By In Ethics Discussions at PEA Soup, Normative Ethics Comments (8)

Ethics Discussion at PEA Soup: Robert Cowan’s “Rossian Conceptual Intuitionism,” with a critical précis by Philip Stratton-Lake

Welcome to what we expect will be a very interesting and productive discussion of Robert Cowan‘s “Rossian Conceptual Intuitionism.” The article was published in the most recent issue of Ethics and is available through open access here. Philip Stratton-Lake has kindly agreed to contribute a critical précis, and it appears immediately below. Please join in the discussion!

Chike Jeffers

In his excellent paper Robert Cowan offers an alternative to a perceptualist version of Rossian intuitionism which he calls Rossian Conceptualist Intuitionism (‘conceptualism’ for short). Perceptualism, as Cowan understands it, is the view that

an occurrent adequate understanding of a self-evident proposition, p, crucially involves standing in a noncausal, nonsensory, but perceptual-like relation to the concepts (which on this view are abstract entities, e.g., Universals) figuring in p, for example, an ‘apprehension’ or ‘acquaintance’ (826).

Although he does not explicitly distinguish them, Cowan lists three problems with perceptualism:

  1. It “requires an extravagant philosophy of mind, namely, a capacity for, or faculty of, nonsensory awareness of abstracta,” and this “jars with the claims that (at least some) contemporary Intuitionists make to the effect that the view only requires modest commitments” (827).
  2. It is committed to “the idea that nonsensory awareness makes us aware of a domain of abstracta” (828)
  3. Perceptualism requires that we have a sort of direct access to non-natural objects” (828)

2 is the objective correlate of 1, as the intuitions that figure in the “extravagant philosophy of mind” are supposed to put us in touch with a distinct domain populated with abstract objects. The existence of such a domain is a controversial metaphysics rather than an extravagant philosophy of mind, and is optional for an intuitionist. For all intuitions do is present some proposition as true. It is an open question how such presentation is to be understood or whether something else is presented rather than the truth of a proposition. The same is true of 3, as the claim is that perceptualism commits us to non-natural properties, and whether there are non-natural properties is not a matter of philosophy of mind.

I’ll make a few quick comments on these problems for perceptualism before moving on to Cowan’s positive thesis. First it is not clear to me that the idea of a distinctive mental state – an intellectual seeming – by itself is that extravagant. By itself it is just the view that there is a mental state that cannot easily be reduced to a belief or judgement. Certain propositions do present themselves to the mind as true when we consider them, and they share many phenomenological features with perceptual seemings, such as persisting even when we know that things are not as they seem.

It is only once one tacks on the idea that these states are a direct awareness of abstracta that an air of spookiness may appear. But that is the second objection not the first, and it is not clear that a non-conceptualist intuitionist is committed to this (although Cowan builds it into the definition of perceptualism). One may think, as Richard Price did, that what intuition puts us in touch with is a particular instance of a moral property – that is a particular way a particular thing or event is. Seeing the distinctive way in which some particular is, Price thought, gives us moral ideas (his term for moral concepts). Ross thought that we have the ability to move from the apprehension of a particular prima facie wrong acts to universal principles, but this ability is not obviously weird. It involves simply an ability to see that the prima facie wrongness of some particular act depends solely on its nature, and that need only involve considering particular cases carefully.

The abstracta the non-conceptualist Rossian is supposed to be commited to are concepts, but on the version of perceptualism defended by Richard Price we acquire these concepts through apprehension of concrete instances, rather than apprehension of abstract objects. We do not even need direct apprehension of the non-natural properties themselves; just their instances, plus an ability to see resemblance amongst instances.

Even if non-conceptualist intuitionists are committed to abstracta, that would not be a huge burden for them. Intuitionists are already committed to non-natural properties, so if they had to allow non-natural objects (abstracta) into their ontology, that would not be terrible news. My own view is that we should allow as many types of things/properties as is needed to offer the best account of how things are.

So I don’t think non-conceptualist Rossians are committed to an extravagant philosophy of mind. (Perceptualism is, but that is just because that is stipulated in the definition of perceptualism.) There is to my mind nothing extravagant about intellectual seemings. Extravagance only seems to come if we go on to posit a world of abstracta to which these seemings are connected. But a) more needs to be said to show that non-conceptualists are committed to this, and b) even if they were they would not be letting a new type of thing into their ontology, as they already buy into the non-natural.

