Quantum Paradoxes

by Luke Muehlhauser on September 24, 2010 in Science

drescher good and real smallI’m blogging through Good and Real by Gary Drescher, perhaps the best book on naturalism I’ve read yet. (See the series index.)

So far I’ve summarized Drescher’s views on the mind, consiousness, and the nature of time. Now we turn to quantum mechanics.

The mechanistic, clockwork universe of 19th century physics had no need of a divine artificer and was thus radical in its own way, and yet its simple determinism was comprehensible and comforting. Even Einstein’s reformulation of space and time could be accepted as the evidence mounted. But quantum mechanics presents a universe that is, frankly, insane.

Not only is the quantum world indeterministic and probabilistic, it has a spooky, observer-dependent nature. For example:

Of several states that a particle might be in, it turns out that all may co-exist. It is as though there are several simultaneous versions of the particle, each in a different state… Bizarrely, however, whenever we observe the particle in this so-called superposition of states, we see just one version in just one of the previously co-existing states. And thereafter, the previous superposition vanishes, with no further trace of the other superposed states, as though they had never been present in the first place. (Which of the states we observe is unpredictable in principle – hence the apparent nondeterminism.)

At least, that is the usual interpretation of our observations in quantum mechanics.

In 1957, Hugh Everett proposed an alternative view. He thought the mathematics of quantum mechanics did describe

a straightforwardly deterministic system after all. There are quantum superpositions, but we observers are among the physical objects that can be in a superposition of states. Each such superposition pairs a particular state of the observed object with the corresponding observer-state, so each superposed version of the observer observes only one of the superposed states of the object, even though the superposition persists. Because there are multiple superposed versions of the observer too – each observing just one superposed state – the fact that (each version of) an observer sees only one such state does not imply that the superposition has (randomly) collapsed to just one such state.

The very concept of a conscious observer in a superposition of states – some versions observing one thing, other versions observing another – is anathema to the view of consciousness as a unitary, transcendent, extraphysical, extramechanical phenomenon. But if consciousness is just an ordinary physical process, its superposition is no stranger than that of any other physical process…

Thus, the common argument [that the "observer dependency" of particle states shows that consciousness has special powers] gets things exactly backward. When a particle that had been in a quantum superposition is observed, the seeming collapse of the superposition into a single random state stems from failing to conceive of the observer herself as a physical object subject to quantum superposition. This failure smuggles the notion of nonmechanical consciousness into the very interpretation of quantum mechanics. If the smuggling goes unnoticed, it creates the false impression that quantum mechanics itself provides independent evidence for nonmechanical consciousness…

Commitment to a nonmechanical concept of mind distorts our intepretation of physics; thus distorted, physics seems to support a nonmechanical concept of mind.

Wave-particle duality

Consider also the wave-particle duality revealed by double-slit experiments. The double-slit experiment is well-illustrated by this clip from an otherwise bullshit movie:

This is all very strange:

According to the Copenhagen interpretation, no physical phenomenon is real until it has been observed. Nothing real passes through both slits of the apparatus. Instead, there is a potential for a real particle to pass through either slit, but that potential is not realized unless the passing-through is observed; at that point, the particle settles, at random, into one of its potential positions. The potential itself is wave-like, exhibiting interference effects.

…Thus, quantum mechanics seems to challenge not only the world’s determinism, but the very objectivity of the world’s existence. Indeed, the Copenhagen interpretation provides no way to express the state of the universe as a whole, since a system’s state is real only with respect to an external observer, and the universe as a whole has no external observer.

But let us return again to Hugh Everett’s interpretation. The Copenhagen interpretation says that none of the yet-unobserved states of a particle are yet real. Everett’s interpretation says that all of them are real.

In Everett’s formulation… superposed states remain in superposition even after observation (whether by inanimate objects or by conscious observers)…

…observing a superposed state results in different versions of the observer in different versions of the universe, each version of the observer seeing a different outcome to the exclusion of all other outcomes. (Of course, it makes no difference whether the observe is animate [or, say, a recording device].) Thus, versions of the observers themselves are in superposition. But they are mutually isolated, so each sees a seemingly unique outcome. Following Everett, I argue here that this interpretation (which may sound desperately implausible on its face) is in fact the far more parsimonious one – but it takes a formal model to demonstrate that claim.

Drescher them embarks on an ambitious question to formally illustrate why Everett’s interpretation is superior the the Copenhagen interpretation, but in a way that laymen can understand. This is no small feat, and may be one of the best reasons to buy the book. In any case, I will not spoil his illustration by summarizing it here. :)

The upside is this: If Everett’s interpretation is correct, the universe is indeed deterministic, there is no “quantum collapse” dependent on conscious observers, and quantum non-locality is explained.

In the next chapter, Drescher explains how choice works in a deterministic universe.

Previous post:

Next post:

{ 23 comments… read them below or add one }

thepowerofmeow September 24, 2010 at 6:29 am

I suppose what I do not understand is why, if quantum occurrences are still determined in a more traditional way, we are able to perceive particles in superposition? Shouldn’t the superpositions of our consciousness match up with the superpositions of particles in a way to where we are unaware of the other positions?

