[Textop-en-phil] Top level outline headings

[Larry Sanger]

Sorry for the delay replying.  But here goes!

> From: Dr. Clea F. Rees [mailto:cfrees at imapmail.org]
> I just think there ought to be some way in via the labels - 
> however crude and misleading these may be. It may actually be 
> more important for non-philosophers than philosophers, for 
> example. Suppose somebody has come across the term 
> 'metaphysics' and wants to know what that is about. Whereas a 
> specialist might be able to find the relevant sections by 
> studying the outline (and what is beneath the outline), 
> somebody whose question is of this kind will not be able to do that.
> 
> Maybe disciplinary maps could be included with brief
> explanations of terms and relevant links. (Not all the links, 
> obviously, but good "starting places".)

This is a good idea.  We can have what I would call "outline overlays,"
in somewhat the same way there are overlays on Google Maps.  For that
matter, you could take the table of contents of any book and link to the
relevant nodes in the outline.  Could be very useful for education and
research: as you read through Hobbes, or whomever, you can see what
others said about exactly the same topics.  I can easily imagine coders
going wild over what can be done with the data, once it's compiled.

But I think we need to have a single "master outline" to begin with.
I'll explain more about this when I get around to responding to
Brittney.

> I think it is important to think about "ways in" for both
> specialists and non-specialists. If the outline really 
> requires extensive study in order to find anything, few 
> people will take the trouble - at least on the web.

Well, there will be multiple "ways in."  Not only can there be multiple
overlays, there can be multiple types of search (of outline headings, of
summaries, of chunk text, and combinations of these).

This said, I really doubt that the outline, when mature, will require
extensive study for people who are already familiar with a field.  For
others, no doubt, going through the outline will be an education unto
itself.  You shouldn't expect to understand the details of an outline of
a subject you're not familiar with.

> They will
> quickly move on to the next link Google threw up for them (or 
> whatever non-Google engine they use). This is partly why I 
> asked about the intended audience.

The essential purpose of the Collation Project is to analyze texts
deeply and, as an outcome, group closely related texts together.  That's
what people who arrive from Google will see: a heading, such as
"felicity or eudaimonia," together with several texts that talk about
that topic:

http://www.textop.org/wiki/index.php?title=Felicity_or_Eudaimonia

(Eventually, that will be only the top level node of many more nodes
that go into the details of the topic.  We'll add more children nodes as
we collect more texts on a single topic.)

Once the software is written and running, there will be links from this
page that invite the visitor to explore more of the outline and to view
the chunks in context.  The result will be enormously interesting and
useful to scholars and students, as well as the average intelligent,
curious person (the same sort of person who spends hours and hours on
Wikipedia).

> How accessible is it
> supposed to be? It could be accessible, I think, even if one 
> needed to study the outline a great deal to get the most out 
> of it.

I think the best way to answer your question "How accessible is it
supposed to be?" is "About as accessible as the texts and the concepts
they contain." Which is to say of quite variable accessibility.  Parts
of the outline will be obvious.  Other parts will be total mysteries to
everyone except specialists.  Who other than people familiar with
Leibniz should know what to make of "monads" as a heading?

Also, again bear in mind that we can have simplified overlays if
necessary. It's just that I would like the actual work of chunking and
original outlining to be done on a "master outline."  (Again, to be
explained better in response to Brittney's mail.)

> I suspect it may also matter in terms of recruiting
> specialists. I'm not quite sure how this is meant to go with 
> these nodes since it seems unlikely that it will be possible 
> to get people who are really specialists for a particular 
> node (or even sub-sub-sub-node). I'm not sure about this - it 
> is more of a question than anything. I'm trying to fit what 
> you're saying here together with your picture of how the 
> editing may end up working.

I'm not sure what you're asking here.  I'm not sure we will be
recruiting people to manage subjects, i.e., individual nodes or branches
of the outline (although eventually we might).  We'll be recruiting
people to chunk texts and thus will hope to find experts on those texts.
The hope is that, to begin with anyway, people who are familiar with a
text are also adequately familiar with the conceptual structures
contained in the text.

> > As a result, for many months now I've been thinking that
> students will
> > eventually actually *study* some mature version of the outline,
> > because they will find such study to unravel the structure 
> of various
> > topics, and because it will make their research easier.  The people
> > who work on the outline will eventually be "experts" on the outline 
> > because they'll know right away "where" something can be found (or 
> > placed).  I don't think there is any way to avoid this sort 
> of initial
> > confusion.
> 
> I think there has to be or (most) people just won't use it.
> But I do think it might be provided via multiple "ways in". 

