I'm curious what relationship, if any, this argument has with the debate about whether philosophy makes progress like natural science does or simply generates new perspectives like literature does. One reason we don't assign Newton's writing isn't just that there are more clearly written accounts of Newtonian mechanics available. It's that there have been important advances in our understanding of mechanics since the time of Newton. By contrast, we wouldn't think of Pynchon's novels as improving on Homer's epics--he's just telling different stories. So do today's cutting edge variants of consequentialism constitute *progress* over Mill's account of utilitarianism? Or are they just articulating different positions one might take on enduring questions? One thing I like about this blog is that I know some of our readers have thought about these issues more deeply than I have, so please enlighten me!
Curious to hear what others think. I'm definitely torn. I do think there's some stuff that deserves to be called progress in philosophy, and I think utilitarianism.net vs Mill's utilitarianism is a good example. We're not going to make progress by settling whether utilitarianism is true. But I do think someone who only reads utilitarianism.net will have a clearer understanding of the options open to the utilitarian, and of how they can respond to the most pressing objections to their view, than someone who only reads Mill.
And I'm tempted to say that isn't super rare. With other old debates--e.g., about the relationship between the mind and the body, or free will--I think philosophers today have a clearer sense of the different possible views one might adopt and their various implications than the people writing a long time ago did, even if we can't hope to make progress by settling the big picture questions.
I suppose this is why I both don't despair about contemporary philosophers plugging away at trying to publish cutting edge stuff in journals--this may well be a rationalization, take it with a huge grain of salt--while I also think for the purpose of broader intellectual culture, it makes sense for most people who aren't going to spend their lives thinking about philosophy to get some exposure to (typically old) Schelling points.
The older stuff also has the tremendous advantage of existing outside the current cultural milieu.
If someone comes to cultural relativism by reading and understanding works from different cultural contexts, I would say they have done so honestly. Likewise if they settle on Catholicism/Spinoza/Kant/utilitarianism.net as the best of the lot, in their view.
There is a “This is Water” issue in only reading contemporary textbooks and wikis.
—I'm curious what relationship, if any, this argument has with the debate about whether philosophy makes progress like natural science does or simply generates new perspectives like literature does.
Whether a discipline makes progress depends on what we mean by progress. What is it that's supposed to progress? Technology? Wisdom? Some skill? Winning an argument? And who or what is making progress?
It also depends on whether consensus is required. I would say philosophy doesn't build on consensus and it doesn't simply generate new perspectives (I'm not sure literature does either). Philosophy addresses the underlying ideas of literature and art directly. It builds on an evolving dialectic that spreads—often silently—into culture. If philosophy makes progress, it's only for those who have the patience to listen closely to the slow-motion exchange of ideas. Those who don't won't know where the worldview they believe in comes from, or whether it's one they would have chosen had they been aware they had a choice. It's an unfortunate sign of our times that what used to be considered a basic education and crucial for developing as citizens, as human beings, now needs to be justified against a materialist standard of scientific progress and boiled down to textbook knowledge. Anyway, what would be the point in that? That's what AI is for!
A student could spend many many years reading the dead white guy philosophers, which would be ideal, but a single balls-to-the-wall survey course necessarily requires carefully picking and choosing, so it makes perfect sense to identify quintessential philosophers whose most important works add the next logical counterpoint to the ongoing dialectic, the entire thrust of which should hopefully feel like a conversation. People often make the mistake of thinking a survey course is just a variety of viewpoints presented in chronological order. It's not. It's really not possible to understand Aquinas without Aristotle, Kant without Hume, or all of the above without Plato. It would be insane to start with Heidegger.
And students need to come to grips with the reality that history—especially in philosophy—was not even close to nice. Had history been different, the curriculum would likely have been different. But. So I would tell those students to be the ones to make history instead of asking professors to misrepresent the past. Become the next quintessential philosopher whose ideas can't be excluded from a survey course. The one thing all philosophically-minded people have in common: they would rather know the ugly truth than have their feelings spared and remain in ignorance.
I've found my occasional accidental many-hour binges on TVTropes.com to be helpful to understand the value of a shared cultural canon. It's amazing as you read that site how often you recognize weird little arbitrary things that have become cultural tropes, and then click on that and see more information about all the variants you also recognize (as well as weird things that make no sense because you just aren't familiar with the shared canon).
Short disclaimer: I applied to DS this year, did not get in, and the associated sourness is probably going to drive this response a bit.
But! I think this is nowhere near a good enough argument for devoting a whole year (or 2/3s of or whatever) to the Canon. The Schelling points exist, sure, but it's not like you have to read The Republic to understand what a noble lie is! It's honestly kind of obvious just from the name.
I think this is true probably for most concepts: If I want to engage in conversation / debate with some former DS kid, I can go on Wikipedia or Utilitarianism.net or wherever else beforehand and get a pretty good summary of the main ideas, without reading a thousand 18-subordinate-clause sentences that a purist Kantian translator insisted on putting in the English version too.
This seems pretty close to the "let's just have a textbook" idea, and I don't really understand why it wouldn't work? Sure, fine, writing the one textbook that everyone agrees on is tough — probably the Wiki Talk pages are a total mess — but is it any worse than "everybody read the very confusing text, and then come to your own conclusions about it"? (Or maybe, "everybody read the very confusing text, then come to seminar and hear my conclusions about it.")
If that system works well enough to get common-knowledge about the Schelling points, then surely Wikipedia can do a pretty good job too!
Now, I think there are good things to say about DS specifically in the sense of "reading a lot of pages and then writing a paper about them every week" is a great thing to know how to do, and to be forced to do in freshman year. And obviously conversations & debates about philosophy are straight-up fun to have.
But I really don't buy this discursive utility argument. In my experience so far, it's not been super necessary — if a guy references a Schelling point you don't know, you can just go online later, or even in the moment ask him to explain it in a few words, and then start thinking and reasoning and arguing about it.
Hey Ari, as someone who took Directed Studies 22 years ago, my advice is--take Shelly Kagan's Ethics and Ian Shapiro's Moral Foundations of Politics. They hit a lot of the highlights! And then take my co-blogger for something else!
Lame! I can promise I didn't read your application. For what it's worth, I applied to a similar program as an undergraduate--the intensive humanities sequence at Princeton--and also didn't get in.
On the substantive point, I'm skeptical that the Schelling points would survive in a world where we only got them secondhand. It's all a matter of degree, but I think having some subset of people who focus on them helps keep them salient; as a tempting as your attitude is, I worry that in the long run it would lead to the sort of fragmentation I described.
Hi Ari, I saw your comment yesterday and wanted to come back to it, but you've since received a number of thoughtful responses.
First, as someone who is (much) closer in age to your profs than to you, but who has definitely been in your shoes, let me say that it's nice that you start off acknowledging that being a little salty about not having gotten into this program may somewhat color your views. Not wrt the humanities, exactly, but I’ve been there!
@Pelorus has said much of what was somewhat on my mind, more thoroughly and thoughtfully than I would have. What I would add is mostly a moral distillation of his points.
>If I want to engage in conversation / debate with some former DS kid, I can go on Wikipedia or http://utilitarianism.net or wherever else beforehand and get a pretty good summary of the main ideas
I love, love, love the sorts of informal debates one has w/peers in undergrad, in the dining hall or dorm lounge, or with a good friend or two late into the night. That’s one of the great things about going to someplace like Yale — lots of voracious and energetic minds to discuss and debate with, to bounce ideas off of, to have lots of low-stakes intellectual fun. These conversations naturally evolve over time (yes, “with age”, if you wish), and become fewer and further in between, but no less fun. (That is why I hang out on philstack!)
