Popper’s World 3

I am ready to discuss with you induction, or world 3 for example. (But not the history of my ideas.)

Karl Popper first publicly voiced what would become his world 3 theory in his famous London School of Economics (LSE) seminar on November 15, 1960. He proposed that the products of the human mind can be considered a third world, both autonomous of the physical world (world 1) and of the world of subjective experiences (world 2), and real, because it can produce effects on both the other worlds. Even colleagues and students deeply steeped in his way of thinking were taken utterly by surprise. He had already hesitated for years before introducing the idea, and the bewildered and unwelcoming reaction of his seminar held him back several more years from addressing a wider audience. Yet, with the advantage of hindsight, we can see how so much in Popper’s work and thought led up to what seemed to be his startlingly new proposal.

Popper has been quoted as saying his notion of world 3 was “the philosophy of my old age” (Miller 1997, 395). What he actually wrote was that it “plays such a role in the philosophy of my old age” (Popper 1969 [1974], 47, sec. 13). Far from being the philosophy of his dotage, as some sneered, his world 3 proposal, once fully shaped, became a keystone of his thought, holding its arches together, unifying and extending his ideas, allowing a comprehensive vision of our place in the universe. Hans-Joachim Niemann (2012a, 12-13) goes so far as to propose, not without reason, that Popper’s work on knowledge and the body–mind problem, which revolves around world 3, constitutes his uncollected and unrecognized third masterpiece, after his breakthrough works in the philosophy of science, Logik der Forschung (Popper 1934), and in social philosophy, The Open Society and Its Enemies (Popper 1945). ²

In the “Replies to His Critics” section in the Library of Living Philosophers series, Popper responded to John Watkins’s proposal that the unity of his philosophy lay in its indeterminism: “I should be inclined to regard my emphasis on criticism . . . as being more important than indeterminism is to my theoretical and practical thinking” (Popper 1971 [1974], 1053). We can fruitfully reconcile the two views. What drives Popper, I propose, is the desire to argue for an active role for open minds in an open universe. He therefore emphasizes the free invention of ideas and makes the free discussion, criticism, and testing of ideas central. We may crave sure foundations leading inexorably up to impregnable results, but, apart from this being impossible, it would be a forced march, not a free exploration. Instead, we venture forth freely, often starting from where others have reached, only for reality so often to kick back at us. The world of physical reality, not designed by or for us, cannot point us to the truth, but it can be the standard against which we assess what we propose. Or more pithily: bold conjectures tempered by stiff criticism is Popper’s method; open minds discovering more about an open universe is his goal.

Popper writes in his autobiography: “I think that I was always a Cartesian dualist (although I never thought that we should talk about ‘substances’)” (Popper 1969 [1974], 149, sec. 39). He adds, “and if not a dualist, I was certainly more inclined to pluralism than to monism.” From the time he became conscious of the issues, then, he rejected the denial of either matter or mind, both the mentalist monism that prevailed among philosophers in the world into which he was born and the physicalist monism that would prevail in his adult years. But he did not feel it was a problem to whose solution he could contribute: “My earliest encounters with the body-mind problem made me feel, for many years, that it was a hopeless problem” (Popper 1969 [1974], 149, sec. 39). Only once he could clearly formulate the theory that human minds engage (in world 2) with the products of human minds (in world 3), with the force of theories, proposals, arguments, criticisms, and more, did he feel that he could answer the body–mind problem in terms of interactions among the physical, the mental, and the cultural worlds.

When Popper was 17 or 18, in 1920, attending classes at the University of Vienna without being formally matriculated, his early thinking in this area took a decisive new step. He began to formulate for himself a distinction between an “objective” attitude to music, which he saw in Bach, and a “subjective” one, which he saw in Beethoven. As he later wrote, “The distinction between an objective and a subjective view of one’s work became most important for me; and it has, I may say, coloured my views of the world and of life, ever since” (Popper 1969 [1974], 53, sec. 13). In the objective attitude to one’s work that he saw ideally in Bach, the artist could learn and grow in the process of composition, transcending himself through encountering difficulties in the work as it was being written down and through trying to overcome them, through accepting self-criticism, criticism from others, and criticism, as it were, from the prior structures, traditions, and values of music: “There will be a constant give-and-take between the artist and his work rather than a one-sided ‘give’” (Popper 1969 [1974], 50, sec. 13).

