On the Reduction of Psychology to Neuroscience
The following paper was my midterm examination for a graduate level Philosophy of Neuroscience course.
Jerry Fodor’s argument for the autonomy of psychology and Paul Churchland’s argument for eliminative materialism engage with reductionism’s potential success. Reductionism logically purports that higher-level theories are reducible to lower-level theories such that the higher-level can be explained by the lower-level. Both arguments elucidate the potential for psychology to be reduced to neuroscience albeit through distinct, interrelated channels. Fodor examines the reduction of the special sciences (scientific fields excluding physics) to physics; Churchland analyzes the reduction of folk psychology (a commonsense theory about the mind that facilitates one’s understanding of self and others) to neuroscience. The reduction of psychology to neuroscience might not be feasible according to Fodor and Churchland, but I will contend one of Churchland’s principal reasons for folk psychology’s insufficiency and irreducibility. I will begin by outlining Fodor’s view regarding the autonomy of psychology. I will then introduce Churchland’s eliminative materialism. Finally, I will challenge the criteria for a good theory, positing that folk psychology’s alleged stagnation does not disqualify it from being a good theory.
Jerry Fodor’s Autonomy of Psychology
Fodor argues that reductionism is unlikely to succeed and advance the unity of science because the higher-level laws and properties within the special sciences are irreducible to lower-level laws and properties in physics. The unity of science is a philosophical thesis that describes the interrelatedness of scientific fields as a unified whole. In this section, I will outline the logical construction of Fodor’s reductionism, define token physicalism and natural kinds, and explain how Gresham’s Law serves Fodor’s argument.
Fodor uses a logical relationship to explain how reduction entails that bridge laws, or intermediary biconditional statements, hold between terms in distinct sciences (98). Fodor says the reduction of laws of the special sciences to physics begins with a law S1x → S2y in the special sciences where S1 and S2 are predicates particular to the special sciences that one intends to reduce to physics (98). If reductionism holds true, then this first logical statement should be reducible to some law in physics represented by a logical statement P1x → P2y where P1 and P2 are predicates present in the field of physics (98). For the law in special sciences represented by S1x → S2y to reduce to a law in physics represented by P1x → P2y, there must be bridge laws that explain how things have the property S1 if and only if they also have P1. Similarly, a bridge law must explain how things can satisfy S2 if and only if they also have the property P2. These bridge laws exist in the form of two consecutive biconditional statements S1x ↔ P1x and S2y ↔ S2y (98). The bridge laws essentially explain that there is a law that something holds properties of the special sciences if and only if that thing has certain physical properties. Bridge laws are unlikely causal relationships because of their symmetry, so Fodor conceives of the bridge laws as being identity statements that claim that x with the property S1 is identical to x having P1 (100).
Given that bridge laws are identity statements and that every event subject to laws of a special science falls under a bridge law, then all events that the sciences talk about are physical in nature. Fodor refers to this idea as token physicalism (100). Token physicalism is not as strong as type physicalism, or the idea that all properties and laws are physical properties and laws. Type physicalism assumes token physicalism. Type identity statements in type physicalism might be represented by a statement such as “Anytime I have a desire for milk, I will have X mental state” whereas token identity in token physicalism might be represented by the statement “This particular chair is wood” (100). In the former statement, the corresponding neural state should hold true for any given desire for milk while the latter statement will not hold true for any given chair since chairs can be erected from various materials. Fodor argues that both type physicalism and token physicalism are necessary for a successful reduction of psychology to neuroscience, but that type physicalism is too strong.
Even if token physicalism held up for reductionism, Fodor does not believe that there is a law that explains the necessarily identical or coextensive relationship between psychological natural kinds and physical natural kinds (100). Natural kinds are basic components independent of human construction; they have natural properties that tie the kind together and determine what is part of the kind and what is not part of the kind. A successful reductionism would see that each scientific field identifies its natural kinds and synthesizes them between levels.
Fodor uses Gresham’s Law (an economic law that describes how certain things happen in certain monetary exchanges) as an illustration of identity. Money is a natural kind and any event that constitutes a monetary exchange falls under Gresham’s Law (103). One could describe exchanging dollar bills for quarters using physical nomenclature (some sort of exchange of particles and atoms), but such physical descriptions are disjunctive given the countless means of monetary exchange (one can sign checks, use coins, etc.) (103). There does not seem to be a law that accounts for the physical description of each mechanism of exchange such that the physical predicate (right side of bridge law) expresses a physical natural kind (104). Fodor argues that reducing psychology to neuroscience similarly requires a law that states the requirement for each psychological natural kind to be coextensive with a neuroscientific natural kind, but that such a law is unlikely because of its natural kinds’ similarly disjunctive physical descriptions.