So I think we're agreeing now, as long as we agree that overlays are
"ways in" to a master outline.  No?

> I'm thinking of some sort of set of guides to the outline
> aimed at particular audiences. Searching is good, but a 
> decent guide would let people feel more comfortable with the 
> whole system, I think.

Yep, sounds good.  I have no problem with that, given the above caveats.

> > Now if a philosopher, after more than cursory examination of my
> > outline, does not know what I'm going on about at all, then clearly 
> > that's my fault. But I don't think that's a reflection on the 
> > viability of the enterprise. We just have to accept that 
> interpreting
> > outlines is not as obvious as one might think at first.
> 
> At what level should this be true? Just looking at the top
> level nodes? How many layers of nodes should enable a 
> philosopher to make sense of it? This is a real question and 
> not rhetorical. I'm having trouble understanding the 
> standards by which the outline should be judged. (As with the 
> intended audience question.)

Well, you're a philosopher, so I'm asking you (and others) to help me to
articulate the standards by which the outline should be judged.  If
those are equivalent to the rules by which the outline is to be
constructed, see http://tinyurl.com/o8r2g

The general problem that I'm trying to solve with the outline is this:
what hierarchy of nodes can accommodate a potentially limitless number
of texts (books, essays, etc.) divided exhaustively into pretty
fine-grained chunks (as I've defined them), such that (1) there are not
more than, say, ten chunks per category, and (2) there is a minimum of
nodes duplication or overlap (thus requiring the same chunks to live at
two places in the outline)?

(Note: it's not a bad thing for the same chunk to be filed in multiple
places because *it* concerns several distinct topics; it *is* a bad
thing for a chunk to be filed in multiple places because *the nodes*
have an overlapping coverage.)

> "Human conduct" sounds merely descriptive. In any case, it
> certainly rules out virtue ethics and much of feminist ethics.

We'll discuss this more later.

> ... I actually dislike "human conduct"
> more than "right and wrong" here. At least the latter 
> captures much of Kant's normative theory, for example. The 
> former would mean putting the "top-level" of Kant's view 
> elsewhere since it is supposed to apply to all rational 
> beings. I don't think this is a problem only in the case of 
> Kant, but Kant springs to mind as an especially pertinent example.
> 
> "Human conduct" sound much more to me like
> anthropology/psychology/social science/history/etc. I think 
> any replacement (or partial replacement) for "ethics" needs 
> to capture the normative aspect of the discipline.

What would you suggest?

> I'm also wondering if it doesn't make sense to have a very
> general heading such as "value" and then several very 
> significant sub-nodes. Right now, "several dozen" top level 
> nodes does not sound anywhere near enough to cover multiple 
> disciplines.

Aha!  Perhaps we will do that!  But how would the rest of it go?  Feel
free to propose this on the wiki.

> >> Where will feminist philosophy fit? I don't think that the various
> >> subjects discussed in this discipline can be properly 
> subsumed under
> >> the other headings.
> >
> > Feminist philosophy is obviously an interdisciplinary area, and so
> > feminist approaches are going to be spread throughout the outline.
> 
> I see this as both welcome and problematic. On the one hand,
> I agree that feminist ethics, say, should fall under whatever 
> "ethics" becomes. (Another reason to reject "human conduct" - 
> some feminist ethical theories don't address (all) human 
> conduct or (all) human virtue at all e.g. some kinds of 
> maternal, feminine and lesbian ethics.) On the other hand, 
> people often take this line as an easy way of ignoring the 
> whole thing or, else, as a route to tokenism. I think the 
> same is true of non-Western approaches. Moreover, it isn't 
> obvious to me that these approaches can fit neatly into the 
> outline.

Well, remember that there can be a feminist "overlay," linking into the
outline to the parts that are of special interest to feminists.  Also,
the extent to which feminist thought is represented is a function of the
texts included in the resource.  The only possible danger of "tokenism"
for any kind of thought rests in the selection of texts to include.  But
if the project is *collaborative*, then what texts are included depends
first and foremost on the interests of the collaborators.