Along with the predispositions to dualism or more, and to objectivism, another key predisposition of Popper’s was toward indeterminism and, beyond that, freedom. He was always a determined indeterminist. He became convinced, at least by the mid-1920s, by “strong philosophical arguments, partly logical and partly metaphysical, against determinism,” and, decisively, of “the weakness of ‘scientific’ determinism” (Popper 1982, 42). He was always strongly inclined to emphasize the freedom in human action, especially in action at its most creative, as in artistic or scientific discovery. Indeed, that was surely a motive behind his rejection of induction, which for him implied a kind of automatic stamping of impressions on passively observing minds. But he knew that, like the body–mind problem, the discussion of free will was another philosophical quagmire.

About 1926, he began one-on-one discussions with his philosophy professor, Heinrich Gomperz, at Gomperz’s home. He recalls that he was

at first puzzled and, indeed, shocked when . . . Gomperz pointed out to me . . . that my insistence that I was not interested in subjective beliefs but in scientific or philosophical theories and in the objective arguments for and against them showed that I was not a “nominalist” but rather a “realist” (KPA 134.11)

in the sense of Plato, who believed in the reality of Forms or Ideas—of concepts, and their meanings or essences. (Popper 1969 [1974], 146, sec. 38)

Antiessentialist as he was, Popper “did not like the suggestion” (Popper 1969 [1974], 146, sec. 38), but later—perhaps decades later—he came to recognize that Gomperz was right, and to see that he could accept “the reality and the utmost importance of a third world of theories in themselves, and problems in themselves, besides the world of physical and mental states, without accepting that words, and their meanings, or Plato’s ideas” were part of it (KPA 134.11).

Popper’s breakthrough discoveries between 1930 and 1934, in Die beiden Grundprobleme der Erkenntnistheorie (The Two Fundamental Problems of the Theory of Knowledge, Popper 1930-33 [1978]) and Logik der Forschung (The Logic of Scientific Discovery, Popper 1934), can now be seen to foreshadow key elements of world 3. Both books proposed a shift from a subjectivist basis for knowledge, in individual perception, to an objectivist standard for testing knowledge. Both argued that the key resource for the growth of knowledge was not observation but the expanding, open-ended library of problems and solutions, factual and methodological, in a given area, and the new problems that could be discovered within that library. Both books identified criticism as the sharpest goad to future discovery, through showing flaws in old solutions and perhaps even new proposals. Both stressed that the objectification of knowledge through linguistically formulated theories—preferably written, and, even better, published—was necessary both for the proposer and for others to be able to criticize proposals. Popper himself notes that “the underlying ideas” of world 3 “played a decisive role in my early work.” Long before he introduced the terms of worlds 1, 2, and 3, he recollected, he regarded “as central the fact that world 3 largely controls world 2” (Popper 1971 [1974], 1122), through language, theories, values, technological designs, and all the other public products of human minds.

While he worked on Logik der Forschung, Popper engaged with Bolzano’s (1837) Wissenschaftslehre, and he would later write that “Bolzano’s distinction between statements in themselves and subjective thought processes has always seemed to me of the greatest importance” (Popper 1969 [1974], 143-44, sec. 38). . (“Has always seemed to me,” I might note, offers little help to a biographer or a historian of ideas seeking to date an idea or attitude.) He adds, in his autobiography, that he “had developed certain views of Bolzano’s . . . into a theory of what I called the ‘third world’” (Popper 1969 [1974], 150, sec. 39). When he did this remains unclear, although it was probably not until the 1950s.

It could have been shortly after Logik der Forschung, when he learned from Alfred Tarski, the year after Logik, about Tarski’s rehabilitation of the notion of truth. ³ He writes in a draft of his autobiography: “My own approach to the third world was, largely, through Tarski’s theory of truth” (KPA 137.5; he also notes, in KPA 134.5, “Truth as the first value of world 3”). But he continues in that draft: “and through my own theories of problems, of theories, and of criticism.” Although problems, theories, and criticism were important already in Logik, he developed his “theories of problems, of theories, and criticism” gradually, the next step being shortly after he arrived in Christchurch in 1937, in “What Is Dialectic?” (Popper 1940).

He had gone to New Zealand to keep out of Hitler’s reach. As war broke out, and at a time when many intellectuals and others assumed that whatever the advantages democracy may offer, authoritarian systems must be more efficient, Popper began to write what would become The Open Society and Its Enemies. Here, he turned from science to society: from the advantage of criticism and of the free criticizability of ideas in science to similar advantages in society, and from challenging rather than justifying received authority in science to the same in society. His inclinations were similar, his arguments very different in their details.

