Ideas as Information
This is a difficult post to write, particularly because the ideas it addresses are just plain large. And these large ideas have such permeating ramifications, both towards Spinoza philosophy and contemporary Science it is indeed very difficult to do any justice to them. Instead it must be taken as a kind of rough draft, a sketch, of what may be conceptually possible when bringing the philosophical concepts Spinoza employed into contact with the Information. The thoughts here must be taken as provisional conjecture, but this is not to say that I do not find the comparisons offered here to be valid. Rather, I suspect strongly that what Spinoza was talking about, the relationships in the world that he was attempting to systematize, are very much the same ones that Science today talks about when thinking in terms of information.
All this comes into view with Tom Stonier’s radical Scientific proposal that Information is an essential component of the Universe, found in his speculative book Information and the internal structure of the universe: an exploration (1990). I’ll cite at length from his work below to present the core of his ideas about information, but first I need to make a conceptual leap which will make future thoughts of my application of Stonier’s ideas more clear.
Stonier’s abstraction on the left, Spinoza’s on the right:
Matter = Extension
Energy = Conatus (striving)
Information = Idea
I don’t want to justify these equalizations, but rather just let them remain as starting points, at least until Stonier’s vision of information is made more clear. I will say that the first of these seems obvious. What we mean usually by matter is precisely what Spinoza is attempting to describe through the Attribute of Extension. The second of these is both instinctively appealing but also has some difficulties in translation, mostly due to the much debated theoretical role conatus plays in Spinoza’s philosophy. Perhaps though in reading conatus as energy in the specific context of information theory important aspects of Spinoza’s conatus thinking may come into relief. And lastly, most importantly, the third of these, the equation of 21st century information with 17th century Spinozist idea, is the keystone of the entire comparison, and hopefully will reveal as much about what Spinoza was thinking, as about what he was trying to describe.
But now let us present Stonier’s idea that “information” comprises the universe just as much as matter and energy does.
Information is Real
Stonier spends much of his time hewing out a concept of elemental information from the concept of “energy”, leaving “matter” to remain relatively self-evident. It is in particular the way that we are able to see energy as existing in different forms yet to remain an objective measure of how things are composed, that provides the footing for how information is to be conceived. Much of what Stonier argues is that some of our energy descriptions are better handled as information transformations:
Just as there exist different forms of energy – mechanical, chemical, electrical, heat, sound, light, nuclear, etc – so do there exist different forms of information. Human information represents only one form of information..human information itself, maybe stored and communicated in a wide variety of ways and represent many different forms (9)
Right of the bat we have a very important idea, and one that communicates itself quite well with Spinoza’s notion on the limits of human thinking and epistemology. What we commonly refer to as “meaning” which is ever context bound, is only a form of information, just as mechanical energy is just a form of energy. The ideas we have as human beings are not reducible to the meaning of their expression in language. Rather, as Spinoza sees it, the ideas we have are rather best seen as dispositional relations to really only one thing, the whole of the Universe. The ideas we have are informational or organizational states, what Stonier will call “structure”. Let me quote at length what Stonier describes as the “heart of the concept”. I quote at length both because Stonier’s book is not accessible on-line, and also because he does a pretty good job of expressing himself on what he means:
Information and organization are intimately interrelated.
From this axiom we derive the following theorems;
- All organized structures contain information, and as a corollary: No organized structure can exist without containing some form of information.
- The addition of information to a system manifests itself by causing a system to become more organized, or reorganized.
- An organized system has the capacity to release or convey information.
Let us examine the above theorems, beginning with the first. Any physical system which exhibits organization contains information. Information organizes space and time. The definition of the term “information” becomes analogous to the physical definition of the term “energy”: Energy is defined as the capacity to perform work. Information is defined as the capacity to organize a system – or to maintain it in an organized state. As we shall discuss later, it becomes impossible to perform “useful” work without an input of both energy and information. Conversely, all work brings about a change in organization, hence information.
Organization is a reflection of order. A structure or system may be said to be organized if it exhibits order. Order is a non-random arrangement of the parts of the structure or system. Randomness is the opposite of order, keeping in mind that certain forms of apparent randomness exhibit significant order, eg, a perfectly uniform distribution. For this reason, the terms chaos and disorder are preferable. Any quantitative analysis of information must be based, at least in part, on measuring either the order, or the chaos of the system.
Analyzing the information content of a chaotic system is made more problematical by the fact that a system may only appear to be chaotic: That is, such a system actually is responding to a simple algorithm – the apparent unpredictability reflects the fact that trivial variations in initial conditions may have a major impact on the system’s final behavior.
Organization and information are, by definition, closely interlinked. However, they are different: One cannot have a shadow without light, but a shadow and light are not the same thing. A shadow is the manifestation of light interacting with an opaque object. Likewise, organization is the manifestation of information interacting with matter and energy.
It is important to emphasize the conceptual necessity for an abstract term such as “information”. Information is a quantity which may be altered from one form to another. Information is a quantity which may be transferred from one system to another. This is not true, at least to the same degree, for the more concrete terms “order”, “organization”, “pattern”, or “structure”. The matter parallels the difference between the terms “energy” and “heat”. Energy is being capable of being transformed from one form to another, as well as being transferred from one system to another. In contrast, the limitations of the less abstract concept “heat” (a quantity directly perceptible to our physical senses), cannot explain how heating a boiler causes a locomotive to move, or a light bulb to light up in response to the electricity generated by a steam turbine.
Likewise, “information” maybe transformed from one form to another, as for example, when dictating a manuscript: Patterns of sound waves end up transcribed as words on a printed page. It’s easy to understand that the information was transformed via the stenographer and printer, from the spoken to the written word. It is not clear how the oscillating molecules of air comprising the sound pattern end up as apparently unrelated patterns of dye molecules on a printed page. The matter becomes even more mysterious when one eliminates the human intermediaries and speaks into a voice-to-print device. The structure of the phonemes making up a word is not the same as the structure of the printed syllables making up the same word. The information content, however, may be considered the same for both.
Information, like energy, is an abstract quantity. Communications engineers have recognized since Hartley’s time, over half a century ago, that information may be treated as an abstract quantity. What the present work proposes is more than that, viz, that information, like energy is a physical reality.
To be more precise, heat (involving uncorrelated photons in a crystal or randomly moving molecules in a gas) is the product of the interaction between matter and energy. Structure is the product of the interaction between matter and pure information. Energy, in pre-relativity physics, was considered as the more abstract quantity which, when added to matter, manifests itself as structure (organization).
As will be discussed in a later chapter, such a conceptualization of information leads to a different quantitative definition from that of the communications engineers. Such a definition also differs from the standard dictionary definition which defines information as, for example: knowledge, news, or what is told. Dictionaries go on to define knowledge as all that is, or maybe known. Knowing is defined as: recognizing, perceiving with certainty, being aware (of), being acquainted with. There are other, more specialized meanings provided by dictionaries, but the gist is that information is either a form of knowledge, or equivalent to it. Dictionaries define knowledge and information purely in implicitly human terms. This is in marked contrast to the principle that information is a property of the universe – that it comprises the “internal” structure of the universe.
Human information may involve the perception of that “internal” structure. Every time scientists define a constant such as the gas constant, Avogadro’s number, Boltzmann’s or Planck’s constant, etc, they have discovered another aspect of the organization of the universe. Each such discovery represents the human perception of the information contained within physical systems.
Aspects of human information systems, including the terms knowledge, meaning, significance, intelligence, etc will be explored in a future work, Beyond Chaos. The present work is concerned with the physics of information systems – systems whose reality is independent of human perception and which therefore transcends it.
To sum up: All regular patterns contain information. The mathematics of chaos had demonstrated that even apparently highly irregular patterns, may be the product of some rather simple algorithm which underlies the chaos. To the argument that what we are really talking about is “patterns” and “organization”, the answer is that “information” is a more abstract generalization which, in the long run, one needs in order to measure it by some universal measure such as “bits”. It becomes as difficult to measure quantitatively a pattern or a structure in terms of bits without the aid of the abstract concept “information”, as it is to measure in joules the output of light by a lamp without the more abstract concept of “energy”.
Information is an implicit component in virtually every single equation governing the laws of physics. (25-28)
The first thing that needs to be addressed if we are to make a successful comparison between Stoniers concept of information and Spinoza’s notion of Idea is the thought that information can be “transferred”. I think that this is related to the way in which we view energy as some form of primal substance that can be poured into (or drained out of) various containers. I’m not sure how helpful this image is in either the case of energy or information. The addition of energy to a system is a transformative one. The system itself is changed. And I think Stonier is onto this with his idea that information itself, when added, changes the structure of what it is added to. There is, therefore, something of competing images here, images that have to do with how we view the boundaries of things. From a Spinozist point of view, therefore, when Stonier says:
The addition of information to a system manifests itself by causing a system to become more organized, or reorganized.
An organized system has the capacity to release or convey information.
