "well, a complicated measurement apparatus probably effectuates a collapse, but we can't show it".
The first problem is that, no matter how complicated, the time evolution (which you correctly identify as being deterministic) is given by a unitary operator, which can never give rise to a collapse (it being an invertible, linear operator). But this problem could be fixed by saying that we don't have an absolutely exactly unitary time evolution operator (nevertheless, this is a very basic postulate of QM, so we are changing QM fundamentally here).
The second problem is with locality. EPR situations show that, if collapse is an ontologically happening process, it is non-local and (hence) not Lorentz invariant.

There are 3 ways out:
1) the positivist one: there is no "reality", QM just gives us rules that give us relations between observed probabilities. There is a variant: Copenhagen, which denies reality to objects that are ruled by QM, but gives reality to "classical" objects, without specifying where there is a boundary.
2) MWI - style: (my preferred view) we keep all of QM and SR, and we claim that the full wave function has ontological status, while the Born rule tells us how subjective experiences are derived from it: collapse is not an objective phenomenon, but a perceived, subjective experience.
3) the "new physics" realist one: what we observe corresponds to an ontological reality, and we then throw overboard as well QM and the superposition principle, as locality (and hence SR).

cheers,
Patrick

 

[vanesch]

And back to 'the observer' if I may? honestrosewater wrote: In the case of the poor old cat, does it die (or live) only when a human views a videotape of what went on in the box, 10 million years later? Or is the cat 'an observer' (so the dead/alive superposition can never arise in the first place)?

The MWI view tells you that there is as well a dead cat as a living one, and your observation of a tape or whatever just places you in the case that corresponds to observation.
From the cat's point of view, something similar happens: its consciousness is made to choose between a dead body or a living body to be associated with. From that point on, several different flavors arise: there are those that say that the consciousness can never be associated with a dead body, so every consciousness will always be "lucky and live" for ever (although it will never observe OTHERS to be "lucky and live for ever" because that's highly improbable according to the Born rule). There are those that say that there will be a consciousness associated with each of the new "living" states (but this is not necessarily the continuation of the consciousness that was there before: in which case, it dies). You can also say that there is only one consciousness: yours (solipsist viewpoint), and QM only has to explain YOUR subjective experiences.
But all this speculation is secondary in fact. The important point, shared by all MWI views (which, let us recall, only state that quantum theory is applicable to all of the universe, and that there is no transition to a classical world) is of course that the wave function evolves deterministically and unitarily and is not affected by any "measurement" more than the simple interaction of the measurement apparatus (including the body of the observer) with whatever it observes, as just any other interaction (described by deterministic, unitary evolution). A "measurement" (Born rule + projection postulate) just extracts information from the objective wavefunction into the subjective world of experiences.

cheers,
Patrick.

 

[Tournesol]

The difference between other (former) physical theories and quantum theory is that in quantum theory, the observer plays a central role, while that was much less the case in others.

observation plays a central role.
There is nothing in the formalism you can point at and say "that is an observer"

 

[vanesch]

observation plays a central role.
There is nothing in the formalism you can point at and say "that is an observer"

Yes, and that is the main "difficulty". If the formalism did point out to "that is an observer" and "that is not" (meaning: in the first case, projection and the Born rule applies, in the second: unitary evolution applies), observation wouldn't be a central issue: you would just have different rules applying to well-defined different cases ("that's a conductor, and that's an insulator").
Copenhagen tried to do so, by distinguishing between "macroscopic" and "microscopic" but you run into a lot of inconsistencies that way.
So, if I say that "observation" is whatever an "observer" does, we are in agreement, right ?

cheers,
Patrick.

 

[selfAdjoint]

Let's just raise a philosophical point here. Physicsists assert strongly that the uncertainty principle is not an epistomological result but a deep ontological principle; that particles truly can't manifest both momentum and position at the same spacetime event. But if quantum reality is totally observer dependent, what warrant have quantum physicists to say this? Isn't everything their science predicts ultimately epistomological?