Anyway, the main point I wanted to make here is that the case has not been made that intuitionists have to choose between conceptualism and perceptualism as Cowan defines these. So even if conceptualism has some advantage over perceptualism, it is not clear that that makes it the best theory for intuitionists. Some other form of non-conceptualist intuitionism may be better.

Let me move on to Cowan’s conceptualist alternative to perceptualism. Rossian intuitionism claims that a certain list of moral principles are self-evident. On the account of self-evidence Cowan favours, this means that they can be known solely on the basis of an understanding of them. Intuitionists thus need an account of understanding that enables such knowledge.

Just as an aside, my own view is that it was a mistake to think that adequate understanding provides evidence or justification for belief in self-evident propositions, and so to define self-evidence in a way that relies on that claim. Intuitionists already have intuition as a source of non-inferential justification. They do not need to add understanding as a separate source, and I think it was a mistake to do so. If we reject the idea that understanding is a source of justification, the pressure to come up with an account of understanding that enables it to play that role diminishes.

Cowan argues that Peacocke offers such an account. According to this view concepts are individuated by their possession conditions, which involve a tacit grip of the relevant first order normative theory.

A Rossian Conceptualist thinks that the implicit conception which individuates the concept moral reason encodes the Rossian Principles. Individuals who possess that concept are in possession of informational content such that their judgmental and inferential dispositions reflect a tacit commitment to the Principles. (829)

Given this account of understanding the concept MORAL REASON it is no mystery how someone can know the principles that are included in it. “When one comes to believe on the “basis” of adequate understanding, one is simply drawing— in some sense—on the informational content associated with the implicit conception.” (830) Rossian principles are built into that very understanding.

For example, the implicit conception associated with the concept MORAL REASON can be partially cashed out in terms of the Rossian Promissory Principle and the semantic value of MORAL REASON and, e.g., PROMISE-KEEPING, is such that the principle there is always an overridable but ineradicable moral reason to keep promises that one has made comes out as true. (831)

Actually Cowan has doubts about whether that is true, as he claims that there are insufficient data to support the Rossian list. So his Rossian conceptualist intuitionism may turn out not to be very Rossian. But that doesn’t matter, as he has a schema for knowing whatever principles do fall under that concept if it turns out they are not Rossian ones.

Cowan claims that this account has a prima facie advantage over perceptualism as it does not require knowledge of a third realm. “Subjects with adequate understanding simply possess/or have articulated an implicit conception that corresponds to the implicit conception constitutive of possession of the relevant concept” (831).

I have two concerns about conceptualism. The first relates to how it deals with moral disagreement, say between Rossian intuitionists and consequentialists. It seems there are only two options, neither of which is satisfactory. The first option is to say that both sides of the debate have adequate grasp of the concepts they use, but that they have different concepts with different application conditions and a different conception built into each – one Rossian, the other consequentialist. But that would have the consequence that they are simply talking past each other rather than really disagreeing, say, about whether it is right to keep a promise when doing so doesn’t maximise the good. They are talking past each other because the conception that determines the application conditions built into the concept of a moral reason each of them uses is very different. If that is correct then conceptualism can’t capture simple disagreement between consequentialists and deontologists.

The second option is to say that they really are deploying the same concept of a moral reason (let us say it is the one with the Rossian principles built into it). So there is genuine disagreement. But the problem with this solution is that we will have to say that the consequentialist’s disagreement stems from the fact that they lack an understanding of the concepts they are using – that really they are rather confused. I disagree with consequentialists, but I do not think they are confused, or lack a grip of the concepts they are using.

So it seems to me that conceptualism cannot adequately capture this sort of disagreement. It is either no disagreement at all, or makes a very implausible assumption about the one side of the debate. This debate is a substantive one, and that needs to be retained.

That brings me on to my second worry. It is a sort of inverted substantivity objection (ISO). The substantivity objection (SO) that Cowan lays out is as follows:

P1. The Rossian Principles are substantive propositions.

P2. If a proposition is substantive, then it is not self-evident.

C. The Rossian principles are not self-evident.(832)

Cowan seems to follow Väyrynen in identifying the substantive/non-substantive distinction with the analytic/conceptual synthetic/non-conceptual distinction (833). So this objection assumes that only analytic/conceptual propositions can be self-evident, and Cowan seems to agree. He gets around SO by denying P1. According to conceptualism Rossian principles are conceptual truths, so they can be self-evident (834).

