Epistemic Stances and Rational Obligation - Parts One and Two
In a recent paper, Anjan Chakravartty discusses the concept of epistemic stances. This idea is not new, but Chakravartty provides a good description of what such a stance would be:
For Chakravartty, examples of families of epistemic stances include:
Deflationary stances, which are cautious about whether or how we can describe a mind-independent world;
Empiricist stances, which question whether theorizing beyond the observable phenomena should be a basis for warranted belief;
More metaphysically inclined stances, which regard scientific method as justifiable, and allowing deductions, laws, and generalizations of the familiar type used in traditional science in short, scientific realism.
And perhaps there are more. But Chakravartty asks, Do we want to say that (at least) all three of these stances are rationally permissible (subject to some minimal constraints of internal consistency and coherence)? He characterizes such a position as stance voluntarism.
According to this view, there are various rational positions we could choose to take, various different weights we could give to the factors that comprise an epistemic stance. In particular, we want to chart a course between the poverty of excessive caution and the pitfalls of excessive zeal (Chakravartty 2017) in deciding what warrants our belief. And, crucially, there is no further court of (strictly) rational appeal. Having chosen an epistemic stance, one can deploy reason, logic, evidence, et al. to form judgments within that stance, and to continue to clarify the implications and use of that stance. But choosing the stance itself can never be a matter of rationality alone. This is why a stance is not so much believed (which would imply assenting to its truth) as adopted. This is also why, in Chakravarttys view, disputes between realism and antirealism are rationally unresolvable. Its important, though, to note that the realism/antirealism debate (which is the focus of Chakravarttys paper) is only one place where the question of dueling epistemic stances arises. Similarly, the focus on scientific realism should not lead us to ignore the implications that the stance-voluntarism question has for realism more generally.
This voluntarist position has to answer the objection made most recently by Christopher Pincock here (and in other places by other philosophers, of course). It is, in brief, that an epistemic stance that can, for instance, endorse scientific realism is made obligatory by a certain understanding of rationality. And this understanding (which, its proponents have to claim, is the only legitimate or defensible version) can be shown to rule out other epistemic stances as irrational. It is not the case that an epistemic stance is truly voluntary, as in up to me, on this view or at least not a stance that claims to be rationally defensible. It is possible to show that choosing a non- or antirealist stance involves the chooser in self-contradiction and/or incoherence. This is unacceptable to anyone who wants to say that their chosen stance is rational. So, Pincock concludes, scientific realism is obligatory, because the epistemic stance that undergirds it is also obligatory. (Its an interesting question whether he would agree that it is equally incoherent for a scientific antirealist to claim that their stance is voluntary.)
Lets clarify some terminology, starting with obligatory and voluntary. In this debate, an obligatory stance is one that must be chosen if the chooser wishes to be guided by reason for it is reason itself that makes it obligatory. How might we recognize such a stance? As Pincock points out, some form of reductio could be appealed to, which purports to show that taking the opposite position, and regarding ones (rational) stance as voluntary, leads to pragmatic incoherence. The attractiveness of this position, should it prove true, is that it asks for a very minimal agreement about what being rational consists of: coherence and non-contradiction. But more of this later.
In contrast, then, to an obligatory epistemic stance, a voluntary stance would be one that can be either chosen or not, without violating some obvious constraint on rational procedure. I believe it is key to the argument that there are supposed to be no such stances. To be rational at all is to be obligated to one and only one epistemic stance.
Continuing with terminology: Pincock uses the term theoretical reasons to contrast with practical reasons for choosing an epistemic stance. He reads Chakravartty as saying that there are no theoretical reasons for such a choice; all stance choices must be motivated by practical reasons. Here is one argument Pincock makes about this:
But is it clear that the stance voluntarist must rule out theoretical reasons for their choice of stance? Notice that Pincock has reduced the factors Chakravartty lists as part of an epistemic stance to a single term: values. Are we sure that Pincock is not begging the question against the stance voluntarist, by insisting that the only candidates we might adopt to serve as non-obligatory value reasons are non-rational? I myself dont read Chakravartty as saying that. He isnt declaring that a choice of stance is just what I like. He says that the stance voluntarist does have rational reasons, but the reasons are not therefore obligatory for everyone, nor are they the only reasons for choosing a particular stance. The coherence of that position -- essentially as I described it above in talking about voluntary stances -- is what must be investigated. Can there be rational reasons that dont obligate everyone?
This is a deep question. In some fields, valid reasons rationality itself, it might seem are indeed obligatory. If I demonstrate valid reasons for concluding that there is no highest prime, you had better assent to this, if you want to be rational. And I can claim that this has nothing to do with choices or what I like -- there is only one acceptable rational conclusion. Is committing to an epistemic stance like this? Are the theoretical reasons similarly constraining? Again, remember that once were operating within a particular epistemic stance, all kinds of things become rationally obligatory. The question here is a level up question what about the adoption of the stance itself?
How could Pincock show that the non-voluntarist position is the only rationally acceptable one? To do that, he would first have to begin by arguing for a certain specific view of what it means to be rational, again dealing with the question of what being rational obligates us to. Second, he would need to show that any theoretical reason produces obligatory positions on epistemic stances. It isnt sufficient merely to demarcate theoretical from practical reasons, and claim that Chakravarttys quite varied list of ingredients that go into choosing a stance are all non-theoretical. Pincock needs a tougher definition of theoretical reason than that. He would also have to show that a theoretical reason is always a compelling rational reason, not merely a non-practical reason, and this must be understood in terms of Pincocks particular claim about what rationality entails (call this P-rationality). Once this identification of theoretical with P-rational is made, he could go on to show why theoretical reasons become obligatory.
Chakravartty, in his turn, is almost certainly going to respond that Pincocks version of rational is too constricted, and begs the question against him; that we shouldnt equate all theoretical reasons with this constricted P-version; and thus there is ample room for voluntary theoretical choice of epistemic stances. The very idea that a given stance must be rationally obligatory to be rationally chosen is precisely what stance voluntarism denies. . . . On the voluntarist view, rational choice and rational obligation are distinct concepts and cannot be run together. (1312) That said, I believe Chakravartty would have to agree that there is no version of rational that would permit incoherence and self-contradiction recall his minimal constraints of internal consistency and coherence. So if Pincock can indeed show that this must result from stance voluntarism, then Chakravarttys position may be in trouble.
Ill end this first part with a few questions about the level up problem: reasons for reasons.
Is the argument for something being true, and worthy of belief, within a given epistemic stance the same kind of argument wed give for the stance itself being rationally obligatory? (In the debate between Chakravartty and Pincock, this is framed as a contrast between doxastic and epistemic stances.)
Does one version of rationality pull itself up by its own bootstraps without recourse to a pragmatic-incoherence argument? (Ill lay out Pincocks incoherence argument in Part Two.)
Does a rejection of stance voluntarism lead directly to scientific realism, or merely to the demand for obligatory reasons for adopting any stance?
The debate about epistemic stances mirrors many familiar features of the debate about objectivity itself. So, for instance, when Thomas Nagel argues that reason must be given the last word, is he also denying stance voluntarism? When Nagel defends a rationalist answer against what I shall call a subjectivist one (The Last Word, 3), is he also defending a single version of rationality that it is obligatory to adopt?
Pausing here, I invite comments about the debate so far: Is it clear whats at stake? Are there further terminological questions we need to sharpen?
Chakravartty, 1308-9:An epistemic stance is an orientation, a collection of attitudes, values, aims, and other commitments relevant to thinking about scientific ontology, including policies or guidelines for the production of putatively factual beliefs . . .
A stance is not a claim about the world. Stances are not believed so much as adopted and exemplified in assessments of evidence, producing interpretations of scientific work that yield claims about scientific ontology, and claims regarding matters about which it would be better to be agnostic instead.
For Chakravartty, examples of families of epistemic stances include:
Deflationary stances, which are cautious about whether or how we can describe a mind-independent world;
Empiricist stances, which question whether theorizing beyond the observable phenomena should be a basis for warranted belief;
More metaphysically inclined stances, which regard scientific method as justifiable, and allowing deductions, laws, and generalizations of the familiar type used in traditional science in short, scientific realism.
And perhaps there are more. But Chakravartty asks, Do we want to say that (at least) all three of these stances are rationally permissible (subject to some minimal constraints of internal consistency and coherence)? He characterizes such a position as stance voluntarism.
According to this view, there are various rational positions we could choose to take, various different weights we could give to the factors that comprise an epistemic stance. In particular, we want to chart a course between the poverty of excessive caution and the pitfalls of excessive zeal (Chakravartty 2017) in deciding what warrants our belief. And, crucially, there is no further court of (strictly) rational appeal. Having chosen an epistemic stance, one can deploy reason, logic, evidence, et al. to form judgments within that stance, and to continue to clarify the implications and use of that stance. But choosing the stance itself can never be a matter of rationality alone. This is why a stance is not so much believed (which would imply assenting to its truth) as adopted. This is also why, in Chakravarttys view, disputes between realism and antirealism are rationally unresolvable. Its important, though, to note that the realism/antirealism debate (which is the focus of Chakravarttys paper) is only one place where the question of dueling epistemic stances arises. Similarly, the focus on scientific realism should not lead us to ignore the implications that the stance-voluntarism question has for realism more generally.
This voluntarist position has to answer the objection made most recently by Christopher Pincock here (and in other places by other philosophers, of course). It is, in brief, that an epistemic stance that can, for instance, endorse scientific realism is made obligatory by a certain understanding of rationality. And this understanding (which, its proponents have to claim, is the only legitimate or defensible version) can be shown to rule out other epistemic stances as irrational. It is not the case that an epistemic stance is truly voluntary, as in up to me, on this view or at least not a stance that claims to be rationally defensible. It is possible to show that choosing a non- or antirealist stance involves the chooser in self-contradiction and/or incoherence. This is unacceptable to anyone who wants to say that their chosen stance is rational. So, Pincock concludes, scientific realism is obligatory, because the epistemic stance that undergirds it is also obligatory. (Its an interesting question whether he would agree that it is equally incoherent for a scientific antirealist to claim that their stance is voluntary.)