Fodor’s argument is then one of non-reductive physicalism, or that entities in the world are part of a physical network and may be physical in nature, but that such entities cannot and should not be reduced to physical laws. The reduction of psychology to neuroscience does not satisfy type-physicalism and it does not satisfy the necessary identicality or coextensivity between psychological natural kinds and physical natural kinds. One can reduce each coin or each psychological state to its physical components or physical processes, but it is unnecessary to reduce the entire field of economics of psychology to physics. Fodor concludes by claiming the unity of science only requires token physicalism, not reductionism.
Paul Churchland’s Eliminative Materialism
Eliminative materialism is the stance that a commonsense understanding of psychological phenomena accounts for folk psychology’s fundamental falsity (67). Churchland argues that, since folk psychology provides an insufficient explanation of our mental processes and collapses when challenged by intertheoretic reduction, it is irreducible to neuroscience and must be eliminated (72, 75). In this section, I will describe some of Churchland’s reasoning for folk psychology’s categorization as a theory, outline his three reasons for folk psychology’s falsity, and describe his alchemical response to functionalists.
Churchland first argues that folk psychology is a robust theory that enables people to explain and predict behavior, addresses the problem of other-minds, and clarifies the intentionality of mental states. To explain and predict behavior, humans may construct common-sense laws from everyday folk psychological explanations, if the semantics of familiar vocabulary is understood in the same way as the semantics of other theoretical terms (69). In other words, theoretical terms derive meaning from their respective laws. Furthermore, regarding folk psychology as a theory clarifies the problem of the other minds (69). Given that folk psychology is a theory expressed by laws and the belief that others have minds is an explanatory hypothesis forwarded by the laws inherent to folk psychology, one only needs a knowledge of folk psychology to ascribe particular mental states to another person (69).
Churchland regards intentionality as a structural aspect of folk psychology. Folk psychology’s structural core rests in its involvement with a specific class of mental states, or propositional attitudes (i.e. beliefs, thoughts, desires). Propositional attitudes follow the semantic outline of “X φ that p” where p is the proposition. When compared with numerical statements such as “X has a mass n,” we might understand how both p and n can be substituted to form a predicate (70). In both the propositional and numerical case, one can construct laws by creating sentences such as “for all p” or “for all n''. Thus, theories in both folk psychology and physical sciences are similar in that one can derive truth from sentence structure and meaning from truth. The difference is that folk psychology uses propositions while physical sciences use numbers (71). The elimination of folk psychology entails the replacement of folk psychology’s terms with neuroscientific terms, which suggests the nonexistence of propositional attitudes.
Churchland argues that eliminative materialism’s explanatory failures, history of retreat and stagnation, and incongruence with science highlight its vulnerability to abolition. First, folk psychology fails to explain phenomena such as mental illness, imagination, sleep, intelligence differences between individuals, or even the ability to catch a fly ball and excludes any group that lacks linguistic capacities since propositional attitudes stem from semantic roots (73). Second, primitive cultures harnessed folk psychology as a mechanism to explain a range of phenomena in the world but now the domain of folk psychology is restricted to higher, more complex animals (74). Folk psychology is also characterized by stagnation as it has not experienced significant development for thousands of years. The theoretical version of folk psychology that the Greeks used only marginally varies from its modern manifestation (74). Third, Churchland posits that folk psychology is incompatible both presently and futuristically with the emerging scientific worldview that explains human cognition and behavior without any sort of appeal to folk psychology (75).
Finally, Churchland uses an alchemical analogy to counter non-eliminative functionalists who defend folk psychology as an abstract theory. Alchemy historically accounted for the properties of matter using alchemical kinds, such as spirits (79). However, the development of elemental chemistry spurred the dissolution, or elimination of this theory (79). The reduction of alchemy to elemental chemistry would have been nearly impossible given the two practices’ distinct classifications of matter. Even if the spirits were redefined by functional states (i.e. referring to the process of ensoulment from mercury as the proclivity to reflect light) and functionalists accepted that their theory is abstract and atomically irreducible, Churchland still maintains that, like alchemy, folk psychology is wrong (80). Displaced theories such as alchemy or folk psychology are eliminated and displaced because they are incorrect.
