«University of Connecticut DigitalCommons Articles Philosophy Department 7-1-2000 Physicalism and the Fallacy of Composition Crawford Elder ...»
University of Connecticut
Articles Philosophy Department
Physicalism and the Fallacy of Composition
University of Connecticut Department of Philosophy, email@example.com
Follow this and additional works at: http://digitalcommons.uconn.edu/philo_articles
Elder, Crawford, "Physicalism and the Fallacy of Composition" (2000). Articles. Paper 3.
http://digitalcommons.uconn.edu/philo_articles/3 This Article is brought to you for free and open access by the Philosophy Department at DigitalCommons@UConn. It has been accepted for inclusion in Articles by an authorized administrator of DigitalCommons@UConn. For more information, please contact firstname.lastname@example.org.
Crawford L. Elder Department of Philosophy, U-54 [Published in Philosophical Quarterly, 50 (2000), pp. 332-43.] University of Connecticut Storrs, Connecticut 06269-2054 U.S.A.
Physicalism and the Fallacy of Composition A mutation alters the hemoglobin in some members of a species of antelope, and as a result the members fare better at high altitudes than their conspecifics do; so high-altitude foraging areas become open to them that are closed to their conspecifics; they thrive, reproduce at a greater rate, and the gene for altered hemoglobin spreads further through the gene pool of the species. That sounds like a classic example (owed to Karen Neander, 1995) of a causal chain traced by evolutionary biology. But a view now nearly universal among philosophers maintains that such biological causation is always shadowed, perhaps even rivaled, by causation on a different level.1 That the subgroup of antelopes forages in areas closed to the conspecifics is a state of affairs embodied or realized, notes this view, in certain movements and state changes done by certain physical microparticles—untold billions of microparticles and movements, but a finite and determinate (more on this below) collection nevertheless. That the subgroup reproduces at a greater rate is likewise realized by a huge collection of microparticle movements, a different collection. And the microparticle happenings comprised in the first collection are causally responsible, strictly in accordance with the laws of microphysics, for the microparticle happenings in the second. Biological causation is always shadowed, perhaps even rivaled, by causation on the level of microphysics.
The view I mean is general: any case of causing uncovered by any of the special sciences can be recaptured at the level of microphysics. This view is I think what most philosophers mean by “physicalism”; in any case, “physicalism” is the label I shall use. Physicalism comes in two forms.
Modest physicalism holds that any causal transaction reported by the special sciences can be retraced by microphysics.2 Hegemonic physicalism holds that retracing such a transaction at the level of microphysics is actually reporting the only causation that is going on: the happenings identified by the special sciences only appear to cause something, and the real causing is done by large collections of microparticles.3 This paper focuses on the weaker thesis of modest physicalism. But I will bear in mind that it is perhaps less stable,4 and certainly less exciting, than hegemonic physicalism.
This paper argues that modest physicalism lacks supporting argument; a fortiori, hegemonic physicalism lacks it as well. There is no reason to suppose that causings reported by the special sciences are in general shadowed by causal transactions between one collection of microphysical events and another. This paper attributes the impression that physicalism is undeniable—modest physicalism, at least—to the fallacy of composition. It is undeniable, let us allow, that each microphysical happening comprised in a “cause” collection causes some further microphysical happening, and we can temporarily grant (though see below) that each such microphysical happening causes, by virtue of a causal chain which it launches, some individual microphysical happening comprised in the corresponding “effect” collection. It does not follow that the cause collection causes all the happenings in the corresponding effect collection, or causes a large subset of them, or causes anything. Separate argument is needed to establish those claims. None is available, I shall suggest.
This paper trades on three assumptions about causation which there is not room here to defend— but they are fairly common assumptions, and physicalists especially should be friendly towards them.
The first is that the relata of causation are finely-individuated entities. I have argued elsewhere that they are states of affairs (Elder forthcoming). For the purposes of this paper, it would be just as good if they were finely-individuated events, of the sort which Jaegwon Kim (1969, 1980) discusses or the (different) sort Stephen Yablo (1992b) defends. What they cannot be, for the purposes of this paper, are coarsely individuated events such as Davidson (1967, 1969) affirms. The second assumption is that every individual causing instantiates some general causal law. The third is that for a state of affairs (or Kimevent) c to be a cause of effect e is for it to be what Bennett (following Mackie) calls “an NS condition” of e (Bennett 1988, Ch. III). That is, c must be, in a way set by the causal laws of nature, a Necessary (i.e.
indispensable) component of developments Sufficient, given the surrounding circumstances, for occurrence of e.
Physicalists often elect the hegemonic variant of their position, and to this extent they must be favorably disposed towards the first two of these assumptions. Without the first assumption, the collection of microparticle happenings which realizes the wider foraging by antelopes in the subgroup, and that wider foraging itself, are really one and the same event, just differently described. So whatever the former causes, the latter causes. Distinguishing these events (or states of affairs) is what opens the door to the claim of hegemony. What substantiates that claim, for many physicalists, is the further thought that the laws which underwrite causation at the level of the microparticles are far better, as laws, than the laws (putative laws, perhaps) which underwrite causation by the happenings identified by the special sciences. The laws of the special sciences, hegemonic physicalists argue, are riddled with exceptions unless protected by ceteris paribus clauses. Yet the ceteris paribus clauses seem in principle to resist precise specification, unless as “except where this laws fails to hold”. This suggests, however, that the individual causings reported by the special sciences are less than genuine only if causings need to be underwritten by causal laws. So physicalists tend, and should tend, to be well disposed towards my second assumption as well.
