Gregory Chaitin is at the IBM Watson Research Center in New York. In the mid 1960s, when he was a teenager, he created algorithmic information theory (AIT), which combines, among other elements, Shannon's information theory and Turing's theory of computability. In the four decades since then he has been the principal architect of the theory. Among his contributions are the definition of a random sequence via algorithmic incompressibility, his information-theoretic approach to Gödel's incompleteness theorem, and the celebrated number ?. His work on Hilbert's 10th problem has shown that in a sense there is randomness in arithmetic, in other words, that God not only plays dice in quantum mechanics and nonlinear dynamics, but even in elementary number theory. His latest achievements have been to transform AIT into a theory about the size of real computer programs, programs that you can actually run, and his recent discovery that Leibniz anticipated AIT (1686). He is the author of nine books: Algorithmic Information Theory published by Cambridge University Press;Information, Randomness & Incompleteness and Information-Theoretic Incompleteness, both published by World Scientific; The Limits of Mathematics, The Unknowable, Exploring Randomness and Conversations with a Mathematician, all published by Springer-Verlag; From Philosophy to Program Size, published by the Tallinn Institute of Cybernetics; and Meta Math!, published by Pantheon Books. In 1995 he was given the degree of doctor of sciencehonoris causa by the University of Maine. In 2002 he was given the title of honorary professor by the University of Buenos Aires. In 2004 he was elected a corresponding member of the Académie Internationale de Philosophie des Sciences. He is also a visiting professor at the Computer Science Department of the University of Auckland, and on the international committee of the Valparaíso Complex Systems Institute

Turing's remarkable 1936 paper On computable numbers, with an application to the Entscheidungsproblem marks a dramatic turning point in modern mathematics. On the one hand, the computer enters center stage as a major mathematical concept. On the other hand, Turing establishes a link between mathematics and physics by talking about what a machine can accomplish. It is amazing how far these ideas have come in a comparatively short amount of time; a small stream has turned into a major river. I have recently completed a small book about some of these developments, Meta Math! It is currently available as an e-book on my personal website, and also from arxiv. org, and is scheduled to be published next year. Here I will merely give a few highlights.

My story begins with Leibniz in 1686, the year before Newton published his Principia. Due to a snow storm, Leibniz is forced to take a break in his attempts to improve the water pumps for some important German silver mines, and writes down an outline of some of his ideas, now known to us as the Discours de métaphysique. Leibniz then sends a summary of the major points through a mutual friend to the famous fugitive French philosophe Arnauld, who is so horrified at what he reads that Leibniz never sends him nor anyone else the entire manuscript. It languishes among Leibniz's voluminous personal papers and is only discovered and published many years after Leibniz's death.

In sections V and VI of the Discours de métaphysique, Leibniz discusses the crucial question of how we can distinguish a world in which science applies from one in which it does not. Imagine, he says, that someone has splattered a piece of paper with ink spots determining in this manner a finite set of points on the page. Nevertheless, Leibniz observes, there will always be a mathematical equation that passes through this finite set of points. Indeed, many good ways to do this are now known. For example, what is called Lagrangian interpolation will do.

So the existence of a mathematical curve passing through a set of points cannot enable us to distinguish between points that are chosen at random and those that obey some kind of a scientific law. How then can we tell the difference? Well, says Leibniz, if the equation must be extremely complex ("fort composée") that is not a valid scientific law and the points are random ("irrégulier").

Leibniz had a million other interests and earned a living as a consultant to princes, and as far as I know after having this idea he never returned to this subject. Indeed, he was always tossing out good ideas, but rarely, with the notable exception of the infinitesimal calculus, had the time to develop them in depth.

The next person to take up this subject, as far as I know, is Hermann Weyl in his 1932 book The Open World, consisting of three lectures on metaphysics that Weyl gave at Yale University. In fact, I discovered Leibniz's work on complexity and randomness by reading this little book by Weyl. And Weyl points out that Leibniz's way of distinguishing between points that are random and those that follow a law by invoking the complexity of a mathematical formula is unfortunately not too well defined, since it depends on what primitive functions you are allowed to use in writing that formula and therefore varies as a function of time.

