Spring 2013 Edition
In the philosophy of mind, the multiple realizability thesiscontends that a single mental kind (property, state, event) can berealized by many distinct physical kinds. A common example is pain.Many philosophers have asserted that a wide variety of physicalproperties, states, or events, sharing no features in common at thatlevel of description, can all realize the same pain. This thesis servedas a premise in the most influential argument against early theoriesthat identified mental states with brain states (psychoneural identitytheories). It also served in early arguments for functionalism.Nonreductive physicalists later adopted it (usually without alteration)to challenge all varieties of psychophysical reductionism. The argumenthas even been employed to challenge the functionalism it initiallymotivated.
Reductionists have offered numerous responses. Initial responseseither attacked the argument from the multiple realizability premise tothe anti-reduction/identity theory conclusion or proposed revisions toclassical reductionism that accommodated the premise. More recently,some reductionists have questioned the truth of the multiplerealizability premise itself.
The multiple realizability thesis about the mental is that a givenpsychological kind (like pain) can be realized by many distinctphysical kinds: brain states in the case of earthly mammals, electronicstates in the case of properly programmed digital computers, greenslime states in the case of extraterrestrials, and so on. Correctlycharacterizing the realization relation remains a contentious matter inanalytic metaphysics (Gillett 2003, Polger 2004). But whatever thecorrect account turns out to be, the multiple realizability thesisabout the mental is that a given psychological kind (like pain) canstand in that relationship to many distinct physical kinds.
In a pair of examples illustrating multiple realizability in specialsciences (economics and psychology), Jerry Fodor (1974) implicitlydistinguished between two types of the relation. Call the first type,illustrated in the examples provided at the end of the previousparagraph, multiple realizability “over physical structure-types”:creatures with distinct physical structures realizing theirpsychological states can nevertheless entertain the same psychologicalstates. A more radical type of multiple realizability would obtain if atoken physical (e.g., nervous) system can realize a single mental kindvia distinct physical states of that same system at different times.Call this second sense multiple realizability “in a token system overtimes.” (These terms are from John Bickle 1998, Chapter 4.) This secondsense is more radical because there could be a disjunction of physicalstates realizing each mental kind for every existing cognizer. Theimportance of the more radical type is discussed further (section 1.5below).
In a series of papers published throughout the 1960s, Hilary Putnamintroduced multiple realizability into the philosophy of mind. Againstthe “brain state theorists,” who held that every mental kind isidentical to some as-yet-undiscovered neural kind, Putnam (1967) notesthe wide variety of terrestrial creatures seemingly capable ofexperiencing pain. Humans, other primates, other mammals, birds,reptiles, amphibians, and even mollusks (e.g., octopi) seem reasonablecandidates. But then for the “brain state theory” to betrue, there must be some physical-chemical kind common to this widevariety of pain-bearing species, and correlated exactly with eachoccurrence of the mental kind. (This is a necessary condition of thehypothesized type-identity.) But comparative neuroanatomy andphysiology, facts about convergent evolution, and the corticalizationof function (especially sensory function) as cortical mass increasesacross species all speak against this requirement.
In addition, early mind-brain identity theorists insisted that theseidentities, while contingent, hold by virtue of natural (scientific)law. But then any physically possible cognizer (e.g., pain-bearer) mustalso be capable of possessing that physical-chemical kind. Here thewell-known philosophers' fantasies enter the discussion. Silicon-basedandroids, artificially intelligent electronic robots, and Martians withgreen slime pulsating within their skulls all seem to be possible painrealizers. But they lack “brain states” comparable to oursat any level of physical description. Further still, these mind-brainidentity theories were supposed to be completely general.Every mental kind was held to be identical to some neuralkind. So the critic needs to find only one mental kind, shared acrossspecies yet realized differently at the physical-chemical level. Putnamacknowledges that the early identity theories were empiricalhypotheses. But one of their consequences was “certainly ambitious” andvery probably false.
Stated in canonical form, Putnam's original multiplerealizability argument draws an anti-identity theory conclusion fromtwo premises:
In this simple form, this is a deductively valid argument.
Fodor (1974) extended Putnam's initial argument by arguingthat reductionism imposes too strong a constraint on acceptabletheories in special sciences like psychology. According to Fodor,reductionism is the conjunction of “token physicalism” with the claimthat there are natural kind predicates in an ideally completed physicscorresponding to each natural kind term in any ideally completedspecial science. He characterized “token physicalism” in turn as theclaim that all events that science talks about are physicalevents—a weaker thesis than reductionism or type-typephysicalism. Consider the following string of numerals:
1 1 2.
This string contains two types of numerals (1 and 2), but threetokens of the two types (two tokens of the numeral type 1 and one tokenof the numeral type 2). Mental states admit of a similar ambiguity.When you and I both entertain the belief that Fodor advocates aLanguage of Thought, one type of mental state is entertained, but twotokens of that type (your belief state and my belief state). Type-typephysicalism insists that types of mental states are identical to typesof physical states; this view runs afoul of multiple realizability. Buttoken physicalism only insists that each token occurrence of each typeof mental state is identical to some token occurrence of a physicalstate type—not necessarily an occurrence of a token of the samephysical state type on each occasion.
Fodor gave reductionists the best-developed theory of reduction atthe time: Ernest Nagel's (1961) “derivability” account ofintertheoretic reduction. Nagel's account “connects” disparateelements of the reduced and reducing theories' vocabularies via“bridge laws” (not Nagel's term!) and claims a reduction when thelaws of the reduced theory are derived from the laws of the reducingand the bridge laws. According to Fodor (1974), if reductionism is toestablish physicalism, these cross-theoretic bridge laws must assert(contingent) identities of reduced and reducing kinds. But givenmultiple realizability, the only way this can obtain is if the physicalscience constituent of a psychophysical bridge law is a disjunction ofall the terms denoting possible physical realizations of the mentalkind. Given the extent and variety of actual (not to say possible)physical realizations, it is overwhelmingly likely that the disjunctivecomponent will not be a kind-predicate of any specific physicalscience. It is also overwhelmingly likely that the disjunctivecomponent will not appear in any genuine law of a specific physicalscience. Multiple realizability thus demonstrates that the additionalrequirement of reductionism (beyond token physicalism) is empiricallyuntenable.
