Archive for July, 2012

The Portable Rainbow

7 July 2012


Under the Veil
Chapter 15 (“Explanation and Phenomena”) of Bernard d’Espagnat’s On Physics and Philosophy continues the previous chapter’s exploration of causation and explanation.

With quantum mechanics relying on observation and denying a naive realism, is an empirical explanation good enough?

For d’Espagnat there’s a need to postulate a “Veiled Reality” of which we may only be able to sneak some peeks, if at all.

Nonetheless he believes it’s there.

Prediction vs Explanation
If you measure the state of quantum particles and find a correlation-at-a-distance how do you explain it?

Quantum mechanics is a “recipe” to predict observations from initial conditions. It’s an “explanation” on the level of empirical reality.

If you need some “deeper” explanation about what’s “really” going on you might add “hidden variables” to extend the standard theory.

But then you run into problems with Bell’s Theorem and the experimental results of Aspect and others.

If you don’t have knowledge of a deeper reality, how can an empiricist justify using induction to create laws?

Just because a law summarizes certain observations on certain days, why should we think it’s universal?

Induction vs Unknowable Explanation
For d’Espagnat, a belief in the existence of a deeper reality is enough to ground our use of induction.

We may be incapable of comprehending this deeper reality, but our belief that there is one suggests a connection between empirical reality and an underlying reality.

That belief is enough for d’Espagnat to accept induction without having to justify it every time it’s used.

Even if we don’t know anything much about this deeper reality, there’s still no “logical inconsistency,” d’Espagnat says, in using its presumed existence to justify induction.

Furthermore, this deeper reality—whether “veiled” or even entirely unknowable—will not be an arbitrary reality, d’Espagnat says.

Rainbows vs Quantum Concepts
Although rainbows can’t be directly grasped and manipulated, they’re explained in classical physics.

A description of rainbows might illuminate how we speak about quantum systems.

A rainbow (including its two “bases”) will look different from different locations.

Hence, the particular rainbow someone sees is observer-dependent.

The same reliance on location is true if you set up automatic cameras.

Hence you can’t say that just because we’ve taken a picture of a rainbow that this rainbow “really” existed before that observation.

Similarly, out tendency to “reify” (seeing something as concrete and real) means we jump from an observation to assuming what was observed somehow pre-existed.

If we can argue that a rainbow doesn’t pre-exist, we should be able to argue that a quantum object doesn’t pre-exist either.

Dinosaurs vs Humans
However, surely dinosaurs existed before humans ever walked the earth. No observation was required to bring them into existence.

D’Espagnat says that dinosaur bones are like the pointers of an experimental set-up. We see something and conclude it’s real.

Though d’Espagnat says it’s real, he specifies it’s real in the realm of “empirical reality.”

However, this empirical reality is hardly an arbitrary production. Its qualities are severely constrained, and in the end observers tend to see mostly the same thing.

Explanations vs The Final Key
Classical physics can still provide us with “explanations” as long as we don’t presume they derive from a deeper reality.

D’Espagnat adds that we should not conclude that these explanations are the “final, ultimate key” to understanding the world.

D’Espagnat vs Other Views
I’ve concentrated above on d’Espagnat’s ultimate positions, but here are some examples of how he explains his disagreement with other people’s positions (real or conjectured).

D’Espagnat vs Cassirer
If you see correlations in a quantum experiment then d’Espagnat has trouble imagining Cassirer’s “logical necessity” could explain each particular observation in a sequence.

True, Cassirer could choose (or could have chosen, as he’s now dead) hidden variables, but d’Espagnat says that’s too “metaphysical” for Cassirer, and the Aspect-type experiments have refuted them anyway.

Maybe Cassirer equates “logical necessity” with a pre-existing logos, a primary notion of absolute existence.

D’Espagnat says that whole idea is something the neo-Kantians were trying to get away from, so again it doesn’t sound like Cassirer.

Nonetheless, d’Espagnat says his own position is consistent with considering the “Real” (with a capital R) to consist of such a logos.

D’Espagnat vs Carnap
Carnap says scientists should be more modest. They shouldn’t try to explain the “why” but just the “how” of phenomena.

Carnap’s position is that simply producing entities, such as Driesch’s “entelechy” as an explanation for tissue regeneration, is irrelevant as there are no “laws” connecting conditions and observations.

So what about d’Espagnat’s “Real”? Is it just a meaningless entity?

It doesn’t help us predict anything, so maybe it’s not an explanation at all.

D’Espagnat responds by saying scientists long ago were implicitly believing in the realism of a world ruled by classical physics even if explicitly they concerned themselves with just the laws of observation.

Even some realists nowadays, says d’Espagnat, acknowledge that there could be an underlying reality, not attainable through “discursive knowledge,” that nonetheless grounds our empirical reality.

Furthermore, if laws relate just to our known observations, then what happened before we made those observations?