 

[Tournesol]

Originally Posted by Tournesol
observation plays a central role.
There is nothing in the formalism you can point at and say "that is an observer"

Yes, and that is the main "difficulty". If the formalism did point out to "that is an observer" and "that is not" (meaning: in the first case, projection and the Born rule applies, in the second: unitary evolution applies), observation wouldn't be a central issue: you would just have different rules applying to well-defined different cases ("that's a conductor, and that's an insulator").
Copenhagen tried to do so, by distinguishing between "macroscopic" and "microscopic" but you run into a lot of inconsistencies that way.
So, if I say that "observation" is whatever an "observer" does, we are in agreement, right ?

cheers,
Patrick.

No, no. There is still a very big issue to do with observations. It just
doesn't help to ascribe magical properties to "observers". I would
rather say an observer is whoever is doing the observation -- at least
there is something we can point to in the formalism and say "that is an
obervation".

 

[Tournesol]

Let's just raise a philosophical point here. Physicsists assert strongly that the uncertainty principle is not an epistomological result but a deep ontological principle; that particles truly can't manifest both momentum and position at the same spacetime event. But if quantum reality is totally observer dependent, what warrant have quantum physicists to say this? Isn't everything their science predicts ultimately epistomological?

No, ontological realism is the default position in science. The only sense
in which anything is "observer dependent" is in the sense that how
you set up your apparatus will affect the kind of result you get.
Which is true of everything, not just QM *. And which is really
about observations not observers.

*OK it works in a weird way in QM, particularly with regard to time**.

** oh gaawd, now someone is going to say: "but time is subjective..."

 

[honestrosewater]

Thanks for the thoughtful replies; It will take a while for me to really make sense of them all. It isn't at all what I was expecting. I was expecting someone to say that QM talks about particles and such and that intelligence, consciousness, and human behavior are way outside of its domain. I was also expecting that QM would say things that were, you know, testable and falsifiable.
How does a physical theory model intelligence, consciousness, and human behavior at the subatomic level? What experiments have been performed that measure intelligence, consciousness, and human behavior at the subatomic level? Since when does physics study these things?

I thought the biggest problem with my example (which I have seen before), "The universe must have been created by an intelligent being because QM says there must be an observer for anything to happen." was that QM wouldn't say anything about an intelligent being, which they have taken "observer" to mean.

 

[honestrosewater]

I thought my question was just like asking whether classical logic says that everything in the world either happens or doesn't happen but not both. I already know the answer to that: No, logic says no such thing- period. I thought my question was just as straightforward; I wasn't looking for or expecting speculation. (How) Can I ask this question so it doesn't get moved to philosophy?

 

[Nereid]

I thought the biggest problem with my example (which I have seen before), "The universe must have been created by an intelligent being because QM says there must be an observer for anything to happen." was that QM wouldn't say anything about an intelligent being, which they have taken "observer" to mean.

But aren't the folk you're having a discussion with a little off base? I mean, why pick on quantum physics? Why not Newtonian physics (do apples fall only because there's a Newton to conclude 'gravity!')? or fluid dynamics (does convection in the Earth's core require dolphins and other critter who 'understand' before it can begin?)?

Reminds me of something about a tree falling ...

 

[honestrosewater]

But aren't the folk you're having a discussion with a little off base? I mean, why pick on quantum physics? Why not Newtonian physics (do apples fall only because there's a Newton to conclude 'gravity!')? or fluid dynamics (does convection in the Earth's core require dolphins and other critter who 'understand' before it can begin?)?

Reminds me of something about a tree falling ...

If I did have a discussion about it, I could only question their information, because I don't know what QM says or what actually counts as an observer. That's what I'm trying to find out. Does QM itself actually say anything about intelligence or consciousness? Knowing this would cover many situations. I can't even begin to imagine how QM could say anything about intelligence or consciousness.
And the idea that an observer does need to be intelligent or conscious is precisely the argument many make.

 

[El Hombre Invisible]

How does a physical theory model intelligence, consciousness, and human behavior at the subatomic level? What experiments have been performed that measure intelligence, consciousness, and human behavior at the subatomic level? Since when does physics study these things?

These are emergent properties, and as such cannot be measured at a subatomic level. However, the role-players and button-pushers of such phenomena can be studied at this level. But haven't been by me, I'm afraid. Roger Penrose wrote a book, Shadows Of The Mind, about quantum mechanics within the brain. Haven't read that either. Oh, what the --- am I doing here?

