It’s Not All in Your Head

Not Just in Your Head

Veiled Threads
Another chapter down, three more to go in Bernard d’Espagnat’s On Physics and Philosophy.

Chapter sixteen, “Mind and Things,” is (relatively) straightforward.

Having spent much of the book undermining physical realism and its kin, he focuses next on the excesses of empiricism and idealism.

Much less combative in this chapter, d’Espagnat seems sympathetic to many of the approaches he describes.

Sympathetic, yes. In total agreement, no.

Ultimately, he’s laying the groundwork for exploring his “Veiled Reality” in some detail as the book draws to a close in the chapters to follow.

Empiricism vs Metaphysics
Empiricism’s guiding principle: all of our knowledge comes from our senses.

It started by discarding metaphysical sources of knowledge.

It then emphasized the role of “elementary sensations.”

Experience vs Reality
Early empiricists seem to have believed primary qualities were real aspects of real objects.

Even if our knowledge can’t exceed our experiences, if properly used our experiences produce a good picture of reality.

A large number of modern-day scientists hold this view, which is a kind of “physical realism.”

Empiricism vs Knowledge
Initially the Vienna Circle epistemologists more energetically attacked Kantian views than scientific realism.

Nowadays, though, logical positivism is understood in an antirealist sense.

But this creates a problem. If the empirical connection between experience and reality is questioned, then are we back to Kant or the neo-Kantians?

D’Espagnat thinks that this quandary sank the logical positivist agenda, but contemporary physics can still learn from some of their ideas.

Knowledge vs Phenomena
“Phenomenalism” has various definitions.

One version states that “knowledge is strictly limited to the (physical and mental) phenomena,” says d’Espagnat.

Phenomena are just the objects of our (unanalyzed) perceptions or introspection.

So phenomenalism is at best consistent with “open realism.”

This is the “weak” version of realism that d’Espagnat favours.

Phenomenalists vs Physical Objects
Unfortunately phenomenalists are often vague about the reality of physical objects.

The Vienna Circle positivists were suspicious of counterfactuals: “If you put this lump of sugar in water then it would dissolve.”

It’s hard to assign properties such as “soluble” to an object without using counterfactuals.

Phenomenalists vs Paths of Knowledge
Another problem was pointed out by Bertrand Russell.

Unless a solipsist, a phenomenalist accepts the existence of other observers and their assertions.

But then why not accept the existence of sound waves, since they convey messages?

But then you’re getting into back into physical realism.

Private Sensations vs Public Science
It’s hard to get around this problem since we have no direct access to other people’s sensations.

They’re private.

But science relies on communicating knowledge.

That’s public.

Object vs Method
A third problem:

We describe objects by describing how to get sense data about them.

But then as objects get smaller and smaller the description gets longer and longer.

How do you describe an electron as a “construct”?

You could describe a cloud chamber, how it has to be prepared, and then the probability that the set-up will produce the hoped-for observation.

Stability vs Instability
But that gets at a problem when you move from phenomenalism to contemporary physics.

To a phenomenalist, an object of knowledge is a stable pattern of perception.

In quantum physics there are probabilities of state vectors not some “inherent stability” of perception.

Classical Instruments vs Quantum Systems
To address Russell’s problem we can assume a measuring instrument is classical.

That way various observers can agree on a measurement.

Furthermore, our observations agree with the rules of quantum physics, so our sensations aren’t entirely private.

It’s a kind of “mutual agreement” between our perceptions and quantum rules.

However, d’Espagnat notes this solution is a “hybrid” one.

And, he notes, in chapter eight he looked at the problems with saying an experimental apparatus is classical.

Operationalism vs Phenomenalism
D’Espagant likes operationalism, a modified version of phenomenalism.

It deals with how to make observations, and quantum rules for predicting observations are unquestionably accurate.

Conventional vs Radical Operationalism
There’s a conventional version of operationalism that’s “moderate.”

It talks as if there are real properties that are measured.

But as seen in chapter seven this leads to ambiguity.

“Radical” operationalism is more content to just describe measurements.

However, that’s hard to do without specifying the objects being measured.

Intrinsic vs Convenient Elements
“Radical” operationalism may consider some perceived forms to be “elements” of empirical reality connected by empirical laws.

But unlike traditional operationalism, it doesn’t consider these forms to have intrinsic significance.

Perceived forms aren’t the “constitutive” bricks of anything.

The radical operationalist is prepared to discard one set of predictive rules for another if they work better.

And if two sets of rules produce the same predictions then we should accept both.