I think the idea that probability states are fundamental to reality still seems more likely to me – but I am no expert!


Reginald Selkirk September 24, 2010 at 6:38 am

There are quantum superpositions, but we observers are among the physical objects that can be in a superposition of states.

Except that uncertainty is inversely related to mass. A human being is much much larger than an elementary particle, and thus has a much much smaller uncertainty.

Thus, the common argument [that the "observer dependency" of particle states shows that consciousness has special powers] gets things exactly backward.

No, that is wrong for quite another reason, as covered later in your post:

(Of course, it makes no difference whether the observe is animate [or, say, a recording device].)

The mistake is to think that “observer” implies “conscious human being.” The basic error seems to be based on a simple confusion of different definitions of “observer.” The word “conscious” is simply not a part of the Copenhagen interpretation, and never was.
To quote John Wheeler, who is often, and perhaps unjustly, cited as a prominent source of this “conscious observers are special” idea, “The process whereby the macroscopic world reacts to a quantum event—the process that makes reality—can, in my view, be accomplished with inanimate matter. Following Niels Bohr, I like to call this process “registration” rather than observation (which too strongly suggests human involvement).


Reginald Selkirk September 24, 2010 at 6:50 am

In other words, if Everett and/or Drescher have an alternative to the Copenhagen interpretation, they should put it up against the real thing, not a strawman.


Bill Maher September 24, 2010 at 8:32 am

Tegmark has a cool article on the Many Worlds Interpretation. I have also heard some neat stuff by Krauss and Weinberg on the subject.


lukeprog September 24, 2010 at 8:59 am


There’s no straw man here. A great many people have extended Copenhagen to say that it’s a conscious observer that leads to wave form collapse. Where I mention that theory, that is who I’m responding to. Where Drescher and I make points about Copenhagen in general, then Copenhagen in general is what we’re responding to.


Dan Nelson September 24, 2010 at 10:32 am

This post as evidence just increases my probability estimate that reading this blog is a relatively time efficient means of acquiring an understanding of concepts relevant to the nature of the reality we find ourselves in. You are interested in these topics, read about them, understand them, and then write about them in a style that is committed to the ease of understanding by others above the signaling of intelligence to others so often accomplished by others through writing that is more difficult to understand, AND you happen to be good at writing in this easy to understand style whether by practice or natural talent or some combination thereof. Congratulations. :)


lukeprog September 24, 2010 at 10:53 am


Thanks! That is indeed what I’m trying to do. I hope you also enjoy my upcoming series on special relativity and the philosophy of time.


Chip September 24, 2010 at 11:15 am

“Consciousness Causes Collapse” (CCC) has won the support of some real giants in the history of quantum mechanics (e.g. Eugene Wigner). This mystifies me to no end. It’s not the act of observation that causes “collapse”, but the physical interaction which is necessary for all observation — how could it be plainer? I can’t for the life of me understand the appeal of this interpretation of QM.

I used to scorn Everett’s Many-Worlds Interpretation (MWI). Now that I understand it a bit better, it doesn’t seem nearly so objectionable. I merely disagree with it (and only weakly, at that). All my scorn is reserved for CCC.

Lately, I’ve begun to wonder whether MWI and Consistent Histories (CH, or “Copenhagen done right”) are really saying basically the same thing, disagreeing only on the ontology of the “extra worlds”. I wonder whether CH just is MWI, when you want to ask questions about the world that contains this version of you.

Oh, and FWIW, I’m officially a physicist now. :) (Defended last Tuesday.)


Chip September 24, 2010 at 11:16 am

Forgot to mention:

“The double-slit experiment is well-illustrated by this clip from an otherwise bullshit movie:”

Nailed it with this quote =)


JS Allen September 24, 2010 at 11:32 am

This was a good chapter. It inspired me to brush up on the competing theories, and even code up my own “quantish” simulation. IIRC, Drescher didn’t talk about “Quantum Suicide Machine”, which is an interesting paradox with MWI; but I like the theory.

As far as I can tell, MWI would throw a spanner in the works for my theory about being able to use the current state of the physical universe to reconstruct previous states. Perhaps it’s still possible, I don’t know. But if it was, it would require vastly more computational power.


ildi September 24, 2010 at 12:42 pm

Congrats, Chip!


lukeprog September 24, 2010 at 12:58 pm

Congrats, Chip!


Al Moritz September 24, 2010 at 3:04 pm

Congratulations, Chip!


I liked the previous chapter 3 of the book a lot and found myself in full agreement. It was marvelously argued and well explained. Chapter 4, however, was more of a headache, conceptually. It was well explained though, but I believe it was above the level of many general readers, even though Drescher seems to think otherwise.

