From a recent contribution by Jason Brennan to the ongoing polemic between Michael Huemer and Kevin Vallier on the history of philosophy (as posted at BHL at 2:45 pm, February 11, 2020):
Context: Michael Huemer claims that the “great” philosophers are usually bad thinkers. They defend implausible ideas with bad arguments.
Vallier responds that the great philosophers are like architects. Their great achievement is that they build coherent systems of thought.
I’m not much convinced by Vallier’s response in part because, when I studied the history of philosophy or read papers in the field, it seems that the “greats” often have incoherent systems. A large number of published papers on the greats, and good number of the classes, take the form of “Great Thinkers says X here and Y here, but X and Y are seemingly incompatible. Let me try to figure out a way to spin X and Y to render then coherent.”
Don’t really see how the intended conclusion follows.
If X and Y are seemingly incompatible claims, then either they’re actually inconsistent or not. Either disjunct is possible.
Suppose not. Then if someone tries to “spin a story” making them consistent (aka “engage in scholarship”), then either she succeeds or fails.
Suppose she succeeds. Then she’s succeeded at resolving an apparent inconsistency. So the thinker is not incoherent, at least as far as X and Y are concerned.
Suppose she fails. It still doesn’t follow that she’s succeeded at demonstrating an actual inconsistency. She’s just failed at resolving an apparent but in-principle resolvable inconsistency. Same result as before.
If X and Y are actually inconsistent, it follows trivially that they’re inconsistent, and perhaps that the thinker’s system is incoherent (depending on the essentiality of X and Y to the system, a complicated issue of its own). But it’s an empirical matter whether that’s “often” so, and if so, how often. As far as I can see, the only empirical proof that would suffice is an enumerative (or quasi-enumerative) induction going through all of the relevant philosophers and all of their arguments. But I’ve never seen one, and doubt I ever will.
So I don’t share Brennan’s confidence in the conclusion: “the “greats” often have incoherent systems.” Even when they contradict one another, their systems may end up being internally consistent, or else involve large swatches of local consistency short of “incoherence.” In any case, there’s no clear inference from “there are apparent incompatibilities” to “there are actual inconsistencies,” much less good inferences mediated by the sheer fact that historians spend time and effort resolving the apparent inconsistencies.
Policy of Truth 
Sorry, the point was that people are always pointing out tensions in thought in the big thinkers, as they apparently have lots of contradictory ideas–or at least seemingly contradictory. This is evidence against Kevin’s claim that what’s great about them is their ability to construct coherent systems.
LikeLike
I don’t see how that addresses what I wrote. There’s a big difference between someone’s “apparently” having lots of contradictory ideas and actually having them, and between being “at least seemingly contradictory” and actually contradicting himself, or more to the point, contradicting himself in systematic and essential ways. Your argument only goes through if the apparent contradictions are actual contradictions across the board. But nothing you say suggests why that would be so. The appearance of contradiction can arise through defects in the text or through defects in the reader. Appearances of the latter kind aren’t evidence against Vallier’s claims, or in favor of Huemer’s.
One of PoT’s bloggers, David Riesbeck, recently published a book on Aristotle’s Politics, Aristotle on Political Community. The book argues that various apparent tensions in Aristotle’s political theory can in fact be resolved. Either he’s largely right or not. If he’s right, or largely right, or even right about a couple of important claims, the apparent tensions in Aristotle are illusory. Aristotle is more coherent than has been realized. It’s Aristotle’s readers who have put the apparent tensions there by misreading the text, or not reading it in a sufficiently imaginative (or integrative, or rigorous, etc.) way. Alternatively, the tensions are there but can be resolved by appealing to higher-order principles of Aristotle’s own theory. If the higher-order principles are coherent, then the lower-order incoherence isn’t all that important.
What Riesbeck did with Aristotle’s politics has been done in thousands of other historical contexts–on Plato, on Aristotle, on Locke, on Kant, etc, etc. One can’t dismissively wave all this away by saying that there are lots of apparent tensions there, people argue over them, hence the systems are incoherent. The last inference is a non-sequitur.
LikeLike
Pingback: Nightcap | Notes On Liberty
So, some scholars think that there are lots of contradictory claims in, say, Plato’s thought on the relationship between virtue and pleasure. (Say, e.g., an apparent conflict on this point between the view suggested by Socrates’s line of questioning in the Protagoras and the view suggested by his line of questioning in the Gorgias.) (1) Some suggest that the contradictory claims are genuinely contradictory but innocently explained (say, Plato believed P at one time, and he believed ~P at one time, but not at the same time; or in one case he’s expressing what he took to be Socrates’s own view P, in another place he’s putting his own view ~P into the mouth of the character of Socrates; or whatever). (2) Some suggest that the contradictory claims are only apparently contradictory, but not really so, and often they add that the appearance of contradiction is the fault of bad readers reading without sufficient care to the structure of the claim or of the nuances of the argument or whatever. (3) Some suggest that the contradictory claims are genuinely contradictory and not innocently explained (his views on this point just end up being inconsistent, that sucks, so a consistent thinker will have to throw at least one out).
In case (1), the genuine contradiction is not ipso facto evidence for the claim that the philosopher is bad at reasoning consistently or building coherent systems of thought. If the contradiction is due to a genuine change of mind over time, then it’s evidence the philosopher was wrong at least once, either before or after. But that seems clearly compatible with Vallier’s defense. If it’s due to a change in something else (say, the purpose in using the name “Socrates” for a character in a dialogue), then we don’t even have to go as far as discussing a change of beliefs. In case (2), the seeming but non-genuine contradiction is not ipso facto evidence for the claim that the philosopher is bad at reasoning consistently or at building coherent systems of thought. It’s evidence for the claim that readers (maybe including other clever philosophers) are bad at understanding coherent systems of thought when written out in prose by that philosopher. Maybe sometimes this is because the philosopher’s good at thinking things out but bad at writing them down; maybe sometimes it’s because they’re good at both but many readers are bad at reading. But either outcome seems clearly compatible with Vallier’s defense.
In case (3), a genuine contradiction of that sort is evidence that the philosopher in question has failed at reasoning consistently or building coherent systems of thought in that respect. But lots of people fail at reasoning consistently or building coherent systems of thought. If I got some local Baptist youth group leader to write up a book on My Thoughts on the Relationship Between Pleasure and Virtue, And Further Considerations on Its Bearing on the Arts of Rhetoric and Politics, by Cooper T. Cobb IV, then I dare say that that book might also include some views in tension with one another and might also have some genuinely contradictory ideas contained in it. Let’s say that the works of Plato have N(p) genuine, non-innocently explainable contradictions in them, and that the works of young Cooper have N(c) genuine, non-innocently explainable contradictions in them. Then the relevant question for Vallier’s defense isn’t whether N(p) > 0; it’s more like whether N(p) << N(c). (If so, that might be one reason why it's worthwhile to spend a lot of time on the contradictions that you do seem to find in Plato: if a pile of garbage on the floor has one piece that doesn't fit with the other pieces, that's not surprising, and often not very interesting to talk about; but if you see a relatively well-ordered engine, and one part seems to be moving out of step, then the exceptionality is itself a reason to spend some significant amount of time trying to figure out whether it actually is out of step with the mechanism, and if so, what that means, what the effects are, etc.)
LikeLiked by 1 person
All of that sounds right. Actually, Plato is probably an unnecessarily complicated example to pick, considering the difficulties involved in interpreting dialogues as opposed to treatises (i.e., the difficulty involved in distinguishing his view from that of a given character).
LikeLike
Appreciate you blogging thiss
LikeLiked by 1 person
Thanks!
LikeLike