In the years between completing The Open Society and introducing the three worlds idea to his LSE seminar in 1960, Popper published no major new books, only the book version of The Poverty of Historicism and the English translation of Logik der Forschung. But many separate impulses moved him toward the disclosure to his seminar.

In 1944, not long after completing The Open Society, Popper wrote to a young student, Marianne Fillenz, who had left Christchurch, where they had become friends, for Otago University College in nearby Dunedin. When she reported a lecture on esthetics by John Findlay, Popper responded by elaborating his old contrast between a subjective and an objective outlook in art in an astonishingly Platonic way:

Do you remember Plato’s “Theory of Ideas?” You know that I do not like Plato, but in this matter, I believe in a kind of Platonism. . . . There are different ways in which an artist may look at his work. The one attitude, which I believe to be the only true one . . . I shall call the “Platonic attitude.” . . . A composer [like Bach] who begins to write a fugue [for example] “knows” that this fugue already exists; its divine original exists in heaven. He tries to copy it, though he knows he can never fully succeed; for his attempts mean interference, mean labour; it is not quite easy to copy the “Idea,” and any mistake made in copying may confuse or even destroy his grasp of the “Idea.” Thus his attitude must be this: he knows that anything added by himself—for example, the covering up of a place where he did not succeed in copying the original—must be bad. If possible, they should be excluded altogether. . . . You will see that the “Platonic attitude” is one of simplicity, modesty, humility. ⁴

For Popper, this is an unusually Platonic sense of world 3, quite different from the active groping through trial and error, through proposal and criticism, that we find in his discussions of scientific discovery. The idea of copying from an ideal, already fully formed original is something of a commonplace among artists, even artists who dislike Plato. Popper here conflates this traditional idea of the artwork as somehow preceding its actualization with his long-standing sense of the ideal artist as objectivist rather than subjectivist, to stress, on one hand, the secondariness, the unimportance of self, and, on the other, the primacy of the work, what we can with hindsight call the world 3 object.

Marianne Fillenz was already a student of the neuroscientist Jack Eccles, the new professor of physiology in Dunedin. She joined her voice to those prompting Eccles to invite Popper to give some (unpaid) lectures at Otago in May 1945. Popper and Eccles hit it off immediately, intensely, unreservedly. Not only was Eccles a scientist of the first rank, whose flagging confidence in his own work was revived by Popper on methodology, he was also fascinated by questions of the freedom of “man” and the independence of the soul, and by whether neuroscience could throw light on these questions. Within the month, Popper was prompting and challenging him on both the neurological and the philosophical fronts. ⁵

Popper sailed to London at the start of 1946 to take up a position at the LSE. He later recalled:

after the interlude of The Open Society and The Poverty of Historicism, I turned to other issues: first to logic (the logic of natural deduction), and then to the problem of the history of rationalism and its present dangers. Only a little later in 1953 I discussed the body-mind problem. (Popper 1971 [1974], 1068)

But his July 1948 paper for the Rationalist Press Association, “Towards a Rational Theory of Tradition” (Popper 1949) led the way to his contribution to solving the body–mind problem. In it, he developed a hint in The Open Society (Popper 1945, I.188) on the importance, for a rationalist, of a tradition of knowledge, of growing knowledge, and for a tradition of criticism:

we cannot start afresh; . . . we must make use of what people before us have done in science. If we start afresh, then, when we die, we shall be about as far as Adam and Eve were when they died. (Popper 1949, 129)

Science is not mere accumulation; if it were, we could start accumulating all over again easily if the scientific tradition were lost. If, as he proposes,

science advances by the tradition of changing its traditional myths, then you need something with which to start. If you have nothing to alter and to change, you can never get anywhere. Thus you need two beginnings for science: new myths, and a new tradition of changing them critically. (Popper 1949, 130)

World 3 as the expanding world of problems and proposals, as the air in which knowledge can breathe and grow, is not too far away. Popper does not yet aim directly at the body–mind problem, or at indeterminism and free will, or as he would put it, at the openness of worlds 1 and 2 to world 3, but he wants first to address his old problems in the nature of science and the growth of knowledge. Where standard empiricist theories of science need world 1, the physical world, and world 2, as the world of observations, he asserts that the growth of knowledge needs not only the physical world and the world of thought processes but also the traditions of knowledge he will come to call world 3. Here, in 1948, Popper has significantly advanced his “own theories of problems, of theories, and of criticism” (KPA 137.5), which, along with Bolzano and Tarski, he noted as shaping his approach to world 3. The first stage of his proto-world 3 had been just a world of statements, like Bolzano’s, but at some stage, he realized he

had to people my third world with inmates other than statements; and I brought in, in addition to statements or theories, also problems, and arguments, especially critical arguments. For theories should be discussed always with an eye to the problems which they might solve. (Popper 1969 [1974], 145, sec. 38)