I think it is better said that an organized system has the capacity to improve the organization of (the adequacy of the ideas of) systems outside of it. Information does not pour out of a system, into another, but rather communicates itself, interactively, through the improvement of the organization of things beyond it. In this way the physical object of a book does not “release or convey” but rather through interaction, re-organizes the materiality of the reader. Key to changing the metaphor we use to describe informational relations is to see that when there are such interactions nothing is being passed back and forth, but rather what is involved is the substantive change in the relational capacities of each distinct thing, in the context of something larger than each (be it a larger system, or the Universe itself).
Stonier in his re-imagining of information uses the concept to address itself to the problem of entropy. He works to show that entropy is not strictly equivalent to “heat” (which is one of its manifestations), a difference that actually marks out the need for an information science as structural changes in matter do not exclusively follow heat changes. As such he places organization and heat at odds to each other (heat, the move towards randomness, works against the move towards organization), but energy and information are actually part of a triangle of universal elements:
The application of energy expresses itself as heat which causes particles (molecules, photons, plasmons, etc) to vibrate and to move at random. In contrast, the application of information causes particles to be bound into fixed patterns and ordered motion. In that sense, heat may be considered as the antithesis of organization.
If heat is the antithesis of organization is heat, and by implication, energy the antithesis of information, that does not preclude the possibility that energy and information may interact to provide a mix which might be viewed as “energized information”, or alternatively as “structured energy”. INFORMATION and ENERGY must not be viewed as the opposites of a bipolar system, rather, they must be considered as the two angles of a triangle, with MATTER comprising the third (74-75)
This is problematic to a Spinoza/Stonier comparison, and I think Spinoza actually helps out here. Stonier wants to see something like a crystal at very low temperature as possessing an ideal of information, a structural coherence with very little entropy (heat). I think that this is a mistake in his visualization. Because I view the conatus as equivalent to energy, actually all things that exist possess both informational structure (what I want to call informational or ideational lean towards the Universe), and also the energy (tendency) to maintain that lean (entropy will be handled at another time). In fact the informational and energy dispositions are mutual expressions of each other. The introduction of heat (randomness) is actually an informational transformation from the outside. Instead of thinking of information as merely the internal structure of a thing, it is both the internal and relationship organization of a thing.
We can see this on the most fundamental level in examples of “energy” transfer, reconfigured to reflect exchanges of information. Stonier uses the classic of billiard balls:
…consider two billiard balls, one red, one white, rolling along on a billiard table at equal speed. The red one is moving in a north-easterly direction, while the white one is moving in a south-easterly direction. Let them meet in such a way, that the collision results in a reversal of direction: The red one now traveling south-east, while the white one travels north-east.
The question that one may ask is whether the two balls exchanged energy, or whether they exchanged information. Certainly the collision, involving a glancing blow, seemed not to affect appreciably the energy content of the system as a whole. Nor did the energy content of the individual balls appear to be affected appreciably since they continued moving at virtually undiminished speed. What was altered however, was the direction…To restate the question: Is the conservation of momentum a reflection of the fact that the two bodies merely exchanged information? (81)
Instead of seeing energy as conserved and “transferred” between objects, one can also describe such an interaction as an exchange of information. In fact, I suggest it is not the exchange of information so much as the informational re-orientation of each. The ideas of each ball, its informational properties, has changed through interaction. We can see the foundations of Spinoza’s panpsychism wherein each thing “thinks” (is made of ideas that make a difference in its capacities in the world).
Stonier himself provides an interesting example of the primary dichotomy he would like to set up between heat and organization, with an implicit tension between energy and information, that if warm-blooded mammalian brains. This is more than a mere exception I would suggest, but rather points to the problem of Stonier absolute contrast between energy and information itself. As he writes of the mammal and heat (randomness):
Present-day biological systems, with minor exceptions (eg, certain chemosynthetic bacteria), obtain their energy from the sun. Light, as we shall discuss later, is a form of energy with a high information component. In general, biological systems eschew heat – either as an energy input, or as a product. When heat is generated, it is the by-product of metabolic reactions and usually reflects an inefficiency in the system. The one clear exception is the production of heat to maintain efficiency of advanced metabolic systems operating in highly organized environments. To maintain the very high levels of structural information in the system, the changes in entropy associated with changes in temperature must be kept to a minimum. The most advanced information processing system known is the mammalian brain. When the temperature rises only slightly above a critical threshold (as with a high fever), the system begins to fail as the individual hallucinates. A relatively slight drop in temperature, on the other hand, leads to narcosis. Thus even relatively minor (heat-inducing) changes in entropy, change the delicate organization of the system so as to interfere with effective information processing.