Argument Against Stagnation and Implications for the Reductionism Debate
Having outlined the positions of Fodor and Churchland, I will now contest Churchland’s argument that we should reject folk psychology because of its historical stagnation (Churchland, 74). My claim is that stagnation is insufficient criterion for the rejection of a theory. I have two reasons for this: good theories do not require historical evolution and folk psychology’s stagnation supports its existence as a folk theory.
To elucidate my first reason–the strongest objection to the inadequacy of stagnant theories–one might consider a fundamental mathematical theorem such as the irrationality of the square root of two. This theory emerged around 2500 years ago from the Pythagorean School of Greek mathematicians and has not been improved upon. Yet, we do not value this theorem less because of its stagnation since inception as it is a fundamental theory in any higher-level mathematics course. Similarly, even if we assume that folk psychology has remained stagnant since its formulation, its stagnation does not disqualify it from being a good theory. Thus, it seems a good theory does not require development across time. However, I concede that there is a greater abundance of mathematical counterexamples where static theories are valued than there are scientific counterexamples.
Churchland might respond to this argument by stating that only perfect theories may remain stagnant and folk psychology is not perfect (75). Yet, ‘perfect’ here seems to refer to a theory having an extensive domain of application. I would respond that theories, similar to laws of physics, are not intended to have extensive domains of application. If theories in physics represent the standard for perfection and not all theories in physics have extensive domains of application, then the standard for perfection does not require theories to have extensive domains of application. For example, Boyle’s Law only applies to ideal gasses (theoretical gasses made of disorganized particles) under a specific set of temperature and pressure conditions. This gas law provides a successful account of the relationships between pressure, volume, and temperature within this set of conditions and does not provide a sufficient explanation of gas behavior outside of such conditions. Its failure to explain gas behavior outside of the given set of conditions does not diminish its adequacy in explaining gas behavior within the given set of conditions. Similarly, folk psychology’s failure to explain domains such as mental disorders and sleep should not discredit its potential success in explaining domains such as thoughts, beliefs, and desires. Theories that do not have extensive domains of application are not necessarily disqualified from a characterization of ‘perfect’ and might entail stagnation without sacrificing its standing as a good theory.
My second reason is that the criteria for a good theory as one that is not stagnant is a criteria formed in direct comparison to scientific theories and potentially neglects folk psychology’s simultaneous categorization as a folk theory. Churchland does not explicitly say that folk psychology is stagnant because it is less developed than the hard sciences, but the description for folk psychology’s stagnation is integrated into a discussion about the comparably developing hard sciences, suggesting that folk psychology is stagnant when compared to scientific theories. Yet, comparing folk theories with scientific theories does not seem productive because folk theories are intended to be accessible to the common folk and explain people’s everyday understandings of behavior. Scientific theories do not need to appeal to the common folk. If we assume that the common folk do not follow complex revisions to psychological theories, stagnation might help everyday people access a more rudimentary folk psychology. This objection does not succeed because Churchland would likely reply that, yes, folk psychology is a good folk theory but its explanatory capacities are still limited to the extent that they will eventually be displaced by a completed neuroscience.
In summation, neither Fodor nor Churchland believe psychology will reduce to neuroscience. Fodor believes mental phenomena exist but are not essentially neural while Churchland believes propositional attitudes are merely illusions sustained by modern linguistics. Fodor believes a developed scientific psychology will arise and vindicate folk psychology while Churchland argues folk psychology must be eliminated because there will ever be a developed scientific psychology. Therefore, the survival of psychology as a theory and the existence of propositional attitudes is at stake in the reductionism debate. Churchland ought to address the alleged necessity for folk theories to elude stagnation in order to provide a more complete argument against folk psychology’s irreducibility and insufficiency as a theory.
References
Churchland, P. M. (1981). Eliminative Materialism and the Propositional Attitudes. The Journal of Philosophy, 78(2), 67–90. https://doi.org/10.2307/2025900
Fodor, J. A. (1974). Special Sciences (Or: The Disunity of Science as a Working Hypothesis). Synthese: The Language and Thought Series, 28(2), 97–115. https://doi.org/10.4159/harvard.9780674594623.c10