The multiple realizability premise has also been used, albeit moreindirectly, in early arguments for functionalism. Functionalism in thephilosophy of mind individuates mental states in terms of their causesand effects. Pain, for example, is caused by tissue damage or trauma tobodily regions, and in turn causes beliefs (e.g., that one is in pain),desires (e.g., that one relieves the pain), and behaviors like cryingout, nursing the damaged area, and seeking out pain relieving drugs.Any internal state that mediates a similar pattern of causes andeffects is pain—regardless of the specific physical mechanismsthat mediate the pattern in any given case. Ned Block and Jerry Fodor(1972) note that the multiple realizability of mental on physical typesshows that any physicalist type-identity hypothesis will fail to besufficiently abstract. Functionalism, on the other hand, seems to be atthe next level of abstraction up from explanation of behavior based onphysical mechanisms. In addition, it seems sufficiently abstract tohandle multiple realizability. Block and Fodor also note that multiplerealizability at the level of physical description is a commoncharacteristic of ordinary functional kinds, like mousetraps and valvelifters. Characterizing mental kinds as functional kinds thus appearsto be at exactly the right level of abstraction to handle multiplerealizability. It is a reasonable empirical hypothesis in light of thisfeature of mental states.
Notice that this argument for functionalism is explicitlynon-deductive, in contrast to the deductive (and valid) nature ofPutnam's original argument against identity theories. It isimportant to keep the anti-identity theory argument separate from thepro-functionalism argument, as some criticisms of multiplerealizability may be telling against one but irrelevant against theother.
Many contemporary nonreductive materialists deny that mental kindscan be identified with functional kinds. Some of their criticisms offunctionalism hinge on issues about individualism in psychology. ButPutnam has used multiple realizability to argue against functionalismitself. In specifying the nature of mental kinds, many functionalistsfollowed Putnam (and Fodor) by adopting “Turing machine functionalism”:mental kinds are identical to the computational kinds of a suitablyprogrammed universal Turing machine. Putnam (1988), however, has arguedthat mental kinds are both “compositionally” and “computationally”plastic. The first point is his familiar multiple realizabilitycontention of the mental on the physical. The second contends that thesame mental kind can be a property of systems that are not in the same(Turing) computational state. In this work, multiple realizabilitystrikes back at the very theory of mind it initially motivated.
Psychologist Zenon Pylyshyn (1984) appeals to multiple realizabilityto ground a methodological criticism of reductionism. He described apedestrian, having just witnessed an automobile accident, rushing intoa nearby phone booth and dialing a 9 and a 1. What will this person donext? Dial another 1, with overwhelming likelihood. Why? Because of asystematic generalization holding between what he recognized, hisbackground knowledge, his resulting intentions, and that action(intentionally described).
We won't discover that generalization, however, if we focus on theperson's neurophysiology and resulting muscular contractions.That level of explanation is too weak, for it cannot tell us that thissequence of neural events and muscular contractions corresponds to theaction of dialing a 1. A given physiological explanation only links oneway of learning the emergency phone number to one way of coming to knowthat an emergency occurred to one sequence of neural events andresulting muscular contractions producing the behavior(nonintentionally described). However, the number of physical eventsconstituting each of these cognitive classes—the learning, thecoming to know, and the action of dialing—is potentiallyunlimited, with the constituents of each class often unrelated to eachother at the physiological level of description. (This isPylyshyn's appeal to multiple realizability.) So if there is ageneralization at the higher level of description available forcapturing (and in the pedestrian example there surely is), anexclusively reductionist approach to psychological explanation willmiss it. Thus because of multiple realizability, reductionism violatesa tenet of scientific methodology: seek to capture all capturablegeneralizations. (Fodor 1975, Chapter 5, and Terence Horgan 1993 raiserelated methodological caveats about reductionism resting ultimately onmultiple realizability. Bickle 1998, Chapter 4, responds to these.)
Recent anti-reductionists have stressed the more radical type ofmultiple realizability, in a token system over times. As far back asthe late 1970s, Block (1978) insisted that the required narrowing ofpsychological kinds due to the more radical type of multiplerealizability renders psychology incapable of capturing whatevergeneralizations hold across species. Ronald Endicott (1993) givesBlock's reply an empirical twist by noting detailed facts aboutplasticity in individual human brains. The capacity for distinct neuralstructures and processes to subserve a given psychological functionowing to trauma, damage, changing task demands, development, and otherfactors is extensive. These facts count further against any reductionof or identities between psychological and physical kinds. Horgan(1993) clearly appeals to this radical sense of multiple realizabilitywhen he writes:
Multiple realizability might well begin at home. For all we now know(and I emphasize that we really donot now know), theintentional mental states we attribute to one another might turn out tobe radically multiply realizable at the neurobiological level ofdescription,even in humans; indeed, even inindividual humans; indeed, even in an individual humangiven the structure of his central nervous system at a singlemoment of his life. (p. 308; author's emphases)
This radical sense has become to default position for nonreductivephysicalists, whose solution to the mind-body problem still dominatesAnglo-American philosophy of mind. Putnam's original multiplerealizability remains central to this solution, with the second premisenow replaced with:
(2′) If mental kinds are multiply realizable (in the radicalsense), then psychology cannot be reduced to a physical science;
and Putnam's original conclusion is replaced with:
(3′) Psychology cannot be reduced to a physical science.
Robert Richardson (1979) suggests that the Putnam-Fodor challenge toreductionism results from a misunderstanding of Ernest Nagel's actualaccount of intertheoretic reduction. Although Nagel's detailed examplesof historical cases all involve biconditional cross-(reduced andreducing) theory “conditions of connectivity,” one-wayconditional connections expressing sufficient conditions at thereducing level are all that his “principle of derivability” requires.Richardson even cites passages from Nagel (1961) indicating that Nagelhimself saw this point. Multiple realizability only challengesnecessity (and nondisjunctive) reducing conditions, and so is not achallenge to even a projected Nagelian reduction of psychology to thephysical sciences.
David Lewis (1969) argues that the inconsistency between thereductionist's thesis and multiple realizability evaporates when wenotice a tacit relativity of the former to contexts. A common senseexample illustrates his point. The following three claims appearinconsistent:
(1) There is only one winning lottery number.(2) The winning lottery number is 03.
(3) The winning lottery number is 61.
These three similar claims likewise seem inconsistent:
(1′) (the reductionist thesis) There is only onephysical-chemical realization of pain.(2′) The physical-chemical realization of pain is C-fiberfiring.
(3′) The physical-chemical realization of pain is …(something else entirely).
((2′) and (3′) reflect the multiple realizabilitycontention.) But there is no mystery in how to reconcile (1) –(3). Append “per week” to (1), “this week” to (2), and “last week” to(3). Similarly, append “per structure-type” to (1′), “inhumans” to (2′), and “in mollusks” to(3′). Inconsistencies evaporate. Lewis's point is thatreductive identities are always specific to a domain.