But I would urge you not to be so nakedly utilitarian in your approach to reading material. Their value lies far beyond whether (and how “efficiently’) they can allow you to hold your own in debates w/friends (informal or even in a more structured setting). Take it from someone who could not give less of a f about her humanities courses back in the day, but lingered a bit on the snippets of required (non-textbook non-wiki) reading that piqued her interest — the ways they can enrich your life may not be easily articulated, but they are most certainly real.
There is also much to be said about the curation of reading material, and which aspects of the assigned material instructors choose to highlight. Not to mention how they (grad-student teaching fellows?) focus discussions in small-group settings. When people want to diss “elitism”, it’s stuff like this that they fail to understand about a place like Yale: these are classroom circumstances that are not easily replicated elsewhere. And you’re still benefiting even if you’re not in the humanities-on-roids program!
I hope you have a fantastic time in the humanities courses you take. Make the most of them, so that you don’t regret it later the way I do!
I think there are even more arguments for reading ideas in their original formulations rather than in summary or reformulation:
1. If I read Mill's Utilitarianism, I'm exposed to the first counterarguments to the common objections. It's important that people know that the authors of ideas almost always prefigure the most obvious things you'll want to say against them. So engaging with the ideas at that deeper level will give a superior understanding than reading, say, a wikipedia-level summary.
2. Reading a variety of books exposes one to a range of discursive styles. This improves one's own style, and improves one's ability to read and understand others. For example, in a modern day write up of Utilitarianism, you're going to see things in an orderly sequence of headings and subheadings. In Mill's work things are often presented more dialectically: you're going to wade through a lot of objections before getting to the meat of the argument.
3. Reading an original text is a richer experience than reading summary notes, you're exposed to vivid analogies, asides, speculations and so forth. The turns of phrases are going to be stronger. You're going to be surprised because they always say more than the crystallised point that a summarised "best version of the argument" could extract, and often these extra details are going to help reframe some other unrelated thing. Some of these parts are going to stay with you.
On the last point, one of my favorite examples of this is from Alan Turing's 1950 paper on artificial intelligence. There's a tangent about ESP that is somewhat shocking to modern sensibilities. It certainly wouldn't go in a textbook presentation of the argument. But that would be a sad loss, as I find it a fascinating reminder of how much can change even in just seventy five years.
I agree with these benefits, but the question I come back to is opportunity cost. Time spent reading A is time not spent reading B. In wading through Mill (as you put it), is the juice worth the squeeze?
Fair questions. In our sequence, we definitely make tradeoffs. I think Mill is a relatively easy case, in part because he's pretty easy to read. But Kant's critique of pure reason is very much not. While we make them read the whole Mill book (it's short), they only read excerpts of Kant's critique, which is a kind of concession to opportunity costs, I think.
The way I interpret the "canon wars" of the 80s and 90s, the revolutionaries were partially successful in that the cultural canon was no longer quite as central as it was, but the clearest effect was probably just the addition of Toni Morrison to the canon. My guess is that it's easier to see new works enter the canon than it is to see old works disappear, because (apart from St. John's College, and the few remaining Great Books programs) there aren't many syllabi that officially follow a canon, but there are lots and lots and lots of syllabi that are shaped by it. If something catches on, lots of people are likely to copy it and put it in, but they'll each likely choose a different work to remove, and so the old ones will fade out very gradually.
I'd love to see some serious quantitative work on this. My sense is similar to yours. I'd also expect there are fewer official Great Books programs than there used to be, but even if you don't cram it all into one program, I'd think you could get something pretty close to the same by taking history of philosophy, political theory, and literature classes. My guess is the sorts of books that I think of as making up the Canon are assigned substantially less in 2025 than they were in 1985, and that there are some notable additions. Definitely Morrison, and also Foucault, and probably more. Though a relevant difference is that I'd bet Morrison is more likely to be read in literature survey classes, while Foucault is read in humanities classes--though not philosophy, by and large--that aren't trying to do anything that looks like a Canon survey.
We have done some empirical research in sociology. It is an interesting case, since the contemporary core of the canon (Weber-Durkheim-Marx) took shape only in the late 1970s/early 1980s. Here is a study using textbooks (broadly understood) to track trends over time and across Anglophone, French, and German books:
That could be hard to do in philosophy, given the textbook culture, though there might be some trends in readers and the like that could be tracked? It is also possible to get a read based on syllabi, for example:
One parallel to the Schelling point idea is what we call functionalist rhetoric, based on reading a lot of textbooks for how they justify their use of the canon ("we read the canon because it provides a shared point of reference for the field").
Along with the historicist/contextualist, the functional is probably the most common argumentative style. Still, it is sometimes joined with a broadly humanist appeal, because, while the functional explanation might be true, it is not necessarily very inspiring (similar to functional accounts of religion's role in producing social stability; even if true, the functional value per se is rarely going to generate commitment, and if it does, the conviction is not so much in the religion as in the society, a la Durkheim).
It'd be very interesting to study empirically how philosophers justify what and who they teach as well.
Yes and no, I think. They probably won't be read in great books courses. But pretty much any intro political theory class will have Rawls and Nozick on it.
I definitely lean towards the science-oriented critic you mentioned at the beginning.
It seems to me that more and more of the undergraduate philosophy courses are evolving in that direction. A course with the title “Philosophy of Language” or “Philosophy of Mind” might have very few works pre-1900. When I taught them, I certainly didn’t use anything pre-1900. Those works are just too primitive compared to other things we could use.
Epistemology courses aren’t at that point yet, but I would guess it’s just a matter of time. Teaching any epistemology using Plato is foolish, given how primitive his epistemology is. If you want to teach both epistemology and its history, that’s a different story: you really should use Plato, if you want to go back many centuries. Similar remarks hold for metaphysics courses: they’re not at the point of philosophy of language/mind courses, but they seem to be on their way.
Even though we have coordination needs, we also need to choose landmarks that are, well, good. Plato’s epistemology just doesn’t measure up. I know it’s mean to say, but I think a lot of philosophy teachers use Plato in their teaching only because they read so much of it when they were students that it’s easy and fun to teach. Hence, we continue to teach Plato because . . . we continue to teach Plato. It’s an exaggeration, sure, but it applies to plenty of our choices for readings.
The other class I'm teaching this term is epistemology, and it's not historically oriented. We read Descartes and Hume at the beginning but after that it's all 20th and some 21st century. So, I think I mostly agree.
This is fantastic. Wonder if you think it extends more broadly than just great books? I've been thinking a lot lately about ways it could help with the current crisis to return to a clearer "core curriculum", not merely with "distribution requirements" but actual classes that are required. If I were choosing, I'd want one of those classes to be designed and taught by economics, and one to be designed and taught by computer science (so, not only great books). Lots of reasons for this (maybe I'll write something about it), but I wonder if it would also give us a bigger library of concepts which, even if not ultimately the "right" ones, could be useful to be able to draw on in public discussion.
Your arguments here feel uncomfortably close to the arguments for only voting Red or Blue in U.S. elections even if a third party might better align with your politics. Maybe something like:
“The Democratic or Republican platforms may not be the best policy bundles, but hey—they’re our political Schelling points and any deviation from would be a waste of a vote. It’s not about building the best society—it’s about being able to engage in the arguments of our time with fluency and practicality!”
And the usual counter-argument resembles the views of your skeptic: if society could collectively agree that searching for the best politics is worthwhile, third parties and their ideas would become viable, we could experiment and learn more, and we might just improve the political and social order while we’re at it.