My ISO to his view runs as follows:

P1. The Rossian Principles are substantive propositions.

P2*.  Any account of Rossian principles that is inconsistent with P1 is false.

P3. Conceptualism is inconsistent with P1.

C*. Conceptualism is false.

Like many other intuitionists and many other rationalists I have no problem with the idea of synthetic a priori propositions (although not all synthetic a priori propositions are self-evident). So I would deal with SO by rejecting P2.

In what sense are Rossian principles substantive? I am inclined to endorse the following account discussed by Cowan:

SUB. A denial of p—or a failure to manifest belief that p—by an agent, S, who has entertained p does not itself constitute prima facie evidence that S fails to understand p. (837)

I was using this conception of a substantive claim/dispute in my first worry about disagreement. Cowan rejects SUB on the ground that (a) it does not support P2, and (b) it converts at least one non-substantive proposition into a substantive one. (a) has no force against those who, like me, reject P2. In any case there is no requirement to tailor an account of the substantive/non-substantive distinction so that P2 comes out true.

The example that illustrated (b) is Cowan’s third example of a non-substantive proposition.

c) If scenarios x and y are identical in all their nonethical respects, then x and y are identical in all ethical respects (833).

But it is not clear to me that this is a non-substantive proposition, and in a footnote Cowan himself notes that the inclusion of (c) as a non-substantive proposition is controversial. He keeps the example, but says that “the argument of this section goes through without this assumption” (833). Given that (c) is not supposed to play a role in the argument because it is controversial, he cannot refer to this to reject SUB. Indeed I think there is very good grounds to reject the idea that (c) is a conceptual claim. But there are bound to be propositions of which it is unclear what side of the substantive/non-substantive distinction they stand, and reference to such propositions should not be used in assessing different accounts of the distinction. I think that the transitivity of better than is a case in point. On the face of it this looks like a non-substantive claim, but that philosophers like Temkin can intelligibly argue against this principle suggests that it is substantive. The concepts involved in the substantive/non-substantive distinction, like so many others, may have fuzzy edges, so there may not even be a truth of the matter whether certain propositions such as the two mentioned fall one side or the other of the distinction. These are best avoided when trying to specify what the distinction is.

That consequentialists disagree with Rossian intuitionists is no evidence that they do not understand the concepts involved. This is a substantive disagreement, and I think we should reject any theory that denies this. That may make the epistemology harder. It may mean that intuitionists have to make less modest commitments than they hoped to, such as a commitment to abstracta. But as I said earlier, since abstracta are a type of non-natural object, and many intuitionists are already committed to non-naturalism, that additional baggage may not be very costly at all.

8 Responses to Ethics Discussion at PEA Soup: Robert Cowan’s “Rossian Conceptual Intuitionism,” with a critical précis by Philip Stratton-Lake

  1. Robert Cowan says:

    Thank you very much to the organisers at PEA Soup for selecting my paper for discussion, and to Philip Stratton-Lake for writing such an engaging and thoughtful précis. I have tried my best to respond to his incisive criticisms.

    1. Perceptualism and Extravagance

    Stratton-Lake casts doubt on my claim that a Perceptualist theory of the self-evidence of the Rossian Principles (and other propositions) involves a commitment to an extravagant philosophy of mind. Stratton-Lake says that (A) there is nothing extravagant about the existence of intellectual seemings. Instead, “extravagance only seems to come if we go on to posit a world of abstracta to which these seemings are connected.” However, he says that (B) it is far from obvious that Perceptualists are committed to that, and (C) such a commitment wouldn’t involve introducing a new type of entity into their ontology since they already buy into the existence of non-natural properties.

    Let me first register my agreement with Stratton-Lake on (A) (note, of course, that some do doubt the existence of intellectual seemings – see, e.g., Timothy Williamson, (2007), The Philosophy of Philosophy). Indeed, I think that proponents of Rossian Conceptual Intuitionism (hereafter ‘RCI’; note that I’ll also use the label ‘RCIs’ to refer to proponents of the view) can – and perhaps should – posit the existence of intellectual seemings (see p. 831 of my paper).