Many reductionist philosophers have elaborated on Lewis's point withscientific examples. Patricia Churchland (1986, Chapter 7), CliffordHooker (1981), Berent Enç (1983), and other philosophers ofscience have described historical intertheoretic reductions where agiven reduced concept is multiply realized at the reducing level. Acommon example is the concept of temperature from classical equilibriumthermodynamics. Temperature in a gas is identical to mean molecularkinetic energy. Temperature in a solid, however, is identical to meanmaximal molecular kinetic energy, since the molecules of a solid arebound in lattice structures and hence restricted to a range ofvibratory motions. Temperature in a plasma is something else entirely,since the molecular constituents of a plasma have been ripped apart.Even a vacuum can have a (“blackbody”) temperature, thoughit contains no molecular constituents. Temperature of classicalthermodynamics is multiply realized microphysically in a variety ofdistinct physical states. Yet this is a “textbook” intertheoreticreduction and cross-theory identification. The reductions andidentifications are specific to the domain of physical state.
Lewis's original insight also underlies Kim's (1989, 1992) appeals tostructure-specific “local reductions.” Kim agrees that multiplerealizability rules out a general reduction of (structure-independent)psychology to physical science. But it permits (and even sanctions) alocal reduction to a theory of the physical mechanisms of a givenstructure-type. (Kim admits that the relevant structure-types herewill probably be narrower than biological species.) Local reductionsinvolve “structure-specific bridge laws” where the mental-physicalbiconditional occurs as the consequent of a conditional whoseantecedent denotes a specific structure-type (e.g., “ifX is a member of structure typeS, thenXis in mental stateM iffX is in physical stateP”). Conditionals whose antecedents denote differentstructure types will typically have biconditionals as consequentswhose mental term- constituents are co-referential but whose physicalterm- constituents denote different physical events. Multiplerealizability forces this much revision to the bridge laws ofclassical reductionism. But according to Kim, local reductions are therule rather than the exception in science generally, and aresufficient for any reasonable scientific or philosophicalpurpose. Kim's approach is another way to express the tacit domainspecificity in scientific reductions.
Kim (1992) suggests and Bickle (1998, Chapter 4) emphasizes thatguiding methodological principles in contemporary neuroscience assumecontinuity of underlying neural mechanisms. This assumption informsmost experimental techniques and theoretical conclusions drawn fromexperimental results. Continuity is assumed both within and acrossspecies. If radical multiple realizability really obtained amongspecies in the actual world, contemporary neuroscientific experimentaltechniques built upon this assumption should bear little fruit. Whystudy the macaque visual system to investigate human visual processing,for example, if we can't safely assume some continuity across species?Why should positron emission tomography (PET) and functional magneticresonance imaging (fMRI) reveal common areas of high metabolic activityduring psychological task performance, both across and withinindividual humans—now down to a millimeter of spatial resolution?Standard neuroscientific experimental procedures and even clinicaldiagnostic tools would be hopelessly naïve in the face ofsignificant multiple realizability. But these procedures and tools dowork (and are not hopelessly naïve).
Kim and Bickle insist that these successes are evidence thatpsychological functions are not as radically multiply realized asfunctionalists and anti-reductionists suggest. Even neural plasticityis systematic. It has a regular progression following damage to aprincipal structure; there are underlying neural mechanisms thatsubserve it. Furthermore, function following damage is often seriouslydegraded. Persons can still talk, manipulate spatial representations,or move their extremities, but their performance is often qualitativelyand quantitatively less than normal. This fact gives rise to trickyquestions about individuation of psychological function. Are thesealternative neural structures realizingthe same psychologicalfunction—the same mental kind—as before? (Thislast response has been further developed. See the next sectionbelow.)
Bechtel and Jennifer Mundale (1999) provide the most extensiveempirical details about hypothesized or assumed brain type-identitiesacross species in neuroscientific practice. Their explicit target is amethodological consequence sometimes drawn from the multiplerealizability premise: if psychological states are multiply realizedacross biological species, then neuroscience—the scientificstudy of brains—will be of little use toward understandingcognition. But as details of the neuroscience of vision demonstrate,neuroscientists have successfully used understanding of the brain todecompose cognitive visual function. The neuroscientific goal hasbeen to
show how functional considerations get built intodeveloping the structural taxonomy and how that taxonomy in turn can bea heuristic guide in developing information-processing models. Thisproject has not been impaired by multiple realization of psychologicalstates; rather, it relies on the assumption that there is a commonrealization of mechanisms for processing visual information acrossspecies. (1999, 201)
It is difficult to argue with the empirical successes that haveobtained. So even if one accepts the multiple realizability contention,one should be hesitant to draw strong consequences aboutpsychology'smethodological autonomy from it.
In recent years critics have begun to challenge the truth of themultiple realizability premise. One approach challenges the way thatmental kinds are individuated by multiple realizability proponents.
Nick Zangwill (1992) was the first to suggest that multiplerealizability across biological species has never “beenproven.” The multiple realizability contention assumes atype-identity of mental kinds across species. According to Zangwillthis assumption is problematic, given that the obvious sensory andmotor differences across species by themselves yield differentcause-and-effect patterns at all but the grossest level of description.If successful, this challenge undercuts the multiple realizabilityargument by denying that the same mental kinds obtain across species,to be realized by different physical mechanisms.
Lawrence Shapiro (2000) also contends that philosophers are tooquick to claim that a given kind is multiply realized. Some propertiesof the realizers are relevant to the purposes, activities, orcapacities that define a given functional kind, but others are not.Consider corkscrews. That functional kind can be “multiplyrealized” in two tokens that differ only in their color. Thatphysical difference does not make them genuinely different realizationsof corkscrew, however, because it makes no difference to theirperformance as corkscrews. Similarly for two corkscrews that differonly in that one is made of aluminum and the other of steel. Althoughthat compositional difference might matter for some functional kinds,it doesn't for corkscrews. As Shapiro notes, “steel andaluminum arenot different realizations of a waiter'scorkscrew because, relative to the properties that make them suitablefor removing corks, they are identical” (2000, p. 644).Establishing genuine multiple realizability takesargument—one must point to property differences in therealizers that make for a functional difference.
Shapiro argues that this requirement sets up a dilemma. Considerwhat appears to be a genuine case of multiple realizability, that is,two objects that “do the same thing” but in very differentways. Either the realizing kinds genuinely differ in their causallyrelevant properties or they do not. If they do not, then we don'treally have a case of multiple realizability (like thecorkscrews that differ only in color or composition). If they do, thenthey are different kinds. But then they are not the same kind and againwe don't have an instance of multiple realizability—of asingle kind with distinct realizations.
