Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: SUO: Re: Parse Of Things Remembered




>Yang Yun wrote:
> >
> > > > It also argues for a divide and conquer approach.
> > > > This is why I support Nicola Guarino's recent call
> > > > for a modular approach to reference ontologies.
> >
> > > Modular is nice, of course, if and when you can get it.
> > > There is a catch here though, "triadic irreducibility".
> >
> > Google search for "triadic irreducibility"
> > did not match any documents, whatever ..
> > yy
>
>¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
>
>Y^2,
>
>FAST Search:
>
>http://www.alltheweb.com/cgi-bin/advsearch
>
>gave this very interesting one from the CG List:
>
>http://www.virtual-earth.de/CG/cg-list/msg02192.html
>
>If you use the exact phrase option then you will probably also
>need to try "triadically irreducible" and "irreducibly triadic".
>Or you could always try reading about it in Peirce -- his work
>is in all of the big libraries -- you remember libraries dontcha?

Allow me to interject a note of caution, before we waste yet more 
time on this old chestnut.

Peirce proved a result which he thought very important and 
fundamental. It is indeed correct, but it isnt deep or very 
important, so he was both right and wrong about it. The result was 
this. Consider nodes in a graph and classify them by their 'degree', 
ie the number of arcs that link to them. Then in order to build 
'arbitrary' graph structures one must have at least one node of 
degree three. (Roughly, if you only have nodes of degree 2 then all 
you can do is make long rows of nodes; but once you have degree-three 
nodes, you can build arbitrarily branching structures by imitating 
degree-n nodes by sequences of n trinary nodes linked together, as in 
LISP or RDF.) This is Peirce's 'irreducible triadicity', and he 
thought that it meant that there was something fundamentally trinary 
about identity. However, the result is very fragile. If you allow 
hyperlinks in the graphs, it fails immediately. If you take Peirce's 
own algebra of identity and close it under inverses, there is an 
operation that merges two binary links into a trinary one, so that 
irreducibiity fails. And in any case the logical conclusion only 
follows if you restrict yourself to algebraic notations: as soon as 
you allow quantification into the language, everything can be said 
using binary and unary relations, as has been well-known now for 
about 50 years or more. So for this particular issue, don't bother 
with the library: just ignore it. It has absolutely nothing whatever 
to do with modularity.

Peirce was misled by an analogy between logic and chemistry, which 
wasnt a bad idea in 1885, but seems kind of daft in hindsight.  He 
seems to have thought that the associations between relations and 
their instances, which he was encoding as arcs in his graphs, were 
like valency in chemistry, so that relations (including that of 
identity) had a kind of intrinsic associated 'atomic number', from 
which it follows that the relation of two things being identical has 
to be different from that of three things being identical. This 
particular analogy was similar in some ways to Kepler's idea that the 
planetary distances from the sun arose from packing the regular 
solids into nesting spheres: neat, ingenious, apparently successful; 
but wrong.

---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes