What do you think?
Rate this book
Representation and Reality
Hilary Putnam, who may have been the first philosopher to advance the notion that the computer is an apt model for the mind, takes a radically new view of his own theory of functionalism in this book. Putnam argues that in fact the computational analogy cannot answer the important questions about the nature of such mental states as belief, reasoning, rationality, and knowledge that lie at the heart of the philosophy of mind.
"Representation and Reality is one of the most thorough and careful criticisms of reductionism in the philosophy of mind that we have yet seen, and all future discussions of the computerhuman analogy will have to take account of it." --Richard Rorty, University of Virginia
"This clear, powerfully argued, and thoroughly accessible book is fascinating, and no one with a serious interest in the philosophy of mind or the philosophy of language can afford not to study it." --Stephen Schiffer, City University of New York
Hilary Putnam is Walter Beverly Pearson Professor of Mathematical Logic at Harvard University.
"Representation and Reality is one of the most thorough and careful criticisms of reductionism in the philosophy of mind that we have yet seen, and all future discussions of the computerhuman analogy will have to take account of it." --Richard Rorty, University of Virginia
"This clear, powerfully argued, and thoroughly accessible book is fascinating, and no one with a serious interest in the philosophy of mind or the philosophy of language can afford not to study it." --Stephen Schiffer, City University of New York
Hilary Putnam is Walter Beverly Pearson Professor of Mathematical Logic at Harvard University.
136 pages, Paperback
First published August 1, 1988
About the author
Hilary Putnam
155 books142 followersHilary Whitehall Putnam was an American philosopher, mathematician, and computer scientist who was a central figure in analytic philosophy from the 1960s until his death, especially in philosophy of mind, philosophy of language, philosophy of mathematics, and philosophy of science. He was known for his willingness to apply an equal degree of scrutiny to his own philosophical positions as to those of others, subjecting each position to rigorous analysis until he exposed its flaws. As a result, he acquired a reputation for frequently changing his own position. Putnam was Cogan University Professor Emeritus at Harvard University.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 10 of 10 reviews
March 31, 2022
A critique on functionalism as a model of how the mind works.
This is a short philosophical text that argues we cannot assume there is anything like a "common mental representation" for ideas that is consistent between people, then explores the consequences that follow from that. Putnam discusses how vocabulary can't possibly be innate (in contrast to Chomsky's thinking), and how any attempt to provide a mathematical account of the meaning of language or of how that meaning is constructed will lead to unsatisfying results.
I found this book useful because it made two important concepts vividly clear and intuitive for me:
1) Humans do not perceive or talk about reality, only their mental approximation of reality.
2) Interpreting language is an arbitrarily complex problem requiring the fullness of human reasoning.
I quite like Putnam's style which is precise without being too dense or obtuse, and lightly humorous especially when he gets a little sassy critiquing other philosopher's work.
If you're interested in philosophy of mind and language, this book has some useful ideas, but it's probably best understood in the larger context of Putnam's work and the field generally.
This is a short philosophical text that argues we cannot assume there is anything like a "common mental representation" for ideas that is consistent between people, then explores the consequences that follow from that. Putnam discusses how vocabulary can't possibly be innate (in contrast to Chomsky's thinking), and how any attempt to provide a mathematical account of the meaning of language or of how that meaning is constructed will lead to unsatisfying results.
I found this book useful because it made two important concepts vividly clear and intuitive for me:
1) Humans do not perceive or talk about reality, only their mental approximation of reality.
2) Interpreting language is an arbitrarily complex problem requiring the fullness of human reasoning.
I quite like Putnam's style which is precise without being too dense or obtuse, and lightly humorous especially when he gets a little sassy critiquing other philosopher's work.
If you're interested in philosophy of mind and language, this book has some useful ideas, but it's probably best understood in the larger context of Putnam's work and the field generally.
April 20, 2024
Curious how this fits into the broader literature from the time and how it is viewed by the field now. Reading it from a mathematician’s perspective, it is full of dubious assumptions, massive logical leaps (which the author sweeps away using “obviously”, “clearly”, “of course”, etc), and self-contradictions (often with the second part of his argument contradicting the first).
The main issue is that the author operates at an absurd level of abstraction, but then only makes arguments using trivial examples. He heavily implies that these examples suffice to show his ideas apply at the level of abstraction they are introduced, but this is a nasty fallacy to be trapped by. Having been trained to spot this error in mathematical writing, I found this almost impossible to get past.