Lets clarify some terminology, starting with obligatory and voluntary. In this debate, an obligatory stance is one that must be chosen if the chooser wishes to be guided by reason for it is reason itself that makes it obligatory. How might we recognize such a stance? As Pincock points out, some form of reductio could be appealed to, which purports to show that taking the opposite position, and regarding ones (rational) stance as voluntary, leads to pragmatic incoherence. The attractiveness of this position, should it prove true, is that it asks for a very minimal agreement about what being rational consists of: coherence and non-contradiction. But more of this later.
In contrast, then, to an obligatory epistemic stance, a voluntary stance would be one that can be either chosen or not, without violating some obvious constraint on rational procedure. I believe it is key to the argument that there are supposed to be no such stances. To be rational at all is to be obligated to one and only one epistemic stance.
Continuing with terminology: Pincock uses the term theoretical reasons to contrast with practical reasons for choosing an epistemic stance. He reads Chakravartty as saying that there are no theoretical reasons for such a choice; all stance choices must be motivated by practical reasons. Here is one argument Pincock makes about this:
Pincock, 5:The key problem is that [Chakravarttys] argument assumes that there are no theoretical reasons for ones values. This is a contentious and problematic assumption. Most people assume that at least some values can be supported by theoretical reasoning. . . . I would maintain that the same point applies to epistemic values, such as the value one attaches to having a scientific explanation. That is, just as people should adopt some moral values (and not others), people should also adopt some epistemic values (and not others).
But is it clear that the stance voluntarist must rule out theoretical reasons for their choice of stance? Notice that Pincock has reduced the factors Chakravartty lists as part of an epistemic stance to a single term: values. Are we sure that Pincock is not begging the question against the stance voluntarist, by insisting that the only candidates we might adopt to serve as non-obligatory value reasons are non-rational? I myself dont read Chakravartty as saying that. He isnt declaring that a choice of stance is just what I like. He says that the stance voluntarist does have rational reasons, but the reasons are not therefore obligatory for everyone, nor are they the only reasons for choosing a particular stance. The coherence of that position -- essentially as I described it above in talking about voluntary stances -- is what must be investigated. Can there be rational reasons that dont obligate everyone?
This is a deep question. In some fields, valid reasons rationality itself, it might seem are indeed obligatory. If I demonstrate valid reasons for concluding that there is no highest prime, you had better assent to this, if you want to be rational. And I can claim that this has nothing to do with choices or what I like -- there is only one acceptable rational conclusion. Is committing to an epistemic stance like this? Are the theoretical reasons similarly constraining? Again, remember that once were operating within a particular epistemic stance, all kinds of things become rationally obligatory. The question here is a level up question what about the adoption of the stance itself?
How could Pincock show that the non-voluntarist position is the only rationally acceptable one? To do that, he would first have to begin by arguing for a certain specific view of what it means to be rational, again dealing with the question of what being rational obligates us to. Second, he would need to show that any theoretical reason produces obligatory positions on epistemic stances. It isnt sufficient merely to demarcate theoretical from practical reasons, and claim that Chakravarttys quite varied list of ingredients that go into choosing a stance are all non-theoretical. Pincock needs a tougher definition of theoretical reason than that. He would also have to show that a theoretical reason is always a compelling rational reason, not merely a non-practical reason, and this must be understood in terms of Pincocks particular claim about what rationality entails (call this P-rationality). Once this identification of theoretical with P-rational is made, he could go on to show why theoretical reasons become obligatory.
Chakravartty, in his turn, is almost certainly going to respond that Pincocks version of rational is too constricted, and begs the question against him; that we shouldnt equate all theoretical reasons with this constricted P-version; and thus there is ample room for voluntary theoretical choice of epistemic stances. The very idea that a given stance must be rationally obligatory to be rationally chosen is precisely what stance voluntarism denies. . . . On the voluntarist view, rational choice and rational obligation are distinct concepts and cannot be run together. (1312) That said, I believe Chakravartty would have to agree that there is no version of rational that would permit incoherence and self-contradiction recall his minimal constraints of internal consistency and coherence. So if Pincock can indeed show that this must result from stance voluntarism, then Chakravarttys position may be in trouble.
Ill end this first part with a few questions about the level up problem: reasons for reasons.
Is the argument for something being true, and worthy of belief, within a given epistemic stance the same kind of argument wed give for the stance itself being rationally obligatory? (In the debate between Chakravartty and Pincock, this is framed as a contrast between doxastic and epistemic stances.)
Does one version of rationality pull itself up by its own bootstraps without recourse to a pragmatic-incoherence argument? (Ill lay out Pincocks incoherence argument in Part Two.)
Does a rejection of stance voluntarism lead directly to scientific realism, or merely to the demand for obligatory reasons for adopting any stance?
The debate about epistemic stances mirrors many familiar features of the debate about objectivity itself. So, for instance, when Thomas Nagel argues that reason must be given the last word, is he also denying stance voluntarism? When Nagel defends a rationalist answer against what I shall call a subjectivist one (The Last Word, 3), is he also defending a single version of rationality that it is obligatory to adopt?
Pausing here, I invite comments about the debate so far: Is it clear whats at stake? Are there further terminological questions we need to sharpen?
Comments (63)
As the saying goes, to a hammer, everything looks like a nail. A corollary might go - to a metaphysician, everything looks like metaphysics. I'll now stretch the truth by calling myself a metaphysician.
What you and Chakravartty call "epistemic stances," I might call "metaphysical positions." Here's what R.G. Collingwood says about metaphysics in "An Essay on Metaphysics."
A presupposition is an assumption that establishes the context for a philosophical or scientific discussion. Here's what Collingwood says about absolute presuppositions:
Reading your post, I was trying to figure out how what you are describing is different from what I am. They don't seem exactly the same, but very similar. I don't want to clutter up your thread with a discussion of the differences or similarities, but there is one aspect I think is worth mentioning.
Quoting J
A lot of what you've written here is consistent with what I think Collingwood might agree with some qualifications. I especially like "This is why a stance is not so much believed (which would imply assenting to its truth) as adopted." First, I don't think we typically choose our metaphysical positions, or whatever you call them. Unless real effort is put into them people are usually unaware of them.
Also, to go back to my quote from Collingwood - "...absolute presuppositions have been made by this or that person or group of persons, on this or that occasion or group of occasions, in the course of this or that piece of thinking." You don't have to choose. It is appropriate to use different absolute presuppositions in different situations depending on what is the most useful point of view. Beyond that, they don't have to be, and usually aren't, adopted based on rational criteria.
As I noted, I don't want to disrupt you discussion, so I won't take this any further.
Quoting T Clark
I agree, this is in the same family as "epistemic stance," as used by Chakravartty and Pincock.
(@tim wood, above, also noticed the resemblance to Collingwood.). One difference may lie in the idea of an "absolute presupposition," which I think is too strong. For Chakravartty, at least, an epistemic stance is tentative, flexible, and dependent on a lot more than what I think you're calling metaphysics. Really, as I read him, it isn't an assumption at all, but a carefully chosen "best practice." In the context of doing science, he wants an epistemic stance to walk the line between filtering out weak knowledge claims while not squelching ones that deserve a hearing. Pincock's interest in rational obligation may be a little closer to something absolute. But of course Collingwood points out that different absolute presuppositions may apply in different situations, so perhaps the difference is small.
The debate I'm focusing on, though, is precisely whether this is a good characterization of what an epistemic stance (or presupposition, if you like) entails. Chakravartty would agree with Collingwood that such a stance is not chosen exclusively based on rational criteria, nor can it be defended that way. But Pincock disagrees. So would many others who believe that realism and objectivity are not optional but rather obligated by a certain understanding of rationality.
I hope you find time to read the two papers. Your input would be welcome. But if not, I'll go on to fill out more of the debate within a day or two.
I read the paper you linked and I enjoyed it. The guy writes really well, which isn't the same as saying I understood everything he wrote. What he writes is familiar, but it's different enough from the way I usually talk about it that I get a bit lost. It feels like he's mixing metaphysics and science while I think of them as entirely different things. A metaphor I like is that metaphysics is the traffic laws, science is driving your car.
Some thoughts.
Quoting J
I was thinking something like this when I first read your OP. You quoted Chakravartty as saying "An epistemic stance is an orientation, a collection of attitudes, values, aims, and other commitments relevant to thinking about scientific ontology, including policies or guidelines for the production of putatively factual beliefs . . " That does seem broader and less stringent than what I usually think of as metaphysics. That's one of the things I was wrestling with as I tried to figure out how they compared.
As for the article itself, what bothered me the most is that realism and antirealism are set up as mutually exclusive and incomprehensible. Fact is, you can use both. I come from science and engineering. When I deal with those types of issues, I generally toe the realist line, but I recognize that approach has limitations. One of the first discussions I started six or seven years ago addressed whether the idea of objective reality is a useful one.
From a broader philosophical perspective, I'm definitely an anti-realist. I don't know if you've ever read any of my posts about Taoism. In Taoist cosmology, we bring the everyday world into existence by naming things, making distinctions. Boiling that down into a less mystical and esoteric sounding message, the world we humans see and know is constrained by our human nature - the physical, biological, genetic, and neurological nature of our bodies, established by evolution and development; and the cognitive/psychological nature of our minds, established by learning and the structure of our nervous system. There is significant overlap between those factors.
I don't feel any conflict between those two ways of seeing things. I can have an intelligent, if limited, conversation about hard physics and also a philosophical discussion about Taoist principles.
I'm not sure where to go from here or whether I have anything else to offer.
Quoting T Clark
This occurred to me too. If I had to make one overall criticism of analytic philosophy, it's the tendency to separate ideas into convenient binaries for purposes of discussion, ignoring what is actually done with those ideas in the real world. Pincock makes a strong case for why realism and anti-realism do have something fundamentally incompatible about them, however.
Quoting T Clark
I think both Chakravartty and Pincock would agree that it is useful. Pincock, though, would add that it has the additional virtue of being real or true.
