Argument from Multiple Realizability

Overview

The Argument from Multiple Realizability is a central argument in the philosophy of mind against type-identity theories, which claim that specific mental state types (such as pain, belief, or desire) are identical to specific brain state types. The argument maintains that the same kind of mental state can be realized or instantiated in very different physical systems—such as human brains, animal nervous systems, or even artificial intelligences—so no one specific physical state can be identical to that mental state type.

Instead, the argument suggests that mental states are best understood in more abstract, functional terms, characterized by what they do (their causal role) rather than the specific material from which they are made.

Historical Background and Formulation

The core idea of multiple realizability emerged in the mid-20th century, particularly through the work of Hilary Putnam in the 1960s and Jerry Fodor in the 1960s–1970s. They were reacting to the then-dominant mind–brain type identity theory, defended by philosophers such as J.J.C. Smart and U.T. Place.

Type-identity theorists held claims like:

• Pain = C-fiber firing (in humans)

Putnam and Fodor argued that this kind of identity is too narrow and anthropocentric. They invited philosophers to imagine:

• Creatures with very different biology (e.g., octopuses, Martians with liquid-ammonia physiology),
• Or even artificial systems, such as computers or robots,

all of which could, in principle, feel pain or have beliefs, even though their internal physical organization would differ radically from that of the human brain.

Putnam’s early examples involved Turing machines and hypothetical silicon-based organisms. The core claim was that there is no reason, in principle, to restrict mental states to one particular physical medium. Just as the same software can run on different hardware architectures, the same mental state type can be implemented in different physical substrates.

This line of thought supports functionalism: the view that mental states are individuated by their causal role (inputs, outputs, and relations to other states) rather than by their material composition. On this view, “pain” is characterized by, roughly, its typical causes (e.g., tissue damage), its typical effects (e.g., avoidance behavior, aversive experiences), and its relations to other mental states, not by any single specific neural pattern.

Philosophical Significance

The Argument from Multiple Realizability has had wide-reaching implications:

1. Critique of Type-Identity Theory
   The argument challenges the claim that mental kinds (e.g., pain, belief) are identical to neural kinds (e.g., C-fiber firing). If beings with very different physiologies can be in pain, then “pain” cannot be strictly identical to any particular kind of neural event found only in humans or mammals.

2. Support for Functionalism
   Proponents argue that multiple realizability fits naturally with functionalism, which treats the mind as analogous to software. Mental states are abstract patterns of causal organization that can be realized in different physical systems, as long as those systems share the same functional structure.

3. Defense of the Autonomy of the Special Sciences
   Fodor, in particular, used multiple realizability to argue that higher-level sciences—such as psychology, economics, or biology—cannot be neatly reduced to physics. If higher-level kinds (e.g., “being a predator,” “being a dollar bill,” “having a belief”) can be realized by many different physical types, then laws framed in higher-level vocabularies cannot simply be rewritten as laws about specific physical configurations.

4. Influence on Cognitive Science and AI
   The idea that mental processes might be realized in artificial systems (e.g., digital computers, neural networks, robots) draws directly on multiple realizability. It underpins much of the conceptual framework of early cognitive science and symbolic AI, where minds are viewed as information-processing systems not tied to any one biological substrate.

Major Objections and Responses

Despite its influence, the Argument from Multiple Realizability remains controversial. Several key criticisms have been advanced:

1. Disunity or Coarseness of Mental Kinds
   Critics argue that mental kinds like “pain” or “belief” may be too coarse-grained. If what we call “pain” in humans, octopuses, and AI systems differs substantially in internal organization and maybe even in phenomenology, then they may not, strictly speaking, be the same mental kind. On this view, identity theory might still hold at a more fine-grained level of classification.

   • Proponent response: Supporters maintain that we already treat many scientific kinds as heterogeneous at the micro-level. What matters is that they play the same functional role and participate in similar explanatory patterns, not that they share identical microstructure.

2. Local Reduction and Improved Neuroscience
   Some philosophers propose “local” or “domain-specific” reductions, where mental kinds are reduced not to a single universal physical kind, but to distinct physical kinds within each species or system (e.g., human pain = human neural state N; octopus pain = octopus neural state O). From this angle, multiple realizability is compatible with a patchwork of local type-identity claims.

   • Proponent response: Defenders argue that such piecemeal reductions fail to preserve the cross-system generality of psychological laws and undermine the explanatory unity of mental kinds.

3. Empirical Challenge: Limited Actual Realizations
   Some object that many purported cases of multiple realizability are merely hypothetical. Actual empirical cases—across known species—may show more neural similarity than originally assumed, weakening the inference from possibility to genuine, robust multiple realizability.

   • Proponent response: Advocates reply that the argument is partly modal (about what is possible in principle) and that ongoing AI developments provide increasingly concrete, non-biological realizations of complex, mind-like functions.

4. Fine-Grained Physicalism
   A further response from physicalists is that every realized mental state is still token-identical to some physical state, even if no type-identity holds across systems. Thus, multiple realizability is compatible with token physicalism: every particular mental event is physical, even if mental types do not line up neatly with physical types.

   • Proponent response: Proponents typically accept token physicalism but insist that the failure of type-identity has important philosophical consequences, especially about the autonomy of psychology and the need for higher-level, functional explanations.

In contemporary philosophy of mind, the Argument from Multiple Realizability is widely regarded as a powerful challenge to simple mind–brain type identity theories and a major motivation for functionalist and non-reductive physicalist accounts of the mind, though its ultimate force and scope remain subjects of ongoing debate.