A Philosopher’s Sustained Engagement with the Real, the Normative, and the Limits of Scientific Naturalism
Introduction: The Return of the Philosopher
We begin with Putnam’s own orientation to the collection. He is writing a new Introduction for a volume of papers selected from the last decade and a half of his work, and he immediately flags a central preoccupation: the relationship between truth and justification. He tells us that the essays in this book represent a continuation of his earlier turn away from the metaphysical realism he once defended, and toward a position he calls natural realism — a view he associates with the later Wittgenstein, John McDowell, and William James.
“I have come to think that the deepest problems in philosophy are not separated from the problems of life, and that the great philosophers were trying to describe something of great importance about our relation to the world — a relation which, as I have argued, cannot be understood in purely causal terms.”
Putnam explains that his earlier work on internal realism (the view that truth is idealized rational acceptability) was an attempt to steer between metaphysical realism and relativism. But he eventually came to find internal realism unsatisfactory. The problem, as he now sees it, was that internal realism still operated within a framework that treated the mind and the world as separate domains to be “matched up” — it merely changed the criteria for matching. He now wants to articulate a “direct” or “natural” realism about perception and thought that does not interpose representations between us and the world. This involves rehabilitating the notion of normativity as an ineliminable feature of both thought and action. The essays that follow, he tells us, explore these themes from various angles: philosophy of mind, ethics, metaphysics, and the philosophy of science.
What matters here is that Putnam is not offering a disconnected set of occasional pieces. He is presenting a unified philosophical project, one that challenges the dominant scientism of contemporary philosophy — the view that only the methods of the natural sciences deliver genuine knowledge. He argues that normativity — our capacity to be guided by reasons, to judge correctly or incorrectly, to be responsible to standards — cannot be reduced to or explained away by any natural science. Understanding how normativity fits into a naturalistic picture of human beings is, he claims, the central difficulty.
What is Naturalism?
The first essay proper, “What is Naturalism?”, opens with a direct confrontation with the dominant philosophical orientation of the Anglophone world. Putnam observes that “naturalism” is a term of allegiance: most philosophers want to call themselves naturalists, but what they mean by it varies enormously. He distinguishes two broad strains. One, associated with Quine, holds that philosophy is continuous with natural science — there is no first philosophy that could judge science from outside, and the only legitimate methods are those of the sciences. The other, which Putnam associates with John Dewey and the later Wittgenstein, sees naturalism not as the extension of scientific method into every domain, but as the rejection of supernaturalism and the insistence that human beings are natural creatures engaged in natural transactions with their environment.
“The naturalism to which I am drawn, and which I shall describe in this essay, is not a scientistic naturalism; it is, if you like, a humanistic naturalism.”
Putnam presents the content of his own position by contrasting it with Quine’s. Quine’s naturalism, he argues, is built on a disenchanted view of nature: nature contains only what physics and the other sciences describe. This leaves no room for meaning, normativity, or intentionality as primitive features of reality. Quine attempted to naturalize epistemology by making it a branch of psychology — we study how human beings actually form beliefs, and the norms of epistemology become merely descriptive generalizations about our cognitive habits. Putnam insists this misses the point: the question “What ought I to believe?” cannot be answered by a descriptive account of how I do believe. Normativity is not a feature of the world that science can discover; it is something we bring to our inquiries.
Here Putnam introduces a distinction crucial to the whole volume. He argues that the naturalist who rejects the supernatural is not thereby committed to rejecting everything that is not describable in the language of physics. Meanings, reasons, values, and norms are as real as tables and chairs — they are part of the lived environment of human beings. To deny their reality because they do not figure in the physicist’s ontology is to commit what he calls the scientistic fallacy: the mistake of identifying reality with what science (at some idealized limit) would describe. The proper naturalist attitude, he suggests, is to begin from the natural world as we experience it and inhabit it, not from a metaphysical picture derived from science.
What this section of the essay accomplishes is to clear the ground. Putnam is not arguing against naturalism per se; he is arguing for a more generous, less reductive form of naturalism. The key move is to separate naturalism from scientism. A naturalist can believe that everything real is part of the natural order without believing that only the natural sciences can tell us what that order contains. Normativity, on this view, is natural in the sense that it arises from our actual practices of giving and asking for reasons — it is not a supernatural addition to the world. But it is not reducible to anything the natural sciences describe. It requires its own mode of philosophical investigation.