If you want to know about knowledge within contexts, read here: https://thephilosophyforum.com/discussion/14044/knowledge-and-induction-within-your-self-context/p1 We create our own contexts and knowledge is essentially deduction within that context. With that we can rationally also figure out induction. I think this explains what you're asking far better than this theory here.
Or perhaps I should ask, do you think there's a single version of what is rational -- and hence what should inform our epistemic stance -- when doing science?
Yes. I note it in the paper I linked. First, use deductive knowledge based off of context, resources, and time. Where there are limitations, use inductions based off of hierarchy of induction. So in order of cogency, probability, possibility, plausibility, and irrational.
It depends on what you are trying to achieve. If you're trying to achieve the most logical outcome, then you should. If your goal is to justify something you want and you don't care about it, then no. For example, lets say I want to believe in God. Being truly rational may not allow me to justify that belief. So instead of choosing the most rational outlook, I choose something more subjective and emotional.
Can I voluntarily decide to be rational vs less rational? Of course. There is nothing holding a person down requiring them to be rational. Lots of people hate it. They want what they want and they'll use any tactic to get it. There are of course consequences for being less rational in life, and a person has to decide if paying those consequences for being less rational is worth the outcome they want.
This, I think, is close to Chakravartty's sense. He specifies "minimal constraints of internal consistency and coherence -- so, broadly speaking, logical. In that sense, then, you're saying that such a stance is not voluntary or optional; we should choose it. How would you argue for that? Or do you think Pincock's position basically sets out that argument?
You should choose the most rational stance in X context if you are in X context. If you don't, then don't. Just clarifying there is nothing innate forcing anyone to choose anything.
Quoting J
Again, its contextually dependent. Its irrational to be rational if you want a conclusion that is not rational. If you want a conclusion to be rational, then it is rational to be rational. And what is rational is based on the context of the knowledge and tools you have available to you at the time. Quoting J
I think I have set it out straight clear and I really have no interest in what Pincock thinks. Comparing and contrasting two philosophers is way below my interest at this point. I care about arguments, not about what people might think or how they would defend something. If you think I'm in line with Pincock or against Pincock, great either way. Does the argument and point work? That's all that matters to me.
Boy, I tried. I found it impenetrable.
Quoting J
As I understand it, the idea of objective reality is metaphysics. It's not our only choice of the way to see things, even for science. This is something I've discussed often since I joined the forum. I'm not the only one who sees things that way - it's not goofy philosophy. I'll take it further, I think reality and truth are also metaphysical concepts.
Perhaps we've taken this as far as we can. Focusing on metaphysics rather than epistemology is taking us farther and farther from the OP.
I don't agree. What you say here is critical to the issues the OP raises. After all, Chakravartty himself refers to a realist commitment to scientific method as a "metaphysical stance." When you write, "It's not our only choice of the way to see things, even for science," you're coming to grips with the key point that separates Chakravartty and Pincock. So please stay on the thread if you're interested!
Great OP and interesting paper. Before I go off half cocked, I was wondering if Chakravartty or Pincock have any writings about how one might adapt one's values, or change stance, given evidence? I want to say roughly what I understand of stances so you can tell me if I'm hopelessly wrong.
The construal of an epistemic stance, and indeed of epistemic values, seems to be {in the paper's words} "upstream" of matters of fact and questions of ontology, rather than "alongside" or possibly "downstream" of matters of fact and question of ontology.
I think this is highlighted by how to conceive of defeaters for claims in relationship to stances. By a defeater I mean any fact, whether empirical or a matter of reason, which has a modus tollens impact on a claim - by which the defeater may "defeat" the claim. A stance is "an orientation comprising attitudes and policies relevant to assessing evidence", which if I read it right, construes stances as hodge-podges of attitudes, intentions, implicit principles, explicit principles, social norms, behaviours and so on. It's some knowledge praxis and its rules. A stance doesn't judge matters of fact, it is a means by which matters of fact are judged - much like an assembly line for bikes can't be ridden as a bike. In that regard a commitment to a stance is an enactment of it.
You could conceive of a defeater D for a claim E in a stance S as triple {D,E,S}, which says that "D would refute E in S were D true". That comprehends defeaters as relative to systems of belief and principles.
As an example, consider a study of sample size 20 with a conclusion D that acts as a defeater for the claim E. Alice is quite conservative, and has "I will be agnostic about the conclusions of any study with sample size of 20 or less" as a principle. Bob, a bit of a cowboy, has the principle "I will be agnostic about the conclusions of any study with sample size 19 or less". Bob could revise their beliefs connected to E using D, Alice could not.
You could also conceive of defeaters as having a more pervasive impact, that they might penetrate the supposed "upstream"-ness of epistemic stances from matters of fact and ontology. It could be that in light of some defeater D for E, any stance which allows belief or agnosticism toward D is compellingly refuted but not necessarily through modus tollens impact - D may "defeat" stances permitting E by providing generic reasons for updating conduct that preclude E. An upheaval of the epistemic terrain, as it were.
An example there might be conservative Alice, who would never trust the scientific use of partially uninterpretable AI in scientific publications - and thus be agnostic about the conclusions of any paper using them. And cowboy Bob, who believes in the potential of AI and does not withhold belief on that basis. Alice and Bob would react differently to the relatively recent {almost total} solving of protein geometry given their base pair sequence by an AI, Alice would withhold belief, Bob would not.
Then, the applications of that technology happen, and new effective antibiotics are developed with these quick to press designer proteins. If everyone ought act in accordance with Alice's prohibition on trusting any fruits of AI, no one would have jumped ahead to produce the antibiotics, and we would live in a world with more death and pain as well as less scientific discovery. Alice's beliefs would have hampered the discovery of more truths, and that would be one fact among others.
Nevertheless Alice's beliefs have not been formally refuted in accordance with only the logical principles of their connection, she would need to change a stance defining principle - trust AI more. Which would be a belief about which methodologies are admissible. But that would render discoveries, facts, results - methodology - as potential changes for the admissibility of methodologies, and thus undermine a stance's construal as "upstream" from facts and matters of ontology.
If I am not talking out of my arse, I see an angle of attack in the debate that uses something like an undermining of the fact/value distinction, only this case it's an undermining of the distinctions between fact/method/methodology/meta-methodology, by construing fact, method, methodology and meta-methodology as inferentially related. The flexibility that goes into defining what allows one to adopt or enact a stance seems to give such wiggle room.
I'm glad it engaged you.
Quoting fdrake
I don't know; I only know of these two philosophers from reading the cited papers. An important question, though. I would imagine that a stance voluntarist could give a convincing account of how they switched from one stance to another, given evidence. Someone like Pincock will have a much harder time, as they may have to actually deny that cogent evidence could arise. But notice what Pincock says toward the end of his paper: "A realist should not endorse . . . dogmatic loyalty to IBE [inference to the best explanation]. . . The only viable form of scientific realism is a cautious realism that responds appropriately to the historical record of success and failure for various modes of inference." We're left with wondering exactly how to interpret "appropriately", but the overall tone is not inflexible.
Quoting fdrake
Yes.
Quoting fdrake
Yes. Moving on to "defeaters":
Quoting fdrake
Your entire discussion of defeaters is very good, and I think puts the "upstream" problem in the right context, but let me zero in on the conclusion here. If an epistemic stance is supposed to tell us what we ought to believe, does that mean it has to be a one-way street? Or to stay with the river metaphor, does the justificatory stream flow in a single direction?
Well, how do we cobble together our epistemic stance in the first place? If Chakravartty's characterization is largely correct, it's a combination (hodge-podge?!) of factors, many of which are undoubtedly traveling "upstream" from the downstream events and evidence of our lives. At a certain point, we find we have a stance, however tentative. So the question is, does this stance now put up a kind of dam against any pesky evidentiary salmon that wants to swim upstream with new information that could put the stance itself into question?
I think this depends on how deep the epistemic commitments go. Someone like Pincock probably wants to say that some elements that comprise the commitments are irreversible, on pain of irrationality or incoherence. Your example of conservative Alice and cowboy Bob, however, doesn't seem nearly that bedrock. Couldn't a scientific realist make room for both Alice and Bob in the Temple of Reason? After all, neither one is questioning realism per se. They just have different risk tolerances when it comes to beliefs. The stance voluntarist will say that such tolerances are (largely) unresolvable by rational argument, which is not to say they aren't motivated by theoretical reasons. The stance-obligatorist (if that's a word) will deny this, and perhaps argue that even the difference between 19 and 20 represents the difference between what is rationally obligatory, given a realist stance, and what is incorrect.
This of course relates to your question about how a stance might change. My stance provides certain criteria for what counts as evidence tout court. Does that include evidence for the stance itself? Can the very stance that certifies item D as evidence in good standing be changed as a result of D, when what D defeats is some element of stance E? But then, that might mean D wasn't evidence in good standing after all, if the new stance no longer recognizes D as valid. This is a truly headache-producing circle, and I don't know the answer, other than to say that it motivates my question in the OP, "Is the argument for something being true, and worthy of belief, within a given epistemic stance the same kind of argument wed give for the stance itself being rationally obligatory?"
Quoting fdrake
Yeah, that's what it would be if there's no "rigid rational" epistemic stance that can trump all others, and travel both upstream and downstream is permitted.
Quoting fdrake
I'm leaning that way too but let me give Pincock his say in Part Two.
I think the following is an option - upstream, downstream and alongside relations are allowed between stances and evidence, it just so happens that there is One True Dialectic that correctly links them. The One True Dialectic would have to fully understand how it related to all of its own principles, and conditions of revising them. I don't believe such a thing exists, but I would want an argument to rule it out.
Quoting J
Quoting J
I don't have a good answer in terms of the paper. I just want to throw things together and look at the muck they make. I think this works as a criticism of the paper, because its argument rests on making a few sharp distinctions that instead seem quite blurry. Namely, a stance distinguishes itself from object level factual claims, and that it does so by being "upstream" from them. And also it distinguishes stances from collections of attitudes [hide=*]{or at least doxastic attitudes?}[/hide] toward object level claims. They're construed more as means of properly assigning attitudes toward object level claims.