In November of the same year, 1948, Popper delivered a paper arguing for “Indeterminism in Quantum Physics and Classical Physics,” and championing objectivity by challenging the subjectivism of the Copenhagen interpretation of quantum mechanics (Popper 1950). He wrote to Eccles that it showed that “there can be no conflict between physics and the conviction that we are not completely determined” (Popper to Eccles, January 21, 1950, Eccles Archive, EAD 2NZ-1107).

Early in 1952, on sabbatical from Australia, Eccles visited Popper several times. On September 29, Popper wrote to his friend:

I have thought very intensely, ever since your visit, about the Body-Mind problem, and . . . I have got a simple and important result. I think I can show conclusively that there are aspects in human language (as against so-called animal languages), viz., its descriptive and argumentative functions, which cannot be completely treated by any behaviouristic or physicalistic approach . . . The analysis thus establishes the super-natural character of all reasoning. (EAD 2NZ-1107)

He had often lectured on the body–mind problem and its history, and had long been fascinated by it (Popper 1971 [1974], 1068), but here, he not only expanded on his teacher Karl Bühler’s expressive, communicative, and descriptive functions of language, adding a fourth, argumentative function, as he had done previously (Popper 1949, 134-35), but now for the first time emphasized in a short 1953 paper that the truth of descriptions and the validity of arguments could not be explained within a physicalist framework (Popper 1953).

By this time, Popper was well into writing the Postscript to the Logic of Scientific Discovery, which he had begun in 1950 but was still rounding off in 1956 (Popper 1969 [1974], 119, sec. 33). The whole Postscript stressed indeterminism, objectivity, and realism. The section that became The Open Universe: An Argument for Indeterminism, written before 1956, “was not intended,” in Popper’s later words, “to discuss human freedom and human free will, even though these were really the problems that stood behind it” (Popper 1982, xix). He added that, as The Open Society and The Poverty of Historicism showed, he was “deeply interested in the philosophical defence of human freedom, of human creativity, and of what is traditionally called free will” (Popper 1982, xxi), and that The Open Universe was “a kind of prolegomenon to the question of human freedom and creativity, and makes room for it physically and cosmologically in a way that does not depend on verbal analyses” (Popper 1982, xxi-xxii). It develops the 1953 paper’s argument for the irreducibility of the truth of a description or the validity of an argument to any physical cause (Popper 1982, 83-85).

From the vantage point of 1969, Popper (1969 [1994], 22) wrote, “I had this world 3 for many years and did not in a sense dare to publish anything about it because it was too absurd and too abstruse.” Extending Bolzano, he had arrived by at least the 1950s at the idea of a “world of theories and arguments” and problems but “it all seemed so very abstract, philosophical, and vague.” “I am allergic to hot air, and I did not feel confident that there was more to it than hot air” (Popper 1969 [1994], 33).

But as he would later write,

I think one has to accept this very interesting and serious world 3, but also one has to take away the hot air from it. And so we do up to a point by showing that there would not be such a thing if there were no human beings. (Popper 1969 [1994], 37)

What may have given Popper the confidence that he could take the hot air from his “third world” proposal was the idea he had been gestating since at least 1948, and had returned to in 1953, that the descriptive and argumentative functions of language evolved in humans from simpler expressive and communicative functions in other animals. His world of theories and arguments was not timeless, in the manner of Plato’s world of Forms or Bolzano’s world of statements, but had a historical origin in the evolution of human language. And far from being a world of perfection, the prominence of argument, falsity, and invalidity in his third world reflected the fallibility of a creature not endowed with an unevolvable organ for truth.

Popper was certainly becoming keen to engage with biology. In a letter to Eccles on January 21, 1959, he had asked his friend whether he would review the Logic of Scientific Discovery, and suggested that if he did, this would answer “those who say that I am speaking only about physics, and that my theories do not apply to biology, for example” (Eccles Archive EAD 2AU-1109). ⁶ He would leap at the chance to speak about biology when, a few months after the LSE seminar, he was invited to deliver the Herbert Spencer Lecture at Oxford, and to choose “some such subject as ‘The Method of the Biological Sciences’” (Popper 1961 [1972], 256).