    The extravagance I attributed to Perceptualists is connected to their account of the non-inferential knowledge-grounding role that intuitions (understandings or intellectual seemings) are supposed to perform. [Regrettably, I don’t provide sufficient detail on this point in the paper]. To better understand my point, note that there are, broadly-speaking, two ways in which intuitions might ground knowledge. One is that they ground knowledge in virtue of their etiology. This is roughly what Conceptualism claims: knowledge-grounding intuitions are the upshot of the adequate grasp of concepts, where this doesn’t require awareness of abstracta. The alternative is that intuitions ground knowledge in virtue of what they present or make us aware of (plus some other conditions). This is roughly what Perceptualism claims. More specifically, Perceptualism (as I characterise it) is the view that knowledge-grounding intuitions (at least those concerning self-evident propositions) make subjects aware of the truth-makers for those propositions, and that those truth-makers are abstracta (my thinking on these matters is heavily influenced by Elijah Chudnoff’s (2013) Intuition). This sort of non-sensory perception of abstracta is, I suggest, extravagant.

    In support of (B) Stratton-Lake suggests that intuitions (presumably, knowledge-conferring intuitions) may simply present the truth of propositions, and not truth-makers. So even if the truth-makers are abstract, intuitions need not make subjects aware of that in order to ground non-inferential knowledge. However, if an intuition merely involved a self-evident proposition seeming true and didn’t involve presenting truth-makers for it, then I suspect that the account of the non-inferential knowledge-conferring power of the intuition would require telling a similar story to that told by Conceptualists and RCIs, i.e., it would appeal to grasp of concepts and reliability.

    Even if a presentationalist theory of knowledge-conferring intuitions required that subjects were aware of truth-makers, perhaps intuitions only need to “put us in touch” with particular instances of moral properties (as opposed to abstracta). Stratton-Lake suggests this, and attributes something like the view to Richard Price. However, even if we concede that this model doesn’t require awareness of abstracta (but it does seem natural to think that the subject matter is abstract in the case of self-evident propositions), I am sceptical that such intuitions could ground non-inferential moral knowledge independently of perceptual-like awareness of the moral properties themselves (assuming a commitment to a presentationalist/awareness theory). Given that Stratton-Lake takes moral properties to be non-natural, a commitment to such awareness (and other features, e.g., informational encapsulation) would, I suggest, be extravagant.

    Finally, regarding (C), I don’t think that positing non-sensory awareness of abstracta would be ontologically cost-free for non-naturalist Perceptualists. Although they may already be committed to abstracta, they would be introducing an extra perceptual-like relation between ordinary agents and abstract into their ontology. Compare this with the more modest commitments of Conceptualism: there are abstract objects (roughly: concepts) but ordinary agents are not perceptually aware of them.

    2. Anti-Rossian Denials and Disagreement

    Stratton-Lake suggests that RCIs are committed to endorsing one of the following unattractive options:

    (i) Anti-Rossians possess different concepts from Rossians, OR,
    (ii) Anti-Rossians fail to understand or grasp one or more of the concepts that figure in the Rossian Principles.

    Option (i) allegedly entails that there is no genuine disagreement between Rossians and anti-Rossians, as they are simply talking past each other. Given that their disagreement seems genuine – the debate is “a substantive one” – this doesn’t seem plausible. Option (ii) entails that anti-Rossians are “rather confused” or “lack a grip of the concepts they are using”. But this also seems implausible.

    In response, consider first the following principle:

    Belief: In order for some S to be credited with possession of the concepts MORAL REASON, PROMISE, etc., S must believe the Rossian Principles.

    If RCIs were committed to Belief, then they would indeed either have to say that anti-Rossians possess different concepts, e.g., MORAL REASON*, or, they lack a grip of the concepts that they are using in a fundamental sense. [an aside: it’s not clear to me that endorsing the former option would entail that Rossians and anti-Rossians are talking past each other, so long as the distinctive concepts were similar enough]. However, as I discuss on pp. 840-842 of the paper, I don’t think that RCIs must endorse Belief. Instead, it is more plausible that they are committed to something like the following:

    Disposition-to-Believe: In order for some S to be credited with adequate grasp of the concepts MORAL REASON, PROMISE, etc., and adequate understanding of the Rossian Principles, S must have a disposition-to-believe the Rossian Principles.

