[sc34wg3] Reification or Representation?

Murray Altheim sc34wg3@isotopicmaps.org
Mon, 23 Aug 2004 22:33:33 +0100

Lars Marius Garshol wrote:
> * Murray Altheim
> | 
> | One of the things I've noticed over the past few years is that the
> | Topic Map community has (seemingly unwittingly) been delving deeper
> | and deeper into epistemological issues surrounding what it is
> | exactly that Topic Maps *do*.
> Well, I to be fair I think I would say that *part* of the community
> has been doing that. Another part of the community has tried to just
> define the standards more formally and then move on to supporting
> standards like TMCL and TMQL.

This makes it sound like one part of the community is off in the
clouds worrying about meaning while the other is busy getting work
done. I would suggest that whether there's any truth to that idea
or not, the entire community has both a need and a responsibility
to clearly define what it is it is doing prior to moving on to
supporting standards like TMCL and TMQL (and let's not call them
standards until they're actually standards -- right now they're just
proposals). Part of the difficulty I believe we all find ourselves
in is that there are entrenched positions (on all sides) based on
various ideas that do not constitute solid foundations of under-
standing and agreement. If at this point, four years after the 2000
publication of ISO 13250, we still don't agree on terminology and
meaning, it seems foolish to believe that anything called TMCL or
TMQL would be viable, unless all they are doing is operating at a
syntax level (i.e., not on the topic map but its syntax or syntactical
structure). Until there is agreement on meaning, TMCL and TMQL are
going to remain at their heart empty.

> | [reification]
> |
> | I would suggest that there are a variety of "original" meanings to
> | the term, and likely few who actually do understand the term and its
> | ramifications. 
> After reading Steve's reply I agree with you. This seems to me another
> argument for not using this term for the topic-subject relationship,
> which is central to all that we're doing.

Alternately, it might be another argument for continuing to use the
original and more appropriate term, and simply defining it carefully.
Muddying the water at this point by introducing or substituting another
term (regardless of its vernacular popularity) does not solve the problem.

> | I think we should focus on being accurate with our use of language.
> Very much agreed.
> | Yes, "representation" and "represent" are less scary. They're also
> | very misleading, i.e., their ontological commitments are unclear to
> | most people. Is it better to use a rather common but erroneous term
> | over an unusual but rare one? How many people have a sufficient
> | background in semiotics to understand the ontological commitments of
> | representation?
> I think people understand "representation" well enough to get what we
> mean. The precise meaning will be given by the standard anyway; this
> is just the single word that encapsulates the precise meaning.

If you asked twenty people what "representation" meant, you would
certainly get twenty answers. And yes, this is not probably true of
"reification."  But this statistical difference hardly supports use
of "representation" in our very specific, precise sense. There is a
difference in the definition of reification and representation, and
if those twenty people don't understand the difference, this hardly
justifies using one over the other. It is again better to use the
accurate term, and define it carefully.

> * Lars Marius Garshol
> |
> |   c) it leaves us without a term for distinguishing "reification" (in
> |      the original sense of the word) from "reification" (in the ISO
> |      13250:2000 sense of the word), and we do need such a term.
> * Murray Altheim
> |
> | Are you prepared to distinguish "representation" from the variety of
> | meanings it has? Or do as most do, just leaving people to believe
> | what they will?
> I'm not sure I understand the question. I'm thinking that we will say
> that topics represent subjects, without going any further, except what
> TMDM already does.

My question was whether or not you were willing to precisely define the
very specific usage of "representation" within the context of the topic-
subject relationship, i.e., to carefully distinguish this relationship
against the other possible interpretations of "represent" (in the
semiotic sense, the sign relation being what "represent" generally means).

In other words, are you comfortable not actually defining what you mean
by "representation"? Just using the term without defining precisely the
relationship between topic and subject? Just stating that "a topic
represents a subject"? That seems awfully simplistic given that it's
not a simple relation (witness this discussion).

> Note that like Steve Newcomb you've ignored the point I was making. We
> *do* need a term for what TMDM calls reification.

I apparently then did not catch this particular point, as I wrote a
great deal in response. I thought I had answered your points in great
detail. But I'm certainly happy to attempt to clarify where I've been
unclear (these conversations can as you know be notoriously difficult).

