Ultimate Homogeneity: A Dialog

Throughout his metaphysical writings, Sellars maintains that current microtheory, with its particulate paradigm, can never depict adequately-even in principle-a universe populated with sentient beings like us. Why not? Experience for us involves the presence of an occurrent perceptual core of ultimately homogeneous secondary qualities. Sellars' "Grain Argument" demonstrates (1) that physical objects qua clouds of discrete particles cannot instantiate such qualities and (2) that they cannot be assigned to an intrasentient realm construed as clusters of discrete, particulate neurons. Neither, contends Sellars, can they simply be eliminated from the inventory of any theory claiming to be both empirical and conceptually independent of common sense. And since common sense fails to provide an adequate picture of reality, our only course is to abandon the particulate paradigm of current microtheory in favor of a process paradigm. This paper traces and develops, in dialog form, these arguments. Dramatis Personae: Wilfrid Sellars, Bruce Aune, Robert Hooker, Paul Feyerabend, Richard Rorty. Scene: Sellars' country house. The guests have gathered for the monthly meeting of the Scientific Realists Club. Sellars: Good afternoon, gentlemen. How wonderful is it of nature to provide such a beautiful setting for our gathering: the sun is shining, the flowers are in full bloom, and the blue jays have already ceased their screaming at my cats. I have taken the liberty of setting up our meeting on the patio; there is coffee, tea, wine, and something to nibble on while we converse. Let's see: Bruce, why don't you sit there next to Robert; Paul, why don't you proliferate theories with Richard on that side, opposite 426 STEPHEN FRIEDMAN Bruce and Robert. I'll sit at the head of the table where I can work the tape recorder, just in case one of us says something we might want to remember later. Rorty: Thanks so much, Wilfrid. As always, you are a fine host. As you know, we believe that all things in the universe and their characteristics, merely material or organic, sentient or otherwise, are properly conceived as systems of atoms and their complex states. At the end of our last meeting, you promised to take us through your argument against our position. Sellars: Indeed! Let's begin, though, at the beginning. My principal aim in this discussion is to demonstrate that a certain depiction of reality is inadequate, even in principle. That depiction has it that all ordinary physical objects and sentient beings can be construed as systems-however complex-Of the basic particles of the most current versions of microphysics. Let us call this thesis Physica~ Reductive Materialism.1 There are two versions of Physi~ Reductive Materialism, the Identity View and Eliminative Materialism. Consequently, my argument against Physica~ Reductive Materialism, gentlemen, will be a complex, two-part affair: the first part seeks to undermine the Identity View while the second aims at refuting Eliminative Materialism. So be patient and bear with me. But interrupt me if you find a certain point of mine overly obscure. Feyerabend: Enough of a preamble, Wilfrid. Let's get down to business already! Sellars: Be patient, my friend, we'll have our chance to lock horns soon enough. Okay, so as I said the argument has two sizable components. The centerpiece for the first component, the "Grain Argument", serves to undercut what I have called the Identity view. Remember the puzzle offered by Eddington's two tables?2 There is the common-sense table, which is substantial, occurrently colored, etc., and there is the scientific table, a cloud of individually colorless atoms in a void. Now there are three possible ways of viewing the relationship between Eddington's two tables. First, the instrumentalist insists that the common-sense table is real while the scientific table is not real; physical theory is a heuristic device rather than an accurate picture of reality. Since we are all scientific realists, we reject the instrumentalist position for reasons I will not go into now. The second alternative provides that the scientific table is real and that the common-sense table is not real. This is the position we share, although I understand it differently from you. The third possibility is that both tables are real: the common-sense table is identical with the scientific table. Now what I have called the Identity View embraces this third alternative and contends that in our universe, populated as it is by sentient creatures like ourselves, two fundamental identities hold. First, ordinary physical objects like tables, and their properties such as being occurrently colored, are identical with systems of atoms and their complex states. Second, a person's central nervous system-the manifest core person so to speak-is identical with systems of particles as described in the most current neurophysiological theory. That is, the brain, nerves, etc., are identical with