Carnap, according to d’Espagnat, said laws could exist before such observations but the truth of the laws could not be judged.

D’Espagnat says this amounts to Carnap’s acknowledging a “human-independent reality” that has a structure we might never know.

Since quantum mechanics only predicts observations and does not “explain” underlying reasons, this implies to d’Espagnat that a “Veiled Reality” has a meaning even if we can’t explore it empirically.

But what if we imagine Carnap meant some kind of “linguistic framework” involving “nature” and “existence” that replaced the usual meaning of those terms?

In a world ruled by classical physics it makes sense to speak of “things” and their qualities makes sense.

In a world ruled by quantum physics it makes sense to speak of “sense-data” rather than “things.”

D’Espagnat says this approach works fine for making sure scientific statements are clear.

But it’s not satisfactory from the philosophical point of view.

Carnap, d’Espagnat says, is just “masking” not “eliminating” the connection we make between an explanation of observations and an explanation of what’s going on in some underlying reality.

Since a linguistic framework is “chosen by us” according to Carnap it sounds a bit arbitrary and not like a genuine explanation.


An Influential Relationship

1 July 2012

Influential Arrows

Just Causes and Side Effects
Chapter 14 (“Causality and Observational Predictability”) of Bernard d’Espagnat’s On Physics and Philosophy examines how, and if, we can use the concepts of causation and influence to explain the world.

Reality vs Observations
Taking a break from examining the “notion of reality” d’Espagnat uses this chapter to argue it’s better to predict observations than predict “things as they are.”

Animism vs Empiricism
Aristotle saw causation as related to human will, and even inanimate objects seemed to have some animistic will—as seen when a falling stone somehow desires to return to its natural resting place.

Empiricists went to the other extreme. Physical laws should just be descriptions of events and their regularities. However, what initial conditions are the “cause”? You end up with too many empirical laws.

Mathematical vs Physical Determinism
If two very close points rapidly diverge then they’re not likely to be “physically deterministic.” It’s too hard to calculate their exact paths.

“Strong objectivists” argue that science accumulates knowledge about an underlying reality, not just our experimental observations.

Some strong objectivists argue that “chaotic” behaviour is an example of indeterminism, and others argue initial conditions have to be repeated exactly for us to say deterministic laws apply.

The first approach implies imperfect observations or calculations show reality is indeterministic, but that’s strange since strong objectivists believe in an underlying reality separate from our fuzzy data.

The second approach is a problem since a strong objectivist can’t be absolutely sure the initial conditions won’t be repeated.

Laws vs Predictions
D’Espagnat also criticizes the claim we’ve seen the “end of certainties” just because some calculations make predictions impossible.

He says that’s too harsh as we can still believe in our laws even if sometimes in practice we can make reliable predictions only for the near future.

Classical vs Quantum Indeterminacy
D’Espagnat cautions against regarding chaos theory as some overriding conceptual triumph as it’s grounded on classical concepts of space and time.

Classical physics falters where quantum physics and its apparent indeterminacy excel, particularly on the microscopic level.

Yet d’Espagnat says the defining feature of quantum mechanics is not its indeterminacy but its “weak objectivity.” The theory confines itself to observations of reality, not claims about reality itself.

Individual vs Statistical Determinacy
D’Espagnat agrees with Kant that “regularity in time”—in which one kind of event is followed consistently by another—is a good way to distinguish the empirically real from, say, the events of a dream.

Kant’s “sin of omission” (understandable because of his time and place) was not to consider statistical regularities in which ensemble probabilities are deterministic.

D’Espagnat emphasizes that quantum mechanics makes reliable predictions for observing ensembles of quantum systems, but these are not probabilities of ignorance about individual systems.

At first glance quantum mechanics may seem indeterministic, but if you keep in mind quantum predictions are about observations of multiple systems then it too is deterministic—if only “statistically.”

Laws vs Facts
D’Espagnat warns against “a variant of nihilism” if you don’t pay enough attention to the difference between laws and facts.

He says even Dirac’s musings that universal constants (such as the speed of light) might change over time don’t threaten that distinction.

The nihilistic danger, d’Espagnat says, comes from sociologists, epistemologists, or “pure philosophers” who see in the history of a changing universe a fundamental lack of stability.

They fail to distinguish between laws and facts, or they fail to appreciate the significance of the distinction.

Causes vs Influences
D’Espagnat imagines a Laplace daemon that can possess total knowledge of events in part of the universe.

The fixed speed of light means, in an Einsteinian world, the daemon need only check events in a point’s past light-cone to predict that point’s future.

However, Bell’s Theorem combined with the experiments of Alain Aspect (and others) proved that the locality hypothesis is false.

Add to that the order (in time) of events can vary by reference frame, and we see that (earlier) cause and (later) effect can be ambiguous.

D’Espagnat thus suggests that faster-than-light influences—or “influential relationships”—do exist.