On Fri, Jul 27, 2012 at 6:20 PM, Dan Bron <j...@bron.us> wrote:
> Raul wrote:
>> Anyways, the definition which is relevant when cap is the left tine of
>> a fork is a passive definition, and not an imperative definition. And
>> making this distinction -- that it's being used passively -- seems
>> worthwhile.
>
> There are several issues with this interpretation. The foremost is that [:
> is _not being used at all_ . Not imperatively, not passively, not as quoted
> code, just - not. [1]
The parser uses the symbol to control its activities. That is all
that ever happens in passive contexts.
> Any attempt to impose some kind of meaning to [: in the context of capped
> fork will inevitably founder, because capped fork is a grammatical rule.
> Grammar precedes semantics ("meaning"). At the point of interpreting capped
> fork, everything is manipulated symbolically; opaquely; meaning does not
> enter in. There is no interpretation or wordsmithing that can find its way
> around this obstacle.
I would say this differently:
Grammar is an inextricable part of semantics. We usually treat it as
"obvious" but grammar is a reflection of the meanings of the symbols
we are using.
In J, we have conjunctions and adverbs whose execution can control
both the grammar and the semantics of the sentence we are evaluating.
In English, we need to treat a wide variety of meanings (since
practically all english words are ambiguous references to numerous
experiences and meanings), refining our interpretation until we have
something that makes sense to us.
> The analogy to gerunds does not hold water. Gerunds are simply quoted code.
That was meant to be an illustration of the concept of "passive
meaning" in a J sentence.
That said, I am not sure that you are drawing a meaningful distinction here.
For example, consider:
(<'[:')`:6 +/ *
> Cap is simply a symbol that triggers an otherwise impossible grammatical
> production. It is not dormant, or passive, or even used.
Here is an example of cap not being used:
(3 :'') +/ *
> It is noticed, and immediately discarded. It has no "definition" when it's
> the left tine of a fork. Not imperative, not passive, just - not.
Instead, I would say: the monadic and dyadic definitions of cap are
ignored when it's the left tine of a capped fork.
I think it's a mistake to declare that only these parts of verb
definitions are valid definitions. That would mean, for example, that
=: has no definition.
For me, a J symbol has a definition if it's described in the
dictionary, and has been implemented by the J interpreter.
--
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