On November 15, 1960, between writing to Eccles and accepting the Spencer Lecture invitation, he told his seminar that if no one had a paper, he would read one (Dimitrakos 1999). His paper, “which spoke of ‘three worlds,’ of biology, and gave qualified support to Hegel’s theory of objective mind, took the members of the seminar”—including John Watkins, Imre Lakatos, William W. Bartley III, Ian Jarvie, and Alan Musgrave—“off guard” (Bartley 1976, 465). The listeners looked at each other in surprise (Dimitrakos 1999) and shock: “But that’s Hegel! That’s Plato!” (Schwartz 1999). Popper (1971 [1974], 1050) himself recalled that his paper “was not too well received,” and Bartley (1976, 465) reported, “The discussion that followed was more bewildered than heated; and Popper, usually one of the most persistent of men, did not pursue the matter that term.”

Presumably, the incomprehension of even those most attuned to his thinking helped keep Popper from publishing on his three worlds theory for another five years or more. He must have felt he did not have his ideas sufficiently resolved. As he would write later, “I did not publish anything about the third world until I arrived at the conclusion that its inmates were real; indeed, more or less as real as physical tables and chairs” (Popper 1969 [1974], 146, sec. 138).

In October 1961, he delivered his first public lecture in biology, “Evolution and the Tree of Knowledge,” the Herbert Spencer Lecture, which emphasized the parallels between trial and error in evolution and in the growth of knowledge. It earned little response. Not until May 1963, when he was giving a lecture at Northwestern, did he meet someone else interested in evolutionary epistemology, the psychologist Donald Campbell. Popper wrote to him the next day, but their letters kept missing each other. And not until February 1964 did he receive a copy of an earlier letter from Campbell that had gone astray: “I was very glad you liked my Herbert Spencer Lecture; your August letter, received only now, was the first reaction from somebody interested in this particular field. I shall now consider publishing it” (letter of February 3, 1964, KPA 282.12). ⁷ Campbell had also written, in the new letter enclosing a copy of the old, that he would be writing a paper on “Evolutionary Epistemology” for the Popper volume in the Library of Living Philosophers. These responses seem to have been the encouragement Popper needed.

Later that year, he started to work out at length the consequences of an evolutionary perspective for human freedom. In April 1965, he gave the Arthur Holly Compton lecture at Washington University, “Of Clouds and Clocks,” one of his major essays (Popper 1966, Popper 1972a ch. 6). Picking up on the physicist Compton’s sense of indeterminism and human rationality, as well as on material from his own still-unpublished Postscript (Miller 1997, 395-96), “Of Clouds and Clocks” addresses physical indeterminism but goes beyond it, as Compton had. Here, Popper offers his first large-scale exposition of the idea of the openness of the physical world first to consciousness (to “hopes, and wishes,” Popper 1972a, 219) and then to “such non-physical things as purposes, deliberations, plans, decisions, theories, intentions, and values” (Popper 1972a, 229). “[I]f we submit to our theories, then we do so freely, after deliberation; that is, after the critical discussion of alternatives, and after freely choosing between the competing theories, in the light of that critical discussion” (Popper 1972a, 219): he introduces, in other words, the idea of human freedom in terms of our interaction with a world of purposes, problems, theories, and arguments—as he would soon phrase it, the openness of world 1 to world 2 and world 3. He offers arguments “against monism and mere dualism,” and especially “against the closedness of [the physical world]” (Popper 1971 [1974], 1074) but he was not yet explicit about the three worlds theory.

Even after working for five months on “Of Clouds and Clocks,” he wrote to Eccles, ⁸ he continued toiling on the body–mind problem for another six months. In a paper in Denver in 1966, he took several important new steps. A few years later, he would write that

For a long time I did not even see that [world 3] can be characterized as the world of our products. . . What made me feel more confident was the discovery that there is something analogous to the human world 3 even among the animals. (Popper 1969 [1994], 33)

In the Denver paper (Popper 1970, Popper 1972a ch. 8), although he still does not mention world 3, he introduces ideas and theories as “organs evolving outside our skins,” “exosomatic artifacts” that he compares with the honey produced collectively by bees, and sustaining them (Popper 1972a, 286). ⁹ He clarifies the distinction between “the world of thought-processes, and the world of the products of thought-processes. While the former may stand in causal relationships, the latter stand in logical relationships” with one another (Popper 1972a, 299). He also begins to introduce the notion of the autonomy of the world of objective knowledge, such as the logical consequences of theories that may not be discovered in them until many years later. He emphasizes emergence, and therefore the need for pluralism: the irreducibility of thought contents or objective knowledge to thought processes, of biology to chemistry, of chemistry to physics.