The paper advances the idea that a selection mechanism might work on stances, and render some of them rationally impermissible and some rationally permissible. Above and beyond that, there is the possibility of there being a single stance which is obligate to hold {about some domain}. I mostly want to focus on the rhetoric in the above paragraph because I think it highlights something about the imaginative background of the argument.
Stances are posited as separate - upstream - from the content their principles concern, and thereby the sentence "To add to this dialogue the assurance that..." works as a rebuke on the back of separating the stance's principles from their content, as such a declaration "adds nothing". It is this "adds nothing", that portrays the declaration of an epistemically privileged stance as extraneous, which pumps the intuition of separation set up prior.
I think that's the core of the article's imaginative background on the matter. It cleaves the enactment of an epistemic stance from what it concerns, which could be read as cleaving how things are done from what's done, even though what's done influences how things are done through learning, and how things are done influences what's done through norms.
Epistemic stances also seem modelled off of relatively static principles, specifically commitments toward certain classes of statements:
For example, "deflationary stances" typically are "wary" toward claims that contain reference to a "mind independent world". That tells you that a whole class of stances can be characterised by their {class of} relationship to a class of claims. I say "class of" relationships because there's going to be more than one way to be "wary". The models of stances above are all principally philosophical stances, which makes sense given the terrain, but it's worthwhile to compare this with the expert witness court comparison in the final paragraph's rhetorical flourish:
while keeping in mind the author's comment about stances
Even though both creationism and evolution are at least bodies of putatively factual claims about the world. It could be that teaching creationism and evolution might be more a matter of principle, but that still raises the question of how a body of putatively factual claims can ever be related to without the resultant interaction becoming in part matter of principle and of fact, thereby ending up in the circular muck we're in.
It would then seem that the stance is secretly a list of propositions and attitudes toward them, rather than a means of assigning propositions to attitudes given a context. But that goes somewhat against the author's method of parrying an accusation of doxastic voluntarism:
What saved the author from the charge was a clean distinction between "an orientation comprising attitudes and policies relevant to assessing evidence" and "bodies of putatively factual claims about the world". But which contained examples of attitudes towards beliefs - deflationists are wary toward claims regarding mind independent worlds.
I don't really know what to do with this, and I might be missing a lot of subtleties, but my suspicion is that the distinctions between stance and doxastic attitudes, and stance and object level claims, aren't as clear as the argument needs to go through.
Starting with your final thought . . . I agree. The more I reflect on both papers, the more I wonder whether Chakravartty and Pincock really have the same conception of what an epistemic stance is. Chakravartty gives a clear enough description, which I quoted, but we can see that, because he wants to present stance selection as a broad process involving many factors, he can't be precise about what is and isn't a stance, compared to a doxastic process within a stance. As we've noticed, merely using the "upstream/downstream" idea doesn't settle all the important questions about how that works. And it matters whether a stance is immune from "downstream" input.
Pincock, in contrast, needs a stance to be largely independent of its subject matter, and determinable by rational ("theoretical") criteria alone. Is this even the same thing that Chakravartty describes? A stance, described thusly, results in a huge meta-commitment such as "realism." Whereas Alice and Bob don't seem to have such a disagreement. Their differing stances look much less philosophically weighty -- and that may be Chakravartty's point, in part.
Quoting fdrake
Great. Such a dialectic would presumably be capable of resolving -- or explaining away -- that nasty circle I described, in which a defeater changes a stance, in turn putting into question whether the defeater was legitimate evidence. I'd be very interested in Pincock's take on this: Does non-voluntarism about epistemic stances mean that there must be such a dialectic?
Quoting fdrake
Yes. It's important to remember that Chakravartty is not an "anything goes" guy. He certainly believes that some epistemic stances would be ill-chosen, on grounds of irrationality.
Quoting fdrake
To use some old language, an epistemic stance is imagined as -- conceivably, if not in practice -- an a priori commitment, an armchair commitment that could be determined without recourse to any questions about "what it concerns." The appeal would be strictly to the "how," the process, rather than what that process is working with. We could even go so far as to call it an analytic understanding of epistemology. (Or is this way too strong for Chakravartty, who is very concerned with contexts?) For a non-voluntarist like Pincock, this becomes an appeal to rationality itself, which on this understanding will dictate our epistemic stance.
But, as you point out, this immediately seems to lead to some inconsistencies about what's a factual claim and what's a criterion for a factual claim. Granting that creationism is incorrect, is this because it is a false factual claim, or is it better characterized as a false conclusion based on an epistemic stance that is much too liberal in what counts as evidence? In other words, do the creationist and the evolutionist even agree on what counts as a factual claim -- do they share the same epistemic stance about this? I would say they do not, meaning that their disagreement is in part about stances, not just the facts on the ground. They will have different understandings of how to determine what a "fact" is. And this is where Pincock's realism comes in. He would of course claim that their understandings are not merely different; one is correct, the other ludicrously wrong.
Quoting fdrake
I know what you mean. If a non-voluntarist is going to claim that they have a rationally/theoretically mandatory epistemic stance, they will be asked their reasons for believing this. Will the reasons they give be the same kind of reasons that two people would give who share an epistemic stance but disagree on a particular scientific interpretation? This is hard to understand. And it tempts us to say that all this talk of stances is really a way of justifying some core propositions about method or process which are believed/disbelieved/held as uncertain, not merely "adopted." We want to link propositions with these same attitudes within an epistemic stance -- that's the whole point of having one -- but where are we standing before the stance? What's the further argument that there are worse and better (maybe even obligatory) reasons for enacting the stance?
There's a lot more in what you wrote that is interesting and worth pursuing, but I'll stop here. Since this question of what might make an epistemic stance attractive or even obligatory keeps showing up as central, I really should write Part Two, which concerns Pincock's incoherence argument. So I'll try to do that fairly soon.
My standard complaint about philosophical thought experiments is that they are usually simplistic and unrealistic. Yes, I mean you Trolly Problem. I don't think the way you've laid this out represents how people actually use data. Here's how it would actually work - at least in my imagination. Alice and Bob work together at a pharmaceutical company. They're doing research to identify new candidates for testing as antibiotics. The goal of their work is to prepare a list ordered in terms of the probability that a cost-effective drug can be developed - which are most likely to make the company money. To do that, they search through relevant publications and other sources and come up with a list of possibilities.
Candidates are then evaluated using various criteria including sample size and whether the data source is prepared using AI input, but also cost to manufacture, existence of similar drugs, potential for innovation, possible patent conflicts, and lots of other things I don't know about. Alice and Bob then get together and negotiate the ordering of the list based on each person's epistemic priorities. Perhaps it would be silly to reject data based on one arbitrary standard, but it might make a lot of sense to apply a standard probabilistically as part of a comprehensive evaluation process.
In my experience, discussions like this one often miss the point by focusing on rigid and unrealistic processes for determining what is a fact and what isn't.
I'm dying to know what your software misunderstood here! :grin:
The passage you offer is very on-point. In the OP, I only devoted a single sentence to Chakravartty's idea that stance voluntarism would explain why the realism-antirealism debate can never be resolved. But this is an important claim he wants to make -- one that Pincock would certainly have to deny.
The reasons offered by Fine et al. are in a similar vein, but not identical. What I think they have in common with Chakravartty's viewpoint is the idea that the desideratum of "resolving a disagreement" between two epistemic stances is, on analysis, incoherent.
This passage in particular fits with what @fdrake and I were discussing. I bolded the phrase about semantic content as independently fixed because it's a version of the question, Can we really separate "upstream" from "downstream" input in a neat way that maintains a distinction between epistemic and doxastic stances? The rest of the passage plays this out: If "what the claim says was not determinable apart from those interactions with the world through which we assess its truth," then interpretation and truth-value are viciously circular. It's a bit like the creationist/evolutionist example. Once the interpretation of terms like "fact" or "evidence" become dependent on an epistemic stance, we have to look for an interpretative truth that is outside the stance itself. How do we find it? Oddly, this could be considered an argument in favor of either the realist or the pragmatist position!
Supposed to be I am drawn to the critique of both realism and anti-realism.
Thats what I get for doing all of my composing on an iphone while hiking.
I think it's upstreams all the way down.
Quoting J
I had missed the possibility that, yes, while stances and beliefs about matters of fact consist of intentions toward claims, they might consist of different classes of intentions toward those claims. A stance might solely consist of intentions toward claims regarding evidence regarding putatively factual claims, and the object level discourse would consist of intentions toward some domain of putatively factual claims.
I think that's quite artificial though. Presumably a claim like "some properties are mind independent" is putatively factual even if it can't be determined as true or false, and a claim like "I am wary of claims involving the phrase "mind independence"" is object level in discourse about realism vs anti-realism, since it's an expression in it. I think this is clearer with Alice and Bob out in the wild, as @T Clark was saying. A dispute between Alice and Bob regarding sample size 19 vs 20 wouldn't just be about whether sample size 19 or 20 was good, it would be about whether it is reasonable to believe 19 or 20 is the minimum one to allow study results to update your beliefs. As in, they would explicitly be negotiating what attitudes are appropriate to hold toward claims, and what means of forming those attitudes are appropriate in context.
An IRL example of a thresholding dispute I have seen regarding sample size was about whether 1, 2 or 3 additional data points was "most optimal" to disentangle two hypotheses which seemed reasonable given prior collected data. Each data point was very costly to elicit, and there were diminishing returns on the discriminatory power. We stopped at 2, since a compelling reason seemed to be that getting 90% of the benefit of 3 for a 33% saving was worth it, and 2 gave twice the discriminatory power of 1. We could equally have said "go for three it's the most likely to make what we've done before worthwhile cost be damned" or "go for 1, we can safely deprioritise testing this hypothesis from the data, and you need funding for the follow up experiment", what do you believe and why do you believe it, what optimised discovery and what did we have resources for were already part of the object level discourse. Which is to say, the IRL dispute went beyond the putatively factual into our values as scientists with quite limited funding. Or - there was no way to organise the putatively factual, the methodological concerns, and our values in a hierarchical fashion. There was no upstream or downstream.
