[Textop] FW: Debate Guide Project

[Larry Sanger]

Matthias Brendel posted the following to the [Textop-en-phil] list, but I
thought this was the better place for it.  My response follows.  I'd love to
get this rolling, too.  Actually it seems that there's at least as much
interest in the Debate Guide Project as in the Collation Project, if not
more.

> -----Original Message-----
> From: textop-en-phil-bounces at lists.dufoundation.org
> [mailto:textop-en-phil-bounces at lists.dufoundation.org] On
> Behalf Of Brendel Matyas
> Sent: Monday, August 07, 2006 7:47 AM
> To: Textop English Language Philosophy List
> Subject: [Textop-en-phil] Debate Guide Project
>
>
> Hi all,
>
> I was thinking about the debate guide projcect. It is similar to the
> Outline project, but it has some additional structure.
>
> In The Oultine projects some chunks of texts are inserted as
> nodes and
> they are organized according to topics and according to the source.
>
>
> In the debate projects there are also chunk of texts, whivch may be
> organized also according to topics and according to source. Further
> there is an argumental structure also.
>
> There are:
>
> 1)  Question
>
> 2) thesis nodes.
>
> 3) Argument nodes.
>
> (i) For each question there are at least 2 theses, whihc have
> a debate.
> There may be more.
>
> (ii) For each thesis there are arguments. An argument is
> usually also a
> thesis, which can be debated.
>
> (iii) For each thesis there may be text-chunks, which formulate that
> argument. There may be more, since several philosopher could write
> the same argument in his/her way repeatedly. An argument may also
> be factual, in this case the text chunk may be a reference, which
> describes the fact.
>
> (iv) An argument may belong to more theses. For example one single
> fact may support more theses.
>
>
> How do you find this structure? I was also thinking of another
> structure, where arguments and counter arguments. Are grouped to
> one question, but I found that not so well structured.
>
> Matthias

REPLY:

For my own general vision of the Debate Guide Project, see:
http://www.textop.org/TextAndCollaboration.html#debate

Before I say anything else, let me say that I strongly encourage you,
Matthias, or anyone else, if you feel so motivated, to begin prototyping
different debate guide entries on the wiki.  It's always been my notion that
the wiki could be used for that purpose (see http://tinyurl.com/m2tbt).

I've been wanting to do this for years.  I think my original inspiration
must be a combination of reconstructing zillions of arguments in my work
teaching and studying philosophy, as well as (and maybe even more) my
experience in the high school debate team (my partner and I were ranked #1
in Alaska in our senior year!).

The *purpose* of a debate guide is to serve as a *resource* where
users--students, teachers, debaters, college bull sessioners, policymakers,
lawyers, academics, and in general everyone who cares about the details of
arguments--can find a debate summed up correctly, elegantly, and neutrally,
for a wide variety of purposes, such as simple learning, decisionmaking,
policymaking, and scholarship.  There are a lot of ways of doing this, but,
just as with encyclopedia articles, unless the desired item is carefully
defined against what it is *not*, we will probably end up with garbage.

The question, then, is how so many different parties and so many purposes
can be best served by a debate summary.  What sort of format should the
summary take?  Let me respond to Matthias now.

> In the debate projects there are also chunk of texts, whivch may be
> organized also according to topics and according to source.

Well, I'm not sure that this is necessary.  I grant that to use quotes from
actual texts might in fact be the best way to introduce a debate.  I have
little doubt that the Collation Project might produce huge amounts of source
material for the Debate Guide Project (DGP), if that's how we were to
organize the latter.

However, *if* the purpose of the DGP is to sum up debates for a wide variety
of people and purposes, I think it would be better to start from scratch.

> Further
> there is an argumental structure also.
>
> There are:
>
> 1)  Question
>
> 2) thesis nodes.
>
> 3) Argument nodes.

I think I agree with this.  I do think that a particular summary should be
defined either by a question or by a resolution, and I think I agree with
you that a question would be better.  But we ought to consider the merits of
resolutions (mainly, resolutions are easier to make more definite).

> (i) For each question there are at least 2 theses, whihc have
> a debate.
> There may be more.

Again, I think I agree here.

> (ii) For each thesis there are arguments. An argument is
> usually also a
> thesis, which can be debated.

Rather, the conclusion of a primary argument is itself the thesis defended.

> (iii) For each thesis there may be text-chunks, which formulate that
> argument. There may be more, since several philosopher could write
> the same argument in his/her way repeatedly. An argument may also
> be factual, in this case the text chunk may be a reference, which
> describes the fact.