    One important thing to notice about this principle – in comparison with Belief – is that it concerns adequate grasp and adequate understanding. It is not concerned with grasp of a concept, or what we might call ‘basic’ understanding (see Robert Audi’s The Good in the Right, pp. 49–50 for the distinction between different degrees of understanding). Given this, were RCIs to grant that there is no disposition-to-believe the Rossian Principles for anti-Rossians, then they needn’t be committed to claiming that they (i) possess different concepts from Rossians, or (ii) they lack a grip of the concepts they are using. Admittedly, it would require them to attribute to anti-Rossians the lack of an adequate grasp and adequate understanding. But, depending on how that is cashed out, e.g., if adequate grasp of C involves articulating the content of the implicit conception associated with C, this needn’t be implausible.

    However, as I discuss in the paper (pp. 841-2), RCIs can allow that anti-Rossians have a disposition-to-believe the Rossian principles – and an associated adequate grasp/adequate understanding – but that this is masked by their theoretical commitments (cf. a vase may retain the disposition-to-shatter even when encased in Bubble Wrap – its disposition is masked). One kind of evidence for this would be that anti-Rossians may often be tempted to make judgments about particular cases which would support the Rossian picture. On this approach, Rossians and anti-Rossians are not talking past each other; nor do they lack a grip of the relevant concepts.

    Of course, on this approach anti-Rossians are making a mistake of some sort. Perhaps RCIs must label this as a sort of ‘conceptual’ error. Note, however, that this would only be in the sense that anti-Rossians are misapplying the concept when it comes to considering general principles (and perhaps some particular cases). But that sort of conceptual mistake is compatible with grasping the relevant concepts, and perhaps even an adequate grasp of them. Thus, it is unclear to me that RCIs face a problem.

    Elsewhere in the précis, Stratton-Lake suggests that RCI is incompatible with the plausible thought that the disagreement between Rossians and anti-Rossians is substantive, i.e., about substantive propositions. Stratton-Lake endorses the SUB account (which I discuss on p. 837). On this view, I think that the Rossian Principles would be substantive, even if RCI is true. Thus, RCI isn’t itself incompatible with the debate between Rossians and anti-Rossians being substantive.

    [an aside: Stratton-Lake casts doubt on one of the reasons I give for scepticism about SUB: that it would seem to misclassify (c) as a substantive proposition. Although some people have argued otherwise, I think that it is nevertheless true that (c) is at least a plausible candidate for being non-substantive. In any case, I don’t think that the general thrust of my argument depends upon this claim].

    Consider, however, another view of substantivity (which I discuss on p. 833):

    SUB*: a substantive true proposition, is one which is not encoded in the implicit conception for a concept(s) figuring in the proposition, and one which requires engaging in thought over and above drawing on the informational content of the relevant concept(s).

    The conjunction of SUB* and RCI entails that the debate between Rossians and anti-Rossians isn’t a substantive one. But, once we realise what account of substantivity is in play, and that this doesn’t mean that the participants are talking past one another, or that anti-Rossians are making silly mistakes, I don’t think that this is implausible.

    A brief coda to this section: it is worth noting that Perceptualists also face the task of providing an account of anti-Rossian denial. As far as I can tell, this may ultimately involve saying something like ‘anti-Rossians aren’t (intellectually) seeing things aright’. More would need to be said about this, but on its face, this isn’t obviously a more plausible-sounding diagnosis of denial/disagreement compared to the RCI one.

    3. The Inverted Substantivity Objection

    In the précis, Stratton-Lake poses an Inverted Substantivity Objection for RCI.
    As mentioned above, I don’t think that RCI/Conceptualism, per se, is inconsistent with the substantivity of the Rossian Principles (P1). Rather, insofar as RCIs think that P1 is false, it will be due to their commitment to something like SUB*. So P3 is, strictly-speaking, false.

    Someone might hold P3 on the basis of something like the following idea: if RCIs accept SUB (Stratton-Lake’s favoured account) then this may require them to deny that the Rossian Principles are substantive. Why? I provide one possible explanation on p. 838 of the paper. The idea is roughly this: there is disagreement regarding the Rossian Principles among what appear to be epistemic peers; disagreement with an interlocutor about a proposition, p, who appears to be an epistemic peer of yours with respect to p, constitutes a defeater for knowledge that p; only non-substantive propositions – i.e., propositions for which denial of them constitutes evidence of lack of understanding – are such that we always have reason to think that those who disagree with us are not our epistemic peers. So, if the Rossian Principles are substantive (they fulfil the conditions for SUB) then they are not self-evident (RCI is false).