No doubt the book tackles large, nebulous concepts and there will likely never be a satisfying answer for these. But, rather than come away thinking the author has provided some new insights, I am left thinking (naively, for certain) I could do a better job in many fewer pages with some well chosen notation.
The main issue is that the author operates at an absurd level of abstraction, but then only makes arguments using trivial examples. He heavily implies that these examples suffice to show his ideas apply at the level of abstraction they are introduced, but this is a nasty fallacy to be trapped by. Having been trained to spot this error in mathematical writing, I found this almost impossible to get past.
No doubt the book tackles large, nebulous concepts and there will likely never be a satisfying answer for these. But, rather than come away thinking the author has provided some new insights, I am left thinking (naively, for certain) I could do a better job in many fewer pages with some well chosen notation.
August 16, 2022
I’m sure that this book is historically significant and is probably a great read for those interested in some of the other stuff going on at the time in the philosophy of language and understanding the meaning of ‘meaning’. However, for someone like me who just wanted a critique of functionalism, this book was way too technical on the whole.
There were a few gems along the way but more with respect to semantics and theory change- Putnam argues that meaning survives theory change, e.g. gravity means the same thing, even though theories about gravity have changed over time. He agrees for this view over thinking that each time a theory changes we are simply changing the subject and repurposing the word. So our meaning of gravity persists even when the details change.
There were a few gems along the way but more with respect to semantics and theory change- Putnam argues that meaning survives theory change, e.g. gravity means the same thing, even though theories about gravity have changed over time. He agrees for this view over thinking that each time a theory changes we are simply changing the subject and repurposing the word. So our meaning of gravity persists even when the details change.
Read
June 17, 2022Putnam fue uno de los grandes filósofos contemporáneos no solo por la brillantes de sus ideas individuales y su prolífica trayectoria, sino por su compromiso con la filosofía como disciplina más allá de las etiquetas y los logros personales. Su capacidad para poner en tela de juicio sus propias ideas y reconocer sus fallos es uno de los rasgos que más admiro en cualquier filósofo o filósofa que conozca. Pero es que además sus críticas y alternativas siempre mejoraban lo presente. Ese es sin duda el caso de su autocrítica al funcionalismo y de su cuestionamiento en general de las filosofías de la mente que creen poder explicar y agotar lo mental a base de estructuras internas, esenciales y unitarias que dan perfecta cuenta de todo lo que pensamos, decimos y hacemos. Además de a sí mismo, en Representación y realidad confronta a grandes elefantes de la filosofía de la mente y las ciencias cognitivas como Noam Chomski o Jerry Fodor y propone un modelo explicativo de lo mental que tenga en cuenta la influencia de las otras mentes y por lo tanto de la sociedad y la cultura, así como del entorno físico. No se trata de una teoría fija, sino de una opción de búsqueda menos encorsetada y con miras más abiertas. No son tanto respuestas lo que Putnam nos da, sino nuevas preguntas, a mi juicio más interesantes y ricas.
June 10, 2026
In Chapter 1: Meaning and Mentalism, Putnam argues that mentalism (his previous view of functionalism cannot be right for three interrelated reasons. The first is that meaning is holistic, contrary to the logical positivists who thought that meaning could be defined piecewise and compositionally. The arguments for this point are familiar–the Quine-Duhem thesis and the fact that almost no definitions exist. The second is that meaning is partially a normative notion. When we interpret what other people believe we engage in a charitable practice. For example, we think their “plant” is still our plant. The third point is that our concepts depend on our physical and social environment our brains could not foresee–therefore, with point one and two, there is no plausible way that different representations all have the same syntactic and procedural components across languages, for example.
Chapter 2: Meaning, Other People, and the World: Putnam targets the view that two of the same representations in the language of thought have the same meaning and that two of the same representations that differ have different references (at least including all possible worlds). But none of the common representations we know have it as a property that a difference in representation must result in a difference in meaning. The referent of familiar representations may change with changes in culture. And reference is usually not fixed by what we can differentiate, but by the experts in our society. Hence, mental representations do not fix reference. Aristotle’s idea that words “hook onto” the world because they’re associated with mental representations that “hook onto” the world is false. Additionally, he points out that without being able to point to a specific difference between beeches and elms, one has an identical mental representation of the two, other than their name–hence mental representations do not suffice to fix reference. He also brings out his Twin Earth example to highlight the contribution of the environment to reference.