Maybe I'm taking this too far, but is this another way of saying "We are blank slates, and can only learn from experience" i.e. empiricism? Are there some faculties of mind which start "upstream", and are not created by input from downstream? Huge question, of course, but a serious explanation for how an epistemic stance is chosen must have a tentative answer, I think.
Quoting fdrake
OK. And how do we want to fill in "good"? Presumably with something stronger than "reasonable to believe for purposes of updating beliefs through study results." What might that be? This is a good (sorry!) question to ask a realist.
Quoting fdrake
Very helpful.
Quoting fdrake
Are you sure? Didn't you end up doing precisely that? Or maybe I'm reading "hierarchical" differently from you. I'm thinking you could have (and probably had to) give reasons for what you decided to do, and in explaining them, you'd implicitly be indicating the hierarchy that wound up prevailing. But I could be way off.
I have no idea. If you're just estimating the mean of a data set, sample size 19 and 20 are basically the same thing. It would be really hard to justify one or the other on any purely statistical basis.
Quoting J
Maybe we are thinking of it differently? Though I claim no authority over interpreting this experience.
By hierarchy I meant that there would be direction of influence between things that constitute the stance and things that constitute putatively factual level claims. By denying its existence I meant that a change in the putatively factual level claims may engender a change in what constitutes the stance. I was treating a discovery as a change in putatively factual level claims, specifically the discovery that 2 new data points had the majority of the benefit of 3. And I claim that this triggered evaluating the allocation of resources on that basis, whereas before it was largely a question of scientific accuracy.
We probably had "don't spend too much money" as a value somewhere in the background, but precisely what justification that would manifest in didn't seem fixed. Since we didn't have a rule like "If 2 data points has 90% of the benefit of 3, only gather 2" prior to analysing the data. Adopting that value made sense in light of the evidence.
Quoting J
Yeah I see what you mean. What I'm going to write below has a lot of presumptions in it I've not argued for, and haven't thought through very thoroughly at all.
By "upstream all the way down" I meant that adopting values has a recursive structure as soon as you start blocking influence from one layer of the structure to higher layers - in this case stopping evidence from changing values. I meant that if we start asking "What makes people adopt this attitude toward that?", and we answer with a claim, we can ask of the new claim "What makes people adopt this attitude toward the claim?", and presumably you'd answer with a justification for a putatively factual level claim with some kind of methodological commitment, and if you ask it about a methodological commitment you'd get something about how one assigns attitudes toward evidential standards - which is thus an element of the stance. You could then ask how someone has an attitude toward that and so on, which would be... a stance toward a stance? A meta-stance? Who knows. Notably all of these answers would be inferential, they involve giving reasons.
Quoting J
Whereas, and this is a big complication I think, people may be caused to adopt stances, paradigms of interpretations and so on. The "true reason" that someone values what they value might terminate in describing a cause or telling a story, rather than giving a reason. In that regard if some faculty is "upstream", its relationship to "downstream" inputs would be some kind of feedback relationship. Like if I look at a graph I can see if it has a trend, and that's something I learned, so something "upstream" of my raw ability to access/explore the array of information in my environment visually has conditioned the input into a specific perceptual form based on instruction. I would be able to justify that with "Line go up? Increasing trend", but that justification apes a reconfiguration of my body in accordance with the instruction, which has caused me to be able to see the world differently.
I think the response here might be that this is an overly reductive account of the "stance." The adoption of a stance no doubt involves some propositional beliefs, but on any account leaning towards virtue epistemology, the idea of science as primarily a virtue, a focus on "intellectual habits," the idea of intellectual faculties being more or less developed, etc., "attitudes towards propositions," will fail to fully explain a stance. In part, this will be because attitudes towards key propositions will themselves be determined by the possession of certain intellectual vices/virtues, the strength of relevant faculties, etc.
And even older still. The concerns that dominate much of Wittgenstein's On Certainty are also core concerns of Aristotle's Posterior Analytics, and Aristotle himself is replying to going concerns about "where justification terminates" and "syllogism skepticism" that were apparently common enough to be major philosophical issues back in ancient Athens. Similar concerns show up in ancient Eastern thought as well.
This might be a position that could be added to 's initial list of stances. And it does show how, to 's point, propositional beliefs end up playing a constitutive role in both the adoption and development of stances. If the intellect (and so reason, knowledge, belief, etc.) is defined in terms of language and discursive [I]ratio[/I], we get a different stance where the very intelligibility of any thing is bound up in particular, mutable systems of interpretation.
Interesting. Can you give us a reference in the Analytics?
Quoting Count Timothy von Icarus
Quite possibly, and it raises a subtle question about whether there is a thing called a "realist epistemic stance," or rather whether it's the case that a certain epistemic stance will lead to realism (or the opposite). For instance, the second stance that Chakravartty offers, and I cited, was: "Empiricist stances, which question whether theorizing beyond the observable phenomena should be a basis for warranted belief." Arguably, such a stance could also be framed in @Joshs's terms: We have no basis for believing that an alleged debate between realism and antirealism is even coherent. This reading places realism and antirealism "downstream" from the epistemic stance itself.
It requires a philosophical stance that doesnt axiomatically take the human situation as an end in and of itself, and so is not so solely beholden to the aims of instrumental reason. In other words philosophy that questions existence against a broader context. It must be concerned with what matters, Tillichs questions of ultimate concern. Of course one must not then be so vain as to believe that such an unconditional imperative be the subject of merely propositional knowledge (for the reasons Wittgenstein gives.) And that sounds rather like a belief, doesn't it?
As for realism and anti-realism, I'm generally an advocate for the latter, but I find the peremptory description of 'anti-realist' unsatisfactory. For me, it signifies a stance which recognises the unavoidable subjectivity of judgement, even in the most apparently objective of cases. (This was the main subject of discussion in the thread on Sebastian Rödl.) Anyway, an antirealist may be perfectly realistic in the pragmatic sense of observing conventions, obeying laws, and so on. An anti-realist doesn't necessarily think that s/he can leap from heights and not be affected by gravity. Antirealism simply points out the fact that scientific judgement is always reliant on conditions and exclusions (i.e. the selection of what exactly is the subject of analysis). Even the most universally-applicable of scientific principles pertain to a specific set of phenomena. So drawing conclusions from them about 'the nature of reality' in a more general or philosophical sense, is precisely where realism begins to blur into scientism.
Or, I'm guessing, a purely "rational/theoretical" basis à la Pincock. This is where the specter of strategic rationality becomes a bit frightening. I'm certainly not saying Pincock believes this, but you can't help being concerned if your conception of rationality turns out to be so rigid, so precise, that it could engender an epistemic stance capable of mediating between sample size 19 and 20, and labeling one as obligatory and the other as incorrect. Another way of putting the concern would be: Could there be an epistemic stance so powerfully grounded in rationality that we could predict everything it would and wouldn't countenance "downstream," no matter the feedback? Kinda like the One True Dialectic -- it's obligatory to adopt it, and all subsequent beliefs are also obligatory.
Quoting fdrake
OK, good. I just didn't read/think about this carefully enough, sorry.
Quoting fdrake
That's right. This is what happens when reason-giving is understood as rigidly inferential. So, part of what we want from a rationalist account is a way to either get out of this nightmare, or show why it doesn't matter. To what else must we appeal besides inference?
Quoting fdrake
And here's an answer to my anguished cry :smile:: EDIT: [Should we] talk about causes rather than reasons?
Quoting fdrake
Your use of scare-quotes around "true reason" says it all: Are we willing to accept a cause or a narrative as a reason? It would not be a theoretical reason, as Pincock understands one. And here the question of level is really critical. If you tell me your belief in ghosts is caused by growing up around people who believed in ghosts, I'll say "Thanks much" and completely ignore this as a reason for me to believe in ghosts. From a rationalist perspective, a reason is supposed to be "for everybody." Chakravartty and Pincock both discuss this, and as you'd imagine, Chakravartty believes some reasons can be valid for you but not for me, while Pincock thinks this is loose talk, and that a "true reason" asks for universal consent.
But suppose the explanandum is "Why I have adopted epistemic stance X." Can we opt out at the very beginning of the endless-justification-by-inference loop, by replying with a cause? That would be, in Pincock's terms, a "practical reason." He doesn't think it's enough, because it's not "appropriately connected to the truth." We'll see more of this in Part Two. For now, I think this point you raise about what counts as a "true reason" lies at the heart of the debate. The answer will affect everything, from judgments about upstream-downstream relations, to issues about obligation or voluntarism.
Quoting Wayfarer
Weirdly, this is what I was writing about to @fdrake when your post appeared. In that context, we looked at the idea of "cause" as a way out of strategic (instrumental) reason's circles. What you're talking about, if I understand you, is yet a different form of escape, if it can be found.
Quoting Wayfarer
Just to clarify -- "belief" in this discussion has been used to refer to something like JTB, that is, a warranted belief that would result from propositional knowledge. I think you're using "belief" to mean the opposite, or near-opposite -- something that is held on unconditional grounds that connect to a sense of being that is non-propositional. Neither is right or wrong, of course, I just want to avoid talking at cross-purposes.
Quoting Wayfarer
I think your refinement of it is fine, and we should ask ourselves: Using this understanding of antirealism, would Pincock still be a realist? I believe so. Even a generous and sensible view of how antirealism works in practice (or a commitment to an inevitable Rodelian subjectivity) is probably not going to sway a non-voluntarist about stances.
In his paper, Christopher Pincock gives two arguments that he says demonstrate that someone who adopts a realist epistemic stance must do so, on pain of incoherence.
Quoting Pincock, 7
Ill try to lay out the first argument in this Part Two, while pointing out that Pincock himself does a very good job, and its worth reading this in its entirety rather than my paraphrase.
First, a preliminary point which Ive raised before. We can say that, for Pincock, the term theoretical reasons is essentially equivalent to reasons that are rationally obligatory. And we know that Chakravartty objects to this characterization, claiming that it begs the question against the very idea of stance voluntarism: On the voluntarist view, rational choice and rational obligation are distinct concepts and cannot be run together.
So lets keep this dispute in mind, in what follows.
Pincocks first argument is a reductio designed to show that, if an epistemic realist takes the position that their stance is not obligatory, they will arrive at pragmatic incoherence. Therefore, they should claim that realism is obligatory.
To set this up, Pincock describes a typical instance of inference to the best explanation (IBE), which characterizes the kind of inference he believes realists can and should draw. This involves an account of Benjamin Franklins famous experiment with kite, key, and lightning. The realist will say that Franklin concluded that lightning is an electrical discharge (call this L) because he had a reason for so concluding. He needed to understand that his evidence is evidence for L. He needed to grasp a principle of inference that can link his evidence with L.
Pincock distinguishes Franklins epistemic stance from another one that Franklin might have taken. He might have said, Im disposed to claim to know L when I have this kind of evidence. Its just what I do, or what seems best to me; others may do differently. For Pincock, this wouldnt give Franklin reasons for his claim that L. Pincock asks us to imagine how this non-theoretical Franklin would respond to a challenge to his claim about L: He has nothing at his disposal that would count as a reason for others to adopt, so he would have to be silent in the face of his challengers. The actual Franklin, though, scientific realist that he was, can reply with an account that involves how evidence is connected to knowledge claims. This account will not necessarily carry the day, nor will it have to result in an indisputable knowledge claim, but it does consist of alleged reasons for beliefs that are meant to be convincing for anyone, not just statements about how I proceed when I see X and Y.
Now, here is the core of Pincocks incoherence argument, in which he asks us to imagine a realist who does not believe that their epistemic stance is obligatory:
There are a number of issues raised by this argument, which Pincock discusses carefully. One, inspired by Bernard Williams, concerns whether one can actually acquire a belief without believing it to be true. Another concerns whether such a question applies only at the level of doxastic belief, not the choice of epistemic stance.
But Ill cut to the chase and say that I think the argument as a whole can be defeated simply by denying the characterization of what a stance voluntarist does. Pincocks language includes phrases such as no reason that obliges them, not adopt[ing] their realist stance on the basis of any reasons that reflect the truth, no connection to the truth, and not appropriately connected to the truth. These all-or-nothing characterizations can only hold water if we accept Pincocks idea that a theoretical reason must result in rational obligation. (I should point out that the first phrase, no reason that obliges them, would be conceded by Chakravartty. But he would not concede that there are no theoretical reasons that could have a bearing, or influence the decision merely that they dont result in rational obligation, and that others could have different reasons for their stances, or weight them differently.)
As we know, Pincock maintains that the stance voluntarist has no theoretical reasons of any sort for their adoption of a stance. For Pincock, only desires and values can form the basis for (voluntarily) adopting a stance. Once again, if we look back at Chakravarttys description of how he understands an epistemic stance, this seems to be a misreading:
Or perhaps its not so much a misreading as an interpretation which claims that, if all theoretical reasons create obligations, then everything on Chakravarttys list has to be something else, since Chakravartty is claiming to be a stance voluntarist. It may also be a sort of challenge: If this list is not merely disguised desires and values, then tell us directly what the theoretical element is. What other commitments do you have in mind?
In any case, Chakravarttys response to the incoherence argument is straightforward:
The epistemic realist, of course, wants to say that this argument applies only against other stances; there is something unique about the stance supported by strictly theoretical reasons. Chakravartty says that the only way this could be made compelling is by accepting the conflation of rational choice with rational obligation, which, as weve noted, seems to beg the question against the voluntarist:
And Chakravartty points out what weve alluded to several times now: a non-question-begging argument would have to start with an understanding of rationality that precludes alternative rational standards completely:
In fact, Pincock does offer such an argument, which Ill look at in Part Three. But for now, do you agree that the pragmatic-incoherence argument requires this tall order if its going to go through in a non-question-begging way? I believe Chakravartty is right about this.
I'm going on my own hobby horse here, rather than trying to do any exegesis.
I think generally people accept narratives as reasons. Since they serve as explanations. eg He took a cookie from the jar and ate it, and he ate it since he was hungry. That's a story, it makes sense.
I think people only treat causes as reasons when the causes serve as part of a story. Compare:
The man had a feeling.
He went to the cookie jar.
To
The man was hungry.
He went to the cookie jar.
Even if the sensation in the first story was hunger, you'd only make sense of the story by inferring that the sensation was food related on the basis of the second sentence, which paradigmatically is hunger. Nothing in the story tells you that it is hungry, or food related, except the context of the phrase "the cookie jar" bleeding over into the first sentence.
Contrast to:
The man had a feeling.
He stood up.
Much more ambiguous. You might read that in terms of determinedness, or needing to pee. Absolutely nothing there. The nature of the cause needs to be understood in the context of the event it caused in order to serve as an explanation for it.
I've had a short story in my head for a while now, but I've not figured out a way of writing it. It concerns an argument between two partners that leads to their immediate break up, but they were very happy up until the argument. One partner's shoes had started rubbing their feet that morning. The other partner had put on a tiny amount of weight and felt it in their belt buckle. That lead to them both being irritated all day, which snowballed. The distal cause of their break up is totally and pointlessly irrelevant to their lives. I've found it quite difficult to plan, since the raw contingency of their uncomfortable clothing on that day gets seen as incidental, and how they react to the world around and each other gets given the locus of responsibility due to how it has to be described. If they have an argument, even if they're only saying what they're saying distally because of uncomfortable clothes, we go proximately for character traits. They both look like arses, or it looks like a failed relationship. It's difficult to make something a story beat if it resists any sense of narrative.
I think "universal consent" is a good way of putting it. "Universal comprehension" is also a component of it. people have to understand what they're consenting to. Even ff people agreed that, like in the second story, the man's feeling made him stand up, it serves as a reason for him standing up only by narrative juxtaposition/co-contextualising the feeling and standing up. And that's something I did in writing the sentences, not the hypothetical man standing up. And certainly not the partners with their uncomfortable clothing.
The reason I'm favouring this termination of explanations into causes is that correctly noticing causes doesn't have to make much sense. You don't need the content of a cause to be reasonable, or even explicable, just to notice that it really is a cause.
Which perhaps makes a meta-methodological commitment to believing things that seem to be true regardless of why, but if that's not axiomatically posited as part of a reasonable stance, what's the point in calling some stances reasonable and some not.
It would probably matter whether either partner was aware of the irrelevant distal cause. Matter in terms of how to handle it as a story, that is. If you're within either or both points of view, and if your PoV doesn't include the requisite awareness, then yes, it's a narrative challenge. You'd have to find a way to show the reader what's really going on, while keeping the characters unaware of it. But this can be done. It's a twist on the "unreliable narrator/protagonist" idea. Or of course you could allow yourself an omniscient authorial viewpoint and simply inform the reader what's going on.
In any case, I completely agree that, in our non-philosophical lives, we accept stories as reasons, when they have the appropriate narrative explanatory form. We accept physical causes too, and unconscious motives as well. The peculiarly "rational" reason (a reason for everybody?) seems to come into play when we try to justify beliefs, rather than explain actions. And there's the middle ground of explaining beliefs: Your (narrative) reason for believing in ghosts gives an explanation for the belief, in one sense, but not in the sense of "justifying a true belief." That would require the other kind of reason, which is public and meant to be persuasive.
So terminating at causes is fine, as long as we're not pointing to a cause as the source of a JTB. Problem is, that seems to be what Pincock (and perhaps any rationalist) demands. The inferential loop seems to go on and on, as long as we insist on that special variety of "true reason." And we can't simply say, "Well, then stop at cause here as well," because we can see why it won't explain a JTB, except perhaps coincidentally.
Quoting fdrake
I'm stuck a bit on this. Are you saying that the man, if we asked him why he stood up, would deny the feeling as a cause, or say he wasn't aware of it? Yes, you wrote it that way, but he could agree, couldn't he? Or perhaps this goes back to the "unreliable narrator" question. Let's change the example back to "loses his temper at his partner." And let's say his therapist, who knows him well, realizes that this occurred because the guy got triggered by a certain phrase that connected to a childhood trauma. But the man himself might not know this, and might give a completely different reason. So the therapist is somewhat in the role of you, the author: They know something about this character that the character does not. Is it, then, a reason? Different intuitions are possible here. I'd say it is a reason, and point to the many different ways we use that term. But for the man, as I think you're saying, it can't be a reason unless he goes beyond "narrative juxtaposition" and actually accepts the account.
Quoting fdrake
OK, but as above: You do need the content to be reasonable if we're working toward warranted belief.
Quoting fdrake
Sorry -- regardless of why we believe them, or regardless of why they're true?
I suppose there's a couple of types of content involved. If you established that X causes Y through an experiment, then that's an excellent justification for believing it. But that's far for explaining why X causes Y. So if someone were to say "X happened, that's why Y happened", and someone challenged it: "Why?", you could point to the experiment. But that doesn't tell you the mechanism, it doesn't explicate the why. It demonstrates it. The first type of content would be what suffices to demonstrate truth, the second type of content would be what serves as an explication. They both might work as reasons, but they don't both work as stories or explications, and only attempting to specify a mechanism would tell you why.
So I suppose what I'm saying is that the content of the claim doesn't need to make any kind of sense to serve as an excellent justification, it just needs to be established as true. And in context noting causes, without any further commitment to mechanism or generality, might serve as a terminus of giving reasons. Putting it in -isms, a kind of foundationalism which uses every passing contingency.
Quoting J
What I'm saying is that it doesn't matter whether he was aware of it or not, were it the true and only cause. Whether there was an explication beyond "It's true that his clothing caused the break up" would be irrelevant, even if it made no sense. That's what I was getting at. Extremely narratively unsatisfying explanations that amount to "stuff just happened this way, it was established thus".
Quoting J
Regardless of why we believe them if they are believed because they're true. Or just because they're true, regardless of why we believe them. Like the break up because of the uncomfortable clothes. True, utterly useless as an explication, and no one would believe it because it's not a cromulent story.