The Content and Appeal of Natural Realism
The second chapter moves from the general question of naturalism to a specific application: the philosophy of perception. Putnam presents what he calls natural realism as a way of understanding perceptual experience that avoids the standard philosophical problems generated by the sense-data tradition and the representationalist alternative. He traces the origins of the traditional view to the early modern period, when philosophers like Descartes and Locke argued that we do not directly perceive objects in the world; we perceive only our own mental representations (ideas, impressions, sense-data) and infer the existence of external objects. This generates the familiar problem of skepticism: how can we ever know that our representations correspond to the world?
“The natural realist holds that successful perception is a direct, non-inferential awareness of things in the world, not an awareness of representations that somehow ‘correspond’ to the world.”
Putnam presents the core argument by examining what he calls the interface conception of perception. On this view, the mind is like a user at a computer terminal: it interacts only with the screen (representations), not with the world outside the computer. So long as we think of perception as involving an interface of representations (whether mental images, sense-data, or neural states), we cannot escape the skeptical worry that we might be trapped behind the interface. Natural realism rejects this picture entirely. When I see a tree, I am aware of the tree, not of a mental image of the tree. The experience is world-involving from the start.
But this raises an obvious question: what about cases of hallucination or illusion? If I hallucinate a pink elephant, I am not perceiving any actual pink elephant in the world. Does that not prove that what I am directly aware of is my own mental content? Putnam responds by distinguishing between the veridical case and the non-veridical case. In the veridical case, the object perceived is part of the causal explanation of the experience. In the hallucinatory case, it is not. But in both cases, the *character* of the experience can be described without invoking a common object (a “sense-datum”) that is present in both. The natural realist can say that in the hallucinatory case, the subject is *as if* perceiving a tree — but that does not require that there be a mental object that they are actually perceiving.
What is at stake here is more than just a technical debate in the philosophy of mind. Putnam is defending the idea that perceptual experience is intrinsically intentional — it is *of* or *about* the world — and that this intentionality is not something we add to a brute physical event. Perception presents us with reality directly, and this directness is the foundation of our cognitive access to the world. If we lose this directness, we lose the very thing that makes knowledge possible. The natural realist picture thus supports a conception of truth as the aim of inquiry: we are trying to get things right about a world that is already given to us in perception, not trying to infer our way out of a prison of representations.
Ethics, Normativity, and the Collapse of the Fact/Value Dichotomy
The third major piece in the collection takes up the notorious divide between facts and values. Putnam has long argued that the dichotomy between descriptive statements (factual) and evaluative statements (normative) is philosophically untenable. He now advances a detailed case for its collapse, drawing on the history of philosophy and the philosophy of language. The standard picture, inherited from Hume and fortified by the logical positivists, holds that factual statements are objective (true or false independently of our attitudes) while evaluative statements are subjective (expressions of emotion or attitude, not candidates for truth). Putnam argues that this picture cannot be sustained.
“The logical positivists’ attempt to draw a sharp line between factual statements and evaluative statements collapsed because it is impossible to separate the two: our factual descriptions of the world are always entangled with our values, and our values are always entangled with our factual beliefs.”
Putnam presents his argument by examining the nature of thick ethical concepts. These are concepts like “cruel,” “brave,” “generous,” “selfish” — terms that seem to carry both descriptive and evaluative content simultaneously. To call an act “cruel” is both to describe it (as causing unnecessary suffering) and to evaluate it (as morally wrong). The logical positivist strategy was to split such concepts into a factual component plus an emotional attitude — but Putnam argues this misrepresents their functioning. The evaluative aspect is not detachable; it is part of the meaning of the term. When we learn to use “cruel,” we learn both what counts as a case of cruelty and that cruelty is to be condemned. The evaluation is internal to the concept.
He extends this point to the philosophy of science. Even in the “hard” sciences, he argues, we cannot do without epistemic values — values like coherence, simplicity, predictive accuracy, and explanatory power. These are not merely subjective preferences; they are norms that guide rational inquiry. But they are not themselves empirical discoveries. The claim that simplicity is a virtue in a theory is not a claim that can be tested by experiment. It is a normative commitment about what counts as good science. If we try to reduce these epistemic values to something purely factual (e.g., “simpler theories are easier to work with”), we lose the normative force: we want to say that scientists *ought* to prefer simpler theories, not merely that they do.
What Putnam is driving toward is a pragmatist picture in which facts and values are not two separate realms but two aspects of a single human practice of inquiry and evaluation. We do not first discover the facts, then add our values. We encounter the world already shaped by our interests, our norms, and our standards of evaluation. This does not mean that “anything goes” — we can be wrong about values, just as we can be wrong about facts. But it does mean that the fact/value dichotomy is a philosophical artifact, not a deep feature of reality.
