Philosophy Web—Project Proposal
TLDR: I’m interested in creating an online map of philosophical concepts and their interrelations; which could be used to automatically identify contradictions within, and implications of, given belief systems. I am looking for interested collaborators—especially those with coding capacities – and development advice. I believe there are compelling reasons for rationalists to be interested in this proposal.
[If you’re interested in reading the full Philosophy Web proposal, please see the following link: https://drive.google.com/file/d/1X9fdGUMFase_GGPlXqJH6CcydREaSaDb/view?usp=sharing]
What is Philosophy Web?
Philosophy Web is a proposal to create an interactive online map of philosophical concepts, and the relationships of support and opposition between them. This map would take the form of a node and spoke diagram, with nodes representing concepts, and spokes representing relations of support or opposition.
Users would be able to add these concepts to their own personalised webs of belief. Philosophy Web would then automatically highlight potential contradictions and implications of users’ personalised conceptual maps; helping users expand their intellectual horizons, discover errors in their thinking, and incorporate a broader evidential base in formulating their theories (or do the same for other belief systems they were interested in investigating).
Why Philosophy Web?
Philosophy Web has the potential to assist philosophers in several ways; each of which are expanded upon in the above linked proposal document:
Philosophy Web would facilitate research into the underexplored conceptual space between philosophical specialisms, to pluck the low hanging intellectual fruit which grows there.
Philosophy Web would reveal “long range”, implications of, and contradictions within, philosophical theories; which might otherwise be difficult for supporters (or critics) to discern.
Philosophy Web would support comprehensive philosophical theory building. Comprehensive philosophical theories draw upon—and test themselves against - broader evidential foundations, and are thus likely to constitute epistemic improvements on extant (predominantly very narrow) theories.
Philosophy Web would provide a valuable educative resource for Philosophers.
As a rationalist, you might find this compelling because:
If you intrinsically value knowledge, Philosophy Web could represent a valuable epistemic tool for facilitating new discoveries, encouraging clearer and more synoptic thinking, and generally assisting philosophical progress[1].
If you extrinsically value knowledge, Philosophy Web could constitute a “proof of concept”, which could be translated into similarly useful tools in fields with greater practical relevance, such as political science or economics.
What does Philosophy Web Need?
Presently, Philosophy Web primarily requires technical support. I would guess that the programming required to create Philosophy Web would not be unduly complex, however it comprehensively exceeds my bedrock programming capacity of “absolutely nothing”. If you can code, and might be interested in taking this idea forwards, please reach out to me in the comments or by direct message (or just take the idea and run – I’d be happy with that too). Likewise, please comment if you know a person or group who might be interested in hearing this idea.
Secondarily, I need advice about the best way to present and develop this idea. I currently have three plans for how Philosophy Web could be developed. It could be developed by crowdsourcing, in a manner analogous to Wikipedia. It could be developed by curation, in a manner analogous to the Stanford Encyclopaedia of Philosophy. Or it could be developed by a combination of the two: initial curation to “interesting proof of concept” stage, followed by crowdsourcing to encourage further development. I am especially interested in feedback on the merits of these development options. Any other feedback is of course also appreciated.
Who Are You & How do I Get in Touch?
I am Ben Evans, a solicitor and former philosophy student, interested in rationalism, effective altruism, and intellectual progress. You can get in touch by commenting below or—if you want to reach out in private—by messaging me on LW, or on the following email address: philosophywebmail@gmail.com.
[1] This will be especially compelling if you value “fundamental” knowledge, concerning the base nature of reality, more than other knowledge.
The technology isn’t the problem. There have been between 3 and 6 attempts at solving philosophy using mind mapping applications[*] They pretty much all failed, not as technology, but for reasons of content… the web 2.0 aspect. It’s hard for these projects to attract contributors who know the subject, and easy for them to attract cranks.
There’s already a successful general purpose technology for interlinking any kind of subject, and that’s the wiki. Wikipedia has interlinked philosophy articles along with everything else. Other resources, such as the Stanford Encyclopedia of Philosophy, and the Internet Encyclopedia of philosophy contain higher quality material (they are used as sources by Wikipedia, but not vice versa) which is less interlinked .. but could be made more so, for a reasonable investment of effort. IOW, do it content-first, not technology-first.