When Drescher says on p. 170 that the apparent nondeterminism results from postulating the collapse, this is plainly false and gets things precisely backwards. The apparent nondeterminism is the result of experimental measurement, postulating the collapse is just a (mistaken or not) way of trying to make sense of the hard experimental facts. On the other hand, postulating determinism goes beyond the direct experimental evidence of the individual experiment and seeks support in a particular interpretation of the experimental evidence of the interference patters emerging from many subsequent measurements. Yet this interpretation requires invoking a splitting of worlds that we cannot experimentally observe.

Also, contrary to what Drescher wants us to believe, the many-worlds interpretation raises more questions than it gives answers. How does the universe split into a different spacetime so that it does not interfere with our universe? When a universe splits, does the split-off universe show the same characteristics as ours? If so, then a newly split universe would have a fake history and a fake big bang, fake because it expands and thus requires a big bang in the distant past, but that one never happened since it is a newly generated (split) universe (or am I wrong about this?). Interestingly, the many-worlds hypothesis is not a standard part of the multiverse of chaotic inflation, even though that one also postulates many worlds. In that multiverse, were it to exist, all big bangs would be real.


Chip September 24, 2010 at 6:12 pm

Thanks for the well-wishes, everyone :)

Al: your objections are basically why I was so hostile towards MWI for a long time. But according to my current understanding, it’s much less objectionable.

Basically (if I have it right!), it all comes down to superposition. First, consider the observer as another quantum mechanical system (as we must!). Consider the total wavefunction: system-plus-observer. Everett says (and I’ve no reason to doubt he’s got the math right) that as the system and observer interact, that total wavefunction splits into branches. There’s one branch for each possible outcome, and the branches are weighted according to the standard probabilities calculated from quantum mechanics.

That’s where superposition comes in. If each branch is a solution of the Schrodinger equation, then the superposition of branches is also a solution, and both (all?) will occur simultaneously. Since the Schrodinger equation is linear, the branches won’t interfere. So, “many worlds” is a bit of a misnomer; really only one world is postulated, and the total wavefunction of the universe evolves according to the Schrodinger equation. Perhaps “many histories” would be clearer?

(Keep in mind: the “observer” doesn’t need to be a conscious being, and it’s clearer if we imagine it to be some very simple system.)


Yair September 25, 2010 at 1:15 am

Disappointing, in that I was expecting a good defense of hidden variables, not many worlds (which I agree with anyways). Is there any leading naturalist that doesn’t believe in the MWI nowadays? :sigh:

Congrads, Chip! And I do believe CH and MWI are one and the same. Indeed, I vaguely remember an article on CH by one of its main developers saying as much.


lukeprog September 25, 2010 at 2:47 am

Yair and Chip,

BTW, you could read section 4.3.4 of Drescher’s book an onward, entitled: “Disproving Hidden-Variable Theories: The EPR Experiment.”


G'DIsraeli September 25, 2010 at 5:31 am

“and the universe as a whole has no external observer.” OMFG. But it does, its Yahweh!
Jesus, this book is interesting. Thank you Luke for giver us an easier time with it.


Alexandros Marinos September 25, 2010 at 10:57 am

On the lighter side of things, if consciousness causes collapse, and there are cases where we can observe a superposition, then obviously god is either non-conscious or non-omnipresent :)


lukeprog September 25, 2010 at 11:02 am




Baal September 25, 2010 at 10:21 pm

Not a problem for for the Vedantic conception of God as Nirguna Brahman, ultimately beyond name, form, and attribute, which

By the power of Maya (illusion) the supreme lord (Ishwara) playfully creates multiple worlds and deludes all beings, who are in essence non-different from Him.

Shankara even discovered the MWI. ;)


Yair September 28, 2010 at 2:25 am


I’ll find reading the chapter hard, since I don’t have the book ;) To the best of my knowledge, EPR doesn’t disprove hidden variables. See, for example, Aspect, who as the one who conducted the EPR experiments is somewhat of an authority on these issues:


It is then natural to raise the question of whether one should drop locality — which equates to the impossibility of any influence travelling faster than light — or rather drop the notion of physical reality.

There is no logical answer to that question: one can choose to abandon either concept, or even both.

I and any physicist that I know prefers to drop (local) realism. But this is an aesthetic and philosophical, not scientific, preference. It is critical to keep that in mind.


Robert Oerter October 1, 2010 at 10:15 am

As a naturalist who finds MWI complete metaphysical BS, I suppose I better read Drescher.

Several times in the past weeks I have run across the claim that QM does not require indeterminism. They all seem to refer to MWI as the reason (tho some cite Bohm’s version of QM too). But it seems to me that MWI doesn’t yield determinism in any relevant PHILOSOPHICAL sense. For example, MWI is the ultimate “could have done otherwise” theory: for any action I take, I could have done otherwise, and, in fact, did do otherwise in different branches of the universal wave function.

This point seems to be universally ignored by philosophers.


thepowerofmeow October 1, 2010 at 11:03 am


Thanks for pointing this out. I have thought about it many times as well. Saying that MWI is deterministic is funny – since it admits that all possible realities happen in order to maintain that position. Cake is so much better when you can eat it too!


Leave a Comment