Saturday, July 7, 2012

Loose Ends in Philosophology: The Multiple Realizability of Water

Robert Pirsig coined the term "philosophology" to describe the strange habit philosophy departments have of teaching philosophical history without teaching philosophical method - teaching "what it's been," in other words, without teaching "how to do it."  Many people don't see a difference, but it's as big as the difference between learning music theory from a book and composing your own actual music (again, Pirsig's example).

When I went back to school, some of the "problems" in the history of philosophy struck me as insulting - not because they were stupid questions (they were excellent questions at the time), but because apparently philosophers hadn't been keeping on top of their science in the intervening centuries.  So at least a couple of my papers were dedicated to actually settling these so-called "unanswerable" questions.  This one is dedicated to John Locke's question about multiple realizability - what if what we think of as one thing, and treat as one thing, is actually many different things that merely appear the same to our senses?  In one way, this makes no sense when you're up on your science - but in another equally important way, it's actually the case.

(All citations refer to the 1975 edition of Locke's An Essay concerning Human Understanding, edited by Peter H. Nidditch.)

John Locke defines the meaning of a word as its associated ideas, a superficially innocuous definition which becomes problematic in the details.  Because ideas are conscious states, we cannot define our terms with the unseen corpuscular structures that inform these conscious states, but must instead resort to the conscious states themselves, using words as symbols for them.  However, as there is always more than one way to skin a cat, this may lead us to the apparently absurd conclusion that more than one corpuscular structure could lead to the same set of ideas, and so we may be forced on this analysis to call two things with distinct real essences by the same word.  A more thorough analysis of this kind, with some understanding of modern physics and chemistry, dissolves the problem:  while Locke could not have known quite how, we are in fact able to escape the absurdity while still endorsing a Lockean conception of meaning and essences.

Locke says that words are used as signs of people’s ideas (p. 405).  Ideas, in their turn, are defined by Locke to be any conscious state (p. 104).  Since we are not telepaths, we humans are forced to resort to words in order to try to make our ideas understood by others (p. 404).  So the meaning of “water,” for instance, would be the conscious states I have associated with my experiences of that which I call “water”:  wet, refreshing to drink, colorless, difficult to hold, able to become solid at a certain temperature and pressure, and so on.

But this is only what Locke would call the “nominal essence” of the sort, “water” (p. 439):  it is my own ideas of particular instances of water, with their differences subtracted such that only the similarities remain, that forms my abstract idea of water which constitutes its nominal essence.  This nominal essence is undergirded by the real essence of water, which is the arrangement of substance that causes these instances of water to cause in me the conscious states which satisfy the nominal essence of water (p. 439).  In other words, the real essence of water consists of the unseen properties of water which cause my ideas of water.  The “arrangement of substance,” in Locke’s day, would have referred to a “corpuscular structure,” as the corpuscularian hypothesis provided the most “intelligible Explication of the Qualities of Bodies” available at the time according to him (p. 547).  This poses an interesting question:  could it be that many corpuscular structures might produce one and the same suite of conscious states in a person?  And if indeed more than one real essence resulted in only one nominal essence, might those different real essences be given the same name?  This could not have been a problem in Locke’s time, as he himself points out that the real essences are unseen (p. 439) – they couldn’t see any corpuscules, and so would have had no way to distinguish two substances based on their real essences.

We in our day cannot afford such blissful ignorance, knowing as we do that matter is made up of atoms which are made up of protons and electrons and so forth.  Indeed, Locke gives the example of a child who comes only to associate the color of gold with the idea of gold, and so believes that the golden-colored sections of a peacock’s tail are actual gold (p. 406).  As the child is to an adult, so Locke is to anyone with a modern high school education (with respect to the subatomic structure of matter, that is) – merely lacking key information.  We in our day do not define water as “a clear substance that is wet and blah blah blah,” but by its chemical formula, H2O.  This is, in a “close enough to count” sort of way, the real essence that Locke was pointing at:  the physical composition and structure of H2O is what accounts for its macroscopic properties, causing it to be liquid at most temperatures and pressures encountered on Earth (during human times, that is), and also causing it to occupy more volume when solid than when liquid.  More on this later.

