Wheeler’s Metaphysics of Information
Systems thinking of the twentieth century, combined with doubts about how quantum mechanics could provide an objective understanding of reality, boosted the metaphysical impact of Shannon’s work. His spark of information theory even ignited speculation about information as a foundation for physics. In the twentieth century, it seemed clear to many that quantum mechanical theory did not allow us to fully understand the phenomena described. Scientists could use mathematics from quantum mechanics for prediction and control. But deep questions about what the mathematics represented were discussed without closure. The equations describe a quantum mechanical world, qualitatively different from the classical world. However, we interpret the equations classically. Many asked themselves what this means for our understanding of reality—if there is a layer of classical concepts between us and the quantum world, then in what sense could our judgments be said to be veridical? Antirealism gained in popularity during these times. Some also argued that because our observations of quantum phenomena affect what we measure, we must conclude that there can be no independently existing objective reality. This position was taken up with enthusiasm by postmodernists during the so-called science wars. Postmodernists came to think of reality as being created by language, and some physicists adopted relativist ideas. Did quantum mechanics shatter our picture of a knowable, independently existing world? In this climate of debates over the status of quantum mechanics, new systems-based information metaphysics emerges from within theoretical physics with the work of John Wheeler (1911-2008).
Wheeler, like Turing and Shannon, becomes a technological systems thinker. Wheeler’s picture of reality is based on quantum mechanics and information processing. He imagines a symmetry between how information-processing technology has evolved and the evolution of the universe. Reality, he suggests, is a selfconscious, self-unfolding information-processing system. Wheeler is also attracted to Leibnizian panpsychism and the idea that reality has an ultimate foundation in the mind. He adopts a view based on computer metaphysics and his understanding of the Copenhagen interpretation in quantum physics. Wheeler sees the process of observation in quantum mechanics as mind creating reality. This view is contentious. When Wheeler combines it with computer metaphysics, his task of explaining reality becomes neither easier nor less controversial.
However, the view that reality could somehow be understood as a computer system is not entirely uncommon and has been pursued by other authors. Rather than go through these authors and their views, I will focus on Wheeler. His work has influenced physicists, computer scientists, and philosophers. If any writer has been foundational in “computational metaphysics,” it is Wheeler. As we saw, Chalmers depends on Wheeler’s view for his analysis of consciousness. This is how Chalmers puts it:
Wheeler (1990) has suggested that information is fundamental to the physics of the universe. According to this “it from bit” doctrine, the laws of physics can be cast in terms of information, postulating different states that give rise to different effects without actually saying what those states are. It is only their position in an information space that counts. If so, then information is a natural candidate to also play a role in a fundamental theory of consciousness. We are led to a conception of the world on which information is truly fundamental, and on which it has two basic aspects, corresponding to the physical and the phenomenal features of the world. (Chalmers 2010, p. 26)
To understand what Chalmers is after, we must confront the idea of information processing not just within cognitive science but within Wheeler’s physics. Let’s take a brief look at the development of the field of quantum computation. Wheeler’s graduate student physicist Richard Feynman (1918-1988) pioneered this field. But Wheeler was interested in something beyond Feynman’s work and does not cite him in his 1990 paper (which has 179 references). So let us revisit a keynote speech that Feynman held at a conference on the “Physics of Computation” in 1981. There he introduces his view on quantum computation that got the field started. He asks a simple question:
Now I explicitly go to the question of how we can simulate with a computer—a universal automaton or something—the quantum mechanical effects . . . we can say: Let the computer itself be built of quantum mechanical elements which obey quantum mechanical laws . . . can you do it with a new kind of computer, a quantum computer? (Feynman 1982, p. 474)
The motivation for a quantum computer is clear. Feynman’s interest in simulating physics at a quantum level led him to the idea of a quantum computer. Now why doesn’t Wheeler acknowledge the work of Feynman—one of the greatest physicist we have seen and a founder of quantum information processing? Not only was Feynman his graduate student, but Feynman also commented on Wheeler’s “It from Bit” paper. The reason could be that Wheeler takes on a metaphysical position that is altogether disconnected from Feynman’s concrete aims of simulating quantum mechanics. In contrast to Feynman’s well-defined approach, Wheeler appeals to Leibnizian idealist metaphysics and pancomputationalism. In a paper published one year after Feynman’s keynote speech, Wheeler expresses how he found something surprising in Leibniz’s The Monadology:
One who comes from an older time and is accustomed to the picture of the universe as a machine built out of “atoms” is not only baffled but put off when he reads Leibniz and Leibniz’s conception of the ultimate building unit, the monad. (Wheeler 1981, p. 560)
He then goes on to quote Leibniz at length:
- 1. The Monad, of which we will speak here, is nothing else than a simple substance, which goes to make up composites; by simple, we mean without parts.
- 2. There must be simple substances because there are composites; for a composite is nothing else than a collection or aggregation of simple substances.
- 3. Now where there are no constituent parts there is possible neither extension, nor form, nor divisibility. These Monads are the true Atoms of nature, and, in fact, the Elements of things . . . There is also no way of explaining how a Monad can be altered or changed in its inner being by any other created thing, since there is no possibility of transposition within it . . . The Monads have no windows through which anything may come in or go out . . .
- 9. Each Monad . . . must be different from every other . . .
These words of Leibniz about the “monad” are more relevant to “quantum phenomenon” than to anything one has ever called an “atom.” (Wheeler 1981, p. 560)
Why does Wheeler think monads are more relevant to quantum physics than atoms? Feynman is out to discover how the physical world works, but Wheeler aims for an account transcending it.
Wheeler seeks a world beyond the physical—something more “relevant” than atoms. Feynman has sometimes been considered to be hostile to philosophy and once said “philosophy of science is about as useful to scientists as ornithology is to birds.” I am not sure that Feynman was hostile to philosophy on the whole. His attitude seems to have been that investigations should be done with an open mind, not one shaped by preconceived opinions (whether philosophical or not), and his papers typically have a minimum of references. Wheeler, in contrast, writes to support all claims with references (179 within 14 pages). Fuller, Follesdahl, Smorynski, Quine, Popper, Putnam, Schelling, Parmenides, Berkeley, and Leibniz are some of the philosophers he refers to. While Feynman remains intellectually skeptical as to how philosophy could help science, Wheeler ponders that he and Leibniz are on to something profound—an otherworldly building unit of existence:
The ultimate building unit of existence—call it elementary quantum phenomenon or call it monad or call it what one will—has to be of an intangible and other-worldly character. (Wheeler 1981, p. 565)
Wheeler suggests that Leibniz’s notion of a monad is helpful for understanding the creation of the universe:
How did the universe come into being? Is that some strange, far-off process, beyond hope of analysis? Or is the mechanism that came into play one which all the time shows itself?
Did the genius of Leibniz somehow sense the deep and secret underpinning of existence, the necessity that lies behind the strangeness of the quantum? Did he in the monad anticipate the quantum phenomenon? (Wheeler 1981, p. 564)
But how might Wheeler think that Leibniz anticipated the quantum mechanical universe in The Monadology? Leibniz, as well as other idealists of his time, thought that the only things that could truly exist apart from God would be minds. This gives little room for interpretation of what the ultimate entities could be. They, too, would have to be minds. According to this interpretation then, when Wheeler refers to “the secret underpinning of the universe,” he is thinking about mind over matter in the same way that Leibniz did. Wheeler gives further support to such mental metaphysics when he makes clear that physical reality cannot exist independently of observation. There is no world existing “out there” independent of us:
Useful as it is under everyday circumstances to say that the world exists “out there” independent of us, that view can no longer be upheld. There is a strange sense in which this is a “participatory universe.” Are billions upon billions of acts of observer participancy the foundation of everything? (Wheeler 1981, p. 564)
Here Wheeler appears to make a two-level analysis. Physical reality (the world) depends on observation, and perhaps we are to read the billions and billions of observations as performed by something resembling monadic entities. However, eight years later, in an article titled “Information, Physics, Quantum: The Search for Links,” he writes:
Parmenides of Elea  (~515 B.C.-450+ B.C.) may tell us that “What is . . . is identical with the thought that recognizes it.” (Hey and Feynman 1999, p. 320)
Here he seems to entertain a higher psychological level so that it would apply to humans. He explores this idea in his journal as well:
No space, no time, no electromagnetism, no particles. Nothing. We are back where Plato, Aristotle and Parmenides struggled with the great questions: How Come the Universe, How Come Us, How Come Anything? But happily also we have around the answer to these questions. That’s us. (Overbye 2002)
Our minds determine physical reality. Then, in “Information, Physics, Quantum: The Search for Links,” he brings up the possibility that computers could be conscious subjects:
We, however, steer clear of the issues connected with “consciousness.” The line between the unconscious and the conscious begins to fade in our day as computers evolve and develop— as mathematics has—level upon level upon level of logical structure. We may someday have to enlarge the scope of what we mean by a “who.” (Hey and Feynman 1999, p. 320)
Wheeler suggests that computers could potentially make observations. Supposedly, then, his participatory universe could be one that is “computerized” somehow. Thus Wheeler vacillates between different metaphysical views—the Leibnizian, the Parmedian, and a “computerized” universe. The latter one appears to be his favorite as he begins to examine the evolution of computers:
The evolution from small to large has already in a few decades forced on the computer a structure reminiscent of biology by reason of its segregation of different activities into distinct organs. Distinct organs, too, the giant telecommunications system of today finds itself inescapably evolving. (Hey and Feynman 1999, p. 321)
Wheeler does not explicate but suggests that computer technology evolves along self-synthesizing, evolutionary principles, which govern the universe:
Will we someday understand time and space and all the other features that distinguish phys- ics—and existence itself—as the similarly self-generated organs of a self-synthesized information system? (Hey and Feynman 1999, p. 321)
In his autobiography, he also suggests how the universe—like the computer—is built on yes-no logic:
The computer is built on yes-no logic. So, perhaps, is the universe. (Wheeler and Ford 1998, p. 340)
Here, that “the computer is built on yes-no logic” can mean either that the computer operates according to principles of yes-no logic (which is correct) or that it is constructed out of yes-no logic (which is false). My laptop operates according to principles of yes-no logic, and it is constructed out of plastic, silicon, and other materials. It could not be constructed out of yes-no logic, because logic is not a physically defined notion. Logic can be implemented in physics, but logic is not physical. What about the universe? Here, too, we must read what Wheeler says in two possible ways. When he says that it is built—like the computer—out of yes-no logic, that could either mean that it operates according to principles of yes-no logic (which is a vague, obscure statement) or that the universe is constructed out of yes- no logic (which is false). The universe could—like my laptop—not be constructed out of logic, because both my laptop and the universe are physical and you cannot construct physical entities out of logic. Nevertheless, this is what Wheeler attempts. The fact that he attempts to ground his view in quantum mechanics does not help. On the contrary, it adds confusion, and Wheeler’s explanation remains at an abstract and vague level:
Did an electron pass through slit A or did it not? Did it cause counter B to click or counter C to click? . . . it is not unreasonable to imagine that information sits at the core of physics, just as it sits at the core of a computer. (Wheeler and Ford 1998, p. 340)
Wheeler seems to think of an electron here as having no true reality until it is measured. The measurement collapses the electron probability field, and the counter ticks. According to Wheeler’s interpretation, before the measurement, there was no other reality for the electron than its probability field. Whenever the electron is measured, it comes into being. But what does this measuring have to do with the “it from bit” thesis? What does it have to do with the idea that the universe is built from yes-no logic? Wheeler writes:
Trying to wrap my brain around this idea of information theory as the basis of existence, I came up with the phrase “it from bit.” The universe and all that it contains (“it”) may arise from the myriad yes-no choices of measurement (the “bits”). (Wheeler and Ford 1998, p. 340)
Wheeler then goes on to give an example of how reality is created from measurement:
Information . . . may be what makes the world. An example of the idea of it from bit: When a photon is absorbed, and thereby “measured”—until its absorption, it had no true reality— an unsplittable bit of information is added . . . that bit of information determines the structure of one small part of the world. It creates the reality of the time and place of that photon’s interaction. (Wheeler and Ford 1998, p. 341)
One might read him as suggesting that a bit is, here, simply the absorption of a photon, but that is a physical phenomenon like any other. The bit cannot simply be that absorption. If it was, then the “it” from “bit” thesis would be turned into the “it” from “it” thesis, and the idealist Leibnizian dreams would be crushed. Thus, the bit would have to be immaterial. But then how could the bit give rise to physical reality? To elucidate how information bits could explain physical reality, Wheeler appeals to emergence:
When you put enough elementary units together, you get something that is more than the sum of these units. A substance made of a great number of molecules, for instance, has properties such as pressure and temperature that no one molecule possesses. It may be a solid or a liquid or a gas, although no single molecule is solid or liquid or gas. “More is different” may have something to do with “it from bit.” The rich complexity of the universe as a whole does not in any way preclude an extremely simple element such as a bit of information from being what the universe is made of. When enough simple elements are stirred together, there is no limit to what can result. (Wheeler and Ford 1998, p. 341)
However, it matters neither how many nonphysical bit entities you postulate, nor how you stir them. There is a limit to what can result: nothing physical. It is unclear how Wheeler’s “it from bit” thesis could bridge the gap between information as a nonphysical notion and physics. Any information metaphysics that posits the existence of two ontological worlds—one of information and one of physics—suffers from problems of dualism. However, the emergence of information metaphysics in the twentieth century is not difficult to understand. If I were an advocate of information metaphysics, here is how I would argue:
- 1. According to the Copenhagen interpretation, we cannot understand physical reality objectively.
- 2. If there is an objective reality, it would have to be nonphysical.
- 3. We understand information processing as nonphysical.
- 4. We have objective understanding of information processing in computer science.
- 5. We can view reality as a quantum information-processing system. Its computations may be complex, but the principles of information processing are objective.
- 6. Perhaps we can have objective understanding of the universe through a nonphysical quantum information-theoretical perspective.
Let us go through this argument. (1) The physicist Niels Bohr (1885-1962) suggested that since we can only make classical interpretations of quantum phenomena, we impose a simplification of the phenomena at hand. However, this does not imply that there is no physical reality independent of us. It simply means the universe behaves in, for us, strange nonclassical ways. We might think, with Einstein, that our minds are feeble instruments for the task of understanding the universe, but no matter how feeble our minds are or what strange theories we come up with, it does not lead to the conclusion that the scientific pursuit of objective knowledge of an objectively existing physical reality is misguided. Science presupposes such a reality. (2) The move toward trying to understand ultimate reality as nonphysical has been appealing since Plato and, as we have seen, it has had a late revival with mathematicians and theoretical physicists since the eighteenth century. But it is unclear how postulating such a nonphysical realm could help us explain reality. (3) It is true that information processing can be done on different kinds of machines. If this is what one means by information processing being nonphysical, then it is trivially true. If one thinks that information processing is nonphysical in the sense of existing in a separate realm, then it is false. From the perspective of science, we live in one reality. (4) This is true. We understand how computers work. We have designed them, and we use them for our purposes. (5) It is also true that we can view the universe as a computer, but as Searle pointed out, we can view anything as a computer because a computer is an observer-relative notion. Without us, there would be no computers, just as there would be no station wagons or ballpoint pens—there would just be the brute physics out of which they are made. (6) We end up with a form of dualism of information and physical reality with all its associated problems.
Shannon’s information-theoretical work had a colossal impact on society and cognitive science. It is fundamental for understanding the information-processing revolution and the digital world it brought about—a world of ubiquitous computing and digital communications we use for work and play. Shannon’s contributions provided seemingly ample footing for subsequent grand theories of mind, world, and reality at large—information metaphysics. However, it is unclear how Shannon’s work could give explanatory power to account for consciousness and the nature of reality. The theoretical footing he provides is not about physical entities—it is a mathematics of abstract entities. Instead we need theories with adequate physical footing—theories that explain physical phenomena in terms of, and in relation to, other physical phenomena.
-  Wheeler’s article was received by the journal on May 7, 1981—the day after Feynman’s keynotespeech at MIT, so it was a time when both of these physicists focused on the topic.