The Question of Realism in the Philosophy of Science
The fifth chapter (following further essays on Wittgenstein and James) turns explicitly to the philosophy of science. Putnam here revisits the realism/anti-realism debate that occupied so much of his earlier career. He distinguishes between several varieties of realism. Metaphysical realism, which he once defended, holds that there is a single, true description of the world that is “ready-made” independently of any conceptual scheme. Internal realism, his own earlier view, held that truth is idealized rational acceptability within a conceptual scheme, and that there can be many such schemes. But Putnam now finds both views inadequate. He argues for a position he calls pragmatic realism, which combines a robust commitment to the reality of the entities posited by successful scientific theories with the recognition that our descriptions are always from a human perspective.
“The question is not whether we can know the world ‘as it is in itself,’ but whether we can distinguish better and worse descriptions of the world from within our practices of inquiry.”
Putnam presents the argument by examining the debate about referential success in science. The pessimistic meta-induction (associated with Laudan) argues that since past successful theories (like Newtonian mechanics) turned out to be false about the entities they posited (e.g., absolute space), we should expect present successful theories to be similarly false. Therefore, we should not be realists about unobservable entities. Putnam responds by distinguishing between the *truth* of a theory and its *referential success*. Newtonian mechanics was wrong about many things, but it was referring to real entities (mass, energy, forces) in a way that was refined rather than replaced by Einstein. The world has a causal structure that our theories aim to track, and we can judge our theories by how well they enable us to cope with and predict that structure.
However, Putnam rejects the metaphysical realist’s claim that there is a single, unique way the world is “in itself.” He draws on his earlier work on the multiple realizability of functional states to argue that the world supports many different, equally true descriptions. The description in terms of physics, the description in terms of biology, and the description in terms of everyday objects are all true in their own domains. They are not in competition. The mistake of metaphysical realism is to think that one of these descriptions (physics) is the “really real” one, and the others are mere shadows. Putnam’s pragmatic realism accepts that all successful inquiry gives us knowledge of the real, but it is knowledge of the real *as it shows up for creatures like us*, with our interests and our cognitive capacities.
This section deepens the central theme of the volume: the rejection of a “God’s eye point of view” in favor of a human perspective that is nonetheless genuinely cognitive. We are not cut off from reality; we are engaged with it. But our engagement is always from a standpoint, and that standpoint is shaped by normative commitments — by what we take to matter, what we care about, what we count as a reason. The realism worth having, Putnam concludes, is one that acknowledges the ineliminability of normativity even at the heart of science.
The Varieties of Naturalism: A Critique of Scientism
We are now positioned to examine Putnam’s detailed engagement with the specific forms of naturalism that have dominated late twentieth-century philosophy, particularly in the United States. This section functions as a critical taxonomy, distinguishing between defensible and indefensible versions of naturalism.
Putnam begins by distinguishing between what he calls “modest naturalism” and “scientistic naturalism.” Modest naturalism, which he associates with figures like John Dewey and William James, holds that there is no supernatural realm and that philosophical inquiry must be continuous with scientific inquiry in a broad sense. But it does not reduce all legitimate questions to those answerable by the natural sciences. Scientistic naturalism, by contrast, makes a much stronger claim: it asserts that the only real objects, properties, and facts are those that can in principle be recognized by the natural sciences, especially physics. This position, Putnam argues, collapses into a form of metaphysical realism that is ultimately self-undermining.
“The scientistic naturalist holds that the only things that are ‘real’ are the objects and properties that figure in the scientific image of the world; but what is the status of the normative principles that guide scientific inquiry itself?”
The problem is that scientistic naturalism cannot account for the norms that make science possible. Scientific inquiry presupposes standards of evidence, coherence, explanatory power, and truth. These are not themselves empirical discoveries; they are normative commitments that structure the very practice of inquiry. If we try to treat them as merely psychological or sociological facts about what scientists happen to do, we lose the normative force that makes them *standards* rather than mere regularities.
Putnam draws on the work of Wilfrid Sellars to sharpen this point. Sellars famously distinguished between the “space of reasons” and the “space of causes.” The space of reasons is the domain of justification, evaluation, and normative assessment — where we ask whether a belief is warranted, whether an inference is valid, whether an action is right. The space of causes is the domain of empirical lawful explanation — where we ask what events produced what other events. The core claim of scientistic naturalism, in its most aggressive form, is that the space of reasons can be “naturalized” — that is, reduced to or eliminated in favor of the space of causes. Putnam argues that this project has failed, and must fail, because the very activity of offering a *reduction* or an *elimination* takes place within the space of reasons. The naturalist who claims that norms are “really just” biological or social facts is still making a normative claim: that such a reduction is *correct*, that the arguments for it are *good*, that we *ought* to accept it.