The usual justification for grouping distinct realizers under asingle functional kind is that the classification reveals interestingsimilarities, of the sort we expect to be captured by laws orgeneralizations of higher level science. But according to Shapiro, whenthe realizing kinds differ significantly in their causally relevantproperties for the function at issue, any shared laws orgeneralizations are “numbingly dull” (2000, p. 649): e.g.,all realizers of mouse traps are used to catch mice; both camera eyesand compound eyes have the function to see. Shapiro remarks: “If[functional kinds] share many causally relevant properties, then theyare not distinct realizations … If they have no or only fewcausally relevant properties in common, then there are no or just a fewlaws that are true of both of them” (2000, p. 649). The firsthorn acknowledges a single functional kind but denies that it ismultiply realized. The second undercuts the principal reason forgrouping genuinely different physical kinds under a single functionalkind. Shapiro concludes that taken together these two horns blunt anyclaim ofmultiple realizations ofthe same functionaltype.
Mark Couch (2004) presses a similar dilemma. Defending a claimedmultiple realization involves two steps. Proponents must show (i) thatthe physical states (of the realizers) are type distinct, and (ii)that the functional properties are type identical. Challenges toclaimed multiple realizations can attack either step and, mostimportantly, the step challenged can differ from case to case.(Successfully challenging either blocks the multiple realizabilityargument.) As we saw in the previous section, Bechtel and Mundale(1999) describe cases in which cognitive neuroscientists treat thephysical realizers (brain states) as type-identical across species(attacking step i). In other cases—Couch's example is primateversus octopus eyes—one can appeal to easily-found differencesin functional properties (attacking step ii). The two eyes havedifferent visual pigments in their photoreceptors, different retinas,and different ways of focusing light. These physical differences leadto straightforward input-output (functional) differences: in the opticstimuli the two eyes respond to, in reaction times, and more. Theirfunctions may be similar, but similarity isn't identity and multiplerealization requires the latter. Cross-species functionalsimilarities are often quite superficial, especially across speciesfrom widely differing taxa (a point shared by Couch and Shapiro). Inactual scientific practice, discovered physical (neural) differencestypically incline psychologists to seek out functionaldifferences. Couch's point is that the individuation of psychologicalstates, like the individuation of brain states, is an empiricalissue. Shapiro and Couch hint that claims to multiple realizabilityrely heavily on “folk” psychological intuitions aboutindividuating mental kinds.
Bechtel and Mundale (1999) note that multiple realizabilityproponents appeal to different amounts of “granularity” inindividuating mental and neurobiological kinds. Proponents are contentto analyze psychological states at a coarse-grained level, in whichonly the loosest input-output similarities across species aresufficient for mental kind identities. Yet they insist on veryfine-grained individuation for brain states, in which small differencesacross species are sufficient for neural type-difference. Butpsychological ascriptions admit of finer grains and neural ascriptionsadmit of coarser grains. Bechtel and Mundale insist that when a commongrain is chosen for both, mental-neural type-identities holding acrossspecies are found. In any case, it is unfair to hold neuraltype-individuation to a very fine grain, while adopting a very coarsegrain for mental type-individuation.
Some of these arguments quickly attracted critical attention. Forexample, Gillett (2003) argues that Fodor and other proponents ofmultiple realization assume a ‘Dimensioned’ view ofrealization that allows realizer/realized properties to be instantiatedin thedistinct individuals that bear part-whole relations.Shapiro and other recent challengers assume a ‘Flat’ viewof realization, which demands that realizer/realize properties beinstantiated in thesame individual. Gillett shows first thatcritical arguments like Shapiro's do not go through under theDimensioned view of realization; and second, that the critics have notdefended the Flat view over the Dimensioned view. Gillett concludesthat failing to directly address the nature of realization relationvitiates critiques like Shapiro's and others, who are simply leftbegging the question against original defenses of multiple realizationlike Fodor's.
Besides his appeal to species-specific bridge laws and localreducibility, Kim (1992) offers two additional replies to the multiplerealizability argument. His “denying projectibility” reply starts fromthe familiar fact that the kind “jade” fragments into jadeite andnephrite. Jade is thus incapable of passing the projectibility test fornomicness because of its genuinely disjunctive nature. Multiplerealizability of psychological kinds yields the same consequence.Instead of rendering psychology an autonomous special science, multiplerealizability implies that there is no structure-independent scientificpsychology. There are only “local” scientific psychologies, eachreducible to the theory of the underlying physical mechanisms of thestructure-type in question.
Closely related is Kim's “causal powers” reply. Scientifickinds are individuated by their causal powers, and the causal powers ofeach instance of some realized kind are identical to those of itsrealizer. From these principles it follows that instances of a mentalkind with different physical realizations are distinct kinds. Thus(structure-independent) mental kinds are not causal kinds, and henceare disqualified as proper scientific kinds. Multiple realizabilityyields the failure of structure-independent mental kinds to meet thestandards of scientific kinds. Notice that Shapiro's“dilemma” (discussed in section 2.4 above) is in the spiritof Kim's “causal powers” argument.
Kim's argument has attracted critical attention. As part of theirdefense of the autonomy of the mental, Louise Antony and Joseph Levine(1997) insist that it is not vast multiple realizability that makes aproperty unprojectible—for a property like “having mass ofone gram” is the former but isn't the latter. The projectibilityof a nomic property only guarantees the projectibility of sharedproperties that “areconstitutive of ornomicallyconnected to it” (p. 90, authors' emphases). This rendersKim's appeal to the jade analogy problematic for mentalproperties. Block (1997) argues that kind-ness is both relative andgraded, and so projectibility is always with respect to particulartypes of properties. Specifically, Block distinguishes two types: Dproperties, which are selected (though not necessarily selected for)and whose physical realizations are subject to constraints imposed bylaws of nature; and realization properties, which are due to thepeculiarities of some specific realization. Block argues that Kim'sunprojectibility argument is correct (and important) for realizationproperties of mental kinds; but there are also genuine D properties ofmental kinds (not yet well understood) and these do project from onerealization to others, even in light of vast realizationdifferences.
Fodor (1997) distinguishesdisjunctive frommultiplyrealized properties. The former, like jade, are neitherprojectible nor nomic; but the latter, like mental properties (asconstrued by functionalists) are both. Kim's analogy between jadeand pain breaks down, and thus so does his conclusion about theunprojectibility of the latter. This undercuts the remaining steps inhis argument for reduction. Gene Witmer, on the other hand, (2003)accepts Kim's “linking hypothesis” connecting theunprojectibility of the disjunctive sum of physical realizers with theunprojectibility of the functional kind itself. Instead he challengesthe unprojectibility of the disjunctive sum. There are categories whoseinstances share nothing in common except abstract relational featureswhose denoting expressions occur in generalizations that areconfirmable by their positive instances (the key feature ofprojectibility). Witmer cites examples like “papers written afterbrainstorming,” “products produced by the samecompany,” and “a good night's sleep.” Anargument might overturn our intuitive verdict of confirmability bypositive instances for generalizations containing these terms but theburden is on the denier. This is Witmer's “Moorean”premise: “it is a Moorean fact that we have good reason tobelieve, on the basis of a number of positive instances,generalizations about pain” (67). Now the linking hypothesisturns Kim's argument on its head. By modus tollens, thedisjunctive sum of physical realizers of pain is likewise projectible.(Witmer also provides various readings of Kim's“Inexplicability argument” based the causal exclusionprinciple and argues that each reading fails.)
