In chapter I.2 of The Unity of Nature von Weizsäcker takes up the question that is the basis, or perhaps the thesis, of our information philosophy - "Language as Information." This was the topic of a lecture sponsored by the Bavarian Academy of Fine Arts in 1959.

The lecture asked the difficult question, "Can language be rendered unambiguous in the service of science, using "information" in the sense of information theory?" The short answer is of course "No." The meaning of any information always depends on context.

Disambiguation was the dream of great logicians, from Gottfried Leibniz and Gottlob Frege to Bertrand Russell and Ludwig Wittgenstein, to represent the physical world by "logical atoms." The later Wittgenstein and twentieth-century language philosophers thought that "what there is" could be described with analytically true statements or "propositions." They all were logical failures, but information philosophy builds on their dreams.

Leibniz imagined a characteristica universalis, a universal and formal language able to express mathematical, scientific, and metaphysical concepts, a kind of "natural language." Leibniz also imagined a machine, a kind of computer, that could manipulate the symbols for concepts (the Russell/Wittgenstein "logical atoms"?) in this language. Leibniz called it a calculus ratiocinator.

All these dreams culminated in the school of logical positivism, later analytic language philosophy, which is widely regarded as a dismal failure of twentieth-century philosophy.

Niels Bohr, the great mentor of Heisenberg, von Weizsäcker, and so many other of the physicists who founded the "new quantum mechanics" in the late 1920's, instilled in all his disciples the idea that a language, refined from the concepts of classical physics, is required for explanation of nature, in particular for quantum mechanics. Sadly, this was a product of Bohr's wishful thinking. Information philosophy must go "beyond logic and language."

Von Weizsäcker breaks his lecture into five parts, a) Examples of Information, b) Measurement of information, c) The calculus, d) Information as Form, and e) Can language be reduced to information?

Under Examples of Information, Von Weizsäcker says

I now ask, for the first time: what is information?
Information, in current usage, is the communication of facts..
. The carrier of information is language. Important facts were and are conveyed from mouth to mouth...
The written word has an advantage over the uttered word, however, if language is to serve as information...
Writing, so to speak, “objectifies” information.

In asking for the measure of information, we pose the question “What is information?” a second time, in a more rigorous manner now because of the quantitative setting...
I will first make the concept of information in information theory a bit more precise...
In English- speaking countries a single yes-no decision is called a “bit” of information. “Bit” is meant to be short for “binary digit.” It helps that “bit” also means “small piece.” Information can be measured if one knows the smallest number of small pieces needed to transmit it...
There are, as far as I know, good physiological reasons for assuming that the neural networks in the brain also work with sequences of yes-no decisions (response or non-response of the nerve fiber).

A device can process only information given to it unambiguously...
Opaque though it be to information theory, so-called “natural language,” the language in which ordinary people communicate, cannot be done without, at least not at the start....
Now, not the words, but the meanings of the words are to be put into a form capable of being processed according to a scheme. Is such a step conceivable, perhaps on the basis of some preliminary manipulations? It seems to have been Leibniz who first posed this question with full clarity. Can we find signs to represent, not the linguistic name, but the clearly understood meaning of every concept needed by thought? Can we invent operations with such signs to represent any admissible operation of thought, so that all correct thinking would be mirrored in such operations and controlled by them, even as the operations in algebra mirror and control our thinking about numbers? The calculus of modern logic is the modest execution of a small part of Leibniz’s program. It represents the furthest reduction of language to its unambiguously manipulable informational content. On a new level of abstraction, it turns back from natural to formalized writing.

For an example, let me give you a few formulas taken from the simplest part of logical calculus, the propositional calculus. This calculus treats the relations between entire sentences, the so-called “propositions,” without reference to their internal structure...Can we, to return to our theme, thus transform into information all that language can express?...It would already mean a lot if we could assign an information-theoretical meaning to these challenging, if not completely clearly formulated sentences of Wittgenstein’s:
“What can be said at all can be said clearly; and whereof one cannot speak thereof one must be silent.”