> | I believe you actually mean that the TMDM will change its choice of
> | terms, not the meaning of any given term. We ought not be creating
> | new meanings for existing terms.
> I don't think TMDM *is* creating new meanings for existing terms. Do
> you?

No, I was just reacting against the idea of substituting "representation"
for "reification".

> | For me to accept that "representation" does what you want, you'd
> | first have to explain in greater detail what it is you believe
> | "representation" means, and in greater detail what it is you want to
> | say. Sowa heroicly spends a whole book on the subject, and I dare
> | say he's not particularly clear.
> I'm with you here, but I think the solution is to use a common,
> familiar, and much-used term for this and then leave it at that.

I agree in principle, but I don't agree that "representation" is the
term you're looking for. Its meaning is distinctly different than the
relationship between topic and subject, which is not one of represen-
tation, it is one that mostly closely sounds (to me, and to Steve
Newcomb and others) as reification. The definition of representation
as used in the AI and KR communities is not the one we're looking for.
The topic-subject relation is not a sign relation, it is a reification
or proxy relation. (Steve I thought was very clear on this issue, much
moreso than I.)

> | "Representation" requires the concept that there is some canonical
> | subject to represent, the platonic idea of a subject, e.g., that
> | there is some canonical, platonic concept of "cat". 
> Actually, I don't think that's the case. I could argue this many
> different ways, but I'm not going to, because I think what I wrote
> above is correct, that we shouldn't define this any further than to
> simply use the word "represent".

To represent, in the vernacular, epistemological, semiotic, and AI/KR
senses, is a sign relation [1]. A sign does not act as a proxy for the
thing it represents, it *is* the representation itself. As Steve
correctly pointed out, the sign does not have any properties other
than its own, whereas the proxy (i.e., the reified thing) always has
all of the properties of the thing because it stands in for it.

> | My own reading would suggest an approach to the issue not as one of
> | representation but as one of *expression*, i.e., that these
> | "representations" aren't statements of platonic ideals -- regardless
> | of their formality -- but simply humans expressing themselves,
> | making speech acts. 
> This would be much more in line with my thinking, except that in the
> cause of automatically generated topic maps there need not be a human
> involved at all. That just generalizes what you are saying slightly,
> though.

Regardless of whether humans express themselves directly or write
computer programs to do it for them, it's essentially the same
issue of intercommunication of concepts. Until computers creatively
write the programs absent any human input this will continue to be true.

> | I'm not suggesting any new term here, just that "reification", if
> | thoroughly defined within a framework of a greater epistemology, is
> | a more accurate term than "representation", which while perhaps
> | friendlier, is also because of its familiarity more misleading.
> Okay. I don't agree, but I accept that this is your opinion.

I've tried in detail to explain that there is a real and discernable
difference between these two terms, and that "representation" is not
the appropriate term here. You of course as always have the option of
not accepting what I've said. It sounds like further discussion on my
part is unlikely to alter your opinion, so this is likely an end to
my part of the discussion.


[1] John Sowa begins a discussion on page 81 of "Knowledge Representa-
tion" and continues it in Section 6.6. Without opening up another huge
discussion, C.S. Peirce' solution to this entire problem are his
concepts of Firstness, Secondness, and Thirdness. About a year ago I
spent quite a lot of time on this and in the end agree that Peirce's
solution is likely the correct one epistemologically (and logically),
but that it's damned near impossible to imagine in implementation.
The world is simply much too complex to represent accurately; even the
most simple relationships are in reality recursively very complex, and
our level of unknowning so high. So we reify instead. If anyone thinks
Jon Awbrey is a bit off kilter, it's only because he's actually trying
to crack this nut head on. I wish him well -- he's very heroic.
Murray Altheim                    http://kmi.open.ac.uk/people/murray/
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK               .

   "[The US Dept. of] Justice offered up to $67 million to ChoicePoint
    in a no-bid deal for computer profiles with private information
    on every citizen of half a dozen nations. While the September 11th
    highjackers came from Saudi Arabia, Egypt, Lebanon and the Arab
    Emirates, ChoicePoint's menu offered records on Venezuelans,
    Brazilians, Nicaraguans, Mexicans and Argentines. [...] What do
    these nations have in common besides a lack of involvement in the
    September 11th attacks? Coincidentally, each is in the throes of
    major electoral contests in which the leading candidates have the
    nerve to challenge the globalization demands of George W. Bush."
                                 -- Venezuela Floridated, Greg Palast