This section clarifies why Putnam’s rejection of scientism is not a rejection of naturalism as such. The modest naturalist can recognize the full reality of normative phenomena — of truth, justification, meaning, value — without positing supernatural entities. What the modest naturalist rejects is the idea that something is “real” only if it can be captured by the vocabulary of physics. The world contains many kinds of things — persons, artworks, institutions, obligations, reasons — and to exclude them from “ontology” on the basis of a prior commitment to physicalism is to let metaphysics dictate to experience, rather than allowing experience to inform metaphysics.
The Intertwining of Fact and Value
We now turn to one of the most sustained and influential arguments in the collection: the critique of the fact/value dichotomy. This section occupies a central position in the volume because it brings together Putnam’s work in ethics, philosophy of language, and philosophy of science. The target is the empiricist picture inherited from Hume and amplified by logical positivism, according to which “facts” are objective and “values” are mere expressions of emotion or attitude.
Putnam begins by examining the standard examples used to illustrate the fact/value distinction. Consider the statement “This knife is good.” According to the fact/value dichotomy, this statement combines a factual component (“this knife cuts well”) with an evaluative component (“cutting well is good”). But Putnam argues that this analysis is misleading because it presupposes that we can neatly separate the descriptive and evaluative dimensions of our concepts. Many concepts, which he calls “thick ethical concepts,” are irreducibly both descriptive and evaluative. The concept of cruelty, for example, describes a certain kind of conduct (causing unnecessary suffering) and at the same time evaluates it negatively. To call an act cruel is to say something about what it is and to condemn it. You cannot understand the concept without understanding both dimensions.
“The attempt to split thick ethical concepts into a ‘descriptive meaning component’ and a ‘prescriptive meaning component’ founders on the fact that the two components are not separable in the way the theory requires.”
Putnam extends this argument by examining the case of scientific inquiry itself. The idea that science proceeds purely on the basis of “facts” without any “values” is, he argues, a myth. Scientific inquiry is guided by epistemic values — values like coherence, simplicity, explanatory power, and empirical adequacy. These are values in the full normative sense: they tell us what counts as a *good* theory, what standards we *ought* to apply when choosing among hypotheses. If we try to say that these are not “really” values but merely “functional” or “instrumental” norms, we are still using normative language. The attempt to eliminate normativity from science leads to a vicious regress: the naturalist who claims that all norms can be reduced to facts must, in making that claim, be offering a reason for accepting it — and that reason itself invokes norms of argumentation and evidence.
The section introduces a crucial historical reference: the work of the economist and philosopher Amartya Sen on the “fact-value entanglement” in practical reasoning. Sen argued that many of our factual judgments about well-being, welfare, and justice are themselves evaluatively loaded. One cannot describe a person as “well-nourished” or “literate” without implicitly taking a stand on what counts as a good human life. Putnam uses this to show that the dichotomy is not only philosophically suspect but also practically harmful: it encourages a picture of social science as “value-free” when in fact all inquiry is guided by normative commitments.
This section demonstrates that the collapse of the fact/value dichotomy is not just a point about ethics but about the nature of rationality itself. If epistemic values are genuine values, then the sharp separation between “fact” and “value” collapses, and the project of reducing all normativity to empirical description must fail. The world we inhabit and inquire into is not value-neutral; it is shot through with normative significance.
The Question of Realism in Ethics
The volume now turns directly to the question of moral realism: whether ethical judgments can be true or false in a robust sense, and whether there exist ethical properties and facts that are independent of human attitudes and conventions. This section draws on the prior arguments about thick ethical concepts and the entanglement of fact and value to develop a distinctive position.
Putnam distinguishes his view from two more familiar positions. On one side is ethical non-cognitivism, the view associated with A. J. Ayer and Charles Stevenson, which holds that ethical statements are not really statements at all but expressions of emotion or attempts to influence behavior. On the other side is ethical naturalism, the view associated with Peter Railton and Richard Boyd, which holds that ethical properties can be reduced to natural properties (such as well-being or evolutionary fitness). Putnam rejects both. Against non-cognitivism, he argues that ethical discourse is genuinely truth-apt: we argue about what is right and wrong, we give reasons, and we can be mistaken. Ethical disagreement is not just a clash of attitudes; it often reflects genuine cognitive disagreement about what the facts are. Against naturalism, he argues that ethical properties cannot be reduced to natural properties without remainder because ethical concepts are normative through and through.