Popper continued developing arguments to the point where he could introduce his world 3 theory explicitly (as a “third world,” echoing Frege’s “third realm”) ¹⁰ before a philosophical congress in Amsterdam in 1967 (Popper 1968, Popper 1972a ch. 3). He now overtly included products of our minds other than problems, theories, and arguments, such as houses, tools, and works of art (Popper 1972a, 113). He stressed the autonomy of world 3, using for the first time the natural and prime numbers, which would become his standard quick example of the autonomy of world 3: we invented the natural numbers, but which of them are prime does not depend on us (Popper 1972a, 118). He emphasized also “the close analogy between the growth of knowledge and . . . the evolution of plants and animals” (Popper 1972a, 112). And he stressed especially our capacity both to affect world 1 and to transcend ourselves through interacting with world 3: “our biological considerations . . . are of general importance for the theory of Darwinian evolution: they explain how we can lift ourselves by our own bootstraps. Or in more highbrow terminology, they help to explain ‘emergence’” (Popper 1972a, 119). (As he would say shortly, “we can get more out of world 3 than we ourselves put into it,” Popper 1969 [1994], 32.) He would increasingly argue that

with life there come into existence things entirely new: problems, and problem solving. Nothing of the kind seems to exist in the physical world before the emergence of life. World 3 is merely a belated outcome on the human level of the existence of problems, and of problem solving, which I regard as the most characteristic and specific property of life as we know it. ¹¹

In September 1968, he took the argument still further, in “On the Theory of the Objective Mind” (Popper 1968 [1972], Popper 1972a, ch.4) He emphasized the pervasiveness of dualism of mind and body in Western thought and the weakness of attempts to escape it via monism (Popper 1972a, 153). He asked, how does the theory of objective knowledge deal with the subjective understandings special to the humanities? He stressed that understanding was the goal of both the sciences and the humanities, and that understanding could be perfectly objective in the humanities, if we focus on, and discuss critically, the problem situations of agents.

Three months after this Vienna talk, he led another seminar session at LSE on the three worlds. To clarify the distinctness of world 3 from world 2, he proposed: “It is much better to look upon the third world as a collection of libraries[,] something really material, than to look upon it as in any way resembling human thought” (KPA 382.5). ¹² He placed particular weight on the importance of world 3 for freedom: “does there exist such a thing as understanding arguments and being swayed by arguments, yes or no? The fundamental freedom to make up one’s mind intellectually” (KPA 382.5, p. 21). He also made utterly clear his lack of interest in erecting an ontology. ¹³ To contrast his view with Plato’s perfect forms, he emphasized that his world 3 is “dependent on what I call the argumentative or the critical function of human language,” which is itself “bound up with the fallibility of man” (KPA 382.5, p. 7).

In May 1969, Popper gave a series of six lectures at Emory University in Atlanta, published much later as Knowledge and the Body-Mind Problem: In Defence of Interaction. He was now sufficiently confident of world 3 to reflect on its development in his thought. He introduced the evolutionary argument against epiphenomenalism: “if mind or consciousness is an epiphenomenon, why should it have evolved? Why should it have become more and more prominent with higher animals?” (Popper 1969 [1994], 111; see Lindahl 1997 for a sympathetic critique). And he also introduced the conjecture that “although consciousness may be conjectured to be prelinguistic, what I call the full consciousness of self may be conjectured to be specifically human, and to depend on language” (Popper 1969 [1974], 150, sec. 39) and on world 3 theories of selfhood and continuity in time.