The more radical type of multiple realizability seems to forceincreasingly narrower domains for reductions to be relativized—atthe extreme, to individuals at times. This much “localreduction” seems inconsistent with the assumed generality ofscience. To avoid this problem, some philosophers have suggested morerevolutionary changes to the “accepted” account of(intertheoretic) reduction.
Following suggestions by Clifford Hooker (1981) and Enc (1983),Bickle (1998, Chapter 4) argues that the radical type of multiplerealizability (in the same token system over times) is a feature ofaccepted historical cases of scientific reduction. It even obtains inthe “textbook” reduction of classical equilibriumthermodynamics to statistical mechanics and microphysics. For any tokenaggregate of gas molecules, there is an indefinite number ofrealizations of a given temperature—a givenmeanmolecular kinetic energy. Microphysically, the most fine-grainedtheoretical specification of a gas is its microcanonical ensemble, inwhich the momentum and location (and thus the kinetic energy) of eachmolecule are specified. Indefinitely many distinct microcanonicalensembles of a token volume of gas molecules can yield the samemean molecular kinetic energy. Thus at the lowest level ofmicrophysical description, a given temperature is vastly multiplyrealizable in the same token system over times. Nevertheless, the caseof temperature is a textbook case of reduction. So this type ofmultiple realizability is not by itself a barrier to reducibility.
To accommodate this feature, Hooker (1981, Part III) supplements hisgeneral theory of reduction with an account of“token-to-token” reductions. His supplement builds thepossibility of multiple realizability (including the strong type)directly into the definition of the reduction relation. LetSbe the predicate, “satisfies functional theoryF,”T be the class of systems to which thetoken system in question belongs,S′ be an appropriatepredicate in some lower level theory of T-system causal mechanisms,andT* be the class of systems to whichS′applies. Then, according to Hooker, “systems of typeSof classT are contingently token/token identical withsystems of typeS′ in classT*=df every instance (token) of a typeSsystem externally classified as in classT is contingentlyidentical with some instance (token) of a typeS′system externally classified as in classT*” (1981,p. 504). By “externally classified,” Hooker refers to thesort of cross-classification that holds across differentdeterminable/determinate hierarchies.
To address some acknowledged shortcomings in Hooker'sformulation of his general theory of reduction, Bickle (1998)reformulates Hooker's insights (including his token-tokenreduction supplement) within a set-theoretic “semantic”account of theory structure and relations. The technical details arecomplex and don't bear repeating here, but the basic idea isstraightforward. Bickle's “new wave” accountconstrues intertheoretic reduction as the construction of an image ofthe set-theoretic structure of the models of the reduced theory withinthe set comprising the models of the reducing, modulo a number ofconditions on the resulting mapping. Elements of the sets of modelsinclude token real-world systems to which the theories apply (thetheories' “intended empirical applications”).
Other new conceptions of both reduction and the mind-brain identitytheory have been proposed. Elliott Sober (1999) insists that areductionist thesis actually follows from the multiple realizabilitypremise. He begins by attacking Putnam's (1967)“objective” account of superior explanation, namely thatone explanation is superior to another if the former is more general.According to Putnam, superior explanations “bring out therelevant laws.” But Sober reminds us that explanatorygeneralizations at lower levels bring out more details. Science“aims for depth as well as breadth” and there is no“objective rule” concerning which endeavor is“better” (1999, 550). Both reductionists andanti-reductionists err in privileging one aim at the expense of theother. Sober then notes that multiplerealizabilitypresupposes some form of asymmetric determination: the lower levelphysical properties present at a given time determine the higher levelproperties present. But this assumption commits its proponents to thecausal completeness of physics (a doctrine that Sober sketches towardthe end of his 1999). If one is also concerned with causalexplanation—if one holds that singular occurrences are explainedby citing their causes—then the causal completeness of physics inturn commits multiple realizability proponents to physics'possessing an important variety of explanatory completeness that allother sciences lack. This is “reductionism of a sort”(1999, 562).
William Bechtel and Robert McCauley (1999) develop a version of“heuristic” mind-brain identity theory (HIT) and defend itexplicitly against multiple realizability. HIT insists that identityclaims in science typically are hypotheses adopted in the course ofempirical investigations, which serve to guide subsequent research.They are not conclusions reached after empirical research has beenconducted. Concerning the multiple realizability of psychological onbrain (physical) states, cognitive neuroscience's heuristicidentity claims assert type-commonalities in comparative studies acrossspecies, not type-differences. Bechtel and McCauley illustrate theirhypothesis with case studies: Brodmann's early 20thcentury work mapping the brain into functionally relevant areas;Ferrier's late-19th century work employing electricalstimulation to cortex; and more recent detailed maps of visualprocessing regions in the primate brain. All of these landmarkfunctional anatomical studies used multiple species. As Bechtel andMcCauley remind us,
when they consider theories of mind-brain relations,philosophers seem to forget that the overwhelming majority of studieshave been on non-human brains. … Although the ultimate objectiveis to understand the structure and function of the human brain,neuroscientists depend upon indirect, comparative procedures to applythe information from studies with non-human animals to the study of thehuman brain. (1999, 70–71)
Heuristic psychoneural type-identity claims across species are keycomponents of these standard neuroscientific procedures.
Thomas Polger (2004) handles multiple realizability by developing a“non-reductive mind-brain identity theory.” He insists thatappeals to stronger kinds of multiple realizability are only plausibleunder prior commitment to functionalism, and so beg the question ifemployed against the identity theory. Weaker claims can be handled in afashion akin to Bechtel's, McCauley's, and Mundale'sstrategy: “the fact—if it is a fact—that manydifferent systems can have the same kinds of mental states does notshow that they do not all do so in virtue of having something incommon” (2004, 10). (One can plausibly read Bechtel, McCauley,and Mundale as providing the empirical details for Polger'sassertion of realization-level commonality.) For the remaining(moderate) forms of multiple realizability, Polger insists:“either the [cognizing] thing shares some properties in common[with us] or else it does not have [our] mental states after all”(2004, 11). Polger adopts a reply akin to Couch's tofunctionalist “empathetic” intuitions that we share mentalstates with a wide range of terrestrial creatures: he denies that wereally attribute the same (as compared to similar) mental states toother species (2004, 15).