Therefore, in the one exception where biological systems do produce heat and utilize heat, the function of the added heat is not to provide energy, but to maintain a stable temperature so as to minimize externally induced entropy changes. In other words, heat is used to help stabiliize organization – it is the one instance where the controlled application of heat constitutes an input of information. (66-67)
As I have argued elsewhere when considering Spinoza as a Chaoplexicist, Is Spinoza a Cyberneticist, or a Chaocomplexicist?informational increases cannot be seen solely in terms of an internally defined relation, for instance the structure of crystal. Instead they have to be read as edge-riding properties at the border of chaotic distributions. For instance there cannot be any such heat/organization polarity. If the Universe achieve a degree zero state it would not have reached a state of maximum information. Instead, the heat (randomness) use by mammalian lifeforms is not an exception, but an expression of the informational transformations that make up the structure of the Universe. Organization is best not seen in contest with Energy, but rather Energy expressions are necessarily informational ones. Even a purely random, equilibrium distribution is informational. And information increases (what for Spinoza would be increases in the adequacy of ideas) are not expressed sheerly as “structure” but rather the ability to bestride structure and chaos. This is precisely what lifeforms do with “heat”, not eschewing it, but surfing it.
The locus of this reasoning I believe is found between the two, conflicting theories of Information and its relationship to entropy. Shannon, famously, linked the information content of a message to the surprise factor of its distribution. So if you received absolutely random message (taken to be utterly entropic), its information would be at maximum. Stonier, because he is not dealing with messages, but states, but an absolutely random distribution as the minimum of information structure. Truth be told, the answer lies between these two. A distribution, when seen as a message and measured for information, carries with it its relational capacities found in the reader of that distribution. In keeping with Shannon, the work that must be done in application of informational decoding of a random message is very high, so the message contain maximum information. If one is surprised very little by a message, it is composed of very few differences that make a difference to the reader. Its information is low. With Stonier, a random distribution of gas molecules composes very few differences that make a difference to the observer, so the information is low, but the reader/observer and the system/message have to be taken as a whole. The antithesis between these two perspectives is in their framing. If for instance we were to play a game where the exact distribution of gas molecules in a box near equilibrium state provides clue the game’s aim, suddenly the box is brimming with information, differences that make a difference. In fact, real world information differences, organizational relations that make a maximum of differences in the world, are those that oscilate or rather surf between both Stonier concept of fixed, structural, very low energy information, and Shannon’s very high entropy notion of message information. Maximal information, as lived, rides between this balance between structure and chaos. It is as Spinoza says,
E4p38 – Whatever so disposes the human Body that it can be affected in a great many ways, or renders it capable of affecting external Bodies in a great number of ways, is useful to man; the more it renders the Body capable of being affected in a great many ways, or of affecting other bodies, the more useful it is: on the other hand, what renders the Body less capable of these things is harmful.
To use a Stonier example, for a crystal at low temperature to be in a maximum state of information its constellations of elements would have to be in state in which they can effect or be affected in the greatest number of ways, and one is not sure that this is the case. Such a state is not just useful, I would say that it demarks the greatest adequacy of ideas , or informational orientation, as is historically possible. Part of this is because for Spinoza there is such thing as a state that has no information or organization.
There are additional difficulties to be handled in the equation between Spinoza’s “Idea” and Stonier’s “Information”, for instance the reality of entropy and the ultimately question of whether, or to what degree a “closed system” actually exists, has to be worked through. And there are several other aspects of Stonier’s theory that lend themselves to an elucidation of Spinoza’s thinking, for instance the way in which he re-reads changes in “potential energy” as changes in “information”, the moving of a system into a less probable state. These are things I cannot take up right here, hopefully in the future. It is more that Stonier’s view that information comprises an essential, transformational component of the Universe, just as energy and matter does provides a highly effective backdrop for understanding just what Spinoza means by Idea. What he means by Idea is Information. And it is precisely the distinction between human information and information as an abstraction that best brings out the differences Spinoza meant in both his epistemology and his ontology, the way in which there are distinct limits to what we know, but also that in knowing anything we are changing our informational relationship to both it and the world. Improving the Adequacy of our Ideas is perhaps best seen as improving our Informational organization of ourselves, thinking is position altering. And all things must be regarded as, in some sense, thinking.
I hope that this presentation has not be unfair to Stonier whose theory and book deserves much better treatment. I am not one who enjoys the detailed summaries of positions, and have used Stonier only as a peering into the possibilities of Spinoza’s thinking, both in terms of what he really meant, and how it might help us understand how things are. But Stonier’s theory is beautiful in its own right worth serious study for what he claims. My Spinozist adaptation is at best provisional.