[*] including a version of Arbital.
It seems to me this would work for Analytic Philosophy, but not for other philosophical traditions. For instance:
a. Continental Philosophy has, since Heidegger (or, arguably, Husserl) taken a turn away from conceptual definitions towards phenomenological descriptions, so anything concept-based is subject, as a whole, to all manners of phenomenological criticisms;
b. Classic Philosophy frequently isn’t formalizable, with its nuclear terms overlapping in a very interdependent manner, the same applying to some Modern ones. Splitting them into separate concepts doesn’t quite work;
c. And Eastern Philosophies have a strong tendency to operate apophatically, that is, through negation rather than affirmation of concepts, so that every nuclear term comprises a set of negations, resulting in a kind of mix of “a” and “b”, with inverted signals.
In short, a Philosophy Web, as proposed, would be a specific kind of meta-philosophical effort, and since every meta-philosophy is itself a philosophy, thus subject to being marked as an item among others in alternative meta-philosophical taxonomies, as well as of being refutable from opposite methodologies, it wouldn’t be able to encompass more than a specific subset of philosophical thinking.
Hey Alex, thanks for your thoughts. My response would be as follows:
Analytic Philosophy is probably the preeminent field of philosophical enquiry in the developed world. So, even if Philosophy Web did prove constrained to Analytic Philosophy, it would still possess major epistemic value (the Hubble Telescope is only useful for astronomy, the Hadron Collider is only useful for particle physics, etc.; but that’s not really a problem given the importance of those fields).
Having said that, Philosophy Web ought to be able to capture a wide variety of schools of thought, going well beyond the Analytic:
As regards Classic Philosophy, overlapping concepts ought to be amenable to formalisation in a Philosophy Web type structure: they would simply share some relations of support and opposition (to the extent they overlapped), and not share others (to the extent they did not overlap). Now of course this could create presentational problems (how to show fifty slightly different versions of concept x), but those should also be superable: for instance through sensitivity filters which let you see more or less versions of very similar concepts.
As regards Continental Philosophy, perhaps continental philosophers would object to their ideas being characterised as “concepts” (I am not well read enough in Continental Philosophy to know). However they nevertheless have “theories” (or “ideas”, or however else they might want to characterise their units of thought); and those theories contradict, entail, support, oppose, or otherwise relate to other theories. But this is all that is necessary for those “theories” to be usefully displayed in a Philosophy Web style structure.
As regards Eastern Philosophies, Philosophy Web would indeed initially model Western Philosophy, as it would be easier to model a contiguous tradition, whose theorists are in open dialogue with one another, than to have to make a huge number of guesses at how the concepts of very different philosophical traditions relate. However I’m afraid I don’t quite understand how a negation based term would fail to be representable in a conceptual web—wouldn’t you just include more relations of negation and less relations of entailment?
A final, more general point, is that Philosophy Web would not be intended as a way to definitively prove a given theory, but rather as a tool to assist individuals in identifying fruitful paths for research, potential implications of their ideas, hidden contradictions to explore, etc. Thus Philosophy Web does not need to capture (the connections between) concepts in a manner that is logically irrefutable by detractors of any school; it only needs capture them with sufficient fidelity to materially assist the theorising of those who use it.
My comments:
That’s actually not the case. Analytic Philosophy is preeminent in the US and, to some extent, the UK. Everywhere else it’s a topic that one learns among others, and usually in a secondary and subsidiary manner. For example, I majored in Philosophy in 2009. My university’s Philosophy department, which happens to be the most important in my country and therefore the source of that vast majority of Philosophy undergraduates and graduates who then go on to influence other Philosophy departments, was founded by Continental philosophers, and remains almost entirely focused on that, with a major French sub-department, a secondary German one, some professors focusing in Classic and (continental style) English philosophers. In the Analytic tradition there was exactly one professor, whose area of research was Philosophy of Science.
