from Peter Radford
Keeping the conversation going.
Let’s start with Shannon, from his personal papers published in 1993 …
“The word ‘information’ has been given different meanings by various writers in the general field of information theory. It is likely that at least a number of these will prove sufficiently useful in certain applications to deserve further study and permanent recognition. It is hardly to be expected that a single concept of information would satisfactorily account for the numerous possible applications of this general field.”
So Shannon is fine with an eclectic vision of information, and expects a variety of definitions to emerge as useful. This seems extremely wise. Especially given his own somewhat narrow version.
Weaver, in 1949, expanded on this ecumenical approach when he proposed a three pronged approach to understanding information. He broke the analysis of information down into three basic areas of concern: technical issues to do with the quantification of information [which was where Shannon’s greatest insights lie]; semantic issues relating to meaning and truth; and a final category to do with the way in which information affected human behavior.
All the above information is summarized on page 81 of Luciano Floridi’s “The Philosophy of Information” which is an excellent read. Floridi has also provided us with “The Blackwell Guide to the Philosophy of Computing and Information”, which gives the topic a broad survey and is well worth the effort. For those of you who want to get a taste of the way Floridi surveys the topic I suggest chapter four in the Blackwell Guide, which is simply entitled “Information” and begins thus …
“Information “can be said in many ways” just as being can [Aristotle, Metaphysics] and the correlation is probably not accidental. Information, with its cognate concepts like computation, data, communication etc., plays a key role in the ways we have come to understand, model, and transform reality. Quite naturally, information has adapted to some of being’s contours.”
Information, then, is a tangled web and anyone making assertions of perfect understanding or singular definition is on thin ice.
But I am less interested in the philosophy of information than in its practical impact. I want us to live in that third aspect of Weaver’s analysis. Why? Because, being most interested in how information appears in economic analysis, I am concerned with design.
Economists have for too long ignored the effort it takes to produce the goods and services we consume. Such entities enter the scene fully formed and are then exchanged. As I have argued before, production and the infrastructure needed for production is an immense portion of aggregate economic activity and yet is always — or nearly always — shunted off to one side as uninteresting. Economics concerns itself predominantly with what happens after production, when stuff hits the market. The problem with this focus is that it is far too narrow. Production is just as important, and production is essentially an information process.
Yes, information is embodied into products.
We call that information design. And it is design that allows us to impose order on the previously disordered natural environment so as to extract value from it. Imposing order requires energy, and so we have the three components of fundamental economic analysis: information, the natural resources of the environment, and energy. I have argued that these three are more useful to us than the standard capital and labor because they elevate information onto center stage, which is where it belongs. Besides if you think defining information is difficult try parsing out exactly what we mean by either capital or labor. Shannon’s dictum about the variety of definitions applies equally well, if not more, to them.
For those of you skeptical of this consider a typical product you use. Imagine it being broken. Imagine, as an example, a motor car after a major accident. The car is unusable. It has lost considerable value. Perhaps it can be repaired, but in its damaged state it is worth a lot less. Why? The metal and other natural resources of the car still exist. They have not been lost. So it is not the resource base of the car that is the core of value. We will need to consume more energy to fix the car, so there’s a source of loss. But what is that energy doing? It is allowing us to re-order the material of the car to re-capture its design. It is the design of the far where the largest value resides. And design is information.
Incidentally for the truly pedantic we can also argue that consumption of a good or service is also the destruction of design. The loss of form, the dissipation of order, is the very definition of consumption.
Our ability to impose ever more complex designs onto the material substrate of our environment, by combining energy with design, is the source of our prosperity. Doing that en masse requires an ability to replicate the process, so encoding the design appropriately is the essence of mass production. Increasing our effectiveness in both design and in the combination of energy and materials is where we extract rising productivity.
For those of you still skeptical, I took the above example of the motor car from Cesar Hidalgo’s excellent book: “Why Information Grows. The Evolution of Order, from Atoms to Economics”. Hidalgo uses a Bugatti as the car in his example, which is on page 12 of his book. Of course I have taken liberties with his original story, but the message is the same.
As Shannon indicated in the quote I gave above, we can get drawn into endless and tiring debates as to what information is or isn’t. Clearly it is an elastic term. I have just given a very short description of how I see it being useful in economics. This is beyond the normal use of the word in economic discussions. Economic is not simply about prices and the communication of prices. It is also about discovery and the production of goods and services. Approaching questions of productivity and growth through the lens of information and its accumulation in society is a more fruitful one, especially when we consider the difficulty economists have with the notion of ‘total factor productivity’. Had we been using information in our production functions all along TFP would be a less daunting object.