Deductive vs Inductive Logic
However, radical operationalism relies heavily on induction.

If a rule worked in the past then it must work in the future.

That’s not strictly logical.

Rules vs Explanations
Another problem is that people have trouble seeing these rules as a “genuine explanation.”

So do we need the notion of “cause” beyond the realm of just phenomena?

Phenomenal vs Transcendental Causation
Kantians say that causality is inherent in our a priori understanding, not in the objects themselves.

Abner Shimony notes there are many kinds of causality.

He feels this diversity undermines the universal application of this Kantian claim.

Also, cognitive science has blurred the distinction between the phenomenal and transcendental selves.

Therefore categories of understanding can hardly be limited to just the phenomenal self.

Phenomena vs the Mind
Furthermore, in the past century or two, mathematics and physics have undermined the believe that our understanding of phenomena reflected the ordering principles of the mind.

Therefore Shimony doesn’t believe that only phenomena can be the “causes” of other phenomena.

Shimony’s Causation vs d’Espagnat’s Laws
D’Espagnat focuses on predictive laws in constructing his concept of a “Veiled Reality.”

Shimony puts causality at the root of his ontology.

True, the reliability of predictive laws must have some kind of “cause.”

But d’Espagnat still says he and Shimony have very different views.

Transcendental Uniqueness vs Structure
A basically Kantian approach says a “transcendental object” is the “purely intelligible cause” of various phenomena.

Kant believes that objects exist “per se”—but only in experience.

However, he felt there must be a “cause” of these representations.

The cause will be totally unknown to us, but he still gave it a name: the “transcendental object.”

This unknowable cause is singular.

The phenomena it produces are plural.

Therefore the “transcendental object” is unique.

D’Espagnat already acknowledged (in chapter ten) similarities between Kant’s transcendental object and his own views of extended causality and “ground Reality.”

However, d’Espagnat is willing to accept “some sorts of structures” that end up “implying” our scientific laws.

That structure and that connection to our laws will still be “undecipherable.”

Individual Mind vs Mind in General
Operationalism avoids making ontological statements.

However, someone has to set up and run the experiment, and someone has to be observe the results.

So presumably either an individual “mind” or a “mind in general” exists.

Objective Laws vs Mentalist Consciousness
Jean Petitot says we can be “objective” about the laws of phenomena even if we can’t see what’s behind the phenomena.

Galilean space and time are “mental” concepts.

But these mental forms let us construct the legal rules of phenomena, which become “desubjectivized.”

So “mentalist” or “cognitive” is what’s unique to each person’s consciousness.

Things lying in space are therefore neither ontological nor mentalist.

Objectivity vs Ontology
Petitot says that space and time are the crucial notions in classical physics.

In quantum physics the crucial notion would be probability amplitudes.

But space and time seem independent of us, while probability amplitudes are very much connected to an observer.

D’Espagnat says it’s not a crucial distinction as Petitot separates objectivity from ontology.

Petitot’s approach is what d’Espagnat calls a “weak objectivity” or “intersubjectivity.”

Different observers will get the same measurements under the same conditions.

Assumptions vs Justifications
But d’Espagnat tackles Petitot’s approach on two fronts.

The first objection is that stating a rule and justifying a rule are two different things.

Stating that “reality-per-se” is unobservable creates some interesting consequences.

But the statement is basically an axiom.

Galilean physics can still be explained by the “reality of the accidents.”

The big challenge to realism was quantum physics, not Petitot’s or anyone else’s transcendental claims.

Transcendental vs Individual Subject
The second objection is that transcendentalists create a contradiction when they limit “mentalist” and “cognitive” to individual minds.

Petitot and Kant both believe a transcendental subject is impersonal.

It supposedly conveys a priori sensibility and categories of understanding not limited to an individual.

Kant said his transcendentalism differed from Berkeley’s idealism.

However, Kant’s “empirical realism” is still only empirical.

Objects of experience exist in experience.

But experience requires one or more subjects to have that experience.

D’Espagnat says a “transcendental subject” can’t eliminate the role of knowledge in our experiences.

And knowledge depends on cognition.

That knowledge is then communicated intersubjectively, taking it out of the “private realm.”

Plato vs Galileo
Galileo stressed the mathematical structure of natural laws.

Some people take this as evidence he was a “Platonist.”

Alexandre Koyré said Galilean science started from this belief:

Reason and geometry are enough to acquire “intelligence of the real.”

But Galileo took considerable pains to investigate phenomena.