In response to recovery of function following massive brain trauma,Polger adopts Bechtel and Mundale's line (and perhapsBickle's): “Rather than supporting multiple realizability,these cases suggest that we do not understand how the brainworks—how to individuate brain processes, events, states, andproperties” (2004, 17). In response to “standard”multiple realization claims, he avails himself of Bechtel andMundale's “different grains” response (2004, 21–26).In the end, Polger countenances some multiple realizability, but arguesthat this much does not threaten his “nonreductive” versionof mind-brain identity theory:
Particular kinds of sensations,S1, …,Sr, are identical to particular kinds of brainstates,B1, …,Br. Sensation kinds may cluster into coarser, moregeneral species-specific mental state kinds, … but insofar asthey do, we expect that their members will share physical properties… Creatures that are similar physically … may also haverelatively similar mental state kinds. … We should expect humanbeings and higher primates to have similar conscious mental statesbecause their brains are quite similar to our own. And we shouldexpect the experiences of octopi or aliens to be different from oursto the extent that their brains are quite different from ourown. (2004, 30)
When reduction or identity theory gets reconceived in ways built toaccommodate multiple realizability, are reductionists/identitytheorists and their critics simply talking past one another? It isworth reminding ourselves that many nonreductive physicalists haveemployed multiple realizability to argue againstall forms ofpsychophysical reductionism. If better general accounts of scientificreduction or identity theory make room for multiple realizability,these demonstrations count against this broader challenge. (If“nonreductive” physicalism were to oppose only a specificbrand of psychophysical reductionism, that would weaken the positionsignificantly, so that it would be compatible with other forms of“reductive” physicalism.) In fact, this broader challengeto psychophysical reductionism traces back to Fodor (1974). While hisarguments explicitly targeted a reductionism built on the classicalNagelian account, Fodor suggested in footnote 2 that “what Ishall be attacking is what many people have in mind when they refer tothe unity of science, and I suspect (though I shan't try to proveit) that many of the liberalized versions of reductionism suffer fromthe same basic defect as what I shall take to be the classical form ofthe doctrine.”
In searching for reductive unity underlying the variety of cognitivesystems, Paul Churchland (1982) once recommended descending“below” neurobiology and even biochemistry, to the level ofnonequilibrium thermodynamics. He insisted that finding reductive unitythere was more than a bare logical possibility because of someparallels between biological processes, whose multiply realized kindsfind reductive unity there, and cognitive activity (especiallylearning).
Concerning Pylyshyn's (1984) attack on reductionistmethodology, Patricia Churchland (1986, Chapter 9) suggests thatfunctional theories are constructed in lower level sciences. New levelsof theory thus get inserted between those describing the structure ofthe lower level kinds and those of purely functional kinds: between,for example, the physiology of individual neurons and cognitivepsychology. We might find a common neurofunctional property for a giventype of psychological state across a wide variety of distinct brains.And if the scope of the macro-theory doesn't extend beyond thatof its microfunctional counterpart, then reduction will be achieveddespite vast multiple realizability at the microstructural level.Neurocomputational approaches that have blossomed since the early 1990sgive real empirical credence to Churchland's suggestion.
Bickle (2003) claims that if we leave our neuroscientificunderstanding at the systems level, psychoneural multiple realizabilityseems obvious. Neural systems differ significantly across species. Butneuroscience does not stop at the systems level. As it moved furtherdown, into cellular physiology and increasingly the molecular biologyof nervous tissue, type-identities across species have been found. Manymolecular mechanisms of neural conductance, transmission, andplasticity are the same in invertebrates through mammals. This mattersfor psychology because mechanisms of cognition and consciousness areincreasingly being found at these levels. Bickle's key example ismemory consolidation, the conversion of labile, easily disruptedshort-term memories into more durable, stable long-term form. Work withfruit flies, sea slugs, and mice has revealed the role of the cyclicadenosine monophosphate (cAMP)-protein kinase A (PKA)-cAMPresponsive-element binding protein (CREB) signaling pathway in keyforms of experience-driven synaptic plasticity. Across these verydistinct taxa, this molecular circuitry has also been implicatedexperimentally in memory consolidation. By altering a single protein inthis cascade (using biotechnology and molecular genetics),experimenters have built mutant organisms whose short-term memoryremains intact (as does their sensory, motor, and motivationalcapacities), but which cannot consolidate these short-term memoriesinto long-term form. Bickle quotes with approval statements like thefollowing, from insect biologists Josh Dubnau and Tom Tully:
In all systems studied, the cAMP signaling cascade has beenidentified as one of the major biochemical pathways involved inmodulating both neuronal and behavioral plasticity. … Morerecently, elucidation of the role ofCREB-mediatedtranscription in long-term memory in flies, LTP and long-term memory invertebrates, and long-term facilitation inA. californica [asea slug] suggest that CREB may constitute a universally conservedmolecular switch for long term memory (1998, 438).
Memory consolidation is just one psychological phenomenon, and soits ruthless reduction to molecular events doesn't establish ageneral claim about unitary mechanisms across widely divergent taxa forother shared cognitive kinds. For that argument, Bickle turns toprinciples of molecular evolution. The first principle holds thatevolution at the molecular level—changes to the amino acidsequence of a given protein—is much slower in functionallyimportant (“constrained”) domains than in functionally lessimportant ones. The second principle is that molecular evolution ismuch slower in all domains of “housekeeping” proteins,especially in ones that participate in cell-metabolic processes in manytissue types. These two principles imply that these molecules, theirdomains, and the intracellular processes they participate in willremain constant across existing biological species that share thecommon ancestor first possessing them. (This is what Dubnau and Tullyrefer to above as a “universally conserved” molecularswitch.) In the end, any psychological kind that affects anorganism's behavior must engage the cell-metabolic machinery inindividual neurons. In the brain, causally speaking, that's wherethe rubber meets the road. But that's the machinery conservedacross existing biological species—changes to it, especially itsfunctionally constrained domains, have (almost) inevitably beendetrimental to an organism's survival. So we should expect thatthe molecular mechanisms for any causally efficacious cognitive kind be“universally conserved.” The discovery of these sharedmechanisms of memory consolidation is not some isolated case, butfollows from the core principles of molecular evolution. As‘molecular and cellular cognition’ proceeds, we shouldexpect more examples of unitary realizers (reductions) of sharedpsychological kinds
At present, nonreductive physicalism (probably) is still thedominant position in Anglo-American philosophy of mind. Its proponentscontinue to appeal to the standard multiple realizability argument (seesection 1 above) to challenge all versions of psychophysicalreductionism and identity theory. However, the recent challenges overthe past decade have attracted some notice. Versions of type-identitytheory and reductive physicalism have made comebacks (Gozzano and Hill,2012). Perhaps the nonidentity of mental content properties with anyphysical properties is no longer “practically receivedwisdom,” as Ernest LePore and Barry Loewer called it more thantwo decades ago?