Well, the field that I invented in 1965 and which I call algorithmic information theory provides a possible solution for the problem noticed by Hermann Weyl. This theory defines a string of bits to be random, irreducible, structureless, if it is algorithmically incompressible, that is to say, if the size of the smallest computer program that produces that particular finite string of bits as output is about the same size as the output it produces.

So we have added two ideas to Leibniz's 1686 proposal. First, we measure complexity in terms of bits of information, i. e., 0s and 1s. Second, instead of mathematical equations, we use binary computer programs. Crucially, this enables us to compare the complexity of a scientific theory (the computer program) with the complexity of the data that it explains (the output of the computer program).

As Leibniz observed, for any data there is always a complicated theory, which is a computer program that is the same size as the data. But that doesn't count. It is only a real theory if there is compression, if the program is much smaller than its output, both measured in 0/1 bits. And if there can be no proper theory, then the bit string is algorithmically random or irreducible. That's how you define a random string in algorithmic information theory.

I should point out that Leibniz had the two key ideas that you need to get this modern definition of randomness, he just never made the connection. For Leibniz produced one of the first calculating machines, which he displayed at the Royal Society in London, and he was also one of the first people to appreciate base-two binary arithmetic and the fact that everything can be represented using only 0s and 1s. So, as Martin Davis argues in his book The Universal Computer: The Road from Leibniz to Turing, Leibniz was the first computer scientist, and he was also the first information theorist. I am sure that Leibniz would have instantly understood and appreciated the modern definition of randomness.

I should mention that A. N. Kolmogorov also proposed this definition of randomness. He and I did this independently in 1965. Kolmogorov was at the end of his career, and I was a teenager at the beginning of my own career as a mathematician. As far as I know, neither of us was aware of the Leibniz Discours. Let me compare Kolmogorov's work in this area with my own. I think that there are two key points to note.

Firstly, Kolmogorov never realised as I did that our original definition of randomness was incorrect. It was a good initial idea but it was technically flawed. Nine years after he and I independently proposed this definition, I realised that it was important for the computer programs that are used in the theory to be what I call "self-delimiting"; without this it is not even possible to define my O number that I'll discuss below. And there are other important changes that I had to make in the original definitions that Kolmogorov never realised were necessary.

Secondly, Kolmogorov thought that the key application of these ideas was to be to obtain a new, algorithmic version of probability theory. It's true, that can be done, but it's not very interesting, it's too systematic a re-reading of standard probability theory. In fact, every statement that is true with probability one, merely becomes a statement that must necessarily be true, for sure, for what are defined to be the random infinite sequences of bits. Kolmogorov never realised as I did that the really important application of these ideas was the new light that they shed on Gödel's incompleteness theorem and on Turing's halting problem.

So let me tell you about that now, and I'm sure that Turing would have loved these ideas if his premature death had not prevented him from learning about them. I'll tell you how my O number, which is defined to be the halting probability of a binary program whose bits are generated using independent tosses of a fair coin, shows that in a sense there is randomness in pure mathematics.

Instead of looking at individual instances of Turing's famous halting problem, you just put all possible computer programs into a bag, shake it well, pick out a program, and ask what is the probability that it will eventually halt. That's how you define the halting probability O, and for this to work it's important that the programs have to be self-delimiting. Otherwise the halting probability diverges to infinity instead of being a real number between zero and one like all probabilities have to be. You'll have to take my word for this; I can't explain this in detail here.

Anyway, once you do things properly you can define a halting probability O between zero and one. The particular value of O that you get depends on your choice of computer programming language, but its surprising properties don't depend on that choice.

And what is O's most surprising property? It's the fact that if you write O in binary, the bits in its base-two expansion, the bits after the binary decimal point, seem to have absolutely no mathematical structure. Even though O has a simple mathematical definition, its individual bits seem completely patternless. In fact, they are maximally unknowable, they have, so to speak, maximum entropy. Even though they are precisely defined once you specify the programming language, the individual bits are maximally unknowable, maximally irreducible. They seem to be mathematical facts that are true for no reason.

Why? Well, it is impossible to compress N bits of O into a computer program that is substantially smaller than N bits in size (so that O satisfies the definition of randomness of algorithmic information theory). But not only does computation fail to compress O, reason fails as well. No formal mathematical theory whose axioms have less than N bits of complexity can enable us to determine N bits of O. In other words, essentially the only way to be able to prove what the values of N bits of O are, is to assume what you want to prove as an axiom, which of course is cheating and doesn't really count, because you are not using reasoning at all. However, in the case of O, that is the best that you can ever do!