For him to be a Platonist you’d have to equate what’s “empirically real” with a kind of Platonic idealism.

However, Plato’s cave suggests our pursuit of phenomena will get us only as close as some shadow of the “Real.”

Senses vs Innate Knowledge
Also ambiguous is the notion of what is “innate.”

Both Descartes and Saint Augustine believed we could gain knowledge without use of the senses.

But empiricists believe “reality-per-se” is inaccessible.

So how could we ever experience an independent reality?

Empiricism vs Innatism
If we consider Kantian space, time, and causality then these notions must be innate.

Furthermore, Kant’s categories of understanding are a priori, so they too are “innate.”

But in our “semi-intuitive” world-view we follow sensory evidence as closely as possible, yet interpretation still guides us.

Quantum Mechanics vs Innatism
Quantum mechanics is “weakly objective” hence “antirealist.”

Its view of knowledge is somewhat Kantian, but with a strong dose of operationalism added.

This operationalism prevents quantum mechanics from getting too close to Descartes’ innatism.

Furthermore, the simplicity of quantum rules leads us to infer (unprovably yet irrefutably) a simplicity in the “Real.”

D’Espagnat says this approaches Nicolas Malebranche’s “vision in God.”

Empiricism vs Conventionalism
D’Espagnat says reading between the lines one can see evidence of Henri Poincaré’s “ontological” stance.

Kant believed the axioms of geometry were a priori.

The discovery of non-Euclidean geometries refuted such an idea.

Poincaré saw the axioms as neither a priori nor as experimental data.

They are conventions, he decided.

“One geometry cannot be truer than another one, it can only be more convenient,” he wrote.

The Convenient vs the Real
Poincaré felt the same way about physics.

Experimental data and theories about them are not descriptions of an independent reality.

They are convenient and concise “pictures” to describe observations and connect them.

So, for instance, the ether hypothesis is “convenient.”

Whether or not ether exists is the concern of the metaphysician, not him.

Supporters of “objectivist realism” complain conventionalism favours convenience over truth.

However, both Poincaré and d’Espagnat reply that if rules make the right predictions then we might as well call them “true.”

Knowable vs Underlying Reality
Poincaré says relationships between things are “objective” when they’re “the same for everybody.”

Poincaré says that the relationships between things is “the sole objective reality.”

These relationships cannot be conceived independent of a mind that conceives them.

However, “they are objective nevertheless since they are shared by all thinking beings.”

Poincaré is definitely referring to real, though hidden, objects.

And Poincaré says we can discover true relationships between these real objects.

D’Espagnat says that the only way to make sense of these statements is to believe Poincaré believed some reality that underlies phenomena.

Otherwise it’s hard to imagine how there could be real relationships between real, if hidden, objects.

Separability vs Non-separability
Although d’Espagnat’s viewpoint and Poincaré’s implicit ontology are similar, they differ over “separability.”

Poincaré’s “structural realism” involves “objects-per-se”—unknowable but plural.

D’Espagnat says modern physics does not support separability, and hence there must be “some underlying coherence, or deep unity” to this hidden reality.

Rules vs Ontology
Poincaré believed the equations of classical physics served two purposes.

First, they describe the structure of various laws.

Secondly, they describe the value of certain properties at different points.

Poincaré was happy with the first role.

He had doubts about the second role as he felt equations indicated only what would be observed at those points, not what was pre-existing there.

D’Espagnat wonders if we can give up the second role of an equation’s symbols while retaining the first.

Maybe we could then call this a “structural” realism.

Old vs New Theories
D’Espagnat does note that Poincaré explored ontological issues only with reluctance.

Therefore it would be wrong to attribute this interpretation to Poincaré.

Also, this interpretation has some problems.

As theories evolve, old equations may be seen as merely approximate.

Also, a new theory may have little in common with the old theory.

This would imply the structure of “Reality” is very different under this new theory.

However, normally one could derive the old theory’s equations from the new theory’s equations.

In that case there still might be a meaningful, permanent substratum to reality, despite the objection of radical idealists.

Structural Realism vs Veiled Reality
In the end, d’Espagnat says, structural realism can be justified only after it’s watered down so much it looks like his own “Veiled Reality.”

In the final chapters d’Espagnat says he’ll have to steer between the conceptual difficulties of classical phenomenalism and how physical realism is contradicted by its own science’s results.

Background image: NASA, ESA, J. Richard (CRAL) and J.-P. Kneib (LAM) via


Tags: , , , , , , , , , ,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: