The ACM’s A.M. Turing Award is the Computer Sicence equivalent of the Nobel Prize or the Fields Medal in Mathematics. In 2005, the winner was Peter Naur “For fundamental contributions to programming language design and the definition of Algol 60, to compiler design, and to the art and practice of computer programming.” Naur’s essay “Programming as Theory Building” is one of my personal favourites (try using it as ammunition any time some client starts blathering on about how they’re going to build a “software factory” to build IT Systems just like they build cars—with programmers as completely interchangeable, disposable parts!)
But what’s weird is the Turing Lecture he gave, called “Computing Versus Human Thinking”. In it, he manages to trash Alan Turing, philosophy in general, all of psychology (especially cognitive psychology), and to accuse the ACM (among others) of censorship for declining to publish his papers on his theories of mental life. The lecture itself is a little hard to get a read on. He claims to have created a whole new model for human thinking, based almost entirely on the psychological works of William James (1890) and the neurological work of Charles Sherrington (1906). Everything studied or written since then he implies is hopelessly defective.
Hmmm…. Well, I suppose he could be right. Geniuses often see what others don’t. But I can’t help but be reminded of Stephen Wolfram’s “New Kind of Science”, in which he claimed that every scientist and mathematician who ever lived before him was completely blind and wrong and he alone of all humanity was brilliant enough to discover the truth of the universe. Ummm… ok. Possible, but not too plausible, doncha think?
Update: BTW, you can read the gory details of Naur’s Synapse-State Theory of Mental Life (pdf) and judge it for yourself.
Interesting. I just read the article in CACM myself, and had the same thoughts. Either he is a misunderstood genius, of he suffers from some sort of megalomania. Funny thing is, that things are just not as clear-cut as he seems to perceive them to be. For example, the semiotics of C.S. Peirce (influenced by W. James IIRC), have a strong following among some of the people Naur criticizes. Much of his “Synapse-State” stuff reminds me of Marvin Minsky’s _Society of Mind_, which is an old, but interesting book. But certainly this implies that Naur’s ideas are neither as innovative nor as “heretic” as he would like to think.
I think Naur has always been a man of sharp opinions. I seem to recall talking to CS students at DIKU, who mentioned the way he almost attempted mind-control in the way he imposed his normative vocabulary (“datalogy” for CS, “datamat” for computer, etc.) I have also been reading what he thought of the development of Algol 68, and he certainly wasn’t holding anything back in his attacks on van Wijngarden, to the point that I am suspecting he was trying to hide his failure to understand van Wijngarden’s two-level grammar.
One thing I think is particularly bad is the way he disses psychology and philosophy, without having any significant academic base in those fields himself. He has some good points, but a little less “attitude” would bring them across much better. And sometimes it is painfully obvious that the psychology he criticizes belongs to a tradition that has been out of fashion for a long time. (Behaviourism, for example.)
-Lasse Hillerøe Petersen
Thank you for the thoughtful comment, Lasse. I, too, would have found Naur’s ideas more intriguing if he had presented them with a little less attitude. It’s also interesting to know that he has previously demonstrated a tendency towards dogmatism. Cheers.
I read the article in CACM a couple of days ago. The types of conscious thinking that we experience based on the model he describes (short term memory, train of thought) made sense to me. On the other hand, I was disturbed by his use of “nodes” as a central element of his model. I had not encountered nodes before in the context of neurophysiology, and he did not bother to define them or present physical evidence of their existence.
I also found his description of how “thought goes on” similar to my experience of consciousness, in a way I’ve never felt about, say, a neural network.
Since then I’ve been reading James’ Psychology: The Briefer Course. The longer Principles of Psychology is available online at http://psychclassics.yorku.ca/James/Principles/index.htm
He has a remarkably clear style. Just thinking about thinking I find difficult, but he not only isolates the different activities during though, but manages to communicate it too.
Just read Naur’s “A Synapse-State Theory of Mental Life” (http://www.naur.com/synapse-state.pdf).
In this he does very little but disparage Hebbs and paraphrase James. His addition seems to be a suggestion that there might be 5 different parts to the thought apparatur.
At the end he states the work was submitted to Nature and Neuron, but rejected without being refereed. Of course it wasn’t, there’s nothing novel of worth in it, and I can even tell that.