“The deepest problem with ethical naturalism… is not that it fails to locate ethical facts in the natural world, but that it fails to do justice to the normativity of ethical discourse — to the sense in which ethical claims are about what we ought to do, not just about what we in fact do or desire.”
Putnam’s alternative is a version of what he calls “pragmatic realism” about ethics. This position holds that ethical judgments can be objectively true or false, but that the standards for truth in ethics are internal to the practice of ethical reasoning itself. We do not need to appeal to a supernatural realm of moral facts to make sense of ethical objectivity; we need only to recognize that ethical inquiry, like scientific inquiry, is governed by norms of reason-giving that are not reducible to anything else. The objectivity of ethics is the objectivity of reasons, not the objectivity of objects.
This section clarifies the overall trajectory of the volume. Putnam is not arguing that there is no difference between facts and values, but that the difference is not a sharp dichotomy. He is also not arguing for a radical relativism: there are better and worse ethical views, and we can give reasons for preferring one over another. The realism worth having in ethics is a realism that acknowledges the ineliminability of normative standards while also recognizing that those standards are not discovered by a faculty of “moral intuition” separate from our ordinary cognitive capacities. Ethical knowledge arises from the same kind of reflective inquiry that gives us knowledge in any domain — a process of reasoning, testing, and revising in light of experience and argument.
Realism and Human Understanding
Putnam now moves from the specific case of ethics to a broader epistemological question: what does it mean to be a realist about anything, given the kind of creatures we are? This section functions as a bridge between his critique of fact/value dichotomies and his positive account of what realism should look like when it is not distorted by metaphysical prejudices.
Putnam argues that the traditional problem of realism has been framed in the wrong terms. The standard debate — between metaphysical realism (the view that there is one complete and true description of the world, independent of any conceptual scheme) and relativism (the view that truth is relative to a conceptual scheme) — rests on a shared assumption that Putnam rejects. Both sides assume that if there is no single “God’s-eye view” of the world, then there is no objective truth at all. Putnam calls this the “disastrous” dichotomy, and he insists that we must reject both horns.
The key to breaking out of this dilemma is to recognize that our conceptual schemes are not optional lenses that distort an otherwise accessible reality. They are, rather, part of how we engage with the world. Putnam draws on the later Wittgenstein’s insight that language is not a picture of reality but a form of life — embedded in practices, activities, and shared forms of judgment. This means that the categories we use — objects, properties, causes, values — are not imposed on a passive world by our minds, nor are they simply read off the world by passive perception. They emerge from our interaction with the world, and they are answerable to that world.
Putnam introduces the notion of “internal realism” — the view that the question “What objects does the world consist of?” only makes sense within a conceptual scheme. This is not a denial of an external world; it is a denial that we can step outside all conceptual schemes to compare them with “the world as it is in itself.” The internal realist holds that truth is a kind of idealized rational acceptability — what it would be rational to accept under epistemically ideal conditions. This is not a reduction of truth to what any particular community happens to believe, but an acknowledgment that truth and justification are internally related.
“‘Truth’ is not a name for a metaphysical relation between a representation and what it represents; it is a normative notion, a notion that is inseparable from the idea of rational acceptability.”
This passage captures the core of Putnam’s internal realism. Truth is not a matter of “correspondence” in some metaphysical sense — a word-to-world relation that we can never check — but is tied to the norms of inquiry themselves. When we say a statement is true, we are claiming that it would survive all rational scrutiny, that it is what we ought to believe. This does not make truth subjective or relative: the standards of rational acceptability themselves are objective in the sense that they are not up to us individually, and they can be revised in light of experience. But they are also not given by a metaphysical “view from nowhere” — they are the standards we actually have, developed through our practices of inquiry.
The clarification Putnam offers is that internal realism is not a form of anti-realism. It does not deny that there are facts about the world, or that our claims can be true or false. What it denies is that there is a single, privileged description of the world that corresponds to its “real” structure. The world does not come pre-divided into objects and properties; our ways of dividing it up are partly our own, and they are answerable to our interests and purposes as well as to the world’s resistance. This means that realism about the external world is compatible with a recognition that our conceptual schemes are historically conditioned and fallible.