 

[Nereid]

If I did have a discussion about it, I could only question their information, because I don't know what QM says or what actually counts as an observer. That's what I'm trying to find out. Does QM itself actually say anything about intelligence or consciousness? Knowing this would cover many situations. I can't even begin to imagine how QM could say anything about intelligence or consciousness.
And the idea that an observer does need to be intelligent or conscious is precisely the argument many make.

OK, so in what sense is any 'natural' quantum phenomenon different from any natural 'non-quantum' one?

I mean, sure you may need a jolly good understanding of something like quantum physics to build the latest and greatest PC which Dell has just released (or rather Intel and whoever makes RAM these days), but there are plenty of 'natural' quantum phenomena - what powers the Sun? whence white dwarfs and neutron stars? - why are 'quantum trees' which fall unseen (say, some distant nova) any less deserving of the delicate attention of a certain 18th (17th?) century clergyman than the organic type?

And as to consciousness - well, that's a pretty big coloured finny friend if ever I saw one! I mean, from the POV of your sparring mates, but for their very own 'consciousness', how could they tell? Not just quantum weirdness, but even the least 'abstract' of abstract thoughts?

 

[Nereid]

These are emergent properties, and as such cannot be measured at a subatomic level. However, the role-players and button-pushers of such phenomena can be studied at this level. But haven't been by me, I'm afraid. Roger Penrose wrote a book, Shadows Of The Mind, about quantum mechanics within the brain. Haven't read that either. Oh, what the --- am I doing here?

I have; three gold stars to Penrose for effort, an 'F' for results. Teacher's comment: "if effort were the sole criterion for promotion, then young Roger would surely move up; however, in the balance, we recommend that he repeat this year."

 

[vanesch]

No, no. There is still a very big issue to do with observations. It just
doesn't help to ascribe magical properties to "observers". I would
rather say an observer is whoever is doing the observation -- at least
there is something we can point to in the formalism and say "that is an
obervation".

We need to ascribe magical properties to observers because there is no objective definition of "observation" in quantum theory unless you MODIFY the existing theory! And we already know that that modification is going to cost some dear principles (such as locality), so not knowing HOW to modify it, we just claim it shouldn't be modified (for the moment). But then you need to ascribe "magical" properties to observers in order to make the story fit, which, upon reflexion, is not so "magical" after all. All this of course, if one insists on having something ontological to talk about in QM. Otherwise, it is just a "theory that relates observed probabilities of observations".

The reason why you need to modify something in QM is that if QM is correct, your body ends up in a superposition of position states whenever you observe an event that has more than 1 possible outcome, the reason being that each intermediate step is an interaction of which we know (in principle) the hamiltonian description and hence the associated unitary evolution.
So you say: or we have to change something there (but we don't know how!), or somehow what I subjectively observe is derived from this objective state through the Born rule (in which case we can keep QM in its current form).
The last step, however, means that subjective experience is the only thing that works according to the born rule. You are allowed to call that "magic", but what's wrong with such a claim ? At least until we figured out if QM works in its current form or should be modified (but all experimental results point to its correctness) ?

cheers,
Patrick.

 

[honestrosewater]

Okay, I'm confused again. Can someone give an example of an observation in QM? I guess I'm familiar with the double-slit experiment. When does the observation occur there? When an electron (or whatever) interacts with the screen?

 

[El Hombre Invisible]

Yep, or when an electron scatters light when you're trying to observe which slit it goes through.

 

[El Hombre Invisible]

I have; three gold stars to Penrose for effort, an 'F' for results. Teacher's comment: "if effort were the sole criterion for promotion, then young Roger would surely move up; however, in the balance, we recommend that he repeat this year."

Ah - I have that book. Thanks god I didn't pay for it. I'll donate that to Oxfam then.

 

[honestrosewater]

Yep, or when an electron scatters light when you're trying to observe which slit it goes through.

So the "observer" is just the screen. And personifying the observer is unjustified- period. ??

I mean, I see the role of speculation. But when someone invokes, what- the most successful physical theory ever? and never mentions they are speculating, it bothers me. I don't mean that everyone does it purposefully.