    However, this line of thought is problematic. For example, one might doubt that disagreement among epistemic peers with respect to p always constitutes a defeater for knowledge that p. Also, as I mention in the paper, the line of thought may overgeneralise to cases of paradigmatic non-substantive propositions, e.g., murder is wrong may be debarred from self-evidence. If we think that coherent denial and understanding of these is possible (perhaps due to commitment to a philosophical theory – see again Williamson, 2007), this would undermine one’s evidence that an interlocutor who denies them fails to grasp the proposition. Thus, if we reject the line of thought, but hold on to SUB/its application to Rossian Principles, RCI could be compatible with SUB. Thus it is unclear to me why RCIs should accept P3 of the Inverted Substantivity Objection.

    Suppose, however, that RCIs endorse SUB*, and thus reject P1. I suggest that in order to push back against P2*, they should point out that, despite being non-substantive, the Rossian Principles are, nevertheless, such that (a) reflection (perhaps prolonged) is needed to know them/knowing them may be difficult, (b) knowing them may require subjects to make sincere ethical judgments, e.g. about particular cases where they apply (see p. 834). Indeed, these features may explain why people are tempted to accept P1, despite it being false.

    In summary, I think that RCIs can respond to the objections that Stratton-Lake raises, although there is undoubtedly more that could be said (particularly regarding issue 1 above). However, as I note in the latter stages of the paper, RCI may in any case be problematic for other reasons.

  2. Preston Werner says:

    Hi Robert (and Philip!),
    Thanks so much for your paper.

    I want to pick up on the point about disagreement, and try to extend it. (Hopefully I’m not just making the same point over again.) I think that there is an even deeper worry here that lies behind the issues of disagreement. Suppose the consequentialist and the deontologist don’t share the same concepts. So they both have something resembling ethical knowledge (justification, etc.).
    It looks like, if this is the case, a deeper question immediately pops up: Which one of them has the concept that is latching onto the genuinely normative stuff? And I don’t see how the conceptualist can answer that question without reverting back to claims about awareness of abstracta, and thus the objections you raise for perceptualism.

    But now suppose that the consequentialist and the deontologist share the same concepts, but one of them has a defective grasp on them. One way we could figure out which one has a defective grasp is by looking for holes or inconsistencies in their grasp. But consequentialists and deontologists are clever, so let’s suppose they’ve reflective-equilibrium-ed all of those subtle incoherencies away. So now how are they–how are we–to determine, even in principle, which one’s grasp of the concepts are defective? It looks like, again, we’re going to have to return to talk of one of them being better aware of abstracta (or some such), and thus again to the problems of perceptualism.

    In short, while I think disagreement is a problem, I think it also illustrates a potential deeper problem for Conceptualism–Conceptualism needs to provide an explanation of why the concepts we happen to have will latch up with the normative properties, and that’s going to require more than conceptual grasp (I think, though that’s probably contentious).

    Thanks again for doing this!

  3. Robert Cowan says:

    Thanks for your comment, Preston!

    If Rossians and Consequentialists have different moral concepts, e.g., MORAL REASON, RCIs will say that only the Rossian has a non-defective concept, i.e., it is only for that concept that there are semantic values (entities of the appropriate kind) such that the principles required by the implicit conception come out as true, and such that moral knowledge is facilitated by grasp of concepts. If Rossians and Consequentialists share the same concepts but the latter have an incomplete grasp, then Rossians will say that it is only they who are able to get moral knowledge on the basis of their grasp of moral concepts.

    But, as you say, the question then arises: how do we tell who has the good concept/who has the defective grasp? I’m inclined to think that it might not be possible to tell (but note that, given an externalist construal of Conceptualism, that fact doesn’t undermine the claim that adequate understanding of some propositions can ground knowledge). Perhaps some headway could be made on this issue by engaging in an independent study of concept individuation. Or perhaps some kind of transcendental move could facilitate progress.

    However, even if Conceptualists can’t in principle tell who has a good concept/defective grasp, I’m don’t think that I agree that appealing to perceptual awareness of abstracta would help very much. Let me explain. If Rossian Perceptualism is true then it is only Rossians who have intellectual awareness of abstracta, or it is only Rossians who have clear and distinct awareness. It seems to me that, in this scenario, we still face the challenge of how we tell who has genuine intellectual awareness (it will sure seem to both sides that it is they who have genuine awareness/or that the other doesn’t). Thus, it isn’t clear to me that RCIs are any worse off on this issue. Perceptualists might reply that, on their view, we could make progress on this issue by engaging in an independent study of the nature of non-sensory intellectual perception/the nature of the perceptual-like relation. But, even if that puts them at something of an advantage, it doesn’t seem like a particularly promising route to me!