The Disenchantment of the World Thesis
Putnam now turns to a specific historical narrative that he believes has distorted our understanding of realism and normativity: the claim that the “disenchantment of the world” — the idea that modern science has stripped the world of meaning, purpose, and value — is a necessary consequence of the scientific revolution. This section examines that narrative critically, arguing that it depends on a philosophical interpretation of science rather than on science itself.
The disenchantment thesis is most famously associated with Max Weber, but Putnam traces its philosophical roots to a particular reading of Galileo and Descartes. On this reading, the scientific revolution introduced a distinction between primary qualities (shape, size, motion, number) and secondary qualities (color, sound, taste, smell). The former were taken to be “real” properties of objects themselves, while the latter were regarded as merely subjective — projections of our sensory apparatus onto a world that is, in itself, colorless, soundless, and odorless. Putnam argues that this distinction, and the metaphysical picture it supports, is not a consequence of scientific discovery but of a philosophical decision to identify reality with what physics describes.
Putnam points out that the primary/secondary quality distinction has been deeply problematic since its inception. Berkeley already showed that the distinction is unstable: if secondary qualities are subjective because they depend on the perceiver, then the same argument applies to primary qualities, which are also perceived through the senses. Moreover, the notion that physics gives us a “complete” description of the world is itself a metaphysical claim, not a scientific one. Physics describes certain aspects of reality — those that can be mathematized and subjected to experimental control — but this does not mean that what physics describes is all that there is.
The disenchantment narrative also rests on a particular view of causation that Putnam rejects. On the mechanistic worldview that emerged from the scientific revolution, causation is understood as efficient causation — the push-and-pull of matter in motion. This leaves no room for teleological causation — explanation in terms of purposes, ends, or functions. If the world is nothing but particles in motion, then there is no place for purposes, meanings, or values in the fabric of reality. They become, at best, human projections onto a purposeless universe.
“We have been taught that the world is ‘disenchanted’ — that is, devoid of meaning and value — and that the only knowledge worthy of the name is scientific knowledge.”
Putnam argues that this teaching is itself a form of scientism — the belief that the methods of the natural sciences provide the only legitimate form of knowledge. Scientism is not a scientific claim but a philosophical position, and it is a position that has had enormous influence on how we think about normativity. If the world is disenchanted, then values, meanings, and norms must be either illusions or projections of our own subjectivity. The fact/value dichotomy is thus a direct consequence of the disenchantment thesis: facts are what science discovers, and values are what we project.
But Putnam insists that this is not the only way to understand science. The scientific image of the world is one among many ways of describing reality, and it is not in competition with our ordinary descriptions of persons, actions, and values. The world as described by physics does not contain “tables” or “chairs” or “promises” or “cruelty” — but that does not mean that tables, chairs, promises, and cruelty are not real. They are real in the sense that they are part of our experience and our practices, and they are answerable to norms of inquiry just as much as physical objects are. The mistake is to think that physics gives us the “real” description and all others are merely subjective.
The Collapse of the Fact/Value Dichotomy and Its Implications
Having traced the fact/value dichotomy to the disenchantment thesis, Putnam now devotes a section to showing in detail why the dichotomy collapses and what the implications of that collapse are for philosophy. This is not merely a negative argument — a demonstration that the dichotomy fails — but a positive account of how facts and values are interwoven in our cognitive practices.
Putnam begins by examining the standard argument for the fact/value dichotomy, which he traces to Hume’s claim that you cannot derive an “ought” from an “is.” This claim is often taken to mean that there is a logical gap between descriptive and normative statements: from purely factual premises, no evaluative conclusion follows. Putnam does not dispute this logical point in its narrowest form — he agrees that you cannot deduce an “ought” from an “is” without some normative premise. But he argues that this logical gap does not imply a metaphysical gap between facts and values. It does not show that values are not genuine features of reality, or that evaluative statements are incapable of being true or false.
The deeper problem, Putnam argues, is that the fact/value dichotomy conflates two different distinctions: the distinction between factual and evaluative statements, and the distinction between objective and subjective. The assumption is that factual statements are objective (capable of being true or false) while evaluative statements are subjective (mere expressions of attitude or preference). But this assumption is itself a value judgment — a normative claim about what counts as genuine knowledge. And once we see that, the dichotomy begins to unravel.