There is one special problem re feminism that is worth bringing up in
this connection: most feminist philosophy is still not in the public
domain.  So we will not be able to use it until we have made some
special arrangement, if ever we are able to do so, with copyright owners
& publishers.

> If somebody working in a very different tradition 
> drew up the outline, would it look the same?

This is an empirical question.  I think the answer is, "Eventually,
yes."  Certain conceptual schemes fall rather neatly out of certain
texts and textual traditions.  They may not be organized all the same
way, but a lot of the same categories will eventually appear, simply to
make the most elegant, simple presentation of the texts.

> How static is 
> the outline intended to be?

Not very, especially in the beginning.

> What conditions will need to be 
> met in order for the top level nodes to be changed later? 
> (Adding nodes is one thing, but changing them will be much 
> more complicated.)

Well, sometimes it's relatively easy, and involves just a few
subcategories.  Other times it requires drilling down into particular
parts of the outline, and it's not very easy at all.

> This is one reason I think we need to 
> think about as many fine-grained details as possible ahead of 
> time. Of course, there are limits to this, but it still seems 
> important.

Yes, if what you mean is that the fine-grained, specialized parts of the
outline should determine what the higher-level nodes will be.  I think
I've already discovered that in what work I've done.  Bottom-up
outline-building.  But you have to have some top-level *placeholders*
anyway.

> >> Where does mind fit? Is "Human Being" intended to cover this?
> >
> > Yes--I haven't made a "mind" subcategory but I probably
> will.  Here we
> > have a bit of a problem in that we don't want to prejudice the
> > question whether nonhumans have minds, but most discussions of "the 
> > mind" are about the human mind, after all.
> 
> If "mind" will include psychology, it better not be limited
> to humans. In any case, it seems misconceived to put it under 
> a human-only category. Arguably, not all humans have minds 
> and many non-humans have minds.

Right.

> ... I know nothing
> whatsoever about continental philosophy, though, so I cannot 
> say much. I'd like to know, though, how somebody trained 
> primarily in this tradition would think about the outline and 
> how well texts from the analytical tradition would fit into 
> it. It isn't that I'm especially attached to the distinction 
> - indeed, it is arguably quite problematic. So one way to get 
> at my question would be to ask how somebody specialising in 
> Hegel/Habermas/etc. would think about the outline.

Well, if you're conceiving what we're up to as trying to put, say, "the
paradox of analysis" into a table of contents from Derrida, then by
golly you might have some trouble.  I agree.  But such a conundrum
represents a misconception of where I would take this.  In particular,
it would clearly be problematic to try to merge the tables of contents
of ten books from analytic philosophy and ten books from continental
philosophy, and expect to have anything that makes sense.  It is
*precisely* because the outline headings will be so specialized--*not*
at relatively vague, loose, imprecise level of chapter headings--that it
is possible to collate texts from very different traditions.  To think
otherwise is to think that there are chunks such that we *cannot*
identify any main point of it at all, or, once we have the point
articulated, that we *cannot* say what logico-semantic relationship the
point has to existing outline headings.  And to say that is, I think,
just to say that we fundamentally cannot make sense of the text.  Some
people of course think that (indeterminacy of translation and all that),
but not for any good reasons that I have ever encountered.

Ultimately, I think this is just a matter of empirical testing.  It's
one thing to put stock in incommensurability or indeterminacy of
translation as principles about how concepts and language works, it's
quite another to say that those principles mean we simply *could not*
produce a useful and interesting reference work.

> I think of "enquiry" here in terms of investigating, trying
> to find truth etc. So I'd include epistemology here (e.g. 
> parts of Hume's _Enquiry_). Especially since some parts of 
> philosophy of mathematics are epistemic (and some are 
> metaphysical etc.). Presented with "knowledge" vs. "enquiry", 
> I'd go for the latter if interested in say, what counts as 
> 'knowledge' or how to obtain it. Presented with "knowledge' 
> vs. "methodology", I would probably go for the former, though 
> I'm not sure about this. "Knowledge" sounds as though it will 
> contain facts - things known. I would expect "enquiry" to 
> lead me to links about what standards govern knowledge, how 
> the knowledge is gathered etc. etc. (and more, but I'd think 
> epistemic questions would be here at least).

Well, it all depends on what we are trying categorize.

As to "knowledge" vs. "enquiry," let's return to that later.

Sorry for the length--I hope this helps.

--Larry