So let us suppose that we discover another compound, XYZ, which is macroscopically indistinguishable from H2O.  What do we conclude?  Have we discovered another kind of water, and should we therefore demote H2O to simply one of several real essences giving us the ideas of the sort, “water?”  Or is this XYZ not water, but something else (say, “bwater”)?  Locke might indeed conclude that XYZ is in fact another kind of water – and we have stipulated that these are utterly indistinguishable in all practical ways, so why not?  We have corn flour and wheat flour, so why not Earthling water and Martian water?  There are many alcohols and many polymers, and so having many waters is perhaps confusing but ultimately unproblematic.

However, our current understanding of the physics of water gives the lie to the thought experiment:  water’s structure, mentioned earlier as responsible for many of its distinguishing macroscopic qualities, is a byproduct of the physical properties of its component elements, oxygen and hydrogen.  Oxygen, as elements go, is pretty small – with only eight protons, its orbital “shells” of electrons are only two “deep,” and so the nucleus is relatively close to even the outermost electrons.  This causes the electrons in a molecule of water to probabilistically “cluster” much closer to the oxygen nucleus than in any other periodically similar compound (e.g. H2S or H2Se), giving the “oxygen region” a very strong negative charge and leaving the “hydrogen regions” more or less as exposed protons with a very strong positive charge.  This strong dipolarity governs the shape and conditions in which water may crystallize, and also how it reacts with other substances, giving it its peculiar properties and status as the “universal solvent” – and this is precisely why it would be preposterous to a physical chemist that any compound could perfectly mimic the properties of water:  because its distinguishing properties proceed directly from its uniqueness.  What’s more, water may be reduced to molecular hydrogen and oxygen by electrolysis, which XYZ by definition cannot.  And finally, there cannot be any as-yet undiscovered elements, because we have discovered every distinct element that can exist for more than a few nanoseconds (which X, Y, and Z must be able to do in order to form water) – the strong nuclear force is simply not capable of binding together a nucleus above a certain proton count for any appreciable amount of time because the protons electromagnetically repel each other.  So this thought experiment, with a little grounding in physics and chemistry, is clearly unintelligible as it stands.

But it can be saved!  We can probe deeper into physics, so long as we play a little fast & loose with our definition of “corpuscules,” to preserve both the thought experiment and Locke’s conclusion in a very satisfying way.  Let us say that the corpuscules are not molecules or atoms, but instead the protons themselves which distinguish elements – or even better, the quarks that compose protons.  Instead of having “charge,” quarks have “spin” and “color” (as well as other characteristics which are irrelevant in the matter at hand).  Two “up quarks” and one “down quark” make a proton - one must be “red,” one “blue,” and one “green” – and it doesn’t matter which are up and which is down, so long as all three colors are present.  Neutrons, similarly but distinctly, are composed of the same number and colors of quarks, but two spin down and one up instead.  An “isotope” is a “version” of an element that has the same number of protons (and is thus the same element) but a different number of neutrons:  one proton alone is called protium (or hydrogen-1), one proton and one neutron is called deuterium (or hydrogen-2), one proton and two neutrons is called tritium (or hydrogen-3).  What’s more, oxygen also has isotopes, most of them having eight neutrons, but some of them have seven.  Oxygen’s atomic weight (i.e. number of protons and neutrons combined) is 15.999, meaning that almost all of the oxygen we find has an atomic weight of 16, but some has 15 (it’s actually more complicated than that, but in an unimportant way).  The bottom line is that most oxygen has 48 quarks, while some of it has only 45 quarks, and the spin and color can be in any combination so long as the rules of composition are followed – and what is this, if not a different corpuscular arrangement?  We ourselves still call water with oxygen-15 by the name of “water,” and water with oxygen-16 is also water.  Locke, I believe, would endorse this.

And so we see that Locke’s account of real and nominal essences gets worse before it gets better, but is ultimately unproblematic in the final analysis.  Words mean the conscious states we have associated with our referents for those words, and so “water” means the experiences I have associated with water throughout my life.  This nominal essence proceeds directly as a byproduct from the real essence of water as H2O, which real essence actually comes in more than one configuration.  And thus we see that, perfectly consistent with Locke’s account of meaning and our modern science, more than one real essence may in fact result in one and the same nominal essence.

1 comment:

Alan said...

Well you got me!
In P5 I was thinking exactly what you were leading up to in P6.
But in P7 I have to say that, as a Canadian, I was thinking first of D2O (But isn't density a macroscopic property?)
When you got to quark structure of the proton I was thinking that you were going to use Rd vs Gd vs Bd as your equivalent examples and I suppose that might work even though they probably transition to one another much faster than we can actually look at them.