In the same year, he completed his intellectual autobiography for the Library of Living Philosophers (Popper 1969 [1974], Popper 1976). He ended the autobiography with world 3, and emphasized world 3 as especially the realm of criticism:

Criticism may be said to continue the work of natural selection on a nongenetic (exosomatic) level: it presupposes the existence of objective knowledge, in the form of formulated theories. Thus it is only through language that conscious criticism becomes possible. (Popper 1969 [1974], 112, sec. 31)

Or again, on grasping an idea,

This “it,” this something which I may not have got, which I cannot be quite certain that I have got before I have written it down, or at any rate formulated it in language so clearly that I can look at it critically from various sides, this “it” is the thought in the objective sense, the world 3 object which I am trying to grasp. ((Popper 1969 [1974], 145, sec. 38)

In 1970 and 1971, he worked on his replies to his critics in the two Popper Library of Living Philosophers volumes, clarifying misunderstandings of this new element of his thought, and stressing the link between his emerging focus on biology on one hand, and on the other his developing the distinction between subjective knowledge and the objective knowledge he now called world 3 (Popper 1971 [1974], 1048). (As he had earlier written to Campbell, “the growth of human knowledge is towards greater objectivity,” letter of February 3, 1964, KPA 282.12.) In 1972, in an essay for Encounter, “Indeterminism is Not Enough,” he formulated his world 3 theory concisely and confidently, summing it up with the tight control, and the quiet, efficient awareness of likely misapprehensions that characterized his later exposition of his earlier ideas (Popper 1973).

Late in 1972, in a conference at the Villa Serbelloni, Bellagio, on “Problems of Reduction in Biology,” Popper presented a paper, “Scientific Reduction and the Incompleteness of all Science” (Popper 1974). Arising in part from his recent preoccupation with world 3, it also offered a further argument for world 3’s independence: just as biology could not quite be reduced to chemistry, nor chemistry to physics, nor could world 3 thought contents be reduced to world 2 thought processes, which in turn could not be completely reduced to world 1 processes. Conversely, new and important problems arise at each higher level. Eccles, who was there, was excited by the paper, ¹⁴ and Popper by Eccles’s argument that “reductionism is a necessary strategy in research, but fails when confronted by the brain-mind problem” (Eccles 1982, 232). Eccles (1982, 232) writes, “Karl and I felt very close philosophically, and to my great delight Karl enquired if I would join him in writing a book on the brain-mind problem.” Eccles organized a month-long stay at the Villa Serbelloni in September 1974, where 12 peripatetic discussions between Popper and Eccles were taped to form one-third of the proposed The Self and Its Brain. In the discussions as elsewhere, Popper was consistently more inclined to a naturalistic, evolutionary continuity between animal and human consciousness, and Eccles, because of his religious faith, to a wide gap between them.

Eccles rapidly revised his part of the dialogues and his own solo section. He hoped that Popper’s solo contribution would simply be a revision of his Emory lectures. Popper initially planned to use the material he wrote at the Villa Serbelloni, and some existing lectures, but he had much else to do, and was ill, and depressed by the transcript of his own part of the dialogue—though Eccles thought it brilliant. But the few works Popper wrote as books—the abandoned two-volume Grundprobleme, the Logik and its vast three-volume postscript, and the two-volume The Open Society—were serious and massive, and as he prepared his part of The Self and Its Brain, he could not stop himself beginning to work with something like the same relentlessness. He took up the idea of reductionism, which he had been thinking hard about since 1965, to show that even the materialist program of science had transcended itself (matter was no longer ultimate; energy and fields needed to be brought in to describe the physical world). This idea he would return to even more strongly in 1981, in an addendum to William W. Bartley III’s edition of The Open Universe, where he points out that the program of materialist reduction in science “transcended itself; thanks to the critical tradition of science, which proved stronger than the ideological tradition” (Popper 1982, 171).

In his part of The Self and Its Brain (Popper and Eccles 1977), Popper stressed that mind could no longer be seen as a substance but should be understood, rather, as a process, and that emergence pervades evolution. He also began to reconsider the body–mind problem within the history of philosophy, doing so with all of his customary intensity. Eccles and the publisher, Springer, had expected his material much sooner, and had to defer publication. But Popper could not work on a smaller scale—until in January 1977, the cries from his coauthor and the publisher coincided with the discovery of his wife’s breast cancer. In the face of that news, he was ready to satisfy Eccles and Springer by simply abandoning his plans to continue the body–mind debate from the early nineteenth century up to the present, with the unfortunate consequence that his most significant contribution on the body–mind problem was cut short unfinished.

