saturn return to sender

I wrote another too-long response to a BA blog post. I’ve been complaining about this for years, too:

Which leaves me to say the thing I have said so many times, but which so many people don’t seem to want to understand: there is no such thing as the supernatural. If something exists, then it is real, and it is natural.

But I shoot for the middle path:

Whole-heartedly agreed that the word ’supernatural’ is meaningless. I cry a little inside whenever I see the word used by supposedly intelligent publications or people as if it has a referent.

But with respect to the whole discussion, I wanted to suggest a fundamental (ontological? epistemological?) difference between experience and representation of experience, and the consequences for science.

Let’s say we have the technology to hook someone to a machine that precisely constructs descriptions of her emotions in real time. We have a great deal of faith (ehem) that our model of the brain can convert electrical/chemical activity into these perfect descriptions. I’m not sure what such a description would be. Maybe it’s uncannily well-written, evocative prose. Or maybe a vector in some multi-dimensional emotional space that we know (somehow) is complete across the gamut of possible human emotion.

The person has experience, the machine constructs representations of that experience.

Now, if you could only pick one of those two things to have on a desert island, which would you pick? The exhaustive representation? Or the experience? Which is the more “valid” aspect of existence?

Neither; both. The two—a 1st person experience and a 3rd person representation of experience—are fundamentally separate yet equally valid aspects of existence. Science, no matter how perfect its models of the world are, will never be able to do more than generate representations. Experience can never be more than a private affair. Never the twain shall meet. (Except indirectly through the artifacts of reason and language and art and music and poetry and …)

So what’s the problem? you ask. Well, this places a real limitation on what science, as a method of investigation, is able to do, and raises questions about what “truth” and “validity” mean. Giving it short shrift, does a scientific account of an event always beat out a personal experience account of the same event in the battle for validity? If so, is that assertion “provable”?

The answer is of course, no, it’s not provable. Any claim science makes on validity depends on being able to represent the object of investigation.

So? What’s the problem? Two difficulties spring to mind:

1 - The practical impossibility of a complete model of the universe. Granted, this has little bearing on the everyday, whether your microwave works or not, but if we suppose that the project of science can be “completed,” then there is the problem of the amount of information and computation required to model the entire universe, if we suppose that, for any event, tendrils of causative dependency creep out into the entire universe (or at least to the information horizon of the speed of light), which, for completeness‘ sake, we must suppose.

But okay, forget the universe, how about modeling an afternoon thunderstorm? How about modeling a cup of tea? Completely? Without simplification? I don’t know, perhaps it will be possible some day. Maybe quantum computing will even be able to model the apparently probabilistic nature of the very small.

But, more relevant to the example above, how about modeling human emotion? Can we be sure that the multi-dimensional emotional space we made for our machine truly exhausts the gamut of human emotion? Can we prove it?*

Which brings us to …

2 - Representations must exist in some language, and science’s claims of validity in particular rely upon mathematics. I am no expert by any means, but one interpretation of Gödel’s Incompleteness Theorems is that “truth is a stronger notion than proof”. “Proof” exists only in the idealized, internal world of perfect spheres and such. “Truth” is something else. The philosophical implications for science are that “truth” is (provably!) slippery and the notion that science “explains” things, full stop, a bit naïve.

In fact, as should be well-known (but apparently isn’t), nothing is ever “proven” in empirical science. (I cry at the abuse of that word, too.) What we have in empirical science is a general consensus among experts, and faith, yes faith, that mathematical language maps well enough onto experience to trust the scientific models built from it. The models, the language, and the notion of proof are forever hermetically sealed away from the actual, 1st person experience of the world.

And that’s the part we care about, right? Our experience? Science is cool because it gives us microwave ovens.

So experience and 3rd-person representations of experience are (ontologically? epistemologically?) separate, but valid, aspects of existence. To compare the validity claims of science to the validity claims of personal experience is a bit like comparing apples and oranges. Ken Wilber would call each “true but partial.” It’s best, perhaps, to take each on its own terms.

Reuben Hersh calls mathematics a social activity, and if that’s true for math it’s got be even moreso true for science. Science is a collective representation of the world based on a very mature and useful method of inquiry. It is not, however, the holy grail or the end-all and be-all. I’ll take its results as wonderful things, and will often, but not always, defer to them in the face of a conflicting account of “truth” from my own 1st person experience of the world. Did I see a ghost? Probably not.

But, that’s a matter of judgement, not of mathematical necessity.