I probably agree with the parallel, at least if we're thinking about voting with the aim of affecting the outcome of an election. (You might also vote for expressive reasons, in which case I think third party voting makes perfect sense.) That is to say, I agree that it's fine for there to be a gap between how I'd vote if I thought of myself as voting for everyone (I'd pick the third party), and how I vote given that I know I'm voting just for myself, holding everyone else's votes fixed (I'd pick my preferred of the two).
The line “if society could collectively agree that searching for the best politics is worthwhile” is carrying a lot of weight here! Society is made up of lots of individuals who don’t agree on lots of things and pursue their own ends. The Schellling point argument seems like a more sensible way to navigate society as it actually exists
It really, really warms my heart that these humanities programs are still going strong! I'm painfully aware of the relentless "vocationalization" of higher ed towards STEM over the past almost 20 years, which ofc even the most "elite" institutions have not escaped. I'm going to guess that some of this hardcore optional curriculum used to be required....
As someone who did not go to a tech school but was also never going to major in the humanities, having humanities courses as part of the box-checking graduation requirements was one of the best things that ever happened to me. And one of the best things I ever did for myself was not taking them 100% perfunctorily (let's say only 90%, heh).
As a science person, I must admit I'd never thought about the "why not textbooks?" critique of philosophy pedagogy. I guess I would say that textbooks are great for the kind of material where you (the student) would have to quite rigorously test your understanding at the end of a chapter, via doing a set of exercises that are turned in to be graded. (Not necessarily problems from the textbook -- not all excellent textbooks have problems, and those that do may not be to an instructor's liking.) I know philosophy really does get students to be rigorous in their arguments (often submitted in writing), but very rarely do assignments lend themselves to a "problem set" nature....
And as a second-gen child of immigrants who technically get by culturally (speak English iff it's necessary, etc), I am tentatively going to say that I'm not on board w/the "multicultural" critique. I think every society is going to have its set of Schelling points, and while that set should organically evolve, it will only be sensibly determined by the culture of the society itself. And to decree that material from foreign intellectual traditions be taught (at the opportunity cost of going in depth w/the usual) is to impose a sort of "melting pot" intellectual tradition that I'm not sure is all that helpful. The beauty of liberalism done right is that it enables all sorts of diversity to not just coexist but to flourish together, and that includes what I'd call a deeper and more fundamental form of "viewpoint diversity": diversity in worldviews as informed by different cultural traditions. Homogenization does a disservice to everyone, IMO. Even the French insistence on having French values first and foremost allows for that sort of diversity, as long as its manifestations do not contravene the values of the Republic.
Hot take incoming, but I would venture to say that one cannot fully appreciate works from foreign intellectual traditions anyway, unless one has some firsthand understanding of the culture whence it came. This explains, eg, why many Western foreign policy wonks are pretty Dunning-Kruger-ily clueless when it comes to Asia, to the quiet snickers of Asians in the same field: no amount of reading can truly familiarize you with a culture you just do not and cannot understand at the level of someone who grew up in it and whose mind is simply at home in it. Tl;dr I say we aim for a thorough understanding of our intellectual traditions, and recognize and celebrate the differences that exist in our pluralistic society.
Lots of nice thoughts/observations. And really interesting point in the last paragraph. Here's another hot take coming from a similar place. I think the way we approach foreign language instruction in the US is, at best, pretty odd. The vast majority of people who study a foreign language do not study it in enough depth to learn it, and forget most of what they learned pretty quickly. I think to actually learn a foreign language, you need immersion, which you can't require at scale--at best, people that really care about learning a foreign language can seek it out. It's *especially* hard for Americans to learn foreign languages, because they can get by with English so well in so many situations. I think maybe we should just give up on foreign language requirements, not because learning foreign languages isn't valuable--I think it absolutely is--but because teaching them at the level they're taught in secondary and tertiary education almost never produces enough knowledge to be valuable. Better to just admit that most Americans will be monolingual, and for those who want to learn foreign languages, they'll need to do study abroad, but of course you'd never require everyone to do that.
I think Paul Bloom's characterization of your positions here as a "defence of a sort of conservatism" explains why I'm largely on board: I'm ultimately rather conservative in intellectual disposition. So I'd take a Chesterton's Fence approach to the question of foreign-language instruction: what exactly was it supposed to do for American students (and, consequently, American society)? I'm no historian of these things, but I'd assume it originally had something to do w/notions of "cultural literacy". We used to have lingua francas for the sciences and the arts: notably Latin and French, respectively, since the Enlightenment. So presumably there's this sense that to be monolingual is to be a Philistine, some sort of rube who can't congregate w/those discussing elevated things. Also, "intellectuals" of yore were worldly folks who were relatively comfortable in more than one (European) culture, and so being multilingual presumably means one is worldly in a similar way, or something.
But at some point -- long overdue, I'll submit -- we have to concede that whatever the original purpose of learning a foreign language in the US K-12 system, it's clearly not being served by the methods used. Indeed, no one is learning fluent conversational French by mastering written conjugations of verbs, never mind aspects of French culture from a teacher who (as is often the case in public schools) isn't even a native Francophone. There is no point in being your own personal Google Translate if after 7 years of instruction, you can read and write but are too mortified to speak French even w/shop assistants at CDG, who hear their language mangled on the daily. (And by "you" here, I mean yours truly.) And morally -- circling back to my contention re: culture and worldview -- I'd say the most meaningful reason to learn a foreign language, besides being able to actually converse w/natives, is to encounter ideas best (or even exclusively) natively expressed in the language. The standard manner of instruction ofc falls comically short of that, so yes -- I'd support nap time or Plato for high schoolers in that time slot instead.
And side note -- English is obviously the world's lingua franca for everything now, and most kids outside the Anglosphere apparently attain fluency by imbibing English-language entertainment, and bootstrapping off each other IRL. According to Reddit, there's actually a moral panic in Scandinavia bc kids no longer speak w/o copiously sprinkling in English words! So immersion seems to work fantastically, and probably is really the only viable method.
I'm not sure I agree. One benefit of being exposed to foreign language instruction is that it instills empathy: For the first time in your life you are trying to communicate in a language that you have not mastered! It's hard! It's frustrating! This is an experience everyone should have, because then they'll know: this is what it's like. I have infinite patience for English learners for that reason.
A second benefit (less relevant in the US)....the national unity argument. I'm Canadian, and I've been half-assedly trying to teach myself French for decades... not because there would be any practical payoff that is commensurate with the work involved, but because that's what a good Canadian does. That door was opened with four years of mandatory French in high school (even though i grew up on the West Coast, which is three time zones away from any significant concentration of French speaking Canadians).
(and of course, these benefits are subject to the same "is the juice worth the squeeze?" observation I made below. For this autodidact, it's worth trying to learn French but not worth spending time with Mill's original texts... it depends on what problems society is going to pose for you I guess... I live in a globalizing world and that dictates what issues will be most salient).
One point I'd make in favour of expanding the canon to be more globalist is that since our choice of Schelling point should depend on who we are trying to coordinate with, if we want to facilitate more global philosophical discussion between traditions it would help to include figures from other cultures' philosophical canons in our teaching of "the canon". Our choice of Schelling point/canon works as facilitating who gets included/excluded from the conversation.
A bit like how the rationalist community you began with could have good reason for learning, say, French if there was a large French speaking community nearby that they wanted closer relations with.
I'm sympathetic, but I do think it's hard; at the very least, I don't think the right way to do it is to sprinkle some non-Western texts into great books style classes. Part of it is continuity. One nice thing about these classes is that as you move forward chronologically, the later writers are usually responding, or at least writing with knowledge of, (some of) the earlier ones. If you try to globalize a great books class, you risk breaking that continuity. There are ways of doing it that preserve it--we typically include some medieval Islamic writers in our sequence, and I think that makes sense since they're very much responding to Aristotle--but there are limits on how far you can go in that direction. I really do think the Buddhist tradition has lots of great stuff, but I don't think it would be easy to tack little bits of it on to a Western Canon style class. There's also the problem of *which* non-Western texts you add. There's an embarrassment of riches, which makes coordination difficult.