Criticisms of these new challenges are also starting to amass. CarlGillett and Ken Aizawa have been the most vocal recent defenders ofmultiple realizability. Gillett (2003) develops a precise framework forunderstanding compositionality relations in science generally, and usesthis framework to define property realization and multiple realization,and to distinguish two senses of realization. The first sense,“flat” realization, involves both realized and realizingproperties inhering in a single object. The second sense,“dimensioned” realization, involves realized and realizingproperties inhering in distinct individuals standing in a compositionalrelationship. This distinction is important for two reasons, accordingto Gillett (2002, 2003). First, scientific explanations employdimensioned realizations, as inter-level mechanistic explanationsrelate distinct individuals, Second, Fodor and other proponents of thestandard multiple realizability argument assumed a dimensioned account.But the arguments of Shapiro and other recent critics (see section 2above) challenge the existence of multiple realizability only byassuming flat realization, and no recent critic has defended flatrealization as the correct account involved in the scientific cases atissue.
Applying Gillett's precise framework explicitly, Ken Aizawaand Gillett defend the existence of multiple realization in a varietyof sciences (2009a) and “massive multiple realization”about human psychological properties at every level of organization,from the structure and function of proteins in neurons to socialinteractions (2009b). Their detailed focus in the latter essay isvisual processing. They contend that neuroscientists, unlikephilosophers, are unfazed by massive multiple realization. Multiplerealization has been so contentious in philosophy of mind, they insist,because philosophers both tacitly assume flawed accounts of realizationlike the flat view, and due to an accepted narrative linking multiplerealization to the strict methodological autonomy of psychology fromneuroscience. Aizwa and Gillett (2009b) conclude, however, that theempirical details of vision research shows that a co-evolutionaryresearch methodology is not just consistent with, but explicitlymotivated by massive multiple realization. So this narrative not onlyhelps to blind philosophers to facts that scientists recognize asunproblematic; it is also empirically false.
More recently, Aizawa and Gillett (2011) distinguish two strategiesscientists might adopt to deal with putative cases of multiplerealization. One strategy is simply to take multiple realization atface value. The other is to split the higher-level multiply realizedkind into a variety of sub-kinds, one for each of its distinct lowerlevel realizers, and then eliminate the original higher-level kind, atleast for the purposes of further scientific investigation anddevelopment. Do scientists always favor the second strategy, as recentphilosophical critics of multiple realizability would seem torecommend? With its well-known distinctions between different types orsystems, memory research would seem regularly to have employed this“eliminate-and-split” strategy. Yet Aizawa and Gillettargue that such an assessment oversimplifies the actual scientificdetails in even these much-discussed cases. Here too they sense animportant general methodological lesson: psychology takes account ofneuroscience discoveries, so even taking multiple realization at facevalue does not imply strict methodological autonomy. But the actualdetails of how psychology takes neuroscientific discoveries intoaccount depends both on the nature of the psychological kinds inquestion and the needs of theorizing specific to psychology.
Aizawa has also challenged many of the specific recent challenges tothe standard multiple realizabilty argument. After separating threedistinct arguments in Bechtel and Mundale (1999) (discussed in section2 above), Aizawa (2008) sets his critical sights on their CentralArgument, which argues against multiple realization from the existenceand continued success of brain mapping studies. He argues that Bechteland Mundale misrepresent the actual nature of these studies andmethods employed in functional localization studies. Working withexactly the scientific examples Bechtel and Mundale discuss (mostlyfrom the functional neuroanatomy of vision) Aizawa argues that claimsabout psychological functions do not play the specific role in thesestudies that Bechtel and Mundale insist, and so the success of thesestudies does not imply the falsity of multiple realization. Later inthat paper Aizawa challenges two of Bechtel and Mundale's keyinsistences. He denies that if psychological properties were multiplyrealized, then functional taxonomy of the brain would have to becarried out independently of psychological function. And he deniesthat multiple realization rules out comparisons of brains acrossdifferent species. Hence all the premises of Bechtel and Mundale'sCentral Argument are false. Aizawa (2007) criticizes Bickle's (2003)argument (discussed in section 2 above) that a unitary realization ofmemory consolidation across species has been found at the level ofmolecular mechanisms, despite widespread neural differences in thesebrains at higher levels of neuroscientific investigation. According toAizawa, the protein components of these evolutionarily conservedmolecular mechanisms, and the molecular-genetic components coding forthem, are themselves multiply realized. Finally, Aizawa (forthcoming)presents numerous scientific examples of multiple realization by“compensatory differences.” In such cases, changes to oneor more properties that jointly realize a realized property G arecompensated for by changes in others of the jointly realizingproperties. Although his overall goal in this paper is to bring thisform of multiple realization to wider recognition and study byphilosophers of science, he uses the broader “Gillett-Aizawaframework” to argue that highly specific determinate properties,not just generic determinable properties, are multiply realized inthis specific fashion. Such multiply realized determinate propertiesare indeed exactly similar across distinct realizations, and so answerthe dilemma posed by Shapiro and others (discussed extensively insection 2 above).
Recent critics of the standard multiple realizability argument havelikewise not been quiet. Lawrence Shapiro (2008) raises somemethodological difficulties involved in testing whether a givenpsychological kind actually is multiply realized. (For a relatedargument see Thomas Polger 2009.) Shapiro reminds us of the crucialrole that auxiliary assumptions play in hypothesis testing generally(within a broadly hypothetico-deductive model), and considers acollection of explicit auxiliary assumptions that might be implicitlyused to establish a multiple realization hypothesis. He presents arecent ferret brain-rewiring experiment as a scientific example (inwhich axonal inputs from the primary visual tract were redirected inferret embryos to project to primary auditory cortex—see Sharmaet al. 2000 for the actual scientific details). An auxiliaryhypothesis requiring multiply realized higher-level (in this case,psychological) kinds to be “exactlysimilar”—identical—across distinct realizerswon't help the proponent of the standard multiple realizabilityargument with this purported case.. It is easy to measure better visualperformance in the normally-wired control ferrets compared to there-wired experimental animals. While the experimentally re-wiredanimals have some visual function, it is diminished significantlycompared to controls. “Exact similarity” (identity) ofvisual function is thus not present across these groups. On the otherhand, one might argue for the multiple realizability premise in thisferret re-wiring case using an auxiliary hypothesis that only requiressimilarity in multiply realized higher level properties, yet stillrequires that differences across the realizers should not be limitedonly to the differences that cause differences in the realized (in thiscase, visual) properties. (Shapiro remarks that this auxiliaryassumption seems best to capture the sense of multiple realizationstressed by proponents of the standard argument.) But if we adopt it,again the ferret re-wiring case seems not to provide an empiricalinstance of multiple realization. Shapiro remarks: “thedifferences in ferret brains explain nothing more than differences inferret visual properties” (2008, 523). Shapiro also argues thathis detailed discussion of hypothesis testing difficulties for anymultiple realization hypothesis reveals a flaw in Bechtel andMundale's (1999) influential criticism (discussed in section 2above). Bechtel and Mundale's examples, drawn from thecomparative functional neuroanatomy of vision, only compare homologousbrain structures. But these only have differences that make adifference in their visual properties, nothing else. Instead of theseexamples, Shapiro insists, “one should be looking at differentbrains that reveal similar visual properties despite theirdifferences” (2008, 524)—exactly the kinds of evidence thatBechtel and Mundale's emphasis on homologies doesn'tconsider.