But my conclusion may seem too sweeping, and not just to physicalists. Does it really never happen that an instance of causation at the level of medium-sized objects is underlain by an instance at the level of microphysics? In one particular sort of case, I admit, this does indeed happen. The burden of this paper is just to establish that it does not always happen—not in all individual causings reported by what physicalists call “the special sciences”.
The sort of case in which it happens is the case in which there is a rationale, from the standpoint of microphysics, for collecting together all the microparticles which figure in realizing the event identified as “cause” at the level of medium-sized objects. There rarely is such a rationale, I shall argue, with the individual “cause” events reported by biology or economics (or, for that matter, sociology or behavioral psychology, etc.) To say what such a rationale would look like, and why it matters, will require sections II and III of this paper. But a picture here may be worth many words later. I nudge an irregularly shaped rock towards the edge of a table. By pushing its center of gravity beyond the edge, I cause the rock to fall. So, at least, mechanics—and what some call “folk physics”—reports. Can this instance of causation be recaptured at the level of microphysics? That the rock’s center of gravity is at t1 above nothing but air is a state of affairs realized in the positions and motions of a myriad of microparticles. The rock’s fall at t2 is embodied in a different array of positions and motions, involving very largely the same myriad of microparticles. But plainly it does make sense, from the standpoint of microphysics, to take all the microparticles that figure in realizing the “cause” event together. For the fate of each is intertwined with the position and state of all the rest. Wherever the avalanche of microparticle events begins, that terminates in the “effect” event—in the careers of whichever microparticles—that avalanche ends up embroiling all other microparticles that figure in the “cause” event.
But with typical instances of causation reported by biology, say, or by economics, this synergy— this acting in concert—disappears at the level of the microparticles. Or so I shall argue. Consider, for example, the high-grazing antelopes. Is the microphysical fate of each microparticle that figures in realizing the biologist’s “cause” event intertwined with that of every other? A preliminary question is just which microparticles the question asks about—just which microparticles it is, whose movements and state changes jointly embody the state of affairs that these antelopes now forage in places not open to their conspecifics. Too many, I eventually shall argue, for it to be plausible that a collection so scattered can cause anything at all. But the start of the answer need not be controversial: namely, at the least, all those microparticles contained in the bodily boundaries of the high-grazing antelopes and their conspecifics.
For otherwise the physicalist would run a risk of not circumscribing enough microparticle happenings to capture the fact that it is antelopes which differ in how they are faring. But there will then be familiar problems about just which microparticles to count as falling within these boundaries, and which without (Unger 1980; cf. van Inwagen 1990, p. 214 ff). How, in terms specific enough to generate a list of microparticles, is it determined where a hair in the antelope’s coat leaves off and a drop of water that has begun to permeate that hair takes up? Perhaps the question can be finessed by generosity. Let the physicalist include all molecules that lie within a one-inch zone outside what common sense regards as the antelope’s outer surface; let him also include all microparticles in molecules of gases in each antelope’s lungs; etc. But next note that the physicalist must circle enough microparticle occurrences to embody the fact that the genetically well-favored subgroup is foraging unusually well. Successful foraging involves the ingestion of nutrients. So the microparticles in all the nutrients, contained in all the high-meadow grasses that get ingested, must be included. Then too, what about the jaw movements by which these antelopes tear and chew these nutritious grasses? If the physicalist is to capture the fact that these are not freakishly lucky events, but applications of the antelopes’ normal behavioral repertoire in what is merely a new setting, perhaps enough of the evolutionary history has to be included to capture the fact that the antelopes have foraging in their behavioral repertoire—that neural routines for foraging were at one time selected for (Millikan 1993). Now the cause collection of microparticle movements starts to stretch back in time to ancestor antelopes, and to feature microparticles which no longer are in the bodies of antelopes at all. How likely is it that there will be a rationale, from the standpoint of microphysics, for collecting together the states of all these scattered microparticles as a “cause” event?
If the typical causal transactions reported by evolutionary biology look messy and sprawling, when viewed at the level of the microparticles, so too, famously, do the causal transactions reported by economics. Consider any case in which economics depicts, as the cause of some individual outcome, the fact that money is distributed in a particular economy in a particular way. Money is always physically embodied. But, as Jerry Fodor points out in a different context (Fodor 1975, p. 15), there seems to be virtually no limit to the sort of physical stuff of which money can be composed. Its status as money depends not on its physical composition but on something relational and social—on its being sanctioned to play certain roles by a society or a government. So again there is a hard question about how even to begin assessing the physicalist’s claim—how to determine just which microparticle events fall within the shadow of the monetary distribution which economics depicts as a cause. The physicalist must assemble enough microparticle movements to recapture, at the level of the microparticles, the fact that money in this economy is distributed in such-and-such a way. For this it would evidently not be enough to count just movements of all microparticles in all actual pieces of money. One needs enough microparticle events to capture the fact that those pieces have an assigned social role. So one has to include movements and state changes in at least some microparticles in some persons, perhaps reaching well back in history.
Perhaps also one must include microparticle events which jointly realize the existence of certain legal documents, certain courts of law, certain offices. How likely is it that there will be a rationale from the standpoint of microphysics for collecting all these happenings together as a cause?