[Top] [All Lists]

Re: [ontolog-forum] Body Parts and Early-Learned Verbs

To: ontolog-forum@xxxxxxxxxxxxxxxx
From: John F Sowa <sowa@xxxxxxxxxxx>
Date: Thu, 05 Sep 2013 08:45:52 -0400
Message-id: <52287D00.3020304@xxxxxxxxxxx>
Lars,    (01)

I have no objection to mentioning anyone who has made any significant
contribution to the development of science and technology.    (02)

> I guess when (rightly) mentioning Otto Selz (for schema[tic anticipation]),
> one should also mention Richard Semon (for engram[matic association]).    (03)

But I believe that Selz made a more significant contribution than Semon.
(For a very brief intro, see the Wikipedia entries for Otto Selz,
Richard Semon, and 'engram'.)    (04)

In short, Semon contributed the word 'engram', which continued to be
used for some time.  But the two volume _Principles of Psychology_
by William James (1890) presented an excellent summary of 19th century
psychology without using that word.    (05)

The word 'schema' comes from Greek.  Aristotle used it repeatedly.
Kant used the word 'schema' in a way that influenced most of the modern
uses.  C. S. Peirce said that Kant's brief comments about schemata were
very significant, and that he should have developed them as the main
theme of his philosophy.    (06)

Selz was strongly influenced by Kant, and he did far more than coin
a few words.  In my studies of AI, I read the book _Thought and Choice
in Chess_ by Adriaan de Groot, who was a psychologist and a chess
master.  That book was a 1965 translation of de Groot's 1946 PhD
dissertation (written in Dutch). But it had a chapter that summarized
Selz's theories in a way that seemed very similar to methods of AI in
the 1960s.  That is why Newell and Simon invited him to visit CMU.    (07)

When I was writing my 1984 book, _Conceptual Structures_, I was
intrigued by de Groot's summary of Selz's psychology.  I didn't
read all of Selz's two-volume magnum opus (1913 and 1922), but I
laboriously worked my way through some of the critical sections.    (08)

For one of Selz's horribly convoluted German sentences, I asked one
of my colleagues at IBM for help (Werner Buchholz, a native German
speaker and the person who coined the word 'byte'.)  The two of us
spent 15 minutes disentangling one sentence.    (09)

But Selz did have an excuse for the complexity.  He did not know
anything about modern computers, recursive functions, or AI,
and he was groping for words to express some innovative ideas:    (010)

  1. Schemata, which are similar in structure to the frames and
     semantic networks of the 1960s and '70s.    (011)

  2. Schematic anticipation as a method that was very similar
     to the pattern-directed invocations in many AI systems.    (012)

  3. Top-down backtracking procedures, which were implemented
     in game-playing programs, NLP parsers and logic-programming
     systems such as Prolog.    (013)

For more detail, see the following book, which includes an article
by Herb Simon: "Otto Selz and information-processing psychology".    (014)

Frijda, N. H., & A. D. de Groot, eds. (1981) _Otto Selz: His
Contribution to Psychology_, Mouton, The Hague.    (015)

And by the way, Simon coined the word 'chunk' as a replacement
for 'schema'.  When Minsky published his famous paper on frames
in 1975, Simon wasn't impressed.  He said that frames were very
similar to chunks.  He was right, but he should have cited Selz
(whom he did cite in other publications).    (016)

In summary, Selz should be given credit for inventing or at least
anticipating frames, pattern-directed invocation, and the top-down
backtracking methods.  If some unfortunate circumstances hadn't
intervened (i.e., Hitler), he would probably be famous.    (017)

John    (018)

Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J    (019)

<Prev in Thread] Current Thread [Next in Thread>