There is a long-run trend towards philosophy departments wanting to have a specialist in some non-Western tradition, who will then tend to teach classes in the history of that tradition. I tend to think that's a good thing, and is probably better than trying to tack on the Tao te Ching or the Bhagavad Gita to a mostly Western Canon class.
In Schelling's example, I think it's important that these are travelers from out of town. In 1960 (which I believe is when he was writing), they probably would have gotten there by train. They might have come in on one of the railroads that shared trackage rights with the New York Central, and arrived at Grand Central Terminal, or they might have come in on one of the railroads that shared trackage rights with the Pennsylvania Railroad, and arrived at Penn Station. It's possible that they might have driven, or flown, or taken a boat, but I'm pretty sure rail was dominant.
By the 1950s, the Pennsylvania Railroad was apparently declining more quickly than the New York Central, and they were starting to take down some of the elaborate decorations, and in 1963 they would demolish the entire grand station they had to build the modern one that feels like a combination of malls and mezzanines, with Madison Square Gardens on top. I don't know where in that process they were at the time he was writing, but my guess is that Penn Station no longer had that iconic look to it, which Grand Central still has today, and the fact that these were the two entry points that most people arrived at made them the natural coordination points.
This is super helpful! Part of my puzzlement about grand central was the sense that Penn Station is the one Amtrak uses, and the one I always used when I was taking the train to somewhere outside of New York. I thought of Grand Central as more for commuters (on Metro North). But maybe it was different back the .
In 1968, as all the railroads were failing, the Pennsylvania and the NY Central merged into Penn Central, and a few years later the combined company (which was apparently the 6th largest in the country at the time) had the biggest corporate bankruptcy until Enron, and had its passenger service nationalized a few years later (along with several other failing railroads) to produce Amtrak.
I think before the multiple mergers, Grand Central handled most arrivals from the north, and Penn Station handled most arrivals from west, south, and Long Island.
Re: "Clock in the middle of Grand Central Station" was agreed on because it was a class of (iirc) Yale students, and if in the 60s you got on a train to "New York" from points north of there (such as New Haven or Boston) it would drop you off in Grand Central.
It feels like there's a lesson there about how it's easier to coordinate a Schelling Point if you have less proximity/knowledge of a place.
One might also be intrinsically interested in history and so in, inter alia, what things have been most influential. Astell may have been good, but not nearly so many people read her as read Locke. This too may be canon-revisionary – since who ended up being influential when present canons were largely formed around, say, 1900 were not necessarily those who were influential at other times – but only incidentally in the 'New Narratives' way.
Yes "read what people throughout history read and were influenced by, so you can better understand that history" strikes me as a close cousin--distinct, but related--of the argument I make here, and it's also one I accept.
I don't know if you've seen the show Lost, but there's a famous episode in Season 5, The Constant, in which two of the main characters must agree on a Schelling point located in time rather than space (I will call you on Christmas Eve 8 years from now, and your remembering the date and picking up the phone will reconnect me with my present self). Hopefully most coordination problems don't need to get that trippy!
I stopped watching after I lost faith that the writers had in mind a way of making it all make sense. I think that was after season 2. From what I heard of the later seasons, I was right.
Yeah, very overrated show for sure, at least on paper. I almost quit after Season 2 also. But strangely it starts to suck you in even as it grows more ridiculous, because the scope keeps getting more and more meta. It's like Gilligan's Island as a soap opera, on psychedelics. The Constant episode is fun but you sort of need the context to fully appreciate it. (And yet his followup The Leftovers is one of my favorite shows of all time - ten times better!).
I felt betrayed. I thought the writing in the first two seasons gave the impression of clues and hints pointing towards a bizarre but satisfying conclusion that would tie everything together. Over time it became clear they were just making it up as they went along.
Totally! You have to just roll with it and enjoy the weird twists. They keep adding stuff; it's like the Stone Soup of TV shows. But you definitely should check out his followup Leftovers if you never saw that - it's like night and day.
I grew up in CT in the 70s. 100% if we were totally lost in NYC, we would go to the ticket booth at Grand Central. The only other conceivable rally point would have been the ice rink at Rockefeller Center during the holidays.
Back then, to be in New York meant knowing how to navigate, and navigation for an out-of-towner was Grand Central. Penn Station may have been a terminal, but it was at best a hasty adaptive reuse of one of Dante's circles, so any healthy mind blocked it out as a defense mechanism.
I was just listening to Stephen Pinker on Tyler Cowen's podcast discussing this very example, and they explicitly addressed the decline in salience of Grand Central Station. Apparently polls were done in the 70s and back then Schelling's claim was absolutely true; people were asked what they would do in the example he described and they said they'd meet at Grand Central.
Grand Central is also literally where many trains and travelers cross paths. So maybe in Schelling points there are sometimes heuristic metaphors leveraged in combination with social precedents as a basis for converging?
It is important to note that in many parts of NYC in the 1980s you might die if you hung out too long. It was a VERY different place. When I was in college and we visited NYC, we would dare each other to cross Central Park at midnight.
On the other hand, I also hitchhiked from the West Village to the Upper West Side in 1985, so there is that.
I took DS in the mid-eighties, and I still read the canon for fun (just finishing Selected Essays of Leibniz). I also read Audre Lorde and Tiqqun.
I can buy the Schelling hypothesis, though honestly I know about two people personally who have seriously read the canon, and most “educated” people have read Substack articles or the Stanford Encyclopedia of Philosophy at best.
I would propose an additional value for reading the canon, that it inoculates you against model fixation. For example, reading Leibniz feels like reading an insane person, with his windowless predetermined monads. But if you do the work to understand his terminology and arguments, then you get two things:
1. A coherent (meaning thought-through, not “true”) model that pokes hard problems of determinism, relativity, and identity. Whatever truth is out there cannot be known in any comprehensive way, so a new lens allows us to recognize a little more of the dimensionality of the problem
2. It subverts the cozy familiarity of scientific terminology. It is too easy to throw around terms like relativity, fields, locality, etc. if you are smart and have done a bit of reading. When you read Leibniz you have to grapple with how closely his terms map to the more modern and mathematically determined terms. In that wrestling comes understanding.
The canon gives us connectedness, so when we compare ideas over time we are comparing apples to other sorts of apples, not aircraft carriers.
Our culture has a tendency to equate knowledge to scientific knowledge—everything that isn't science-y is fluff and cotton candy. Did a student really ask you to distill philosophy into "information" in a textbook?
I don't think I've encountered students expressing the view that we should read textbooks instead of primary sources, but maybe they're just being polite. I have come across pretty skeptical takes on the value of studying the history of philosophy out in the wild, such as here:
I'm curious what relationship, if any, this argument has with the debate about whether philosophy makes progress like natural science does or simply generates new perspectives like literature does. One reason we don't assign Newton's writing isn't just that there are more clearly written accounts of Newtonian mechanics available. It's that there have been important advances in our understanding of mechanics since the time of Newton. By contrast, we wouldn't think of Pynchon's novels as improving on Homer's epics--he's just telling different stories. So do today's cutting edge variants of consequentialism constitute *progress* over Mill's account of utilitarianism? Or are they just articulating different positions one might take on enduring questions? One thing I like about this blog is that I know some of our readers have thought about these issues more deeply than I have, so please enlighten me!
Curious to hear what others think. I'm definitely torn. I do think there's some stuff that deserves to be called progress in philosophy, and I think utilitarianism.net vs Mill's utilitarianism is a good example. We're not going to make progress by settling whether utilitarianism is true. But I do think someone who only reads utilitarianism.net will have a clearer understanding of the options open to the utilitarian, and of how they can respond to the most pressing objections to their view, than someone who only reads Mill.
And I'm tempted to say that isn't super rare. With other old debates--e.g., about the relationship between the mind and the body, or free will--I think philosophers today have a clearer sense of the different possible views one might adopt and their various implications than the people writing a long time ago did, even if we can't hope to make progress by settling the big picture questions.
I suppose this is why I both don't despair about contemporary philosophers plugging away at trying to publish cutting edge stuff in journals--this may well be a rationalization, take it with a huge grain of salt--while I also think for the purpose of broader intellectual culture, it makes sense for most people who aren't going to spend their lives thinking about philosophy to get some exposure to (typically old) Schelling points.
The older stuff also has the tremendous advantage of existing outside the current cultural milieu.
If someone comes to cultural relativism by reading and understanding works from different cultural contexts, I would say they have done so honestly. Likewise if they settle on Catholicism/Spinoza/Kant/utilitarianism.net as the best of the lot, in their view.
There is a “This is Water” issue in only reading contemporary textbooks and wikis.
—I'm curious what relationship, if any, this argument has with the debate about whether philosophy makes progress like natural science does or simply generates new perspectives like literature does.
Whether a discipline makes progress depends on what we mean by progress. What is it that's supposed to progress? Technology? Wisdom? Some skill? Winning an argument? And who or what is making progress?
It also depends on whether consensus is required. I would say philosophy doesn't build on consensus and it doesn't simply generate new perspectives (I'm not sure literature does either). Philosophy addresses the underlying ideas of literature and art directly. It builds on an evolving dialectic that spreads—often silently—into culture. If philosophy makes progress, it's only for those who have the patience to listen closely to the slow-motion exchange of ideas. Those who don't won't know where the worldview they believe in comes from, or whether it's one they would have chosen had they been aware they had a choice. It's an unfortunate sign of our times that what used to be considered a basic education and crucial for developing as citizens, as human beings, now needs to be justified against a materialist standard of scientific progress and boiled down to textbook knowledge. Anyway, what would be the point in that? That's what AI is for!
A student could spend many many years reading the dead white guy philosophers, which would be ideal, but a single balls-to-the-wall survey course necessarily requires carefully picking and choosing, so it makes perfect sense to identify quintessential philosophers whose most important works add the next logical counterpoint to the ongoing dialectic, the entire thrust of which should hopefully feel like a conversation. People often make the mistake of thinking a survey course is just a variety of viewpoints presented in chronological order. It's not. It's really not possible to understand Aquinas without Aristotle, Kant without Hume, or all of the above without Plato. It would be insane to start with Heidegger.
And students need to come to grips with the reality that history—especially in philosophy—was not even close to nice. Had history been different, the curriculum would likely have been different. But. So I would tell those students to be the ones to make history instead of asking professors to misrepresent the past. Become the next quintessential philosopher whose ideas can't be excluded from a survey course. The one thing all philosophically-minded people have in common: they would rather know the ugly truth than have their feelings spared and remain in ignorance.
I've found my occasional accidental many-hour binges on TVTropes.com to be helpful to understand the value of a shared cultural canon. It's amazing as you read that site how often you recognize weird little arbitrary things that have become cultural tropes, and then click on that and see more information about all the variants you also recognize (as well as weird things that make no sense because you just aren't familiar with the shared canon).
It's a wonderful site.
Can you share some of your favourites?
Short disclaimer: I applied to DS this year, did not get in, and the associated sourness is probably going to drive this response a bit.
But! I think this is nowhere near a good enough argument for devoting a whole year (or 2/3s of or whatever) to the Canon. The Schelling points exist, sure, but it's not like you have to read The Republic to understand what a noble lie is! It's honestly kind of obvious just from the name.
I think this is true probably for most concepts: If I want to engage in conversation / debate with some former DS kid, I can go on Wikipedia or Utilitarianism.net or wherever else beforehand and get a pretty good summary of the main ideas, without reading a thousand 18-subordinate-clause sentences that a purist Kantian translator insisted on putting in the English version too.
This seems pretty close to the "let's just have a textbook" idea, and I don't really understand why it wouldn't work? Sure, fine, writing the one textbook that everyone agrees on is tough — probably the Wiki Talk pages are a total mess — but is it any worse than "everybody read the very confusing text, and then come to your own conclusions about it"? (Or maybe, "everybody read the very confusing text, then come to seminar and hear my conclusions about it.")
If that system works well enough to get common-knowledge about the Schelling points, then surely Wikipedia can do a pretty good job too!
Now, I think there are good things to say about DS specifically in the sense of "reading a lot of pages and then writing a paper about them every week" is a great thing to know how to do, and to be forced to do in freshman year. And obviously conversations & debates about philosophy are straight-up fun to have.
But I really don't buy this discursive utility argument. In my experience so far, it's not been super necessary — if a guy references a Schelling point you don't know, you can just go online later, or even in the moment ask him to explain it in a few words, and then start thinking and reasoning and arguing about it.
Hey Ari, as someone who took Directed Studies 22 years ago, my advice is--take Shelly Kagan's Ethics and Ian Shapiro's Moral Foundations of Politics. They hit a lot of the highlights! And then take my co-blogger for something else!
Lame! I can promise I didn't read your application. For what it's worth, I applied to a similar program as an undergraduate--the intensive humanities sequence at Princeton--and also didn't get in.
On the substantive point, I'm skeptical that the Schelling points would survive in a world where we only got them secondhand. It's all a matter of degree, but I think having some subset of people who focus on them helps keep them salient; as a tempting as your attitude is, I worry that in the long run it would lead to the sort of fragmentation I described.
Hi Ari, I saw your comment yesterday and wanted to come back to it, but you've since received a number of thoughtful responses.
First, as someone who is (much) closer in age to your profs than to you, but who has definitely been in your shoes, let me say that it's nice that you start off acknowledging that being a little salty about not having gotten into this program may somewhat color your views. Not wrt the humanities, exactly, but I’ve been there!
@Pelorus has said much of what was somewhat on my mind, more thoroughly and thoughtfully than I would have. What I would add is mostly a moral distillation of his points.
>If I want to engage in conversation / debate with some former DS kid, I can go on Wikipedia or http://utilitarianism.net or wherever else beforehand and get a pretty good summary of the main ideas
I love, love, love the sorts of informal debates one has w/peers in undergrad, in the dining hall or dorm lounge, or with a good friend or two late into the night. That’s one of the great things about going to someplace like Yale — lots of voracious and energetic minds to discuss and debate with, to bounce ideas off of, to have lots of low-stakes intellectual fun. These conversations naturally evolve over time (yes, “with age”, if you wish), and become fewer and further in between, but no less fun. (That is why I hang out on philstack!)
But I would urge you not to be so nakedly utilitarian in your approach to reading material. Their value lies far beyond whether (and how “efficiently’) they can allow you to hold your own in debates w/friends (informal or even in a more structured setting). Take it from someone who could not give less of a f about her humanities courses back in the day, but lingered a bit on the snippets of required (non-textbook non-wiki) reading that piqued her interest — the ways they can enrich your life may not be easily articulated, but they are most certainly real.
There is also much to be said about the curation of reading material, and which aspects of the assigned material instructors choose to highlight. Not to mention how they (grad-student teaching fellows?) focus discussions in small-group settings. When people want to diss “elitism”, it’s stuff like this that they fail to understand about a place like Yale: these are classroom circumstances that are not easily replicated elsewhere. And you’re still benefiting even if you’re not in the humanities-on-roids program!