Chapter 3: Fodor and Block on "Narrow Content": Putnam first addresses Fodor’s notion of narrow content which uses perceptual stereotypes. But different languages might have different perceptual prototypes associated with a word with the same meaning. Then he addresses Block’s view that a word’s meaning consists of its narrow and broad content. Block’s view is that a word’s narrow content is defined by its conceptual role in its respective content. But conceptual role can change a lot without a change in meaning. Moreover, it seems that the meaning of natural kind terms depends more on extension. But most importantly, Block’s definition of “conceptual role” does not match the kind that intuitively matters the most.
Chapter 4: Are There Such Things as Reference and Truth?: Putnam thinks it’s a mistake for Stich and Churchland to think that because there is nothing in common between tokenings of the same mental state on the scientific level that those tokenings do not exist at all. In particular, you can apply the same reasoning to chairs which Stich and Churchland do seem to exist. Additionally, if you are eliminationist about propositional attitudes for Churchland’s reasons you also have to reject traditional notions of truth and reference, which is too much. They might try to resort to “disquotational semantics” to defend truth and reference without psychology but Tarski’s semantics do not work because Tarski gives extensionally correct truth predicates but does not analyze our ordinary notions of truth and reference.
Chapter 5: Why Functionalism Didn't Work: Functionalism, construed as the thesis that propositional attitudes are just computational states of the brain, cannot be correct because they are defined externally as well. Just like multiple different physical states could give rise to the same mental state, multiple computational states can as well. Additionally, one cannot just expand functionalism to include environments–deciding whether someone else has the same belief as us is more of an art than a science, and can never be defined functionally. Even if we speak the same language we might have different stereotypes for the same concept, for example. Furthermore, even different species we can communicate meaningfully with might be computationally incomparable with us. However, we can construe functionalism as only requiring two identical mental states be equivalent under some equivalence relation which is itself computable, or at least definable in the language of computational theory plus physical science (Lewis). But this would involve being able to understand so much more than we possibly could.
Chapter 6: Other Forms of Functionalism: Even if we cannot figure out the ideal theory of interpretation, maybe in principle we could. However, this would require we know what kind of thing would make a theory true, which we do not. Then he addresses causal-role functionalism. But then we would need things even more metaphysically contentious than belief–posisble worlds and naturalness–to make sense of his definition of causality. He then surveys his old functionalism which involves isomorphs–but it faces similar problems. Unless the physical states that realize mental states have something in common, they’re just infinite disjuncts. More problems with Lewis’ view is that our platitudes about beliefs might be different for other species. And what about much smarter agents?
Chapter 7: A Sketch of an Alternative Picture: He describes his internal realism. “In sum,
the attempt to survey "meaning" or "reference" fails for the same reason the attempt to survey reason itself fails: reason can transcend whatever it can survey.”
Chapter 2: Meaning, Other People, and the World: Putnam targets the view that two of the same representations in the language of thought have the same meaning and that two of the same representations that differ have different references (at least including all possible worlds). But none of the common representations we know have it as a property that a difference in representation must result in a difference in meaning. The referent of familiar representations may change with changes in culture. And reference is usually not fixed by what we can differentiate, but by the experts in our society. Hence, mental representations do not fix reference. Aristotle’s idea that words “hook onto” the world because they’re associated with mental representations that “hook onto” the world is false. Additionally, he points out that without being able to point to a specific difference between beeches and elms, one has an identical mental representation of the two, other than their name–hence mental representations do not suffice to fix reference. He also brings out his Twin Earth example to highlight the contribution of the environment to reference.
Chapter 3: Fodor and Block on "Narrow Content": Putnam first addresses Fodor’s notion of narrow content which uses perceptual stereotypes. But different languages might have different perceptual prototypes associated with a word with the same meaning. Then he addresses Block’s view that a word’s meaning consists of its narrow and broad content. Block’s view is that a word’s narrow content is defined by its conceptual role in its respective content. But conceptual role can change a lot without a change in meaning. Moreover, it seems that the meaning of natural kind terms depends more on extension. But most importantly, Block’s definition of “conceptual role” does not match the kind that intuitively matters the most.
Chapter 4: Are There Such Things as Reference and Truth?: Putnam thinks it’s a mistake for Stich and Churchland to think that because there is nothing in common between tokenings of the same mental state on the scientific level that those tokenings do not exist at all. In particular, you can apply the same reasoning to chairs which Stich and Churchland do seem to exist. Additionally, if you are eliminationist about propositional attitudes for Churchland’s reasons you also have to reject traditional notions of truth and reference, which is too much. They might try to resort to “disquotational semantics” to defend truth and reference without psychology but Tarski’s semantics do not work because Tarski gives extensionally correct truth predicates but does not analyze our ordinary notions of truth and reference.