Formalization, of any kind, is mostly an Analytic approach. When one formalizes a Continental philosophy, it cease being the original philosophy and becomes an Analytic interpretation of that Continental philosophy, so not the original anymore. And there’s a remarkable loss of content in such a translation.
They have “experiences” and “perceptions”. Husserl’s project, for instance, was to re-fund Philosophy in the manner of a science by insisting that the objects (in the proper Kantian meaning of the word) philosophers work upon be first described precisely so that, when two philosophers discuss about them, they’re talking about precisely the same thing, so as to avoid divergences due to ambiguities in regards to the objects themselves. Phenomenology then, as Husserl understood it, was to focus on developing a full description of phenomena (perceived objects), to only afterwards philosophize about them. Phenomena, therefore, don’t have opposites, since they’re raw “objectively shared subjetive perceptual descriptions”, never concepts. Heidegger was a student under Husserl, so much of his work consists in describing phenomena. And those who then followed both did the same, with so many different emphasis and methods, and mutual criticisms went more about aspects other phenomenologists didn’t notice in this or that described phenomena.
I’ll give an example of how hard that can be. In Buddhist logic there are five truth categories: true, false, true-and-false, neither-true-nor-false, and unitive. In Jain logic, there are seven: true, false, undefined, true-and-false, true-and-undefined, false-and-undefined, true-false-and-undefined. Philosophy Web, as I understand it at least, would focus strongly on opposite categories, that is, this is true therefore those are false, which are seen similarly from the others’ perspectives, so other truth-categories get sidelined. And that’s without entering the topic of the many different Western dialectical methods, such as Hegel’s, who has historically-bound time-dependent truth-variability linked to the overcoming of oppositions.
I don’t mean to imply it wouldn’t be a useful project though. I’m just pointing out its actual scope in practice will be narrower than your original proposal suggests.
You may be forgetting Canada, Australia and New Zealand. When a philosophical field is preeminent in the English speaking part of the developed world; and of significant (but secondary) importance in non-English speaking European countries; it’s a pretty good bet that it’s the largest school of Western philosophy (population of CANZUK+US > population of Western Europe—UK; and I would guess the distribution of funding/size of philosophical faculties would only amplify this trend).
It strikes me as odd to say that Continental ideas couldn’t usefully be “formalised” in any way. When Continental philosophers write books or give lectures, are they not tacitly “formalising” their ideas: setting out what they are, how they relate to other ideas, and so on? If they can do this in their own work, shouldn’t other people come along and present those ideas and their relations in a different, clearer and more useful fashion? Indeed, isn’t that what teachers of Continental philosophy have been doing for the past hundred years or so? Perhaps Continental philosophy aims to be genuinely impenetrable: but that seems a little uncharitable.
I agree with you that, as described, phenomena themselves may not fit into a Web structure, given they may have no relations to other phenomena. However, as your post demonstrates, Husserl also developed theories about phenomena, what they were, what they were relevant to (our process of doing philosophy, our understanding of the world, etc.). That theory is one that presumably supports, contributes to, or opposes other philosophers’ understandings of how philosophy should be conducted/how we understand the world—and thus exhibits all the characteristics needed to integrate it into a web of theories and their relationships (N.B. that the Web wouldn’t be limited to purely logical relations).
I’ll definitely concede that Philosophy Web makes use of “standard” Western notions of truth value at a meta level; although it could integrate theories about the nature of truth etc. into an object level analysis.
Regarding 1 and 3, good points, and I agree.
On 2, when I say formalizable, I mean in terms of giving the original arguments a symbolic formal treatment, that is, converting them into formal logical statements. Much of non-analytic philosophy has to do with criticizing this kind of procedure. For an example among many, check this recent one from a Neo-Thomistic perspective (I refer to this one because it’s fresh on my mind, I read it a few days ago).
On 4, maybe a practical alternative would be to substitute vaguer but broader relations, such as “agrees”, “partially agrees”, “disagrees”, “purports to encompass”, “purports to replace”, “opposes”, “strawmans” etc., to the more restricted notions of truth values. This would allow for a mindmap-style set of multidirectional relations and clusterings.