complex subsystems of neurons, and so on down to the individual neurons, and manifest sensations, features of the core person, are claimed to be identical with complex states of the aforementioned neuron systems.3 And, of course, neurological systems ULTIMATE HOMOGENEITY 427 and their complex states, in the final analysis, are themselves just complex systems of atoms and their states. So according to the Identity View, the catalog of contemporary micro-physics is perfectly adequate to describe completely the actual states of the universe whether or not there are sentient creatures among its furniture. Since common-sense objects and their properties, as well as persons and their sensations, are purportedly identical with the appropriate microtheoretical systems and states of current physics, nothing is lost by thinking of them in the way the physicist does rather than the way in which we think of these things commonsensically-at least in principle. Once current microphysics and neurophysiology have achieved the Peircean ideal of completeness and truth, we will probably not think of the world any longer in common-sense terms but rather the way the physicist in his laboratory would think of them. The primary purpose of my "Grain Argument" is to undermine the Identity View, one of the two versions of Physical2 Reductive Materialism I identified earlier. Anne: Proceed, Wilfrid. This will be quite a show, of that we are sure. I will keep my eye on you. But please, give it to us plainly, with minimal unexplicated philosophical jargon. Sellars: Fair enough, Bruce. You shall be my primary critic, although I am sure I'll hear from the others long before I am through ... Anyway, the "Grain Argument". The "Grain Argument" itself has two parts. In the first part I maintain that ordinary perceptual qualities like common-sense color-you know, the bright red color of tomato skins, the whiteness of teeth, and so on-eannot be identical with the complex qualities or states exhibited by systems of micro-physical particles like atoms. Why not? Because ordinary perceptual qualities like the redness of a tomato are ultimately homogeneous, that's why not (PSI, 35). "And what does that mean?" you might ask (aside: I see you jumping up and down, Bruce. We shall hear from you in a moment). It means at least the following. Consider a transparent pink ice cube that's uniformly pink through and through. The pinkness of this ice cube is ultimately homogeneous in the sense in which we are interested here. That is, every visible region of the pink ice cube is uniformly pink-Or at least uniformly colored-however small those regions may be. So, if we think of the cube as a system of regions, every individual region, however small, is colored and those regions are contiguous. The pink ice cube, then, is a pink continuum. The so-called secondary qualities are all ultimately homogeneous in the sense I have just defined. All ordinary, visible physical objects possess ultimately homogeneous perceptual qualities: every region of these objects must be colored, etc., and these regions must be contiguous. Now is should be quite clear to all of you that a system of atoms cannot possess a state identical with occurrent, ultimately homogeneous, common-sense color. Assume for the moment that the common-sense pink ice cube is in fact identical with a system of atoms in the state Sl' the state purportedly identical with the pinkness of the common sense ice cube. If we conceptually cut up the ice cube into smaller and smaller regions, we will arrive at regions containing single atoms. The following problem then results. According to my physicist friends, no individual atom can in428 STEPHEN FRIEDMAN stantiate occurrent, common-sense pinkness or indeed any such color.4 And so, if Sl is identical with occurrent, common-sense pinkness, then Sl cannot be a property of single atoms. But if Sl is a property of a system of atoms but not of single atoms within the system, then Sl cannot be ultimately homogeneous and therefore cannot be identical with occurrent, common-sense pinkness. It follows generally that the appropriate complex states of atoms cannot be identical with ultimately homogeneous perceptual qualities, and therefore, common-sense physical objects cannot be strictly identical with systems of atoms, since the former can instantiate ultimately homogeneous perceptual properties while the latter cannot. Such is the first component of the "Grain Argument". And if it is correct, the first part of the Identity View, which asserts the identity of common-sense physical objects with atomic systems and their complex states, is untenable. Hooker: I follow your argument, Wilfrid, but I must question a key contention in the first part of your "Grain Argument": your claim that common-sense physical objects are occurrently colored and in an ultimately homogeneous manner. Is this an experiential claim? When I turn my mind in upon my own conceptual scheme-to parody Hume's analytical