Chapter 5: Why Functionalism Didn't Work: Functionalism, construed as the thesis that propositional attitudes are just computational states of the brain, cannot be correct because they are defined externally as well. Just like multiple different physical states could give rise to the same mental state, multiple computational states can as well. Additionally, one cannot just expand functionalism to include environments–deciding whether someone else has the same belief as us is more of an art than a science, and can never be defined functionally. Even if we speak the same language we might have different stereotypes for the same concept, for example. Furthermore, even different species we can communicate meaningfully with might be computationally incomparable with us. However, we can construe functionalism as only requiring two identical mental states be equivalent under some equivalence relation which is itself computable, or at least definable in the language of computational theory plus physical science (Lewis). But this would involve being able to understand so much more than we possibly could.
Chapter 6: Other Forms of Functionalism: Even if we cannot figure out the ideal theory of interpretation, maybe in principle we could. However, this would require we know what kind of thing would make a theory true, which we do not. Then he addresses causal-role functionalism. But then we would need things even more metaphysically contentious than belief–posisble worlds and naturalness–to make sense of his definition of causality. He then surveys his old functionalism which involves isomorphs–but it faces similar problems. Unless the physical states that realize mental states have something in common, they’re just infinite disjuncts. More problems with Lewis’ view is that our platitudes about beliefs might be different for other species. And what about much smarter agents?
Chapter 7: A Sketch of an Alternative Picture: He describes his internal realism. “In sum,
the attempt to survey "meaning" or "reference" fails for the same reason the attempt to survey reason itself fails: reason can transcend whatever it can survey.”
January 25, 2026
10/73
Hilary Putnam’s Representation and Reality is a critical examination of representational theories of mind and language that dominated analytic philosophy in the mid-20th century. The book is not a general introduction, neither is it a systematic replacement of one theory with another. It is an internal critique, written by a philosopher who had worked extensively within the very frameworks he is now questioning. As an amateur reader, I found the book challenging less because of its prose than because it assumes familiarity with debates that are not always reconstructed from first principles.
The central target of the book is the idea that the mind represents the world by forming internal models that correspond to a fully determinate, mind-independent reality. Putnam refers to this cluster of assumptions as “metaphysical realism.” According to this view, the world has a fixed structure, language hooks onto that structure through reference, and truth consists in correspondence between representations and how things are “in themselves.” Much of Representation and Reality is devoted to showing that this picture fails to deliver what it promises.
Putnam’s best-known argument here is model-theoretic. Very roughly, he argues that no matter how completely we specify the physical facts about the world, there will always be multiple, incompatible ways of assigning reference to our terms that preserve truth conditions. If this is right, then reference cannot be fixed purely by the world’s structure. The idea of a single, privileged mapping between language and reality turns out to be unsupported. While the technical details are demanding, the philosophical conclusion is clear enough even without mastering the formalism: metaphysical realism relies on a notion of determinacy it cannot justify.
What is notable is what Putnam does not do in response. He does not conclude that truth is subjective, that science is merely a social construction, or that all conceptual schemes are equally valid. On the contrary, he insists that objectivity, rational criticism, and empirical constraint remain central. His claim is narrower but more disruptive: objectivity does not require, and cannot be grounded in, a “God’s-eye view” of reality that stands outside all human conceptual practices.
Putnam’s alternative position, often labeled “internal realism,” holds that truth is not correspondence with a ready-made world but idealized rational acceptability within a conceptual framework. As a reader, I found this formulation initially unsatisfying, because it seems to weaken the notion of truth just when philosophy is supposed to secure it. Over the course of the book, however, it becomes clear that Putnam’s point is not to weaken truth but to describe how it actually functions in science and everyday reasoning. We argue, revise, and reject claims from within shared practices, not by comparing our theories to an inaccessible reality as such.
Placed within the broader philosophical landscape, Representation and Reality belongs firmly to the analytic tradition in method and reference points. Putnam engages with formal semantics, philosophy of language, and philosophy of mind, and he takes scientific practice seriously throughout. At the same time, the book distances itself from the foundational ambitions that once defined analytic philosophy. In this respect, Putnam’s position shows affinities with Kant’s insistence that knowledge depends on conditions of intelligibility, and with later critiques of representational thinking found outside the analytic canon, even if Putnam does not adopt their vocabulary or historical framing.