He published little more on world 3 after that, ¹⁵ but at almost 90 was ready to accept an invitation to contribute to a special issue of Theoretical Medicine for 1992, even announcing his paper as “A New Theory of the Mind-Body Problem.” ¹⁶ He tried hard to meet the September 1991 deadline, writing late in August a draft, “The Mind Brain Problem: A Conjectured Solution” (KPA 508.3; Popper 1991 [2012]). Here, he extended ideas of the mind as producing forces capable of acting on matter that he had developed in The Self and Its Brain, but he could not finalize the text because of “the unfinished character of the problem . . . simply, we do not know enough about those entities . . . which we call forces.” ¹⁷ In lieu of a written paper, he eagerly accepted to be interviewed by the journal’s editors ¹⁸ late in 1992, and again in July 1994, just under two months before his death (Popper, Lindahl, and Århem 1993, Popper, Lindahl, and Århem 2010). In the interviews, he continued to emphasize that consciousness could not be explained biologically if identity or parallelistic theories were correct. Only an interactionist theory, where the mind’s efforts made a causal difference, could explain why a phenomenon like consciousness would have evolved. Evolution evolves functions, not expensive epiphenomena. To the last, he was true to the idea of active minds engaging with problems that could not be reduced to physicalist terms.

Popper often proposed that we analyze a scientist’s or a thinker’s work in terms of the problem situation the individual faced. I find this enormously helpful, yet in the case of his own world 3 theory, it seems likely that his idea first arrived as an inchoate solution almost before he knew what problems it solved. It reflected his lifelong predisposition to objectivity, and to individuals’ engaging, in their work, with something beyond themselves, and thereby transcending themselves, whether in art or science or thought. But once he reached that solution, with the help of Bolzano, he also felt it was riddled with problems, risking being merely philosophical vapor, like Plato’s or Hegel’s sense of Forms or Spirit. It was not, perhaps, for decades that he began to see his “third world” not solely in the context of philosophy but also in wider contexts: (a) evolution, (b) language as the special niche that human minds have evolved for themselves, (c) the criticism that language makes possible, and (d) problems as something that emerge with life in general and cannot be reduced to the simply physical.

A cloudy intuition that would clarify in Popper’s mind, world 3 was with him for a long time, in a form he was too dissatisfied with to introduce into public discussion. But when he did find a way of articulating it that would satisfy him, during the 1960s, it became a keystone for his thought, linking and deepening many of his central ideas: (a) the idea of objective knowledge, and the epistemological centrality of the growth of knowledge, rather than subjectivity as a supposed foundation for knowledge (already broached of course in the Logik der Forschung); (b) the prime importance to human knowledge, and to the specifically human mind, of active engagement with problems, theories, and arguments, formulable with the help of language, as opposed to the idea of the mind as passively registering and accumulating sense impressions (again, prefigured in the Logik); (c) a world of discussable, criticizable theories and values offering a way beyond violent competition (already anticipated in The Open Society). And, to look still wider, (d) a universe where problems emerge with life and deepen over time as problems compound with older problems and solutions and lead to richer problems; (e) a universe, that is, marked by emergence, and the self-transcendence of life, in the course of evolution, in the species, and in individuals, in which active organisms, active human minds foremost, play the leading creative role by problem-solving; (f) a universe where individual efforts matter hugely, where active minds make a difference by venturing into new problem spaces and proposing their own solutions, but where individual efforts rely on what we can achieve and critique together; (g) an open, indeterminate universe that allows for freedom and values and for the growth of freedom and values, a universe in which the physical is open to the mental and the mental to ideas and arguments; and (h) a realism that accepts the primacy of the physical, or world 1, but not its ultimacy.

How do we evaluate the place of world 3 in Popper’s life and thought? In his biography of the first half of Popper’s life, Malachi Hacohen (2000) argued that Popper’s essential breakthroughs were in the “formative years” of his title. Bartley (1976, 466), who was to have been his first biographer, however, thought that Popper’s late work, his “three worlds” and his evolutionary biology and epistemology, “generalised and unified his philosophy.” I agree with Bartley, and would take his claim even further. World 3 of course was no foundation for Popper—who always challenged the notion of firm foundations—but it was a capstone for his thought, allowing him to integrate and strengthen many of his key ideas.

How have others evaluated the three worlds hypothesis? Mostly, philosophers deride or, much more often, ignore it. A few philosophers and others partial to Popper hail it: the biologist Peter Medawar rates Popper’s world 3 the “most compendious description of the content of exogenetic heredity” (Medawar and Medawar 1985, 97), while the novelist Mario Vargas Llosa (1990, 1020) writes, “In the almost infinite series of nomenclatures and classifications that the wise and the mad have proposed to describe reality, Karl Popper’s categories are the most transparent.”.

Time, and, let us hope, argument—world 3 argument?—will tell.

—University of Auckland