So this is an area in which mathematical truth has absolutely no structure, no structure that we will ever be able to appreciate in detail, only statistically. The best way of thinking about the bits of O is to say that each bit has probability 1/2 of being zero and probability 1/2 of being one, even though each bit is mathematically determined.

So that's where Turing's halting problem has led us, to the discovery of pure randomness in a part of mathematics. I think that Turing and Leibniz would be delighted at this remarkable turn of events.

Now I'd like to make a few comments about what I see as the philosophical implications of all of this. These are just my views, and they are quite controversial. For example, even though a recent critical review of two of my books in the Notices of the American Mathematical Society does not claim that there are any technical mistakes in my work, the reviewer strongly disagrees with my philosophical conclusions, and in fact he claims that my work has no philosophical implications whatsoever. So these are just my views, they are certainly not a community consensus, not at all.

My view is that O is a much more disagreeable instance of mathematical incompleteness than the one found by Gödel in 1931, and that it therefore forces our hand philosophically. In what way? Well, in my opinion, in a quasi-empirical direction, which is a phrase coined by Imre Lakatos when he was doing philosophy in England after leaving Hungary in 1956. In my opinion, O suggests that even though math and physics are different, perhaps they are not as different as most people think.

What do I mean by this? (And whether Lakatos would agree or not, I cannot say.) I think that physics enables us to compress our experimental data, and math enables us to compress the results of our computations, into scientific or mathematical theories as the case may be. And I think that neither math nor science gives absolute certainty; that is an asymptotic limit unobtainable by mortal beings. And in this connection I should mention the book (actually two books) just published by Borwein and Bailey on experimental math.

To put it bluntly, if the incompleteness phenomenon discovered by Gödel in 1931 is really serious---and I believe that Turing's work and my own work suggest that incompleteness is much more serious than people think---then perhaps mathematics should be pursued somewhat more in the spirit of experimental science rather than always demanding proofs for everything. In fact, that is what theoretical computer scientists are currently doing. Although they may not want to admit it, and refer to P ? NP as an unproved hypothesis, that community is in fact behaving as if this were a new axiom, the way that physicists would.

At any rate, that's the way things seem to me. Perhaps by the time we reach the centenary of Turing's death this quasi-empirical view will have made some headway, or perhaps instead these foreign ideas will be utterly rejected by the immune system of the math community. For now they certainly are rejected. But the past fifty years have brought us many surprises, and I expect that the next fifty years will too, a great many indeed.

--------------------------------------------------------------------------------

References Chapter on Leibniz in E. T. Bell, Men of Mathematics, Simon & Schuster, 1937. R. C. Sleigh, Jr., Leibniz and Arnauld: A Commentary on Their Correspondence, Yale University Press, 1990. H. Weyl, The Open World: Three Lectures on the Metaphysical Implications of Science, Yale University Press, 1932, Ox Bow Press, 1994. First article by Lakatos in T. Tymoczko, New Directions in the Philosophy of Mathematics, Princeton University Press, 1998. M. Davis, The Universal Computer: The Road from Leibniz to Turing, Norton, 2000. Play on Newton vs. Leibniz in C. Djerassi, D. Pinner, Newton's Darkness: Two Dramatic Views, Imperial College Press, 2003. J. Borwein, D. Bailey, Mathematics by Experiment: Plausible Reasoning in the 21st Century, A. K. Peters, 2004. J. Borwein, D. Bailey, R. Girgensohn, Experimentation in Mathematics: Computational Paths to Discovery, A. K. Peters, 2004. G. Chaitin, Meta Math! The Quest for Omega, Pantheon, 2005.

Purchase
Anyone may subscribe or purchase back issues.

Advertise
Ideal for advertising jobs, books, software and other products. Here are current rates and deadlines.

Write
Notes on how to submit an article.

Some books by Gregory J. Chaitin. :
The Unknowable. Springer. 1999
Exploring Randomness. Springer. 2001
Conversations with a Mathematician: math, art, science and the limits of reason. Springer. 2002