The desire to think clearly is common to philosophy and the sciences; in literature, too, it can be salutary. Should we surmise that one can think clearly only what can be said clearly; that one can say clearly only what can be unambiguously expressed in terms of information; that one can express unambiguously as information only what can be written in the form of a calculus? As the result of this third section, may we interpret clarity in terms of information, and information in terms of the concept of a calculus?

We can test the truth of these conjectures only if we can be sure of having understood their meaning. Do we know with sufficient clarity what we mean by information?...
I have conjectured that only what we can write down in terms of a calculus counts as information. But I have not said what information really is. Socrates would not have been satisfied with me...[Nor would Plato!]
Let us assume that information is the thought process in the mind of the person thinking the contents of the wire. Then what I thought when sending the wire and what the addressee thought on receiving it is not the same information. Information is not one or the other act of consciousness but what is known by the act of consciousness, something that is common to these conscious persons who are otherwise so different.

Today we are beginning to get used to the idea that information must be regarded as something different from matter and consciousness. What one has thus discovered is an old truth in a new guise. It is the Platonic eidos and the Aristotelian form, expressed in such a way that even twentieth-century man can learn to obtain a glimpse of them.

But much more powerfully, information objectifies objects. Information "represents" a physical/material object in the most minimal and least ambiguous way, totally independent of Bohr's mandatory "classical language."

John Wheeler famously coined the slogan "It from Bit." It is not clear exactly what Wheeler was thinking, but the connection between the discrete number of particles in an information structure and the minimum number of bits needed to communicate the structure is the essential fact. Some physicists jump to the erroneous conclusion that the object ("It") is nothing but information (the "Bits").

This supports popular but simplistic ideas like the body is a machine and the brain is a computer, maybe the whole universe is a computer, etc. These all are not helpful. Reduction to one (monist) or the other side of the dualism is very popular in philosophy, but ultimately explains everything with one idea which by itself explains nothing.

Is von Weizsäcker's concept "unity of nature" a similar simplistic misconception? The observable universe contains an immense, but in principle knowable, number of discrete particles. They are arranged in an enormous number of unique information structures. We must conclude that no monism can ever do justice to such a "pluralistic universe," as William James described it. This "it" will always need a very large number of "bits" to describe , to communicate "what there is" (ontology) between scientists and philosophers so we can know something about it (epistemology).

But let's continue with von Weizsäcker.

One could try to explicate what I have just named the linguistic character of information by means of another concept that does not contain the term “language.” “Communication” offers itself, for example. One could try this: “Information is a form that serves in communication.” The “serves” here must be understood as potential; a book no one has read and the current flow in a computer unperceived by anyone are not actually someone’s communication, but they are the sort of structures that can be considered communication. The term “communication” suggests that language does not relate merely to an isolated consciousness, to a Cartesian res cogitans, but that it is essentially interlocution, communication between persons. I won’t follow up on this remark; it is merely the eyelet through which the threads that connect with the other lectures may be drawn.

I must, however, deal with a possible objection. Modern biologists speak perfectly legitimately of information—in genetics, for example. A set of chromosomes contains in its genes the information that determines the phenotype of the individual (to the extent that the phenotype is in fact genetically determined)...Information-theoretical concepts are clearly applicable here if they are anywhere. But there is no one here who talks, no one who communicates something or understands what is being communicated to him... [No human, but biological entities clearly communicate meaningfully!]

I suspect that I am not really meant to solve this problem with the conceptual means to which, for the sake of achieving preliminary clarity, I am today restricting myself. I therefore now turn back from information beyond human language to language as information...

In conclusion I now return to the question: Is it possible, in principle, to reduce language to information understood in this way? Not that everything expressible in words must necessarily fall under this concept of language; but is the concept itself at least clear enough to enable us to define what Wittgenstein termed “saying clearly”? I am inclined to answer in the negative...

At this point von Weizsäcker spirals down into the "epistemic circle" of language philosophy, anticipating the post-modern position that language is a vicious circle of signifiers. There is no contact with the world outside the text (il n'y a pas hors de texte).

In 1969, von Weizsäcker wrote an important essay called "Matter, Energy, and Information."

