Back    All discussions

2011-02-02
Ancient Philosophy Categories
Before suggesting any category additions/changes here, I thought it might be useful to have some discussion among those with interests in ancient philosophy.

I think that, in general, for the categories to be useful, the Ancient Greek and Roman Philosophy categories need to be more fine-grained.

(1) Specifically, I think that at least Plato and Aristotle and probably Stoics, Epicureans, and some others as well need subject sub-categories such as Ethics, Metaphysics, Epistemology, etc. These subject categories could further be cross-listed under other major categories. For example, a Political Philosophy sub-category under Aristotle could be cross-listed under Social and Political Philosophy as History of Political Philosophy/ Ancient Greek Political Philosophy/ Aristotle. (For some it might also even be useful to have categories for specific works, such as the Republic.)

(2) It also might be useful to add general subject categories under Ancient Greek and Roman Philosophy for articles that treat topics across individuals, schools, or time-periods, as the existing Misc. categories seem inadequate, and it doesn't seem quite right to relegate these sorts of articles to a Misc. bin.

(3) It also might be worthwhile to develop the categories for Roman and later philosophers more fully. It might be useful to have a category for Lucretius or Seneca, for example, apart from the Hellenistic school categories.

(4) Given the close connection of ancient philosophy to other areas of classical studies and classical studies in general, it might also be useful to develop the Other Academic Areas/ Arts and Humanities/ Classics category more fully and to integrate this category more fully with the Ancient Philosophy category.




2011-02-03
Ancient Philosophy Categories
I agree wholeheartedly that we need to subdivide many of these categories, especially those like Plato and Aristotle that currently contain thousands of entries.  I do think we need to do some careful planning of just how to do this.  Topical subdivisions are one very reasonable idea; it might be useful to have a standardized set of default topics not only for ancient philosophers but for historical figures generally, and some thought needs to go into choosing the best ones (but your suggestions here are very useful).  There are also other changes I think should be made: we need to add named-philosopher categories for some other figures (e.g. Theophrastus), some other groups, etc.  I've been wprking through a large accumulation of uncategorized entries (about 25,000 of them) in the top Ancient Philosophy category for the past month or so, and I have about two thirds of those now reallocated to subcategories (or in many cases to the 'Classics' category: a lot of these are not really philosophy material).  I hope to have that done within a few weeks.  One thing that's emerging from this process is a much better picture of just which categories will be useful.  When I've finished the backlog of categorizations, I will be working on a system of finer-grained categories to propose.  A discussion in this forum of how to do that would be a very great help: so, thanks very much for your suggestions.  Let's keep this thread going.  (Robin Smith, editor of the Ancient Philosophy category)

2011-02-03
Ancient Philosophy Categories
Reply to Robin Smith
Thanks, Larkin and Robin.  This raises the more general issue of how to divide historical categories by topic and by work.  Right now we've just subdivided by school/period and by figure, and have put off the further issue because it raises hard structural issues.  But it's getting to the point where we need to face up to this,

Regarding topics: a general thought is that we should use topical classifications used elsewhere in the hierarchy wherever possible.  That way e.g. "Aristotle's Ethics" (or "Aristotle on Ethics", which has the advantage of disambiguating the topic from the work) can be a subcategory of Aristotle (primarily) and of Ethics (secondarily).  It might also be a subcategory of "Ancient Greek and Roman Ethics" (secondarily).  Maybe "Aristotle on Ethics" could be a direct subcategory of "Aristotle", and then maybe "Aristotle on X" (where X is a relevant subtopic of Ethics) could be a subcategory of that.  That raises the further question of how "Aristotle on Ethics" are placed as subcats of "Ethics" and of X.  Obviously they can't all be direct subcats,  I suspect that at the end of the day, we may need, for arbitrary topical categories such as Ethics and X, a direct subcategory such as "History of Ethics" and "History of X", which will then subdivide into "Ancient Greek and Roman Ethics", "Medieval and Renaissance Ethics", and so on, and those will subdivide from there.  These policies have the slightly awkward consequence that parts of the structure of the topic tree will be frequently replicated under various historical categories, and parts of the structure of the history tree will be frequently replicated under the topic tree. 

The obvious alternative would be not to subdivide "Aristotle" topically or "Ethics" historically, and instead rely on a history classification and a separate topical categorization for every historical paper.  But that will lead to huge historical categories and might be less useful and usable for users of those categories.

A third possibility is to use separate history and topic categories (and classification of entries under both) to automatically generate structures of combined history/topic categories along the lines above.  Perhaps there is something to be said for that -- though there would be many tricky issues about just how to do the automatic generation (e.g. maybe only generating categories at certain levels of size).  Presumably there would also be a need for occasional topical subcategories that don't mirror the general topical tree, and that would need manual handling.

