SUO: Re: Re: Sign Relations & Communication
From: "Jon Awbrey" <jawbrey@oakland.edu>
> SR: http://robustai.net/mentography/awbreySignRelation.gif
>
> JA: From my peirce-spective this looks like a genuine improvement!
> In particular, the appearance of degree-3 nodes in your graph,
> to wit, the nodes labeled "L1", "L2", "L3", is a good sign to me.
Well if you liked it, we are making progress, if guess :)) Here is a new
graph showing how this explicit degree-3 node was entailed by the arrows I
drew in my original diagram. Here you will find the (complete?) set of
entailed relationships and only one of the entailments .. I omitted the
others for brevity. I designated the three operant nodes and their
corresponding contexts as unbound variables so that we can apply this form
wherever. Notice that to express any particular signage it is not necessary
to fix all bindings, in fact instantiating the ends of any two arrows will
instantiate the entire L(2) signage.
http://robustai.net/mentography/aSignRelationship.gif
Imagine that that when the variables in the graph are instantiated with some
things in some domains that all the relationships in the model (the graph)
either will (or will not) be deemed true in our interpretation of what
actually obtains. And it is true enough that this fact is not expressed in
that diagram. Expressing that meta fact would necessitate another diagram
and another instantiation of our general SignRelationship form. Are you
gonna make me draw it?
> SR: However I note that I end up seeing a category error in the diagram.
> The arrows that I must draw in the graph based solely from your table,
> above, seem wrong. The arrows labeled object from the sign
relationship
> instances should all be drawn with their sharp end on the box labeled
> "Objects that are dogs" rather than to the individual dogs as objects.
> Otherwise it seems to me that you are mixing signs for plurals with
> signs for individuals.
>
> JA: No, that's okay, but it does raise an issue
> that maybe we have come far enough to tackle.
It is still is not ~okay~ by me :(
> JA: The sign relation L(2) is what what we might call an "object sign
relation",
> in the sense that some people use when they talk about an "object
language".
> In other words, the whole sign relation L(2) is an object of our
discussion,
> and we have it so well pinned down at this point in the dissecting
tray that
> one might well call it a "formaldehyded sign relation", or more
prosaically,
> a fully formalized object of discussion. Of course, in order to
discuss it
> we are obligated to use all sorts of additional verbiage and
video-montage
> that neither of us would be able to completely vivisect all the rest
of our
> lives, even if we should be foolhardy enough or, indeed, even willing
to try.
> And the several total or the unseveral sum of all of that extra
significating
> is what I usually describe as our "meta sign relation", though I'm
sure you'll
> discover that usage varies in respect to this, as it does in most
every regard.
>
> JA: Now, with respect to this somewhat blurry distinction that arises
> from the precipitation of an object system out of its meta-liquor,
> the domains and elements of our focal sign relation can be seen
> to enjoy slightly different statuses with respect to each other.
>
> JA: Thus, it ought to be clear that we are always working on a semantic
platform
> that is at least one story higher than the object system we are
discussing.
> Really, nothing but signs can appear on the page or on the silver
screen,
> and so we are using the signs "Rover", "Specs", "Trixy" to mention
these
> three dogs of our real or imaginary acquaintances, but these signs are
> not part of the object sign relation -- they are parts of the vast,
> unmapped meta sign relation that we are using to talk about L(2).
Ok, I'm with you to this point.
> So, to make a long story come to a semblance of
> a nexus before it goes on to become even longer,
> there are no signs for individuals that achieve
> any proper membership in the sign relation L(2).
> All such signs are signs that we employ in what
> I would dub a "higher order" (HO) sign relation,
> one in which we participate while we talk about
> the internal structure of L(2).
Is that your excuse for the category error I pointed out, or is this a
different topic?
> To say it another way, the actual words that appear in the Object column
> as names of objects in L(2) are not a fixed part of this sign relation's
> specs in the same way that the quoted and bracketed signs that appear in
> the Sign and Interpretant columns are.
I think the variables in the forms bind to their particulars using quite
different mechanisms in all three domains: sign, object, interpertant.
The sign column is easy for it usually exists in the same media and plain as
our diagram, database, or other notation. But the object column usually
gets fixed to its variable by dint of the apprehension of our brains (were
we talking to a human) ... and so, yes I agree, ... any sign we place in our
diagram in the object column is but a stand-in for some apprehended object.
I thought I had fixed this problem in my original diagram by placing the
~object signs~ in a context (Venn diagram circle) such that any mechanism
that attempted to actually access the objects from the database would
necessitate whatever special handling was called upon by dint of the meta
description of that set. I though I had done this quite explicitly enough
that it was visible to a human mathematician comprehending the diagram ...
apparently not ... but nevertheless it does contain (or so I conjecture)
all the necessary information such that a computer could calculate
consistently what actions would be called upon in any such predicament.
See, were you to agree to use the computer's databases as our target
Interpertant, we would always have some razor to decide the consequences of
our notations :)) Alas ... I'll probably never be graced with a direct
answer to that burning question.
> I could put any other equivalent
> information in the Object column that serves to specify the same objects,
> and the sign relation would would not be regarded as different from L(2),
Agreed ... but that ability is a property of all the things that can take up
the role of sign:sign objects ... and proves (I guess your point), that
what is in the object column in your tables is still and all but a sign ...
you current notation not providing any mechanism to differentiate it from
the signs in the other columns.
> but if I change any entries at all in the Sign or Interpretant columns,
> then I have transformed L(2) into a distinct sign relation L(2)'.
Not really, me thinks they are all signs and the same transformation (or non
transformation) could be expressed for each column. I must be too thick to
grok how you have demonstrated a differential towards the object column
here.
Seth Russell