In it he says that "information is neither matter nor energy," Information philosophy agrees with this fundamental insight, but we emphasize that information needs matter to be embodied and energy to be communicated. Information is abstract and immaterial, like gravitational field theory that describes how matter moves, or the quantum wave function that controls (statistically) the positions of particles.

All theories, including mathematics, are "free creations of the human mind," said Albert Einstein. They are all ideas. Their usefulness must be decided by experimental tests.

Like many mathematical physicists, in this essay, von Weizsäcker wants theories to be substantial. "Form as Substance, he titles one section, in which he says

This hypothesis entails the following theses: Substance is form. More specifically: Matter is form; movement is form. Mass is information; energy is information.

Von Weizsäcker bases these strange ideas on his theory of ur-objects which have ur-alternatives. These appear to be our alternative possibilities in information philosophy.

Von Weizsäcker in his final section makes a connection to Parmenides's idea of "the One" through Plato's dualism that privileges "the Ideas" over the materialism that Aristotle will make the origin of all things, since ideas are merely abstractions away from various forms in material objects.

Von Weizsäcker concludes his book with analysis of nature as a unity.

We begin by recapitulating the facts and conjectures in which the idea of the unity of nature has presented itself.

The unity of the law comes first. This is merely another expression for what physicists call the universal validity of a fundamental theory. A “theory” of this type consists of a number of terms, as well as of fundamental propositions which connect these terms and from which additional propositions can logically be deduced. Further, it must be sufficiently clear for practical purposes how the theoretical terms are to be applied in experience, and thus also how the theoretical propositions can be put to the test. A theory has “validity” only if these procedures are available, and if the propositions thus tested agree with experience. We will not recapitulate the methodological problems implicit in these requirements but will rely for the moment on the fact that, in general, physicists agree on these matters among themselves. The validity is “universal” if it extends to all possible objects of a theory; i.e., to all objects covered by the terms of the theory. Here, too, we are satisfied for the moment with practical universality, leaving open the discovery of exceptions or of still more universal laws. We will call a theory “fundamental” if it extends to all possible objects of nature. The universal validity of a fundamental theory means that all objects of nature are subject to one and the same lawful scheme; it is in this sense that we term this validitydie “unity of the jaw.” Let me emphasize that all these terms are merely descriptive. They formulate the approximate ^elf-interjT/etation of contemporary physics, and the following recapitulating reflections will clarifv or revise them.

We do have such a fundamental, theory—namely, quantum theory. Let us examine in more detail what demands should be imposed on a fundamental theory, and in what sense quantum theory fulfills them.

The theory is to apply to arbitrary objects of nature. To this end, it must be capable of characterizing an arbitrary object. It does so by specifying the totality of its possible (“formally possible”) states. The theory must also specify how these states can change in time. These two requirements can be stated from the point of view of classical physics; quantum theory supplements the requirements in its own characteristic way—namely, by fulfilling them.

According to quantum theory, every object possesses, mathematically speaking, the same manifold of possible states; these can be characterized as the one-dimensional subspaces of a Hilbert space. Quantum theory also specifies a universal rule for the composition of two objects into a single object: the Hilbert space of the composite object is the Kronecker product of the Hilbert spaces of the two part-objects. The theory subdivides the question as to the temporal change of the states into two questions. If the state changes without being observed, it does so in accordance with, a unitary transformation of Hilbert space. A particular species of objects(e.g., helium atoms) is characterized by its formally possible unitary transformations, which are mathematically specified by their infinitesimal element, the Hamilton operator H. The Hamilton operator of an isolated object characterizes its internal dynamics and thereby designates certain of its states (for example) as eigenstates of H with particular eigenvalues of the energy. The interaction of the object with other objects is described in terms of the Hamilton operator of the composite object constituted by these objects; this operator can, within certain approximations, be reduced to the Hamilton operator of the original object taken as situated alone in a fixed environment. If, on the other hand, the state is observed, then the state changes in another manner. A particular observation admits of only a subset of the formally possible states of the object as possible results of measurement; this subset is constituted by the eigenstates of the Hamilton operator of the object when the instrument of measurement is specified as part of its environment. If Ψ was the state prior to the observation, then the probability of finding a particular state Φ_n among the manifold of possible results of the observation equals the square of the magnitude of the inner product of the unit vectors in the directions of states Ψ and Φ_n

Because of the mathematical formalism that it requires, this description of quantum theory might seem a bit heavy-handed. From the conceptual point of view, the theory may be said to achieve a certain maximum in possible simplicity. The theory characterizes, in unique terms and by means of universally valid prescriptions, arbitrary objects, their composition, changes in their state when not observed, and the prediction of observations. And yet quantum theory, even if we assume it to be universally valid, does not yet express the full unity of nature.