Shapiro and Polger (2012) build upon their accounts of the complexityof actually testing for scientifically-justified multiplerealization. They insist that it renders the significance of multiplerealization far more dubious than philosophers of mind typicallysuppose. They introduce explicit criteria to capture the commonassumption that multiple realization requires not merely differencesbetween realizing kinds, but “differently the same”-ness:the features of entitiesA andB that lead them to beclassified differently by the realizing scienceS2“must be among those that lead them to be commonlyclassified” by the realized scienceS1 (2012,282, criterion iii). This explicit criterion rules out a common appealto camera eyes with different photoreceptive chemicals in theirretinal cones from being genuine (empirical) instances of multiplerealization. Considered coarsely, such eyes are doing the same thingin the same way, so they're not “differently the same.”Considered finely, the two kinds of eyes are sensitive to differentranges and peaks of spectral stimulation, so they're“differently different, not differently the same” (2012,283–284).
Shapiro and Polger's final explicit criterion captures the“differently the same” intuition in terms of quanitativedifferences: the relevant variation between entitiesA andB inrealizing scienceS2 “must be greater than” theindividual differences betweenA andB recognized by realized scienceS1 (2012, 282, criterion iv). The variationrecognized by the realizing science must not merely map ontoindividual differences between A and B recognized by the realizedscience. The demands in actually establishing multiple realization arethus quite strict. Not any old variation will do. According to Shapiroand Polger, these strict demands show both that multiple realizationin the sense required to fund the standard argument is “arelatively rare phenomenon”—despite the vast variabilityeverywhere in the world—and that a “relativelymodest” mind-brain identity theory has little actually to worryabout from it (2012, 284).
Similar in some ways to Couch's arguments (discussed insection 2 above), Colin Klein recently raises a challenge to scientificcontributions made by multiply realized kinds. Noting the varieties ofthings that materials science classifies as ‘brittle,’Klein (2008) notes that few to none of the many scientific discoveriesabout realization-restricted brittle things—about brittle steel,for example—generalize to other realization-restricted types(like brittle glass). Klein insists that generalizations about genuinescientific kinds should be projectable across instances of those kinds,so this requirement seems not to be met by a significant class ofmultiple realized kinds (the realization-restricted ones). Applyingthis point to psychological kinds, instead of supporting ascientifically-backed nonreductive physicalism, it appears rather thatspecial sciences should abandon multiply realized kinds. Klein notesthat proponents of scientifically-based multiple realizability can findterms in special sciences that figure in legitimate explanations, andso appear to refer to projectable multiply realize kinds. But closeinvestigation of some paradigmatic examples reveals these to beidealizations of actual kinds. Special-science kind-terms are thustypically ambiguous. Sometimes a given term refers to an actual butrealization-restricted kind. Other times it refers to features ofexplanatory but non-actual idealized models. (Klein 2008 illustratesthis ambiguity with his detailed example from materials science.)Neither suffices to provide a kind of actual multiple realization thatthe standard argument requires. However, he insists that his argumentisn't entirely negative for non-reductive physicalists.Idealizations can function in explanations that are autonomous in animportant sense from lower level sciences. And Kim's (1996)assumption, that all explanatory work in science must appeal torealization-restricted kinds and properties (discussed in section 2above), is simply incorrect. Still, Klein insists, there appear to beno actual and projectable—hence genuinelyscientific—multiple realized kinds.
Finally, Bickle (2010) questions whether the “secondwave” of criticisms of the standard argument, those thatchallenge the multiple realization premise itself (discussed in section2 above), give aid and comfort to psychoneural reductionists. Sincepsychoneural reductionism was one of the explicit targets of thestandard multiple realization argument, one might plausibly assume thatthey do. Yet none of the “second wavers” are themselvesreductionists (with the possible exception of Shapiro, and morerecently Bechtel 2009, although his response to the multiplerealization argument figures nowhere in his plumb for“mechanistic reduction”). Some (Polger 2004) are explicitlyanti-reduction. (Though Polger 2004 is also explicitlyanti-anti-reduction. He argues that multiple realization has little ifanything to do with reduction.) This fact alone should give apsychoneural reductionist pause. Second, the direction that the secondwave debates have developed, starting with Gillett's (2003)criticisms—deeply into the nature of the realization relation,and so into the metaphysics of science rather than into scienceitself—should prompt the psychoneural reductionist with ametascientific bent to simply tell the second wavers, thanks fornothing! Does that leave psychoneural reductionism back on its heels,in light of the standard multiple realization argument? Not at all,Bickle (2010) insists. For the “first wave”actual-scientific-history challenge to the first premise of thestandard argument, and the initial critical discussions in section 2above) turns out never to have been rejoined by anti-reductionists. Whynot? Bickle speculates that Kim's more metaphysically-inspiredchallenge to the standard argument was the culprit. Non-reductivephysicalists seem to have assumed that rejoining Kim's argumentdismisses the entire first wave of challenges. It does not. There arenumerous examples of multiply realized kinds that are components ofscientific theories widely acknowledged to having been reduced to othertheories. So multiple realizaation alone is no barrier to actualscientific reduction. The detailed scientific cases that fill in that“first wave” challenge to the standard argument remainunanswered to this day.
So the renewed critical interest in multiple realizability, begunmore than a decade ago, continues to the present day. The assumptionthat multiple realizability “seals the deal” againstreductive physicalism and the type identity theory of mind wasmisplaced initially, and is now even more misplaced after the secondwave of recent criticisms. Proponents of the standard argument need tofollow Aizawa's and Gillett's recent leads, and offer newdefenses and counter-responses. What is at stake here should not beunderemphasized: nothing less than one of the most influentialarguments from late-20th century Anglo-American philosophy,one that impacts not only the philosophical mind-body problem but alsothe relationship between sciences addressing higher and lower levels ofthe universe's organization.
How to cite this entry. Preview the PDF version of this entry at theFriends of the SEP Society. Look up this entry topic at theIndiana Philosophy Ontology Project (InPhO). Enhanced bibliography for this entryatPhilPapers, with links to its database.
functionalism |physicalism | reduction