I hope you have a fantastic time in the humanities courses you take. Make the most of them, so that you don’t regret it later the way I do!
I think there are even more arguments for reading ideas in their original formulations rather than in summary or reformulation:
1. If I read Mill's Utilitarianism, I'm exposed to the first counterarguments to the common objections. It's important that people know that the authors of ideas almost always prefigure the most obvious things you'll want to say against them. So engaging with the ideas at that deeper level will give a superior understanding than reading, say, a wikipedia-level summary.
2. Reading a variety of books exposes one to a range of discursive styles. This improves one's own style, and improves one's ability to read and understand others. For example, in a modern day write up of Utilitarianism, you're going to see things in an orderly sequence of headings and subheadings. In Mill's work things are often presented more dialectically: you're going to wade through a lot of objections before getting to the meat of the argument.
3. Reading an original text is a richer experience than reading summary notes, you're exposed to vivid analogies, asides, speculations and so forth. The turns of phrases are going to be stronger. You're going to be surprised because they always say more than the crystallised point that a summarised "best version of the argument" could extract, and often these extra details are going to help reframe some other unrelated thing. Some of these parts are going to stay with you.
On the last point, one of my favorite examples of this is from Alan Turing's 1950 paper on artificial intelligence. There's a tangent about ESP that is somewhat shocking to modern sensibilities. It certainly wouldn't go in a textbook presentation of the argument. But that would be a sad loss, as I find it a fascinating reminder of how much can change even in just seventy five years.
I agree with these benefits, but the question I come back to is opportunity cost. Time spent reading A is time not spent reading B. In wading through Mill (as you put it), is the juice worth the squeeze?
Fair questions. In our sequence, we definitely make tradeoffs. I think Mill is a relatively easy case, in part because he's pretty easy to read. But Kant's critique of pure reason is very much not. While we make them read the whole Mill book (it's short), they only read excerpts of Kant's critique, which is a kind of concession to opportunity costs, I think.
The way I interpret the "canon wars" of the 80s and 90s, the revolutionaries were partially successful in that the cultural canon was no longer quite as central as it was, but the clearest effect was probably just the addition of Toni Morrison to the canon. My guess is that it's easier to see new works enter the canon than it is to see old works disappear, because (apart from St. John's College, and the few remaining Great Books programs) there aren't many syllabi that officially follow a canon, but there are lots and lots and lots of syllabi that are shaped by it. If something catches on, lots of people are likely to copy it and put it in, but they'll each likely choose a different work to remove, and so the old ones will fade out very gradually.
I'd love to see some serious quantitative work on this. My sense is similar to yours. I'd also expect there are fewer official Great Books programs than there used to be, but even if you don't cram it all into one program, I'd think you could get something pretty close to the same by taking history of philosophy, political theory, and literature classes. My guess is the sorts of books that I think of as making up the Canon are assigned substantially less in 2025 than they were in 1985, and that there are some notable additions. Definitely Morrison, and also Foucault, and probably more. Though a relevant difference is that I'd bet Morrison is more likely to be read in literature survey classes, while Foucault is read in humanities classes--though not philosophy, by and large--that aren't trying to do anything that looks like a Canon survey.
This is all very interesting. Thank you.
We have done some empirical research in sociology. It is an interesting case, since the contemporary core of the canon (Weber-Durkheim-Marx) took shape only in the late 1970s/early 1980s. Here is a study using textbooks (broadly understood) to track trends over time and across Anglophone, French, and German books:
https://www.researchgate.net/publication/373029364_Toward_a_Historical_Sociology_of_Canonization_Comparing_the_Development_of_Sociological_Theory_in_the_English-_German-_and_French-Language_Contexts_since_the_1950s
That could be hard to do in philosophy, given the textbook culture, though there might be some trends in readers and the like that could be tracked? It is also possible to get a read based on syllabi, for example:
https://www.researchgate.net/publication/323199964_The_Institution_of_Sociological_Theory_in_Canada
One parallel to the Schelling point idea is what we call functionalist rhetoric, based on reading a lot of textbooks for how they justify their use of the canon ("we read the canon because it provides a shared point of reference for the field").
https://www.researchgate.net/publication/358906199_The_Rhetoric_of_the_Canon_Functional_Historicist_and_Humanist_Justifications
Along with the historicist/contextualist, the functional is probably the most common argumentative style. Still, it is sometimes joined with a broadly humanist appeal, because, while the functional explanation might be true, it is not necessarily very inspiring (similar to functional accounts of religion's role in producing social stability; even if true, the functional value per se is rarely going to generate commitment, and if it does, the conviction is not so much in the religion as in the society, a la Durkheim).
It'd be very interesting to study empirically how philosophers justify what and who they teach as well.
Have Rawls, Nozick, Sen, Rorty, or Nussbaum made it into the canon?
Yes and no, I think. They probably won't be read in great books courses. But pretty much any intro political theory class will have Rawls and Nozick on it.
I definitely lean towards the science-oriented critic you mentioned at the beginning.
It seems to me that more and more of the undergraduate philosophy courses are evolving in that direction. A course with the title “Philosophy of Language” or “Philosophy of Mind” might have very few works pre-1900. When I taught them, I certainly didn’t use anything pre-1900. Those works are just too primitive compared to other things we could use.
Epistemology courses aren’t at that point yet, but I would guess it’s just a matter of time. Teaching any epistemology using Plato is foolish, given how primitive his epistemology is. If you want to teach both epistemology and its history, that’s a different story: you really should use Plato, if you want to go back many centuries. Similar remarks hold for metaphysics courses: they’re not at the point of philosophy of language/mind courses, but they seem to be on their way.
Even though we have coordination needs, we also need to choose landmarks that are, well, good. Plato’s epistemology just doesn’t measure up. I know it’s mean to say, but I think a lot of philosophy teachers use Plato in their teaching only because they read so much of it when they were students that it’s easy and fun to teach. Hence, we continue to teach Plato because . . . we continue to teach Plato. It’s an exaggeration, sure, but it applies to plenty of our choices for readings.
The other class I'm teaching this term is epistemology, and it's not historically oriented. We read Descartes and Hume at the beginning but after that it's all 20th and some 21st century. So, I think I mostly agree.
This is fantastic. Wonder if you think it extends more broadly than just great books? I've been thinking a lot lately about ways it could help with the current crisis to return to a clearer "core curriculum", not merely with "distribution requirements" but actual classes that are required. If I were choosing, I'd want one of those classes to be designed and taught by economics, and one to be designed and taught by computer science (so, not only great books). Lots of reasons for this (maybe I'll write something about it), but I wonder if it would also give us a bigger library of concepts which, even if not ultimately the "right" ones, could be useful to be able to draw on in public discussion.
I love this idea.
Your arguments here feel uncomfortably close to the arguments for only voting Red or Blue in U.S. elections even if a third party might better align with your politics. Maybe something like:
“The Democratic or Republican platforms may not be the best policy bundles, but hey—they’re our political Schelling points and any deviation from would be a waste of a vote. It’s not about building the best society—it’s about being able to engage in the arguments of our time with fluency and practicality!”
And the usual counter-argument resembles the views of your skeptic: if society could collectively agree that searching for the best politics is worthwhile, third parties and their ideas would become viable, we could experiment and learn more, and we might just improve the political and social order while we’re at it.
I probably agree with the parallel, at least if we're thinking about voting with the aim of affecting the outcome of an election. (You might also vote for expressive reasons, in which case I think third party voting makes perfect sense.) That is to say, I agree that it's fine for there to be a gap between how I'd vote if I thought of myself as voting for everyone (I'd pick the third party), and how I vote given that I know I'm voting just for myself, holding everyone else's votes fixed (I'd pick my preferred of the two).
