SOME HISTORICAL NOTES
From: Stephen Wolfram, A New Kind of Science
Notes for Chapter 12: The Principle of Computational Equivalence
Section: Intelligence in the Universe
Teleology. There is a common tendency to project human purposes onto natural objects and events - and this is for example almost universally done by young children. Ancient beliefs often held that things in nature are set up by a variety of gods for a variety of purposes. By 400 BC, following ideas of Anaxagoras, Socrates and Plato discussed the notion that things in nature might in effect be optimally designed for coherent purposes by a single mind. Around 350 BC Aristotle claimed that a full explanation of anything should include its purpose (or so-called final cause, or telos) - but said that for systems in nature this is often just to make the final forms of these systems (their so-called formal cause). The rise of monotheistic religions led to the widespread belief that the universe and everything in it was created for definite purposes by a single god. But the development of mathematical science in the 1600s - and its focus on mechanisms ("efficient causes") - led away from ascribing explicit purposes to physical systems. In the mid-1700s David Hume then claimed on philosophical grounds that we fundamentally have no basis for ascribing purposes to any kind of natural system - though in the 1790s Immanuel Kant argued that even though we cannot know whether there really are such purposes, it is still often necessary for us to think in terms of them. And in fact the notion that systems in biology are so complex that they must have been intelligently designed for a purpose remained common. In the late 1800s Darwinian evolution nevertheless suggested that no such purposeful design was necessary - though in a sense it again introduced a notion of purpose associated with optimization of fitness. Ever since the 1700s economics had been discussed in terms of purposeful activities of rational agents. In the early 1900s there were however general attempts to develop mechanistic explanations in the social sciences, but by the mid-1900s purpose was again widely discussed, especially in economics. And in fact, even in physics, a notion of purpose had actually always been quite common. For whenever a physical system satisfies any kind of implicit equation, this defines a constraint that can be viewed as corresponding to some kind of purpose. (See page 942.) That something like a notion of purpose is being used has been more widely recognized for variational principles like the Principle of Least Action in mechanics from the mid-1700s. Results in the late 1900s in astrophysics and cosmology seemed to suggest that for us to exist our universe must satisfy all sorts of constraints - and to avoid explaining this in terms of purpose the Anthropic Principle was introduced (see page 1030). What I do in this book goes significantly further than traditional science in getting rid of notions of purpose from investigations of nature. For I essentially always consider systems that are based on explicit evolution rules rather than implicit constraints. And in fact I argue that simple programs constructed without known purposes are what one needs to study to find the kinds of complex behavior we see.
Stephen Wolfram, A New Kind of Science (Wolfram Media, 2002), page 1185.
© 2002, Stephen Wolfram, LLC