Were atoms real? (Interesting discussion of scientific philosophy.) | Hacker News Reader

Were atoms real? (Interesting discussion of scientific philosophy.)

lesswrong.com

35 pointsyummyfajitas16 comments

I think you can take this one step further and talk about models that are actually false, but nonetheless work. Perhaps the best example would be calculus, which treats space as continuous, and in that sense is the polar opposite of the model of atoms. We now know that space is discretized (at the Plank-length scale, ~10^-36ish meters), but calculus nonetheless remains a remarkably effective and efficient model for making testable and verifiable predictions about mechanics (which itself is false but useful). So calculus is an example of the sort of "pedagogical device" that the 1860's philosophers were debating.

The takeaway from this is that models need not be based firmly in reality to be useful, and conversely that models that are useful and make correct predictions may still nonetheless be based on false descriptions. My favorite example of the latter is Lorentz transforms, which the author alluded to in the article. The math behind changing frames of reference in special relativity predates relativity itself, and were actually developed as a mathematical model of the now-defunct theory of "ether," the posited medium that light waves must move through and which happened to magically distort space with some sort of drag force. Despite the fact that its theoretical bases was completely false, Lorentz' (Lorentz's?) work nonetheless was an accurate model, and could be translated whole into a totally different theoretical framework.

Whether that new theoretical framework is accurate (i.e. "real") or not is hard to tell. It seems to be holding up to advanced scrutiny so far, but that's not a sure indicator.

> We now know that space is discretized

We do not. That is merely the case in some controversial quantum gravity theories.

However, I can give another example. In electrodynamics, you often treat a bulk charge as a continuous charge density (for example a charged sphere) rather than a large discrete set of point charges. This is because then you can apply calculus, which makes computations feasible.

It is said that Poincaré, when finally convinced that the evidence of atoms was overwhelming, exclaimed something like "but then it's all false!", meaning that all the continuum-based theories like fluid dynamics and electrodynamics, which had previously seemed like exact descriptions of the underlying reality, had suddenly shown themselves to be mere approximations...

The takeaway from this is that models need not be based firmly in reality to be useful

What, then, is the criterion for being "based in reality?" Isn't it safe to assume that the theories that say space is discretized are also false in some way?

Also, a minor but important quibble: calculus itself is not a physical theory.

The world is flat.

False, but true enough when you are building a bridge.

Or needing to find your way to the next village.

Actually, I think very large bridges do need to account for the curvature of the earth. For example:

http://en.wikipedia.org/wiki/Vasco_da_Gama_Bridge

Calculus itself is rather geometrical. A definite integral is the area under a curve on a 2d graph of the real numbers, and the real numbers certainly are continuous. Obviously one can argue (and many do) about the relation between mathematics and the real world.

As someone who has spent the last 5 years doing a phd in metaphilosophy and metaphysics... please just trust me when I say:

Don't read this article... don't think about the question... Don't go near it!

I can't remember exactly why though. However, I did see admiral ackbar around here somewhere - I think he knows.

It's a trap!

(Sorry, couldn't resist)

This article kind of is a trap, actually. Check out the last paragraph:

> It seems that one disguised empirical question scientists are asking, when they ask “Is X real, or just a handy predictive device?” is the question: “will I still get accurate predictions, when I use X in a less circumscribed or compartmentalized manner?” (E.g., “will I get accurate predictions, when I use atoms to predict quantized charge on tiny oil drops, instead of using atoms only to predict the ratios in which macroscopic quantities combine?".

Rather than trying to answer the question, the author turns it into a slightly different one that's easier to answer, and potentially more useful.

Yeah theorists have been doing this since Plato. They can't understand the question they've asked so they try to re-parse it into one that they can.

I'm not sure why - but it rarely occurs to them to just stop asking the first question, and stop asking whether the second actually has the same semantic content as the first. When I point this out to them they retort that I have no rational argument that could reasonably compel them to accept that these questions can't be answered. I accept the truth of this retort. However, I offer them the prediction that animals that tend toward behaviours such as theirs will get selected out of existence a whole lot sooner than those that don't. It's a purely empirical hypothesis so there is nothing more that can be said on the matter - and so it mercifully ends the conversation.

I never forgot a quote attributed to Liebnitz in my high school trigonometry book, which went something like:

"Although the world may be a dream, I would call that dream real enough if, using reason well, I was never deceived by it"

The whole quantum physics is just in such a dubious, uncertain stage. We don't know for sure whether its models are "real" or merely a useful tool.

I disagree, you can push quantum physics very hard, in weird corners where is is manifestly wrong. Then you do the experiments, surprise surprise, quantum mechanics turn out to be right. E.g. Single Photon Interference

As long as you stick to mathematics and numbers, quantum mechanic is really robust.

Just don't start interpreting that monster. There be dragons.

Reply on news.ycombinator.com