John S and John B:
<snip/>
RC> 1. Paul
goes to the water fountain;
>>
>>
2. Wanda goes to the water fountain;
The only combination that can be derived by strict
deduction
is the conjunction of #1 and #2.
Please forgive me for dropping into metaphor, but I think this is
equivalent to stating that, given a relational table of data, all rows are
true. Therefore the conjunction of all rows is also true, for that
column.
So the point is that the conjunction may be true by definition, or by
construction, of the way we interpret relational tables.
But each entry in the table (of the two S0, S1 in this case) is a
sequence of symbols, lisp-like in my personal projection. And every
symbol came from some symbol producing input source. So why not view each
entry as a conjunct of the symbols in the sequence?
Using the same reasoning I offered earlier about relational tables, I
can analyze sentences as lisp lists. Every symbol in a sequence can eval
to Si := false or to a new sequence Si := f(Si). So why not view it as a disjuncts
of its symbols, any of which can be false or otherwise.
In either case, if I want to look for a match to some third querying
statement, filled with symbols, including some variables that have to be
unified across a binding to S0 or to S1, I can then state a reasonable definition
of EQUALITY among rows in the table. A query row is EQUAL to any row that
can be retrieved and matched against said query row after unification.
But EQUALITY is just a constant value returned from a comparison
function, which could also provide < and > or even just /= functinality.
If I can COMPARE any two rows under unification, that seems to me to be more
basic to the conversion of reality to symbols than the consideration of whether
to start with conjuncts or disjuncts.
Now the Reality => FOL translation has Existence, Disjunction, Comparison
and Equality. That's an excess of riches, so to speak. This is way
more than we need, as a basis, to build the axioms and the deduction
mechanisms. What can be struck from the list?
-Rich
Sincerely,
Rich Cooper
EnglishLogicKernel.com
Rich AT EnglishLogicKernel DOT com