Path: utzoo!attcan!uunet!mcvax!enea!kth!sics!sics.se!alf
From: alf@sics.se (Thomas Sj|land)
Newsgroups: comp.lang.prolog
Subject: The meaning of "declarative"
Message-ID: <ALF.88Oct2184429@hathor.sics.se>
Date: 2 Oct 88 17:44:30 GMT
Sender: news@sics.se
Distribution: comp.lang.prolog
Organization: Swedish Institute of Computer Science, Kista
Lines: 57


This is somewhat philosophical, but it turns up in all Prolog
classes when you try to motivate the language with something
else than its pragmatic virtues. (Please submit explanatory followups only.)

Statement 1, (commonly accepted, it seems, although the meaning is unclear):
	 "Prolog is a declarative language"

Statement 2, (viewpoint held by some experts who agree to statement 1):
	 "functional languages are NOT declarative, while logic programming
	  and equational languages are if they have completeness"

My view: "(declarative programming <-> programming without added intension)"
	  -> Prolog allows declarative programming but doesn't guarantee it.

	Indeterminism of the theorem prover
	which gives completeness has nothing to do with the "declarativeness".
	My intuition leads me to think of soundness rather than completeness
	when I hear the term "declarative".  Functional programs which are pure
	lambda expressions are declarative, even when they are "higher order"
	in contrast to first order logic programs, which are declarative only
	when the intension coincides with the Herbrand interpretation. This
	view rules out the use of difference structures a.o.t. from 
	"declarative logic programs".

	 The phrase "declarative programming" makes sense if your intension
	 of the program is understood as a database for a theorem prover and
	 not as a coded algorithm, i.e. any fact you intended to express in
	 the database can be concluded by the interpreter. Consider the difference
	 between the completeness for first order predicate calculus and
	 the incompleteness of arithmetic as expressed by the Peano axioms
	 as a first order theory. Coded algorithms in logic programs seems
	to me to have a similar relation to the underlying theorem prover
	for horn clauses as arithmetic to the underlying first order theory
	used to code them. A difference is perhaps that logic programming
	assumes a particular model, the Herbrand interpretation.

Question: Does "declarative" refer to the completeness of the theorem prover
	used to run the programming language or does it refer to the intension
	of the program ? 










	 

--
Thomas Sj|land
SICS, PO Box 1263, S-164 28 KISTA, SWEDEN
Tel: +46 8 752 15 42	Ttx: 812 61 54 SICS S	Fax: +46 8 751 72 30
Internet: alf@sics.se or {mcvax,munnari,ukc,unido}!enea!sics.se!alf