For one can speak, secondly, of a unity of nature in the sense of a unitary character of the species of objects. This character expresses itself in quantum theory in the existence of objects with particular Hamilton operators. Today we believe that all species of objects can in principle be explained as being composed of a small number of species of elementary particles. In the case of inorganic nature, we all believe this to be so; in the case of living organisms, it is the hypothesis on which we have based this book. Finally, we hope to reduce the species of elementary particles to a single basic lawful order, which perhaps we ought not to describe as the existence of a single basic species but rather as the law that specifies all of them.

Thirdly, in the context of contemporary cosmology, it makes sense to talk of the unity of nature as the totality of objects. One speaks of the world as if it were a single object. Quantum theory does indeed permit the composition of arbitrary objects into a new object. It even requires this composition, in the sense that it regards the actual state space of a number of coexisting objects as precisely the state space of the total object they compose; the isolation of individual objects is, in the eyes of quantum theory, always a mere approximation. If the totality of objects in the world can, at least in principle, be enumerated, then quantum theory obliges-as in principle to introduce the additional object “world,”, which is composed of that totality. At this point, however, certain conceptual problems that form a principal theme of the present essay appoar. Let me merely name them for now: If the object “world” is to exist, for whom is it an object? How are we to conceive of an observation of this object? If, on the other hand, the object “world” is inadmissible, how are we to describe the coexistence of objects “in the world” quantum mechanical/? Or are we to conclude that quantum theory meets its limits here?

Fourthly, we have tried to base the unity of nature (as conceived under the three preceding aspects) on the unity of experience. We talked, to begin with, of the preconditions of the possibility of experience, and understood “experience” to already be unified in the sense that “every” experience may be thought of as -connected with every other experience in a contexture of interactions that is free from internal contradictions. This unity appears in Kant under the title of “the unity of apperception.” In our own approach, which starts not from subjectivity but from temporality, this unity appears as the unity of time. The unity of time (which in our presentation of course embraces space) is, most likely, the only adequate framework for the problem of the totality of objects. With these latter reflections we have delved into the midst of the fundamental problems of classical philosophy. Before confronting these problems, we must still introduce our last approach, the approach of cybernetics.

Fifthly, the unity of man and nature is part of our conception of the unity of nature. Man, in whose experience the unity of nature is discovered, is at the same time part of nature. We try to describe human experience in terms of a cybernetics of truth, which is conceived of as a process in nature. The philosophical problem that arises here is obvious: if this program can be carried through, at least in principle, then the unity of nature is somehow represented within nature as the unity of the experience of man. What does this “somehow” mean? To put it differently: the subjects, for whom the objects are objects, now form part of the totality of objects. Furthermore, in a cybernetics of truth, human consciousness stands apart from animal subjectivity as a higher-level structure, but the two are also part of a genetic continuum. In the attempt to reduce matter and energy to information, the subjectivity of all substance, if only implicitly and unclearly, is presupposed. The classical formula that nature is spirit which does not know itself as spirit urges itself upon us as a shorthand notation for these problems; but this does not mean that we have understood this formula in the least. As a next step, we therefore explicitly confront our complex of problems with the ideas of classical philosophy, among which we in fact already find ourselves. Aren’t we in the midst of the problems faced by the Eleatic philosopher Parmenides? Hen to pan: One is the totality. The totality is, first of all, the world, “comparable to a well- rounded sphere.” But this world embraces experiencing as much as what is experienced, consciousness as well as Being: To gar auto noein estin te kai einai, for it is the same to see and to be. I translated noein with “to see” to avoid the abstract introversion of “to think.” What can Parmenides teach us?