This I think would be tedious for most readers.  It would be useful for
researchers, perhaps, but this particular reference material should be
available in the Collation Project.  It sounds to me like you are interested
in the part of the Collation Project that represents arguments.  I am too,
but I don't think that that part can really do the work of a finely-crafted
Debate Guide.

> (iv) An argument may belong to more theses. For example one single
> fact may support more theses.

This is interesting, and definitely worth considering.  But I am inclined to
think that, in the long run, transcluding the same arguments in different
debate summaries will make the debate summary less interesting and useful as
a "narrative" of the debate (see below).

> How do you find this structure? I was also thinking of another
> structure, where arguments and counter arguments. Are grouped to
> one question, but I found that not so well structured.

Except as noted, you're thinking about things along the same lines I am.

I think that, as a whole, a debate summary should be very *readable*: as
much of a clear prose narrative whole as a group of collaborators can
possibly put together.  In other words, it should *not* be a mere database
of structured information.  Everything from the question, to what arguments
are included, to what rebuttals are included, etc., must be carefully,
artfully designed and then written in clear, entertaining, and
persuasive--but fair--prose.

BRIEF INSTRUCTIONS FOR CRAFTING A DEBATE SUMMARY: first, settle carefully on
a question, making sure that it does not subtly favor any one side of the
debate.  Second, list out arguments on your side in advance.  The other
side(s) do(es) this at the same time.  Match arguments that directly respond
to each other (i.e., that argue for contrary claims: the Iraq War increases
the global terrorist threat vs. the Iraq War decreases the global terrorist
threat).  Place these first in a two- or multi-column table, followed by "X
side arguments" and "Y side arguments."  Set word limits for each side.
Third, rather than a "rebuttals" section, as arguments grow, divide them
into logical groupings of arguments.  The first argument(s) in each grouping
should be the most general or fundamental, leading to the more derivative or
specialized.  Finally, when the debate summary is in good shape, precede the
whole with a word-limited summary of the arguments.

HOW THIS IS COLLABORATIVE: not only do the sides collaborate among
themselves, wiki-style, but the different sides must agree when to split
apart or consolidate issues.  For instance, in elaborating an argument, a
side might bring up a "sub-argument" that supports a key premise.  If
necessary, the opposing sides must all agree then to elaborate the
sub-argument and responses not there, but elsewhere on the page.  The sides
must also agree that argument on a particular point really belongs under its
own debate summary.

PERSONNEL CONSIDERATIONS: a debate summary can be initiated only when there
are at least two people, one on each side of a debate.  (Perhaps four, or
six, or even more, will be a *real* quorum.  But two, at least, for
prototyping.)  The persons editing one side must not edit the other side(s).
There must also be some sort of moderator: we need to think about how to do
that, because there are very many interesting problems to consider there.
Ultimately (if not at first), the summaries are to be vetted and edited by
experts on each side.  The reason for these personnel restrictions is to
ensure that partisans of any sort do not "take over" the project.  The
maximally useful Debate Guide will be as neutral as possible: the sides will
state their views as best they can, and it will be up to the reader to
decide what to think.  In a collaborative debate project, neutrality
practically requires getting equal representation.

If anyone would like to flesh these various considerations out on the wiki,
that would be fantastic.  I will certainly contribute as and when I can.
There is much more I could say on policy issues; I have a fairly clear idea
of what would be most interesting and useful to most audiences.

One outcome of a lively and useful prototyping collaboration would be a good
sense of how long a debate summary should be.  Generally speaking, I don't
think a debate summary should run *too* long; it is always possible to
create a new question, and new sides, on more specialized questions that
inform broader questions.  For example, a debate summary on the question "Is
the Iraq War [or whatever we'll call it] just?" might go so deeply into Abu
Ghraib that we decide to create another debate summary: "What took place at
Abu Ghraib and does that demonstrate that the U.S. military engages in war
crimes?" (well, we might not want to use that precise question, but you get
the idea).  Then the first summary could, at the relevant place, simply link
to the second one (with a summary, of course). 

Would anyone like to approach an expert, lively, but polite community of
interlocutors to create a prototype?  I'm thinking of an academic mailing
list, with a broad spectrum of views, that deals with a charged topic in
politics or law?

This is extremely exciting to me, and I want to make it happen, but I do
need your help.  And, by the way, we can set up another mailing list for the
project if there is sufficient interest.

--Larry