Quoting J
Yeah. I've invited the reader to juxtapose them, but the man would be utterly insane if he blamed his partner leaving him on his itchy underwear. Even if he's totally correct.
There is an unstated allegory lurking behind this example, or so it seems to me. Here the effect of an electrical impulse on the key is an allegory for scientific explanation in general which relies on reasoning to the best explanation. The best explanation for the particular observation in this case is the effect of lightning on the key. This is a very specific situation with an identifiable causal sequence. It seems to me that the alternative presented by the 'non-realist' Franklin would be more typical of a more general, or a less specific, type of problem. An example would be instrumentalism in atomic physics. As is well known, instrumentalism keeps shtum about what kind of entity is being measured by observation (wave or particle?) Consequently it doesn't offer a thesis about the ultimate nature of what is being observed, only that 'this kind of observation produces this measurement outcome'. In this context, the anti-realist attitude that 'we can't really account for why we get this outcome' is quite reasonable. You could say that it leaves the question open (which is also a commendable scientific attitude in my view). By conflating pragmatic coherence with rational obligation, Pincock oversimplifies the range of legitimate epistemic responses. Instrumentalism, for instance, operates within a perfectly coherent rational framework yet explicitly avoids metaphysical commitmentsa stance that clearly avoids the "pragmatic incoherence" Pincock accuses voluntarists of.
I dont get why people even argue about this kinda shit... "because these epistemic receptors were triggered its obviously gotta be sameness of stimulation between people even though they don't share a homology of nerve endings between the two... to the point its oblogatory for everyone because we all share the same neurons and thought processes..."
Sorry but we're not all one normal person who thinks all the same way...who are exact replicas of the next with the same genetic code and make up and same neurons...
Posterior Analytics 1.2 is the big discussion. It shows up in other parts of the Organon more tangentially. From the SEP article on "Aristotle's Logic:"
But Aristotle reasons:
If the skeptic is right, discursive knowledge is impossible.
But discursive knowledge is possible.
Therefore the skeptic is wrong.
(Note, if the skeptic rebuts this claim, they cannot possibly claim to [I]know[/I] their own rebuttal's truth without contradicting themselves).
Wherefore Aristotle launches into his discussion of noesis, intellectual consideration, although sense knowledge is important here too (Aristotle recognizes many more "types of knowledge" than most other philosophers, but even Plato has 4-5, versus often just 1-2 today). But the case for noesis isn't fully made in the logical works, since it is supported by the psychology, epistemology, and metaphysics of De Anima, the Physics, and Metaphysics, which amounts to more of a "metaphysics of knowledge" than a modern epistemology, in that it doesn't start from skepticism in the way most modern epistemology does.
The recognition of the crucial role of intellectus as opposed to solely ratio (discursive computation-like justification), is, in part, what leads to the dominance of "virtue epistemology" for much of philosophical history, until Descartes introduces the alluring idea of "tearing everything down" and then "building it back up with the perfect method."
That makes sense. The dominant form of "realism" has become loaded with a host of metaphysical assumptions, making the distinction somewhat fraught. So, for instance something like John Deely's semiotics, based in the tripartite semiotics of the Doctrina Signorum, is "realist" but does not hold to the axiomatic assumption of much realism, that consciousness is a contingent, accidental [I]representation[/I] of a sort of "bare noumena" reality. Signs, meaning, are present virtually in things. And while it would be hard to claim that the Thomist position is "anti-realist" it is also not "realist" in the sense of ontological truth lying outside Intellect. I suppose Hegel would be another example where the suppositions "realism" is often loaded with is fraught.
Sure, different philosophers have different interests, and worry about different things. I understand this isn't the thread for you.
Yes, this "circle problem" is very much in the spirit of the voluntarism/obligation debate. One question: by "demonstration," do you take these thinkers to mean inferential or logical proof of validity? Presumably, at any given point we enter the circle, we're equipped with premises that don't (at that moment) require demonstration, so does "demonstration" describe a "downstream" process (to use the language we're developing on this thread)? Starting from premises and reasoning to conclusions?
Quoting Count Timothy von Icarus
Ummm . . . but you don't think this works, do you? How does it not simply beg the question?
Quoting Count Timothy von Icarus
That would depend on whether the skeptic believes that it's discursive knowledge itself which is leading them to conclude that discursive knowledge is impossible. I have little sympathy myself for radical skepticism, but in fairness I think the skeptic can rebut the claim without also needing to claim that the rebuttal is discursive knowledge. Or, if "rebut" is too strong, let's say "show the claim to be highly implausible."
I understand that noesis and intellectus are meant to come to the rescue here, as in so many other places where Greek thought is contrasted with 20th century emphases on strictly (and literally) "rational" thought. Making this rescue attempt attractive is hard, even though I suspect it's correct in some fundamental ways. I'm reminded of this, above:
Quoting Wayfarer
Would you agree this is similar to the desideratum you (and perhaps Aristotle) seek?
I agree with your overall point about the Franklin example being something of an outlier, but I believe a realist could acknowledge this without putting their stance into question. But of course Pincock has chosen the Franklin example with a purpose in mind, and could be accused of picking some low-hanging fruit, I suppose. Not every IBE works out so tidily.
Quoting Wayfarer
I think so. Once again, he'd have to show that instrumentalism can't be understood as "avoiding metaphysical commitments," and that these commitments are at odds with theoretical reasons as he defines the term. As Chakravartty says, it's a tall order, but we haven't seen anything so far that indicates it's impossible. And Pincock does offer such an argument (not about instrumentalism specifically), which I'll discuss later.
I'm sure if you looked at it from the perspective of the values that might cause someone to adopt a particular stance, you'd find some commonality with your interests.
A "demonstration" would generally be a syllogism in this context, although obviously there is a sense in which demonstrations can be less formal.
Points 1-6 are a discursive demonstration. The skeptic is claiming to have demonstrated that discursive knowledge through demonstration is impossible through the use of discursive demonstration. Hence, I don't think it's question begging, it simply shows:
A. The skeptics' argument is self-refuting (they are claiming to be able to demonstrate the insufficiency of demonstration).
B. The conclusion that discursive knowledge is impossible is absurd. But when one reaches an absurd conclusion the first step should be to check one's argument for validity and one's premises for truth. This intuition is often ignored today, and instead we follow out the absurdity and try to build whole "systems" atop it (e.g. instead of questioning Hume's premises, we accept a standing "problem of induction.")
C. It is also the case that the skeptics' argument flows from premises that are less well known (the nature of logic and knowledge) to a conclusion that seems very well known, our capacity for discursive learning and for the productive arts/techne that rely on it.
Perhaps, but Aristotle is responding to those who are claiming to demonstrate their claims through discursive reason. Yet if someone claims to have a noetic intuition that logic cannot yield knowledge or truth, I'd see little reason to believe them.
One cannot justify reason and argument through reason and argument in a non-circular manner. Hence, the misologist can never be [I] refuted[/I] without in some sense begging the question and presupposing the proper [I]authority[/I] of logos. Reason is defenseless. Likewise, proof, demonstration, argument, etc. must [I]presuppose[/I] at least some inference rules (e.g. the laws of thought) to even get off the ground. One cannot justify all of one's inference rules without presupposing at least some. As Gadamer says, prejudices are a prerequisite for inquiry.
I don't think we need a "rescue attempt," but even were this so, it would be attractive at least in principle because the main alternatives seem to be dressed up versions of logical and epistemic nihilism (sometimes in democratized formats), and a radical divorce from reality.
However, "noetic position" is normally badly strawmanned as the hand-waving claim that "some things [I]we just know[/I]," with little further investigation. However, I think there is a strong practical, psychological, and effective argument that can be made for the authority of logos that is developed through Plato's psychology and notions of freedom, and which reaches (IMHO) its greatest formulation in the Patristics, Desert Fathers, and the later Eastern ascetic tradition. Whereas there is also a physical and metaphysical justification that runs through Aristotle up through today (since obviously it is informed by advances in the natural sciences, etc.). Of course, acceptance of such discourses, their "attractiveness," will itself be contingent on the acceptance of the authority of logos, leading back us to the original dispute.
What can be said? "There is no arguing with a misologe" seems like a truism. One cannot expect to convince someone who denies the authority of reason and argument through reason and argument. But this is perhaps precisely why the neglect of the affective/emotional and practical path that runs through Plato and the Patristics, and the neglect of "epistemic virtue" (regulative and faculty) could stand to be remedied. The myopic focus on method presupposes that method can justify itself.
Related, Mark Burgess has a pretty good dissertation on the sorted history of critiques of noesis (what he calls "transcendental apriorism"), and at the very least it's true that engagement with it has generally been quite facile. Actually, it's a great reminder of how "great names" can accidentally poison discourse for generations, e.g. where you have Kant dominating views of a tradition he did not understand, Nietzsche dominating reception of an ascetic tradition he never seriously studied, etc.
The distinction between a justification and an explanation is excellent. I agree with everything you're saying here except whether justification alone can serve as a terminus of giving reasons. Or let me rephrase that: Certainly it does serve; long before humans knew anything about planetary motion, they were absolutely justified in believed that the "sun will rise" tomorrow; they just didn't know why. But . . . is that the same thing as "established as true"? Comes down to usage now, I suppose. We're used to thinking of establishing the truth of something as being able to explain not just our belief in the phenomenon, but how the phenomenon comes to be so constituted as to produce the regularities that result in our true belief. But, as you point out, is that really required? If we say "No," then we also seem impelled to say that scientists have no reason other than curiosity to motivate them to discover explanations for otherwise obviously true phenomena. That feels wrong, but I'll have to think more about it.
Quoting fdrake
This makes me wonder whether the uncomfortable-clothes explanation could be true. I suppose it depends how you phrase it. "They broke up because they had an argument, largely caused by how each was feeling physically" seems believable enough. "She left him because of his itchy underwear" is surely inadequate . . . and is it even true? A bizarre version of the butterfly effect, which is also questionable as a good explanation of anything.
I suppose it would depend on if the nearest possible world without the underwear itchiness was also a world in which they were still together. Not that I like possible worlds much. It just seems also not to care about narrative and explication in the same was as causes don't.