The line “if society could collectively agree that searching for the best politics is worthwhile” is carrying a lot of weight here! Society is made up of lots of individuals who don’t agree on lots of things and pursue their own ends. The Schellling point argument seems like a more sensible way to navigate society as it actually exists
It really, really warms my heart that these humanities programs are still going strong! I'm painfully aware of the relentless "vocationalization" of higher ed towards STEM over the past almost 20 years, which ofc even the most "elite" institutions have not escaped. I'm going to guess that some of this hardcore optional curriculum used to be required....
As someone who did not go to a tech school but was also never going to major in the humanities, having humanities courses as part of the box-checking graduation requirements was one of the best things that ever happened to me. And one of the best things I ever did for myself was not taking them 100% perfunctorily (let's say only 90%, heh).
As a science person, I must admit I'd never thought about the "why not textbooks?" critique of philosophy pedagogy. I guess I would say that textbooks are great for the kind of material where you (the student) would have to quite rigorously test your understanding at the end of a chapter, via doing a set of exercises that are turned in to be graded. (Not necessarily problems from the textbook -- not all excellent textbooks have problems, and those that do may not be to an instructor's liking.) I know philosophy really does get students to be rigorous in their arguments (often submitted in writing), but very rarely do assignments lend themselves to a "problem set" nature....
And as a second-gen child of immigrants who technically get by culturally (speak English iff it's necessary, etc), I am tentatively going to say that I'm not on board w/the "multicultural" critique. I think every society is going to have its set of Schelling points, and while that set should organically evolve, it will only be sensibly determined by the culture of the society itself. And to decree that material from foreign intellectual traditions be taught (at the opportunity cost of going in depth w/the usual) is to impose a sort of "melting pot" intellectual tradition that I'm not sure is all that helpful. The beauty of liberalism done right is that it enables all sorts of diversity to not just coexist but to flourish together, and that includes what I'd call a deeper and more fundamental form of "viewpoint diversity": diversity in worldviews as informed by different cultural traditions. Homogenization does a disservice to everyone, IMO. Even the French insistence on having French values first and foremost allows for that sort of diversity, as long as its manifestations do not contravene the values of the Republic.
Hot take incoming, but I would venture to say that one cannot fully appreciate works from foreign intellectual traditions anyway, unless one has some firsthand understanding of the culture whence it came. This explains, eg, why many Western foreign policy wonks are pretty Dunning-Kruger-ily clueless when it comes to Asia, to the quiet snickers of Asians in the same field: no amount of reading can truly familiarize you with a culture you just do not and cannot understand at the level of someone who grew up in it and whose mind is simply at home in it. Tl;dr I say we aim for a thorough understanding of our intellectual traditions, and recognize and celebrate the differences that exist in our pluralistic society.
Lots of nice thoughts/observations. And really interesting point in the last paragraph. Here's another hot take coming from a similar place. I think the way we approach foreign language instruction in the US is, at best, pretty odd. The vast majority of people who study a foreign language do not study it in enough depth to learn it, and forget most of what they learned pretty quickly. I think to actually learn a foreign language, you need immersion, which you can't require at scale--at best, people that really care about learning a foreign language can seek it out. It's *especially* hard for Americans to learn foreign languages, because they can get by with English so well in so many situations. I think maybe we should just give up on foreign language requirements, not because learning foreign languages isn't valuable--I think it absolutely is--but because teaching them at the level they're taught in secondary and tertiary education almost never produces enough knowledge to be valuable. Better to just admit that most Americans will be monolingual, and for those who want to learn foreign languages, they'll need to do study abroad, but of course you'd never require everyone to do that.
I think Paul Bloom's characterization of your positions here as a "defence of a sort of conservatism" explains why I'm largely on board: I'm ultimately rather conservative in intellectual disposition. So I'd take a Chesterton's Fence approach to the question of foreign-language instruction: what exactly was it supposed to do for American students (and, consequently, American society)? I'm no historian of these things, but I'd assume it originally had something to do w/notions of "cultural literacy". We used to have lingua francas for the sciences and the arts: notably Latin and French, respectively, since the Enlightenment. So presumably there's this sense that to be monolingual is to be a Philistine, some sort of rube who can't congregate w/those discussing elevated things. Also, "intellectuals" of yore were worldly folks who were relatively comfortable in more than one (European) culture, and so being multilingual presumably means one is worldly in a similar way, or something.
But at some point -- long overdue, I'll submit -- we have to concede that whatever the original purpose of learning a foreign language in the US K-12 system, it's clearly not being served by the methods used. Indeed, no one is learning fluent conversational French by mastering written conjugations of verbs, never mind aspects of French culture from a teacher who (as is often the case in public schools) isn't even a native Francophone. There is no point in being your own personal Google Translate if after 7 years of instruction, you can read and write but are too mortified to speak French even w/shop assistants at CDG, who hear their language mangled on the daily. (And by "you" here, I mean yours truly.) And morally -- circling back to my contention re: culture and worldview -- I'd say the most meaningful reason to learn a foreign language, besides being able to actually converse w/natives, is to encounter ideas best (or even exclusively) natively expressed in the language. The standard manner of instruction ofc falls comically short of that, so yes -- I'd support nap time or Plato for high schoolers in that time slot instead.
And side note -- English is obviously the world's lingua franca for everything now, and most kids outside the Anglosphere apparently attain fluency by imbibing English-language entertainment, and bootstrapping off each other IRL. According to Reddit, there's actually a moral panic in Scandinavia bc kids no longer speak w/o copiously sprinkling in English words! So immersion seems to work fantastically, and probably is really the only viable method.
I'm not sure I agree. One benefit of being exposed to foreign language instruction is that it instills empathy: For the first time in your life you are trying to communicate in a language that you have not mastered! It's hard! It's frustrating! This is an experience everyone should have, because then they'll know: this is what it's like. I have infinite patience for English learners for that reason.
A second benefit (less relevant in the US)....the national unity argument. I'm Canadian, and I've been half-assedly trying to teach myself French for decades... not because there would be any practical payoff that is commensurate with the work involved, but because that's what a good Canadian does. That door was opened with four years of mandatory French in high school (even though i grew up on the West Coast, which is three time zones away from any significant concentration of French speaking Canadians).
(and of course, these benefits are subject to the same "is the juice worth the squeeze?" observation I made below. For this autodidact, it's worth trying to learn French but not worth spending time with Mill's original texts... it depends on what problems society is going to pose for you I guess... I live in a globalizing world and that dictates what issues will be most salient).
I really liked this.
One point I'd make in favour of expanding the canon to be more globalist is that since our choice of Schelling point should depend on who we are trying to coordinate with, if we want to facilitate more global philosophical discussion between traditions it would help to include figures from other cultures' philosophical canons in our teaching of "the canon". Our choice of Schelling point/canon works as facilitating who gets included/excluded from the conversation.
A bit like how the rationalist community you began with could have good reason for learning, say, French if there was a large French speaking community nearby that they wanted closer relations with.
I'm sympathetic, but I do think it's hard; at the very least, I don't think the right way to do it is to sprinkle some non-Western texts into great books style classes. Part of it is continuity. One nice thing about these classes is that as you move forward chronologically, the later writers are usually responding, or at least writing with knowledge of, (some of) the earlier ones. If you try to globalize a great books class, you risk breaking that continuity. There are ways of doing it that preserve it--we typically include some medieval Islamic writers in our sequence, and I think that makes sense since they're very much responding to Aristotle--but there are limits on how far you can go in that direction. I really do think the Buddhist tradition has lots of great stuff, but I don't think it would be easy to tack little bits of it on to a Western Canon style class. There's also the problem of *which* non-Western texts you add. There's an embarrassment of riches, which makes coordination difficult.
