Hilary Putnam

2.1.1. Multiple Realizability

Putnam's hypothesis of multiple realizability, articulated in key papers in the late 1960s, challenged the type-identity theory of mental and physical states. He argued that a single mental state, such as pain, could be realized by entirely different physical states of the nervous system in various organisms. For instance, the brain structures underlying pain might differ significantly across diverse animal species, yet they all experience the same mental state of "being in pain." He extended this argument to hypothetical entities like alien beings, artificially intelligent robots, and other silicon-based life forms, asserting that their lack of human neurochemistry should not preclude them from experiencing pain. Putnam concluded that type-identity theorists made an "ambitious" and "highly implausible" conjecture, which could be disproved by even a single instance of multiple realizability. This is sometimes referred to as the "likelihood argument."

Putnam also formulated an a priori argument for multiple realizability based on "functional isomorphism." He defined two systems as functionally isomorphic if "there is a correspondence between the states of one and the states of the other that preserves functional relations." For example, a computer made of silicon chips and one made of cogs and wheels can be functionally isomorphic while being constitutionally diverse. Since functional isomorphism implies multiple realizability, this provided another line of argument.

Putnam, along with Jerry Fodor and others, contended that multiple realizability not only served as an effective argument against type-identity theories but also suggested that any low-level (e.g., microphysical) explanation of higher-level mental phenomena would be insufficiently abstract and general. This led to the development of functionalism, which identifies mental kinds with functional kinds characterized solely by their causal roles, thus abstracting from the microphysical level. Many functional kinds, such as mousetraps and eyes, are multiply realized at the physical level.

Critics of multiple realizability, including William Bechtel and Jennifer Mundale, argued that if it were strictly true, research in the neurosciences would be impossible. They contended that neuroscience relies on the existence or assumption of universal consistencies in brain structures to generalize findings across species. If multiple realizability were an empirical fact, experimental results from one species or organism would lose their meaning when applied to others. Other philosophers, such as Jaegwon Kim, David Lewis, Robert Richardson, and Patricia Churchland, have also criticized metaphysical realism in this context.

2.1.2. Functionalism

Putnam himself presented the initial formulation of functionalism, known as "machine-state functionalism." This theory was inspired by analogies between the mind and Turing machines. The core idea is that the nature of a Turing machine's state is defined by its relations to other states and to its inputs and outputs, irrespective of its material constitution. Similarly, according to machine-state functionalism, a mental state like "being in pain" is defined by its causal role-disposing one to cry out, become distracted, or seek the cause-rather than its specific physical realization.

A simple example of a Turing machine that writes '111' after reading three blank slots and then stops can be represented by the following state transition table:

This table illustrates that if the machine is in State 1 and reads a blank ('B') slot, it writes '1' and remains in State 1. If it reads '1', it moves right and transitions to State 2. The definition of each state is purely functional, based on its causal relations within the system, not its physical makeup.

2.1.3. Rejection of Functionalism

Despite his pioneering role, Putnam later abandoned his support for functionalism and other computational theories of mind. In Representation and Reality (1988), he largely denounced his former philosophical psychology. His primary reason for this shift was the difficulty computational theories faced in explaining intuitions related to the externalism of mental content, famously illustrated by his Twin Earth thought experiment. He also developed a separate argument against functionalism in 1988, contending that functionalism, as a diluted identity theory of mind, still mistakenly identified mental kinds with functional kinds, whereas mental kinds could be multiply realizable over functional kinds themselves.

Despite Putnam's rejection, functionalism continued to thrive, evolving into numerous versions by philosophers like Fodor, David Marr, Daniel Dennett, and David Lewis. It played a crucial role in laying the foundations for modern cognitive science and remained the dominant theory of mind in philosophy during the late 20th century. By 2012, Putnam accepted a modified version called "liberal functionalism," which posits that consciousness and mental properties depend on the right kind of functional capacities, regardless of the specific matter subserving them. These capacities could refer to external factors and need not be restricted to computational descriptions.