No, Aristotle has to say that 1-6 purport to be, but are not, a discursive demonstration. Which upsets the whole apple-cart.
You're reading Aristotle's reasoning as follows:
If the skeptic is right, discursive knowledge is impossible.
But discursive knowledge is possible, because the skeptic has just engaged in it.
Therefore the skeptic [is wrong] has said something incoherent.
I'd respectfully suggest, then, that the argument needs to be expanded with an additional step:
If the skeptic is right, discursive knowledge is impossible.
The skeptic, in discussing this matter, has presented us with a piece of discursive knowledge.
So we see before us an example of discursive knowledge.
Therefore the skeptic has said something incoherent.
And this, I'm afraid, changes a lot. You or Aristotle can no longer avail yourselves of premise 2. You can't simultaneously say that what the skeptic has given us is discursive knowledge, while also denying the truth of what the skeptic says. All you can do is point to the incoherence of the entire chain of thought -- the usual liar's-paradox problems -- but we already knew this, that's precisely what we're trying to find a way out of.
Now if Aristotle could point to some other piece of discursive knowledge that was somehow self-credentialing, that would be different. But of course "self-credentialing" is so close to "question-begging" that I don't know if even that would advance us.
A theoretical win may not be able to be scalable for several reasons which might take a HOW-to approach. Quine details that we can't ever know WHY a something failed in an experiment that should technically work in theory.
Doesn't that amount to demanding that the absurd premise in a reductio be true in order for a reductio to be successful?
Considering this . . . I think you can make Aristotles argument go through if you drop the premise The skeptic has presented a piece of discursive knowledge (or Points 1-6 are a discursive demonstration):
1. The skeptic, in discussing this matter, claims to have produced a piece of discursive knowledge.
2. That piece of allegedly discursive knowledge purports to show that discursive knowledge is impossible.
3. Therefore the skeptic has said something incoherent or self-contradictory.
In other words, make it about what the skeptic says, not what theyve actually done. In fact, I think you have to deny that the skeptic has done what they claim, i.e., present a piece of discursive knowledge. The whole question of discursive knowledge itself becomes weirdly beside the point; the skeptic is wrong because the form of the argument is wrong, not because there is or isnt such a thing as discursive knowledge.
So this would not be a powerful enough conclusion to show that discursive knowledge is possible (one of the original premises of the argument as you gave it). In this version there is no longer a piece of discursive knowledge to point to. So perhaps this doesnt get you (or Aristotle) where youd like to go.
Yes. @J does not understand how a reductio works. As I've pointed out before, a reductio does not prove falsity per se. Instead it proves inconsistency or incoherence.
Nevertheless:
Quoting Count Timothy von Icarus
I don't see Aristotle doing that according to the SEP article you cite {supposing this is meant to represent a reductio}. Indeed, I think Aristotle would see this as a sub-optimal response to the "agnostics," one which fails to address the better part of their challenge. More precisely:
Quoting Count Timothy von Icarus
In Aristotle's precise sense a demonstration is not merely a syllogism, and the "agnostic" is not giving a demonstration. Therefore they do not fall afoul of the reductio. For Aristotle a mere syllogism or argument has "unsecured" premises, so to speak, whereas a demonstration has premises which are first principles (and are therefore securely known in one way or anotherthe way is here the point at issue).* This is why a demonstration produces scientia proper - the highest form of knowledge. So when the "agnostic" says, "Therefore, nothing can be demonstrated," they are not excluding forms of knowledge which are weaker than scientia, and their conclusion would be one such example. The Pyrrhonists are doing a similar thing when they distinguish acceptable from inacceptable qualities of certitude.
Note too that it would be very rare for someone like Aristotle or Aquinas to place an argument in their opponent's mouth which is susceptible to a reductio. With rare exceptions, that is bad form, intellectually speaking.
It's worth noting that a quasi-skeptic like @J could probably find allies among the "agnostics" or the Pyrrhonists.
* More precisely, a demonstration is a syllogism or argument whose premises are first principles (and are therefore "securely" known). Thus syllogisms and arguments are not necessarily demonstrations.
Quoting J
Let's construe Pincock's argument as saying that, "Chakravartty has no reason to adopt one stance rather than another, when choosing among the subset of stances which are rational." This looks to be the most charitable interpretation, and it precludes the response that, "Choosing one stance involves 'rational choice' because one can produce reasons in favor of that stance."
Suppose all possible stances are represented by the set {A, B, C, ..., X, Y, Z}. And suppose that Chakravartty's set of "rationally permissible" stances is {A, B, C, D} (and therefore 4/26 stances are rationally permissible). Given this, my construal of Pincock's argument pertains to "choosing among the subset of stances which are rational," i.e. {A, B, C, D}. Chakravartty can say that he has a reason to adopt C rather than F, and that he has a reason to adopt C simpliciter, but he apparently cannot say that he has a reason to adopt C rather than D (which is what he needs to say if he is to properly answer Pincock).
This way of construing Pincock's argument has much to recommend it, given that it is in line with what is traditionally understood as "voluntarism." Namely, voluntarism posits that the choice in question is traced to the will rather than the intellect, such that one might explain their choice by saying, "I did it because I wanted to, not because I was rationally guided to do so." *
I think Chakravartty tries and fails to address this difficulty in section 3. We can boil it down with the dichotomy, "Either you have a reason for your choice or you don't" (where the voluntaristic answer that "I wanted to" does not count as a reason). Does Chakravartty have a principled (all-things-considered) reason to choose C and reject D? Apparently he can't have a principled reason, because if he did then D would not be "rationally permissible" (for him). The whole rationale for voluntarismincluding stance voluntarismis that the subset of rationally permissible stances ({A, B, C, D}) are equally rational, and are therefore immune to rational predilection. Voluntarism entails that a decision between C and D is not rationally adjudicable.
This constitutes an internal problem for Chakravartty, because at the end of his paper he assumes he is still entitled to the general idea that we should "encourage others... to see things our way":
This is a nice moral sentiment, but it isn't rationally coherent. If the voluntarist claims that the subset of rationally permissible stances are not rationally adjudicable, then he is not rationally permitted to "encourage others" to drop their D and adopt his C, given that there are, by definition, no compelling reasons to choose C over D. He must restrict his stance-disagreements to those interlocutors who hold to one of the 22 stances which are not rationally permissible.
If Chakravartty wants to coherently "encourage others to see things his way," then he must reject his own voluntarism. He doesn't need to be an ass about it, but he must hold that, "My epistemic stance is more rational than yours." If he doesn't hold that then he has no grounds to try to convince his interlocutor to reject D and adopt C. If he is a true voluntarist then he would not argue against the stance of someone who holds to one of the four rationally permissible stances.
* Note that voluntarism signifies choice or will, but if the "values" that Pincock characterizes are inherited rather than chosen then everything I say here still follows. Any non-rational predilection for C will result in the same problem, whether that predilection is based on will, inheritance, or anything else. As long as Chakravartty cannot hold to A, B, C, and D all at the same time, he will be forced to possess one rationally permissible stance rather than another, yet without having a reason that counts as a worthy reason to choose among that subset of stances. Thus Pincock's point about the realist will also apply to Chakravartty himself, who sees himself to hold C rather than A, B, or D, for no good reason at all. This creates a deep incoherence between the non-rational stance and the "rational" effects that flow out of it. @fdrake is correct to note that the stance cannot be cordoned off in this way. In real life when someone notices that they have no good reasons to hold C, they simply stop holding it and end up trying to hold to the four rationally permissible stances equally.
(For the record, I find both authors to be rather confused, especially Pincock. So I'm not throwing in with Pincock. Pincock is using "rational obligation" in a softer sense than Chakravartty recognizes, but given that Pincock is clear about his usage the misunderstanding is on Chakravartty (unless the draft Chakravartty read was substantially different than the published paper). If Chakravartty thinks he possesses some coherent distinction between 'rational choice' and 'rational obligation', then the onus is certainly on him to make that distinction clear. It seems to me that he relies heavily on ambiguous and undefined terms, including "rational obligation.")
Good post. The vagueness of a "stance" strikes me as a big problem, and this point about cordoning stances off from their downstream "effects" is a good example of that.
I would prefer Aristotle's Rhetoric or Newman's Grammar of Assent. In the Rhetoric Aristotle talks about "enthymeme," by which he means a "shooting from the hip" sort of argument (as one would be likely to hear when a politician tries to make a point given a very short bit of time). That sort of argument can hit or miss depending on the background conditions of one's hearers. Even Pincock's abductive reasoning would be a form of "enthymeme" for Aristotle.
The trouble with "enthymeme" is that it is a kind of per accidens argument. It is like tossing a hand grenade into the fray and hoping you hit someone. For this reason the phenomena surrounding that sort of argument isn't scientifically precise or predictable. Chakravartty can only pretend that a study of that sort of argumentation is scientific by talking about "stances" and pretending that he has some relatively precise notion of what he means by a "stance." He almost certainly does not. This tends to make his thesis vacuous, like the certitude that neither Alice nor Bob are irrational in their preferred sample size.
The issue from that other thread is catalogued and addressed more completely in . An excerpt:
Quoting Leontiskos
Chakravartty is claiming that we should "encourage others... to see things our way" even if we both hold to one of the four non-adjudicable, rationally permissible stances. He says that if I claim that my stance is better than yours, this "adds nothing of rhetorical or persuasive power." Obviously he is wrong, given that moving from
Chakravartty's work is helpful insofar as it codifies the incoherence of "pluralism" into a clearer position. Pincock's opposition is lackluster at times, and Chakravartty misrepresents him on things like "rational obligation," but I think Chakravartty's attempt makes it easier to see the incoherence of "pluralism." He is trying to give the clear position which no one on TPF is willing to offer. Once we steelman Pincock the rest is easy enough:
Quoting Leontiskos
The moral of the story is that if someone takes up Chakravartty's stance voluntarism, then they must give up their ability to "encourage others... to see things our way." By definition, the stance voluntarist has no reasons for why someone should "see things his way." More advanced ages could see this fact in the blink of an eye.