Subdivisions by works is more straightforward.  I don't think it would work to have subcategories for specific works by Plato as direct subcategories of Plato, if only because there will be too many.  The obvious thing is to have a category such as "Plato's Works" for every historical figure that we want to subdivide by works, and then have "Plato's *Republic*" and the like as subcategories for every sufficiently significant work (criterion of more than X entries, where X = 10 or 20 or whatever).

Any thoughts are welcome.

2011-02-03
Ancient Philosophy Categories
Please note that I've moved this thread to the Categorization Project forum because it's now drifting towards general categorization issues.

I think the way to go here is to enable browsing of intersections of categories: if you're interested in Aristotle on Ethics, you browse the intersection of the Aristotle and Ethics categories, that is, the works that are classified under both categories. Creating actual in-database categories for all intersections of two categories would swamp the system, and manually adding works in these intersection categories would not be manageable. We're much better off taking advantage of the fact that we can infer membership in virtual intersection categories from membership in the basic categories. 
The challenge here is to implement this in a way that is intuitive for the end user and computationally manageable. One possibility to keep things simple is to limit intersections to pairs of categories involving one area-level cat. So you could browse Aristotle and Meta-Ethics but not Aristotle and the Problem of Evil. I don't know if that would be fine-grained enough. Since there are about 30 categories and 3000 items under Aristotle, on average each such intersections with Aristotle should have about 100 items, but there could be huge variations. 

Limiting intersections to areas would have the advantage that we could provide a drop-down box of intersection options or an expanded list to pick from. If we allow all combinations, we have the usual problem that the full list of categories is unmanageably big. We are pretty much limited to providing a search box like in the category finder. Sometimes its hard for people to find what they are looking for in this way. Another option is to provide a dropdown of areas and a searchbox for other cats. This is probably the best on reflection.

In any case, the widget to create an intersection should probably be located within the 'Related categories' section just below category headers. This would further strain this label, but the option has to be prominent and we probably need more than a single line. The alternative is to put it next to the 'search inside' feature, which constraints us to a line of text. 

The implementation of a single intersection (as opposed to arbitrarily long intersections) should be easy and efficient. We could also allow 3-cat combinations. But I wouldn't want to do arbitrary combinations because that's useless and would really complicate things under the hood. 

2011-02-04
Ancient Philosophy Categories
Reply to David Bourget
User-generated intersections will be useful, but I'm not sure they will be sufficient.  For a start, this will give very little presupplied structure for users: just thousands of entries under "Aristotle", which users can then generate subtopics for if they want to.  This is pretty different from the usual PhilPapers model of giving the user an immediate sense of the important divisions in the literature as it stands and the ability to follow them directly.  I suspect it would be much less useful, especially for nonexperts in a given area, and I suspect that even experts will end up using these subtopics less than presupplied subtopics.

Also, it's clear that there will at least sometimes need to be topical subcategories in historical areas that don't correspond to topics in the main topical tree, which after all have mainly been generated to fit contemporary discussion, not historical discussion.  And it would be odd to have these editor-generated subtopics sitting alone as the only pre-existing topical subcategories under a given topic.  Furthermore, even when historical topics fit into some nonleaf category in the topical tree reasonably well, they often won't fit so well into any underlying leaf categories, again because these have been generated to fit contemporary issues.  So this would lead to entries being in the awkward position of being in a nonleaf category in the topical tree but no leaf category -- or else very large "misc" subcategories for nonleaf categories.  Having "history of X" subcategories under nonleaf X categories would avoid this issue, but this requires a different approach.

One intermediate possibility would be to have automatically generated intersection subcategories which users will see when they go to a category (whether historical or topical), or at least "virtual" subcategories that show up in the user interface even though they are not explicit subcategories in the tree.  These might be generated e.g. whenever the intersection is of a large enough size.  I suspect that this will still end up being too inflexible to meet the varying demands of different historical areas, though.

I suspect that at the end of the day we are going to need editor-generated intersection subcategories, which will reflect accepted structures in various historical areas, and which can stand alongside occasional topical subcategories that aren't intersections per se.  Maybe we could have some sort of tool that makes these easy to create -- the editor for a given historical area can designate a topical category to intersect with (perhaps best to restrict this to nonleaf categories, to avoid having to further subdivide leaf categories), and the intersection category will be created (subject to approval from general editors).  The relevant superstructure might then be automatically generated too (e.g. relevant embedding under the historical and the topical areas), though exactly how much superstructure to include will require some decisions (there's also a question about whether all this structure should be explicitly represented in the tree or generated at the level of user interface).  Of course this will create some extra load, but only as much load as we should expect to have to adequately reflect the structure of historical areas.