Functionalism has faced other criticisms. John Searle's Chinese room argument (1980) directly challenged the notion that thought can be represented as a set of functions. Searle argued that a purely functional system could mimic intelligent action without genuine interpretation or understanding, suggesting that systems operating solely on syntactic processes cannot achieve semantics (meaning) or intentionality (aboutness). Ned Block also advanced several arguments against functionalism.

2.2.1. Semantic Externalism

One of Putnam's most influential contributions is his theory of semantic externalism, encapsulated in his famous slogan: "meaning just ain't in the head." He argued that the meaning of terms is determined by factors external to the mind. His "Twin Earth" thought experiment, introduced in Meaning and Reference (1973) and further elaborated in The Meaning of "Meaning" (1975), serves as the primary defense for this thesis.


In the Twin Earth scenario, everything is identical to Earth, except that its lakes, rivers, and oceans are filled with a substance called XYZ instead of H₂O. Consequently, when an earthling, Fredrick, uses the English word "water," its meaning differs from the Twin Earth-English word "water" used by his physically identical twin, Frodrick, on Twin Earth. Since Fredrick and Frodrick are physically indistinguishable when they utter their respective words, and their words have different meanings, Putnam concluded that meaning cannot be determined solely by what is "in their heads." This led him to embrace semantic externalism regarding meaning and mental content. The philosopher Donald Davidson hailed semantic externalism, developed by Putnam, Saul Kripke, Tyler Burge, and others, as an "anti-subjectivist revolution" that allowed philosophy to move beyond the Cartesian emphasis on subjective experience and directly address the objective realm.

2.2.2. Theory of Meaning

Putnam, along with Kripke and Keith Donnellan, contributed to the causal theory of reference. In The Meaning of "Meaning", he argued that the objects referred to by natural kind terms (e.g., "tiger," "water," "tree") are the primary components of their meaning. He proposed a "linguistic division of labor," akin to Adam Smith's economic division of labor, where the references of such terms are fixed by "experts" in their respective scientific fields. For example, zoologists fix the reference of "lion," botanists fix "elm tree," and chemists define "table salt" as sodium chloride. These referents are considered rigid designators in the Kripkean sense and are then disseminated throughout the linguistic community.

Putnam proposed describing the meaning of every term in a language using a finite sequence of elements, or a "meaning-vector," comprising four components:

• The object to which the term refers (e.g., the object individuated by the chemical formula H₂O).
• A set of typical descriptions of the term, known as "the stereotype" (e.g., "transparent," "colorless," "hydrating").
• Semantic indicators that place the object into a general category (e.g., "natural kind," "liquid").
• Syntactic indicators (e.g., "concrete noun," "mass noun").

This meaning-vector describes the reference and use of an expression within a specific linguistic community, providing conditions for its correct usage and allowing judgment on whether a speaker attributes the appropriate meaning or if its use has changed. According to Putnam, a change in meaning occurs only if the term's reference, not just its stereotype, has changed. However, since no algorithm can determine which aspect has changed in a particular case, considering the usage of other expressions in the language becomes necessary. Given the potentially infinite number of such expressions, Putnam embraced a form of semantic holism.

Despite his frequent changes in other philosophical positions, Putnam consistently adhered to semantic holism. However, this position has faced criticisms from philosophers like Michael Dummett, Jerry Fodor, and Ernest Lepore. They argued that if semantic holism were true, it would be impossible for a speaker to learn the meaning of an expression, as it would require mastering the entire language, which is beyond human cognitive limits. Furthermore, semantic holism struggles to explain how two speakers can mean the same thing when using the same expression, thus undermining the possibility of communication. If a sentence's meaning derives from its relations with all other sentences in a language, then any change in an individual's vocabulary would alter the entire network of relations, causing the meaning of that sentence to change. This would imply that one cannot change opinions about the same sentences, as their meaning would differ at different points in time.

2.3.1. Indispensability Argument

In his 1971 book Philosophy of Logic, Putnam presented what became known as the locus classicus of the Quine-Putnam indispensability argument. This argument, which he attributed to Willard Van Orman Quine, asserts that "quantification over mathematical entities is indispensable for science, both formal and physical; therefore we should accept such quantification; but this commits us to accepting the existence of the mathematical entities in question." While Putnam likely endorsed this version in his early work, he later diverged from some of its underlying views.

In 1975, Putnam formulated his own version of the indispensability argument, drawing from the no miracles argument in the philosophy of science. He stated, "I believe that the positive argument for realism [in science] has an analogue in the case of mathematical realism. Here too, I believe, realism is the only philosophy that doesn't make the success of the science a miracle." According to Putnam, Quine's argument focused on the existence of abstract mathematical objects, while his own aimed for a realist interpretation of mathematics that could be provided by a "mathematics as modal logic" view, which might not necessarily imply the existence of abstract objects.

Putnam also held that mathematics, like physics and other empirical sciences, employs both strict logical proofs and "quasi-empirical" methods. For example, before Andrew Wiles proved Fermat's Last Theorem for all integers n>2 in 1995, it had been proven for many specific values of n. These partial proofs stimulated further research and established a quasi-empirical consensus for the theorem. Although such knowledge is more conjectural than a strictly proven theorem, it was still instrumental in developing other mathematical ideas.

The Quine-Putnam indispensability argument has been highly influential in the philosophy of mathematics, sparking ongoing debate and development. Many in the field consider it the strongest argument for mathematical realism. Prominent counterarguments include Hartry Field's claim that mathematics is not indispensable to science, and arguments from Penelope Maddy and Elliott Sober challenging whether indispensability necessarily commits one to mathematical realism.

2.4.1. Brain in a Vat Argument

Putnam is widely known for his argument against skeptical scenarios, particularly the "brain in a vat" thought experiment, which is a modernized version of Descartes's evil demon hypothesis. Putnam argued that one cannot coherently suspect that one is a disembodied "brain in a vat" placed there by a "mad scientist."

This argument stems from the causal theory of reference. According to Putnam, words always refer to the kinds of things they were originally coined to refer to, based on the experiences of the user or their ancestors. Therefore, if a person, say Mary, were a "brain in a vat" whose every experience was generated by wiring and other machinery, her idea of a "brain" would not refer to a real brain, as neither she nor her linguistic community would have ever encountered one. Instead, to her, a "brain" would refer to an image or signal fed to her through the wiring. Similarly, her idea of a "vat" would not refer to a real vat. Thus, if, as a brain in a vat, she were to say, "I'm a brain in a vat," she would actually be saying, "I'm a brain-image in a vat-image," which Putnam argued is incoherent. Conversely, if she is not a brain in a vat, then saying she is one would still be incoherent because she would mean the opposite. This argument exemplifies a form of epistemological externalism: the idea that knowledge or justification depends on factors outside the mind and is not solely determined internally.

Putnam clarified that his primary target with this argument was not skepticism itself, but rather metaphysical realism, which he believed implied the possibility of such skeptical scenarios. Metaphysical realism posits a gap between how one conceives the world and how the world truly is. Skeptical scenarios like the "brain in a vat" or Descartes's evil demon present a formidable challenge to this view. By arguing that such a scenario is impossible, Putnam aimed to show that the notion of a gap between one's concept of the world and its true nature is absurd. He contended that one cannot have a "God's-eye" view of reality, being limited to one's conceptual schemes, and therefore, metaphysical realism is false.

Putnam's "brain in a vat" argument has faced criticism. Crispin Wright argued that Putnam's formulation is too narrow to refute global skepticism. The possibility of being a recently disembodied brain in a vat, for instance, is not undermined by semantic externalism. If a person lived their entire life outside the vat, speaking English and interacting normally with the world, their words and thoughts (e.g., "tree," "grass") would still refer to the external objects they referred to before being "envatted" by a mad scientist. Another scenario posits a brain in a vat connected to a supercomputer that generates random perceptual experiences, in which case the brain's words and thoughts would refer to nothing, rendering the argument meaningless.

