Boole, Shannon and the Electronic Computer

Photo of George Boole by Samuel Prout Newcombe  Source: Wikimedia Commons

Photo of George Boole by Samuel Prout Newcombe
Source: Wikimedia Commons

In 1847, the self-taught English Mathematician George Boole (1815–1864), whose two hundredth birthday we celebrated last year, published a very small book, little more than a pamphlet, entitled Mathematical Analysis of Logic. This was the first modern book on symbolic or mathematical logic and contained Boole’s first efforts towards an algebraic logic of classes.

6882068-M

Although very ingenious and only the second published non-standard algebra, Hamilton’s Quaternions was the first, Boole’s work attracted very little attention outside of his close circle of friends. His friend, Augustus De Morgan, would falsely claim that his own Formal Logic Boole’s work were published on the same day, they were actually published several days apart, but their almost simultaneous appearance does signal a growing interest in formal logic in the early nineteenth century. Boole went on to publish a much improved and expanded version of his algebraic logic in his An Investigation of the Laws of Thought on Which are Founded the Mathematical Theories of Logic and Probabilities in 1854.

LoT-GB200-news-story-mag

The title contains an interesting aspect of Boole’s work in that it is an early example of structural mathematics. In structural mathematics, mathematicians set up formal axiomatic systems, which are capable of various interpretations and investigate the properties of the structure rather than any one specific interpretation, anything proved of the structure being valid for all interpretations. Structural mathematics lies at the heart of modern mathematics and its introduction is usually attributed to David Hilbert, but in his Laws of Thought, Boole anticipated Hilbert by half a century. The title of the book already mentions two interpretations of the axiomatic system contained within, logic and probability and the book actually contains more, in the first instance Boole’s system is a two valued logic of classes or as we would probably now call it a naïve set theory. Again despite its ingenuity the work was initially largely ignored till after Boole’s death ten years later.

As the nineteenth century progressed the interest in Boole’s algebraic logic grew and his system was modified and improved. Most importantly, Boole’s original logic contained no method of quantification, i.e. there was no simple way of expressing simply in symbols the statements, “there exists an X” or “for all X”, fundamental statements necessary for mathematical proofs. The first symbolic logic with quantification was Gottlob Frege’s, which first appeared in 1879. In the following years both Charles Saunders Peirce in America and Ernst Schröder in German introduced quantification into Boole’s algebraic logic. Both Peirce’s group at Johns Hopkins, which included Christine Ladd-Franklin or rather simply Christine Ladd as she was then, and Schröder produced substantial works of formal logic using Boole’s system. There is a popular misconception that Boole’s logic disappeared without major impact, to be replaced by the supposedly superior mathematical logic of Whitehead and Russell’s Principia Mathematica. This is not true. In fact Whitehead’s earlier pre-Principia work was carried out in Boolean algebra, as were the very important meta-logical works or both Löwenheim and Skolem. Alfred Tarski’s early work was also done in Bool’s algebra and not the logic of PM. PM first supplanted Boole with the publication of Hilbert’s and Ackermann’s Grundzüge der theoretischen Logik published in 1928.

It now seemed that Boole’s logic was destined for the rubbish bin of history, a short-lived curiosity, which was no longer relevant but that was to change radically in the next decade in the hands of an American mathematical prodigy, Claude Shannon who was born 30 April 1916.

Claude Shannon Photo by Konrad Jacobs Source: Wikimedia Commons (Konrad Jacobs was one of my maths teachers and a personal friend)

Claude Shannon
Photo by Konrad Jacobs
Source: Wikimedia Commons
(Konrad Jacobs was one of my maths teachers and a personal friend)

Shannon entered the University of Michigan in 1932 and graduated with a double bachelor’s degree in engineering and mathematics in 1936. Whilst at Michigan University he took a course in Boolean logic. He went on to MIT where under the supervision of Vannevar Bush he worked on Bush’s differential analyser, a mechanical analogue computer designed to solve differential equations. It was whilst he as working on the electrical circuitry for the differential analyser that Shannon realised that he could apply Boole’s algebraic logic to electrical circuit design, using the simple two valued logical functions as switching gates in the circuitry. This simple but brilliant insight became Shannon’s master’s thesis in 1937, when Shannon was just twenty-one years old. It was published as a paper, A Symbolic Analysis of Relay and Switching Circuits, in the Transactions of the American Institute of Electrical Engineers in 1938. Described by psychologist Howard Gardner as, “possibly the most important, and also most famous, master’s thesis of the century” this paper formed the basis of all future computer hardware design. Shannon had delivered the blueprint for what are now known as logic circuits and provided a new lease of life for Boole’s logical algebra.

17228_2

Later Shannon would go on to become on of the founders of information theory, which lies at the heart of the computer age and the Internet but it was that first insight combining Boolean logic with electrical circuit design that first made the computer age a viable prospect. Shannon would later play down the brilliance of his insight claiming that it was merely the product of his having access to both areas of knowledge, Boolean algebra and electrical engineering, and thus nothing special but it was seeing that the one could be interpreted as the other, which is anything but an obvious step that makes the young Shannon’s insight one of the greatest intellectual breakthroughs of the twentieth century.

7 Comments

Filed under History of Computing, History of Logic

7 responses to “Boole, Shannon and the Electronic Computer

  1. telescoper

    Reblogged this on In the Dark and commented:
    A reblog to mark the centenary of the birth of Claude Shandon, pioneer of information theory..

  2. There’s a podcast of a good RTE documentary on Boole at this link, if you can handle the ads
    http://www.rte.ie/player/ie/show/the-genius-of-george-boole-30003666/

  3. You might look a little further into the contributions of C.S. Peirce, who lectured at Harvard University and the Lowell Institute (1865–1866) on what he called the “laws of information” and who is known for his early (maybe the first) proposal of how electrical circuits could be used to compute Boolean functions.

    • Peirce’s existential graphs, a simple graphical formulation of first-order predicate calculus, are also fascinating:

      http://www.jfsowa.com/pubs/egtut.pdf

      “Peirce’s existential graphs (EGs) are the simplest, most elegant, and easiest-to-learn system of logic ever invented. Yet most logicians have never used them or even seen them. “

  4. miss three

    thanks for Boole

  5. Pingback: Whewell’s Gazette: Year 2, Vol: #38 | Whewell's Ghost

  6. Pingback: first read on wordpress! | vamsikrishnab

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s