There is a long-run trend towards philosophy departments wanting to have a specialist in some non-Western tradition, who will then tend to teach classes in the history of that tradition. I tend to think that's a good thing, and is probably better than trying to tack on the Tao te Ching or the Bhagavad Gita to a mostly Western Canon class.
In Schelling's example, I think it's important that these are travelers from out of town. In 1960 (which I believe is when he was writing), they probably would have gotten there by train. They might have come in on one of the railroads that shared trackage rights with the New York Central, and arrived at Grand Central Terminal, or they might have come in on one of the railroads that shared trackage rights with the Pennsylvania Railroad, and arrived at Penn Station. It's possible that they might have driven, or flown, or taken a boat, but I'm pretty sure rail was dominant.
By the 1950s, the Pennsylvania Railroad was apparently declining more quickly than the New York Central, and they were starting to take down some of the elaborate decorations, and in 1963 they would demolish the entire grand station they had to build the modern one that feels like a combination of malls and mezzanines, with Madison Square Gardens on top. I don't know where in that process they were at the time he was writing, but my guess is that Penn Station no longer had that iconic look to it, which Grand Central still has today, and the fact that these were the two entry points that most people arrived at made them the natural coordination points.
https://en.wikipedia.org/wiki/Pennsylvania_Station_(1910%E2%80%931963)
This is super helpful! Part of my puzzlement about grand central was the sense that Penn Station is the one Amtrak uses, and the one I always used when I was taking the train to somewhere outside of New York. I thought of Grand Central as more for commuters (on Metro North). But maybe it was different back the .
In 1968, as all the railroads were failing, the Pennsylvania and the NY Central merged into Penn Central, and a few years later the combined company (which was apparently the 6th largest in the country at the time) had the biggest corporate bankruptcy until Enron, and had its passenger service nationalized a few years later (along with several other failing railroads) to produce Amtrak.
It looks like it was 1970 that all the different intercity services that had terminated at Grand Central were re-routed to Penn Station: https://en.wikipedia.org/wiki/Northeast_Corridor#Reorganization_and_bankruptcy
I think before the multiple mergers, Grand Central handled most arrivals from the north, and Penn Station handled most arrivals from west, south, and Long Island.
Re: "Clock in the middle of Grand Central Station" was agreed on because it was a class of (iirc) Yale students, and if in the 60s you got on a train to "New York" from points north of there (such as New Haven or Boston) it would drop you off in Grand Central.
It feels like there's a lesson there about how it's easier to coordinate a Schelling Point if you have less proximity/knowledge of a place.
One might also be intrinsically interested in history and so in, inter alia, what things have been most influential. Astell may have been good, but not nearly so many people read her as read Locke. This too may be canon-revisionary – since who ended up being influential when present canons were largely formed around, say, 1900 were not necessarily those who were influential at other times – but only incidentally in the 'New Narratives' way.
Yes "read what people throughout history read and were influenced by, so you can better understand that history" strikes me as a close cousin--distinct, but related--of the argument I make here, and it's also one I accept.
I don't know if you've seen the show Lost, but there's a famous episode in Season 5, The Constant, in which two of the main characters must agree on a Schelling point located in time rather than space (I will call you on Christmas Eve 8 years from now, and your remembering the date and picking up the phone will reconnect me with my present self). Hopefully most coordination problems don't need to get that trippy!
I stopped watching after I lost faith that the writers had in mind a way of making it all make sense. I think that was after season 2. From what I heard of the later seasons, I was right.
Yeah, very overrated show for sure, at least on paper. I almost quit after Season 2 also. But strangely it starts to suck you in even as it grows more ridiculous, because the scope keeps getting more and more meta. It's like Gilligan's Island as a soap opera, on psychedelics. The Constant episode is fun but you sort of need the context to fully appreciate it. (And yet his followup The Leftovers is one of my favorite shows of all time - ten times better!).
I felt betrayed. I thought the writing in the first two seasons gave the impression of clues and hints pointing towards a bizarre but satisfying conclusion that would tie everything together. Over time it became clear they were just making it up as they went along.
Totally! You have to just roll with it and enjoy the weird twists. They keep adding stuff; it's like the Stone Soup of TV shows. But you definitely should check out his followup Leftovers if you never saw that - it's like night and day.
I grew up in CT in the 70s. 100% if we were totally lost in NYC, we would go to the ticket booth at Grand Central. The only other conceivable rally point would have been the ice rink at Rockefeller Center during the holidays.
Back then, to be in New York meant knowing how to navigate, and navigation for an out-of-towner was Grand Central. Penn Station may have been a terminal, but it was at best a hasty adaptive reuse of one of Dante's circles, so any healthy mind blocked it out as a defense mechanism.
I was just listening to Stephen Pinker on Tyler Cowen's podcast discussing this very example, and they explicitly addressed the decline in salience of Grand Central Station. Apparently polls were done in the 70s and back then Schelling's claim was absolutely true; people were asked what they would do in the example he described and they said they'd meet at Grand Central.
Grand Central is also literally where many trains and travelers cross paths. So maybe in Schelling points there are sometimes heuristic metaphors leveraged in combination with social precedents as a basis for converging?
It is important to note that in many parts of NYC in the 1980s you might die if you hung out too long. It was a VERY different place. When I was in college and we visited NYC, we would dare each other to cross Central Park at midnight.
On the other hand, I also hitchhiked from the West Village to the Upper West Side in 1985, so there is that.
I took DS in the mid-eighties, and I still read the canon for fun (just finishing Selected Essays of Leibniz). I also read Audre Lorde and Tiqqun.
I can buy the Schelling hypothesis, though honestly I know about two people personally who have seriously read the canon, and most “educated” people have read Substack articles or the Stanford Encyclopedia of Philosophy at best.
I would propose an additional value for reading the canon, that it inoculates you against model fixation. For example, reading Leibniz feels like reading an insane person, with his windowless predetermined monads. But if you do the work to understand his terminology and arguments, then you get two things:
1. A coherent (meaning thought-through, not “true”) model that pokes hard problems of determinism, relativity, and identity. Whatever truth is out there cannot be known in any comprehensive way, so a new lens allows us to recognize a little more of the dimensionality of the problem
2. It subverts the cozy familiarity of scientific terminology. It is too easy to throw around terms like relativity, fields, locality, etc. if you are smart and have done a bit of reading. When you read Leibniz you have to grapple with how closely his terms map to the more modern and mathematically determined terms. In that wrestling comes understanding.
The canon gives us connectedness, so when we compare ideas over time we are comparing apples to other sorts of apples, not aircraft carriers.
Our culture has a tendency to equate knowledge to scientific knowledge—everything that isn't science-y is fluff and cotton candy. Did a student really ask you to distill philosophy into "information" in a textbook?
I don't think I've encountered students expressing the view that we should read textbooks instead of primary sources, but maybe they're just being polite. I have come across pretty skeptical takes on the value of studying the history of philosophy out in the wild, such as here:
https://dailynous.com/2022/09/21/article-spotlight-the-end-of-history-by-hanno-sauer/
That was a great article, thank you! Very refreshing for its honesty. I'll post the direct link to the article here in case anyone's interested:
https://www.tandfonline.com/doi/full/10.1080/0020174X.2022.2124542
And the counterargument as well:
https://www.tandfonline.com/doi/full/10.1080/0020174X.2023.2190771#d1e121
What's missing from both papers is the idea that perhaps the ancients were actually not "wrong about everything" but "right about what counts".
That's such a relief, even if they really would rather have had a textbook. I'll check out the article. Thanks!