Site Meter

Monday, March 19, 2012

A brief History of "Atoms"

The quotation marks aren't scare quotes. I do believe in atoms, but I want to talk about the history of the idea and paraphrase thinkers (from memory).

Disclaimer: The Wikipedia is not reliable. Don't trust the Wikipedia. The Wikipedia once included the claim (since corrected) that if there is Ricardian equivalence, then a temporary increase in government spending won't cause increased nominal demand. But the Wikipedia is a whole hell of a lot more reliable than this blog.

Once there was a hypothesis that matter is made of atoms. Now everyone familiar with the data is convinced that atoms just plain exist the same way, say, Australia exists. I think that most people don't know roughly when the concept of atoms shifted from being a useful concept in a useful model to a (perceived to be) plain fact. Please guess now.

As far as I know, the concept dates back to Democritus who wrote something along the lines of "All that exists are atoms and the void." This was a bit too extreme for Socrates. It was a pure guess. The properties of "atoms" were that there can't be 2 of them in one place, they are indestructible, they cling to each other in solids, and they fly around separately in gasses. importantly a (somewhat contradictory) second principle is that there are universal natural laws (something like "I would raterh discover a natural law than be King of Persia"). So really there are atoms, the void and universal laws. This is key. It means that to guess where an atom will go, we don't need to know of what whole it is a part. It means that the properties of large things made of atoms can maybe be deduced from teh properties of the atoms (solids are solids because they are made of atoms with loops and hooks like velcro) but not vice versa.

The atomic wild guess consists of materialism (matter can be understood without brining gods, minds or ideas into the discussion) and reductionism (parts can be understood without knowing of what whole they are a part). This has been a rather successful approach. Rather more successful, I would say, than any other human effort.

I might add that, along the way, each of the original guesses (which I ascribe to Democritus based on no evidence at all) has proven false. Atoms are destructible. They are not elementary particles (the modern translation of the Greek atom could be "electron, Quark, Gluon or who knows what all else). The elementary particles are all destructible (by anti-particles if by nothing else). The current view is that one atom isn't in one exact place at one time (Heisenberg and all them) and that two can be identical provided the sum of neutrons plus protons plus electrons is even. Finally, molecules not atoms are believed to fly around in gasses. So now atoms refer to something which is not elementary, indestructible and the only occupant of one space. But it also refers to something which, scientists claim, has been observed.

Faith in the actual existence of atoms as a fact not a useful concept is quite new. The question was very definitely open in 1900. One of the people who convinced other scientists that atoms really exist was Einstein who noted that Brownian motion (spores on water jiggle around) can be explained if there are a finite number of molecules of water so the law of large numbers does not apply to their collisions with very small things. This places the water is made of particles hypothesis on roughly the same plane as the light must be considered (in some ways) to consist of particles hypothesis.

In retrospect, the insistence of many that the concept of an atom was useful, but it was unscientific to assert that the just plain exist seems odd. What about Lavoisier ? What about Avogadro ? What about Maxwell, Boltzman, Arrhenius et al ? Well they had equations which fit the data and which could be elegantly explained with drawings of atoms. But that fits the useful concepts view doesn't it ?

I think part of the issue was extreme attachment to scientific rigor (not at all like mathematical rigor) so that all statements are translated to statements about data as in "the data are consistent with a model in which ..." or observations are as if this model were plain true. This view is, of course, irrefutable. The data are consistent with they hypothesis that I am not a butterfly dreaming I am an economist. But no one is really open minded about the actual existence of, say, Australia*.

Boltzman tried to explain thermodynamics as the consequence of the laws of probability applied to complex systems consisting of many many atoms. He managed to fit the data. Some of his contemporaries said roughly "so what, we already knew of the laws of thermodynamics based on observation. Tell us something we don't know already. Not having a scanning tunneling electron microscope at hand, Boltzman despaired.

OK off to Wikipedia (for the first time really)
"Only a couple of years after Boltzmann's death, Perrin's studies of colloidal suspensions (1908–1909), based on Einstein's theoretical studies of 1905, confirmed the values of Avogadro's number and Boltzmann's constant, and convinced the world that the tiny particles really exist."

* I am paraphrases a character in "A Clergyman's Daughter" by Orwell (a book which he hoped had ended up down the memory hole). The character asks the clergyman's daughter "but do you really believe in Hell in the same way you believe in, say, Australia." Overall, the novel is a very bad novel, but it does have its high points. That is one. Another is the page where a character in the novel is described as ""like a character in a bad novel" In my heart I am sure that Orwell meant to communicate to any actual reader that he knew perfectly well that it was a bad novel (written because he was broke and no one paid squat for short stories of which there are several good ones glommed together in the bad novel). He also knew that editors, publishers and reviewers would not read the novel (he also worked as a book reviewer) and type setters wouldn't care, so he could sneak the confession past the people who paid him to the innocent readers if any (and there are many as many of us read every surviving word he wrote including the war time BBC propaganda which he clearly really counted on ending up in the memory hole).


Mordecai said...

Is there really a useful distinction between "belief in atoms" and "accepting that they are a useful concept in a useful model?" Why? I've been using the word "exist" in scientific disciplines to mean the latter since early undergrad.

Robert said...

I know that the standard philosophy of science (largely based on the thoughts of Boltzman's colleague Mach who drove Boltzman to despair) is as you write. I don't believe that this is the way scientists really think. I think plain belief in useful concepts in useful models changes the way people think. I am not convinced that people really can stick with considering something a useful concept without believing it just plain exists or not.

So I claim that the distinction is used and is useful.

This conviction is not based on much of any evidence. It corresponds to my introspection. I can chearfully admit that I don't know that atoms exist, but I can't doubt that they just plain exist.

I recall Sheldon Glashow meeting with students (after I got a terrible grade in his course with a memorably horrible exam). Someone asked if he thought quarks really existed (or maybe challenged a statement which suggested he did). At the time (1982) the useful concept etc view was still strong. He said "of course they do. There is more evidence that quarks exist than that atoms exist. I mean that there was evidence when it was accepted that atoms exist."

He also starred in an article in which someone says that the average auto mechanic is more interested in philosophy than the average quantum mechanic. The journalist was particularly frustrated when he asked Glashow about the collapse of the wave packet or something and he said "you can't think about those questions. That destroys your ability to work." I note this in the context of his refusing to sticmk with "a useful concept in a useful model." I think that trying to consistently think that way destroys ones ability to work.

Will said...

I like it when you delve into history of science. Most popular expositions fall prey to the cursed "great man theory" of history and to a "Whig history" approach, failing to see all of the randomness and humor in the subject.

"the modern translation of the Greek atom could be "electron, Quark, Gluon or who knows what all else"

As a child, I felt a keen sense of betrayal when I learned that molecules were not the smallest units of matter, but comforted myself that now I knew the true smallest unit: the atom. Then the betrayal played out again when I found out that atoms were composed of protons, neutrons, and electrons. And even that wasn't the end of it.