The biggest individual-philosopher categories by far are (unsurprisingly) Plato, Aristotle, and Kant, so I'm especially interested to hear from Robin and from Andrew Chignell (editor for Kant) about this, as well as from other editors in historical categories.

2011-02-04
Ancient Philosophy Categories
Reply to Robin Smith
I like Robin's idea of default subcategories for historical figures, but it's a bit tricky.

One possibility would be using existing level-1 topical subcategories: i.e. Metaphysics and Epistemolology, Value Theory, Science Logic and Mathematics.  Problems: (i) for some philosophers, almost everything will be in e.g. Metaphysics and Epistemology; (ii) "Kant on Value Theory" has a somewhat artificial sound to it, as do many other such categories; (iii) this will lead to a lot of embedded structure.

Another possibility, perhaps better, is to use level-2 categories -- so e.g. "Aristotle on Metaphysics" or "Kant on Aesthetics" as direct subcategories of Aristotle and Kant.  It could be up to the editor which level-2 categories to use for a given figure.  These seem a bit more natural to me, but there will still be oddities.  E.g. I take it that it's not so easy to sort Kant's metaphysics and epistemology into mind/language/metaphysics/epistemology.  And likewise the distinction between meta-ethics, normative ethics, applied ethics, and so on will sit a bit awkwardly with some historical figures.

A third possibility is to allow any intersection with a nonleaf category to serve as a direct subcategory of a historical category, but to encourage certain policies as default.  E.g. for some figure, a level-3 topic such as causation might be resonsible for a large part of the secondary literature, so that "X on Causation" will be a natural direct subtopic.  But the default policy might be to go for level-2 categories -- or some subset thereof (perhaps allowing closely related topics such as "Ethics" in some cases).  As before there will sometimes be a need for topical subcategories "X on foo" where "foo" isn't a topic in the topical trees, but perhaps the policy could be to embed these under "X on phi" where "phi" is in the topical trees wherever possible.

Again, all thoughts are welcome.

2011-02-04
Ancient Philosophy Categories
Sure, we can combine normal sub-categories with the intersection mechanism--which is just an additional search filter. I had in mind that intersection search would be enabled on all categories, so Making Aristotle/Plato/etc a middle cat wouldn't preclude using this. 

We can also display shortcuts for particularly useful intersections. So the widget could look something like this:

Intersect this category with:
  • Value Theory  
  • General Philosophy of Science
  • Metaphysics
  • Epistemology
  • Philosophy of Mind
  • [ pick another category ]

The suggestions would probably have to be set up by editors, though we could automatically generate reasonable defaults based on the cardinalities of intersections. 

The important point I wanted to make is that we can address much of the need for further divisions under historical cats using intersection search/browsing rather new categories. That's a much more scalable mechanism, and it's going to be a lot easier to bring about the refinement in this way, because (quadratically) less categorization is needed. If/when we're missing topical cats to refine a historical cat, it might sometimes be a good idea to add categories to the topical tree. 

As you say, there's the issue that often when an entry belongs under Aristotle and Value Theory, say, it won't be clear where in Value Theory it belongs. But we should be able to use the 'misc' and 'general' subcategories for that in almost all cases. That's where I've been putting the 100% historical phil of mind stuff I've come across already. 

NB: thread is back in Ancient phil



2011-02-05
Ancient Philosophy Categories
Here's my take: I like the idea of being able to browse intersections of categories. I would propose that, for each philosopher, the editor identify certain relevant areas of overlap, the main areas for which that philosopher is known (perhaps a default list can be generated based on the categories with the greatest overlap, and the editor can then modify it). The level of specificity could vary by the philosopher. For instance, the main topics for Berkeley are should probably be idealism and perception, which are pretty specific, as opposed to someone like Aristotle, who should probably be associated with very broad areas, like metaphysics, logic, and ethics. These could be displayed as sub-divisions on the category page.

One complication: there are some really important topics in the history of philosophy on which a fair amount of work is done for which categories don't currently exist, either because they are no longer considered part of philosophy (e.g. 'natural philosophy') or because people are no longer interested in them (e.g. the metaphysics of the Eucharist). Items of the latter sort it  wouldn't be so bad to add categories for, but it would be odd to have categories for the former. I'm not sure what to do about this.

It would be nice if it was also possible for users to generate intersections of arbitrary categories, perhaps from the advanced search interface, but this shouldn't really be part of the main categorization system.