Putnam offers a series of examples to show how facts and values are entangled in our everyday language. Consider the word “cruel.” Calling an action cruel is both a description (it involves causing unnecessary suffering) and an evaluation (it is morally wrong). The description and the evaluation cannot be pried apart: you cannot specify what cruelty is without using normative concepts, and you cannot evaluate cruelty without knowing what it is. The same is true for terms like “brave,” “generous,” “cowardly,” or “dishonest.” These are “thick ethical concepts” — concepts that are at once descriptive and evaluative.
“The idea that we can neatly separate the factual content of a statement from its evaluative content is an illusion that has persisted in philosophy for far too long.”
Putnam extends this analysis to the language of science itself. Even the most “factual” scientific descriptions are permeated by normative concepts. When we say that a hypothesis is “well-confirmed” or that a theory is “elegant” or “simple,” we are making evaluative judgments — judgments about what counts as good evidence, what counts as a satisfactory explanation. These judgments are not arbitrary; they are governed by norms that are themselves part of scientific practice. But they are not reducible to “brute facts” either. The norms of scientific inquiry — such as the norm that we should prefer simpler theories — are not discovered by empirical investigation; they are presupposed by it.
The collapse of the fact/value dichotomy has profound implications for how we understand objectivity. If values are not merely subjective projections, then there is room for a kind of normative objectivity — the idea that some values are better than others, that some ethical judgments are more rational than others. This does not require a Platonist metaphysics of moral facts existing independently of human beings. It requires only that we recognize that our evaluative practices are answerable to reasons, that we can argue about values, and that some arguments are better than others.
Putnam concludes this section by noting that rejecting the fact/value dichotomy does not commit us to any particular ethical theory. It does not tell us whether utilitarianism or Kantianism or virtue ethics is correct. What it does is remove a philosophical obstacle to taking ethics seriously as a domain of rational inquiry. If values are not mere projections, then ethical reasoning is not a sham — it is a genuine form of inquiry into what matters, what is good, and how we should live.
The Significance of the Fact/Value Entanglement for Metaphilosophy
We are in the culminating movement of the work, where Putnam draws out the metaphilosophical consequences of the collapse of the fact/value dichotomy, applying the lessons learned to the practice of philosophy itself.
Putnam argues that the fact/value entanglement has profound implications for how we understand the task of philosophy. The traditional view, which he has spent the entire book dismantling, holds that philosophy must model itself on natural science. On this view, scientific facts are objective and value-free, while values are subjective and beyond rational adjudication. Philosophy’s job, then, is either to become a branch of science (naturalism) or to restrict itself to the logical analysis of scientific language (scientism). But if the fact/value dichotomy is untenable, this whole picture collapses. Putnam writes: “The dichotomy between fact and value is not a distinction that can be drawn cleanly; but the denial of the dichotomy does not commit us to the view that there are not genuine differences between statements that we take to be factual and statements that we take to be evaluative.” The point is not to eliminate the distinction, but to recognize that it is a matter of degree and context, not of sharp metaphysical kinds.
This realization changes what philosophy can and should be. If facts and values are entangled, then philosophy cannot simply be a handmaiden to science. It must take seriously the normative dimension of human life — the realm of reasons, meanings, and evaluations. Philosophy becomes, in Putnam’s view, a discipline that must grapple with the full complexity of human experience, including the ethical, aesthetic, and conceptual dimensions that natural science cannot capture. The philosopher’s task is not to reduce these dimensions to something simpler, but to understand how they hang together in our actual practices. This is what Putnam means by “realism with a human face” — a realism that acknowledges that our cognitive and evaluative practices are irreducibly human activities, answerable to reasons that are not reducible to physical laws.
The implications are also negative: many philosophical problems that seem intractable are artifacts of a bad metaphysic. The problem of free will, for example, only seems insoluble if we assume that causation must be either deterministic or random, and that moral responsibility requires a special kind of causation that fits into the scientific image. Once we recognize that our normative concepts — responsibility, freedom, obligation — are not competing with scientific explanation but operating in a different register, the problem dissolves. Similarly, the problem of moral objectivity disappears once we see that ethical reasoning does not need to mirror scientific reasoning in order to be objective. It has its own standards of evidence, internal coherence, and responsiveness to experience.
Putnam is not arguing that philosophy should abandon rigor or rationality. On the contrary, he is arguing for a more demanding conception of rationality — one that does not restrict itself to what can be modeled by formal logic or empirical science. This is a rationality that includes interpretation, judgment, and the weighing of incommensurable values. It is the kind of rationality we exercise when we try to decide what kind of life to live, what kind of society to build, or what kind of person to be. Philosophy, on this view, is continuous with this broader human rationality. It is not a specialized technique but a reflective engagement with the fundamental questions of life, conducted with the utmost care and honesty.
