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

SUO: RE: ontology as science




John:
 
Okay.  I agree that lexicographers can treat language use as data and be as objective about its data analysis as any other scientist.  I'll also accept that lexicographers can account for and predict semantic drift.  But what are the ramifications of this on ontology design or engineering?  (Specifically of the SUO.)  This raises a methodological characteristic of ontology development that hasn't been (or I haven't seen) discussed in the SUO group yet.  My perception has been that the sole goal as be the development of discrete ontologies with mappings/relationships between them and - perhaps if we're lucky - an "upper" ontology that relates all the others.  Yet this characteristic/feature - semantic drift - requires some method be included with the SUO to govern its evolution over time in response to changing requirements.  (Likewise, some method is needed to account for language and contextual differences.)  Perhaps we are not yet at a stage to discuss these, but I haven't!
  seen even a suggestion that such work is planned.  (I'll again admit that I haven't followed ALL of the SUO discussions in detail, so I might have missed such a reference.  :-)
 
Bill

	-----Original Message----- 
	From: John F. Sowa [mailto:sowa@bestweb.net] 
	Sent: Thu 6/19/2003 11:11 AM 
	To: Burkett, Bill; sowa@bestweb.net; John F. Sowa; standard-upper-ontology@ieee.org 
	Cc: 
	Subject: RE: ontology as science 
	
	

	Bill,
	
	I believe that the following point contains the crux of
	the misunderstanding:
	
	WB> However, #2 and #3 will always be subjective and therefore
	> of questionable testability from a scientific perspective.
	
	Yes, the way people use words depends on their subjective
	inclinations.  But other people can study the data about
	how people use words in a purely objective way.
	
	For example, the behavior of any species, from bacteria to
	cats to monkeys, can be studied in purely objective ways,
	even though that behavior is based on those organisms' internal
	(i.e., subjective) processes.
	
	As an outside observer, a lexicographer can be just as
	objective as any other scientist, and the results are every
	bit as objectively testable and repeatable as the results
	of any other science.
	
	In fact, linguists can reconstruct ancient languages, such
	as Egyptian hieroglyphics, from purely objective data without
	ever meeting or talking with an ancient Egyptian.  And the
	reconstructions can be and have been independently verified
	by other linguists.
	
	WB> With respect to semantics, however, it [lexicography]
	> cannot take into account the semantic drift of words in a
	> language as a language evolves over time.  Nor can it take
	> into account different languages except as a distinct data
	> sets.
	
	What???  Lexicographers analyze semantic drift in exactly
	the same way that geologists analyze continental drift.
	They can even make predictions about which kinds of terms
	are most likely to drift and which kinds tend to be stable.
	
	For example, value judgments, such as "good", "bad", or
	"beautiful" tend to drift most rapidly, while references to
	physical objects tend to be preserved over much longer
	periods.  The Indoeuropean word for salmon (lachs) is still
	common in modern languages, as are the words for mother,
	father, sister, brother, beech tree, etc.  Since there were
	no salmon in India, the word "lachs" was transferred to the
	the color from which we have the English "lacquer".  And the
	Indoeuropean word "sneg" from which we get English "snow"
	was transferred to a word for "slippery" in India.
	
	The German word for "good" (gut) is cognate with the English,
	but bad (boese) and beautiful (schoen) are not.  And the Latin
	"bonus", "malus", and "pulcher" have no relation to the
	Germanic terms, the Greek terms, the Slavic terms, or the
	Sanskrit terms.  Even "pulcher" was replaced by "bellus"
	in modern Romance languages.
	
	The words for abstractions, such as legal and sacred terms,
	have also been preserved for thousands of years (including
	the words "legal" and "sacred", which are derived from
	Indoeuropean roots of similar meanings).
	
	All of this is objectively testable and repeatable by
	independent observers.
	
	John