2.5.1. Changes in Realism

In the late 1970s and 1980s, influenced by mathematical logic and some of Quine's ideas, Putnam abandoned his long-standing defense of metaphysical realism. Metaphysical realism is the view that the categories and structures of the external world exist independently of the human mind's conceptualizations. Instead, he adopted a different position he termed "internal realism" or "pragmatic realism." Internal realism posits that while the world may be causally independent of the human mind, its structure-its division into kinds, individuals, and categories-is a function of the human mind, meaning the world is not ontologically independent. This general idea was influenced by Immanuel Kant's concept of the dependence of our knowledge of the world on "categories of thought."

According to Putnam, the flaw in metaphysical realism lies in its inability to explain the possibility of reference and truth. Metaphysical realists suggest that our concepts refer because they mysteriously align with the pre-structured categories and individuals inherent in the external world. Putnam questioned how it could be possible for the world to "carve itself up" into certain structures, for the mind to carve the world into its own categories, and for these two "carvings" to perfectly coincide. His answer was that the world is not pre-structured; rather, the human mind and its conceptual schemes impose structure upon it. In Reason, Truth, and History, Putnam equated truth with "idealized rational acceptability," meaning a belief is true if it would be accepted under ideal epistemic conditions.

Nelson Goodman articulated a similar notion in Fact, Fiction and Forecast (1956), stating, "We have come to think of the actual as one among many possible worlds. We need to repaint that picture. All possible worlds lie within the actual one." While Putnam rejected Goodman's form of social constructivism, he maintained the idea that multiple correct descriptions of reality can exist. None of these descriptions can be scientifically proven to be the "one, true" description of the world. He thus accepted "conceptual relativity"-the view that it may be a matter of choice or convention whether, for example, mereological sums exist or whether spacetime points are individuals or mere limits.

Curtis Brown criticized Putnam's internal realism as a disguised form of subjective idealism, arguing it falls into the trap of solipsism. If existence depends on experience, as subjective idealism maintains, then if one's consciousness ceased to exist, the rest of the universe would also cease to exist. In a response to Simon Blackburn, Putnam later renounced internal realism, believing it assumed a "cognitive interface" model between the mind and the world. Under the growing influence of William James and the pragmatists, he adopted a direct realist view of this relation, according to which perceptual experiences directly present one with the external world, without mental representations or other intermediaries. By 2012, however, he further refined this to "transactionalism", a view that accepts both world-involving transactions in perception and their functional describability, including the involvement of qualia. Although he abandoned internal realism, Putnam still resisted the idea that any given thing or system can be described in exactly one complete and correct way. He ultimately accepted metaphysical realism in a broader sense, rejecting all forms of verificationism and the notion of humans "making" the world.

2.6.1. Interpretation of Quantum Mechanics

Throughout his career, Putnam held various positions on the interpretation of quantum mechanics. In the 1960s and 1970s, he contributed to the quantum logic tradition, proposing that the apparent paradoxes of quantum theory could be resolved by modifying the logical rules used to deduce the truth values of propositions. His early work on this topic included "A Philosopher Looks at Quantum Mechanics" (1965) and "Is Logic Empirical?" (1969). He developed different versions of quantum logic over the years but eventually moved away from it in the 1990s, influenced by critiques from Nancy Cartwright, Michael Redhead, and others.

In 2005, Putnam stated his rejection of the many-worlds interpretation of quantum mechanics, finding no clear way for it to yield meaningful probabilities. He considered both de Broglie-Bohm theory and the spontaneous collapse theory of Ghirardi, Rimini, and Weber to be promising but ultimately unsatisfying, as it was unclear if either could be made fully consistent with the symmetry requirements of special relativity.