The End of Philosophy as Epistemology
In the final sections of the book, Putnam addresses the fate of epistemology itself — the branch of philosophy that has traditionally been concerned with the nature, sources, and limits of knowledge. He argues that the collapse of the fact/value dichotomy, combined with his defense of natural realism, signals the end of epistemology as it has been practiced since Descartes.
Traditional epistemology, in Putnam’s diagnosis, is built on a foundationalist model: the mind is a repository of representations (ideas, sense-data, beliefs), and knowledge consists in having these representations correspond to an external reality. The problem that has haunted this tradition is how to get from the representations to the world — how to justify the claim that our ideas actually represent what is out there. This is the problem of skepticism, which Descartes took as his starting point: if all I have access to are my own mental states, how can I know anything beyond them? Putnam argues that this entire problematic rests on a mistake — the mistake of thinking that perception is a process of forming internal representations that must then be linked to external objects.
The alternative, which Putnam has been developing throughout the book, is a direct realism or natural realism about perception. On this view, when I see a tree, I am directly presented with the tree itself, not with a mental image of the tree. The idea that there is a “veil of perception” between the mind and the world is a philosophical construction, not a phenomenological given. Once we accept that perception is a direct relation to objects, the skeptical problem dissolves — or rather, it is shown to be an artifact of a bad theory. We do not need to infer the existence of the external world from our inner experiences, because our experiences are already experiences of the external world.
However, Putnam is careful to note that this does not mean skepticism is refuted once and for all. It means that the epistemological project of proving that we know the external world is based on a misunderstanding of what perception is. The skeptic’s question — “How do you know that you are not dreaming?” — only seems pressing if we assume that our perceptual experiences are internally indistinguishable from dreams or illusions. But in real life, we have criteria for distinguishing dreams and reality — criteria that involve coherence, context, and ongoing interaction with the world. These criteria are not infallible, but they are good enough for the purposes of ordinary life and science. The demand for absolute certainty is itself a philosophical artifact.
Putnam concludes that the task of philosophy is not to provide a foundation for knowledge that is immune to doubt. Rather, it is to understand the actual workings of our cognitive practices — to describe how we acquire knowledge, how we correct errors, and how we justify our claims — without imposing an artificial standard of certainty. This is a pragmatist and naturalist epistemology, but one that is naturalist in the Deweyan sense: it starts from our actual practices and asks how they can be improved, not from a metaphysical picture of what knowledge “must” be.
The Return to the Everyday
The book ends with a reflection on the relationship between philosophy and ordinary life. Putnam argues that philosophy has too often alienated itself from the everyday world, treating ordinary experience as naïve or pre-scientific. This is the attitude he calls scientism: the belief that science, and only science, tells us what the world is really like, and that our ordinary ways of talking about tables and chairs, persons and values, are at best approximations to scientific truth, at worst illusions.
Against this, Putnam insists on the authority of the everyday. The world we live in is the world of middle-sized objects, of persons with intentions and values, of art and morality and love. This world is not an illusion that science will someday explain away. It is the world within which science itself is practiced. The scientific image is a refinement and extension of our ordinary ways of knowing, not a replacement for them. As Putnam puts it: “The world is not a product of our minds; but it is a world that we are in, and it presents itself to us through our forms of life.” The task of philosophy is to do justice to this world — to capture its richness, its complexity, its normative dimensions — without reducing it to something that fits into a pre-defined metaphysical schema.
This does not mean that philosophy must accept commonsense uncritically. It means that commonsense is the starting point and the testing ground for philosophical reflection. When philosophy tells us that we do not really perceive objects, or that our values are mere projections, or that our sense of free will is an illusion, it has lost its bearings. The burden of proof lies on the philosopher to show why we should abandon our ordinary understanding. And as Putnam has argued throughout the book, the arguments for such revisions almost always rely on a false dichotomy — typically, the fact/value dichotomy or the dichotomy between the scientific image and the manifest image.
The final note of the book is one of integration. Philosophy must become a discipline that respects both the authority of science and the authority of the normative. It must recognize that our lives are shot through with value, that our knowledge is always situated and fallible, and that the search for truth is inseparable from the search for meaning. This is not a retreat from rigor but a demand for a richer, more human form of rationality. It is, in the end, an invitation to see philosophy not as a technical exercise in conceptual analysis or a branch of cognitive science, but as a reflective engagement with the deepest questions of human existence — questions that cannot be evaded by declaring them meaningless.