One limitation of the book, especially for non-specialists, is that Putnam often moves quickly over background debates. Readers unfamiliar with logical positivism, Quine’s critiques, or computational models of mind may struggle to see exactly which assumptions are being dismantled at each step. In addition, Putnam’s refusal to present a fully worked-out alternative theory can feel unsatisfying. The book ends without a new architecture in place of the old one.
That refusal, however, appears deliberate. Representation and Reality is not an attempt to rebuild metaphysics on safer foundations. It is an argument that certain philosophical expectations about representation, reference, and realism should be abandoned. From that perspective, the book succeeds. It does not tell the reader what reality ultimately is like. It shows why some familiar ways of asking that question fail, and why philosophy may need to lower its ambitions without abandoning rigor.
Hilary Putnam’s Representation and Reality is a critical examination of representational theories of mind and language that dominated analytic philosophy in the mid-20th century. The book is not a general introduction, neither is it a systematic replacement of one theory with another. It is an internal critique, written by a philosopher who had worked extensively within the very frameworks he is now questioning. As an amateur reader, I found the book challenging less because of its prose than because it assumes familiarity with debates that are not always reconstructed from first principles.
The central target of the book is the idea that the mind represents the world by forming internal models that correspond to a fully determinate, mind-independent reality. Putnam refers to this cluster of assumptions as “metaphysical realism.” According to this view, the world has a fixed structure, language hooks onto that structure through reference, and truth consists in correspondence between representations and how things are “in themselves.” Much of Representation and Reality is devoted to showing that this picture fails to deliver what it promises.
Putnam’s best-known argument here is model-theoretic. Very roughly, he argues that no matter how completely we specify the physical facts about the world, there will always be multiple, incompatible ways of assigning reference to our terms that preserve truth conditions. If this is right, then reference cannot be fixed purely by the world’s structure. The idea of a single, privileged mapping between language and reality turns out to be unsupported. While the technical details are demanding, the philosophical conclusion is clear enough even without mastering the formalism: metaphysical realism relies on a notion of determinacy it cannot justify.
What is notable is what Putnam does not do in response. He does not conclude that truth is subjective, that science is merely a social construction, or that all conceptual schemes are equally valid. On the contrary, he insists that objectivity, rational criticism, and empirical constraint remain central. His claim is narrower but more disruptive: objectivity does not require, and cannot be grounded in, a “God’s-eye view” of reality that stands outside all human conceptual practices.
Putnam’s alternative position, often labeled “internal realism,” holds that truth is not correspondence with a ready-made world but idealized rational acceptability within a conceptual framework. As a reader, I found this formulation initially unsatisfying, because it seems to weaken the notion of truth just when philosophy is supposed to secure it. Over the course of the book, however, it becomes clear that Putnam’s point is not to weaken truth but to describe how it actually functions in science and everyday reasoning. We argue, revise, and reject claims from within shared practices, not by comparing our theories to an inaccessible reality as such.
Placed within the broader philosophical landscape, Representation and Reality belongs firmly to the analytic tradition in method and reference points. Putnam engages with formal semantics, philosophy of language, and philosophy of mind, and he takes scientific practice seriously throughout. At the same time, the book distances itself from the foundational ambitions that once defined analytic philosophy. In this respect, Putnam’s position shows affinities with Kant’s insistence that knowledge depends on conditions of intelligibility, and with later critiques of representational thinking found outside the analytic canon, even if Putnam does not adopt their vocabulary or historical framing.
One limitation of the book, especially for non-specialists, is that Putnam often moves quickly over background debates. Readers unfamiliar with logical positivism, Quine’s critiques, or computational models of mind may struggle to see exactly which assumptions are being dismantled at each step. In addition, Putnam’s refusal to present a fully worked-out alternative theory can feel unsatisfying. The book ends without a new architecture in place of the old one.
That refusal, however, appears deliberate. Representation and Reality is not an attempt to rebuild metaphysics on safer foundations. It is an argument that certain philosophical expectations about representation, reference, and realism should be abandoned. From that perspective, the book succeeds. It does not tell the reader what reality ultimately is like. It shows why some familiar ways of asking that question fail, and why philosophy may need to lower its ambitions without abandoning rigor.
April 29, 2025
I was reluctant to read this and I was right. It is written for a functionalist or computationalist theorist of the mind who is a little uncomfortable with his Fodorian beliefs. So there’s tens of pages of Putnam explaining to such people why believing that there’s a lot of cats in the neighborhood can’t be reduced to one or several computationally implementable brain states. (Yawns)
May 19, 2013
I hard time connecting some of the dots.
November 5, 2016
Functionalism and stuff
Displaying 1 - 10 of 10 reviews