    There is, however, a slightly different point that I *think* you were making in your comments, and which I think may indeed be specific to Conceptualism (and thus RCI). As I mention in footnote 38 of the paper, some (specifically: Carrie Jenkins in her 2008 book) have raised the worry that there is a lacuna in Conceptualist accounts of a priori knowledge like Peacocke’s. Very roughly, her thought is that we ordinarily explain knowledge in terms of a process which begins with some kind of input from the external world. In the case of perceptual knowledge we can appeal to something like causal interaction with macrophysical objects and their properties. In the case of a priori knowledge things are trickier. Although I think their view is extravagant, Perceptualists do have a story about input: it involves awareness of abtracta. Conceptualists, on the other hand, don’t obviously have a story about the input stage. We possess concepts, and some of these are good ones, such that we are able to form true beliefs about the world on the basis of our grasp. But that might seem a bit too accidental for knowledge (as opposed to mere true belief). Hence, there may be a lacuna in Conceptualist views regarding the input step of a priori knowledge.

    Now, I’m not entirely sure what I think about that line of thought. In particular, I’m not sure whether the allegedly accidental connection involved in the Conceptualist story would be knowledge-threatening. Jenkins – who, by the way, is interested in arithmetical knowledge – thinks that it is and that there needs to be a story about the input stage. Very roughly, her own view is that the input is supplied by non-conceptual sensory experience: conceptual reflection is able to generate knowledge because and only when concepts are empirically grounded.

    Back in the moral case: someone who was sympathetic to Jenkins’ critique might claim that the input in the moral case would need to be supplied by something like non-conceptual intuitive perception, i.e., they might think that we need to endorse a view along the lines of Perceptualism. But that move isn’t straightforward, and would need to be argued for.

    I hope at least some of that speaks to your comment!

  4. Anonymous says:

    Hi Robert.
    I am not sure of the etiquette of these discussons but I thought I’d present a few thoughts on some of the things you say in your reply to my Precis.

    “However, if an intuition merely involved a self-evident proposition seeming true and didn’t involve presenting truth-makers for it, then I suspect that the account of the non-inferential knowledge-conferring power of the intuition would require telling a similar story to that told by Conceptualists and RCIs, i.e., it would appeal to grasp of concepts and reliability.”

    There are I think other options. One might, for example, maintain that the simplest explanation of why A seems to be f is that A is f. This would apply to perceptual as well as intellectual seemings. Of course this explanation can be undermined, but it seems to be a perfectly respectable explanation, albeit one lacking philosophical sophistication. If that is right then, absent undermining defeaters, that some proposition seems true would constitute evidence that p is true, and so could justify one in believing p. That justification might be strong enough to ground knowledge. On this view the role of adequate understanding is not to justify the proposition understood, but to get it clearly in view so that it can present itself as true (seem true). But it is the seeming rather than the adequate understanding that does the justifying.

    “I don’t think that RCIs must endorse Belief. Instead, it is more plausible that they are committed to something like the following:
    Disposition-to-Believe: In order for some S to be credited with adequate grasp of the concepts MORAL REASON, PROMISE, etc., and adequate understanding of the Rossian Principles, S must have a disposition-to-believe the Rossian Principles.”

    In my precis I did not always use the phrase ‘adequate grasp’ and sometimes used the term ‘grasp’instead. But I meant the argument to run with the notion of adequate grasp/understanding. I did however omit the important role of a disposition to believe in your account. On this view whether or not critics of Rossian intuitionism have an adequate understanding of Rossian principles will depend on whether they are disposed to believe the view they (occurently) think is false. Why must we think that is always true? You suggests that evidence that they are disposed to believe the Rossian view is that they are often “tempted to make judgments about particular cases which would support the Rossian picture”. They don’t give into this temptation as their consequentialist theory prevents them from doing so. But they are tempted nonetheless, and so we may assume they are disposed to believe.

    This does not, however, fit with the account of a disposition to believe you assume in your paper. There you assume “a simple view according to which S (who adequately understands p) has a disposition to believe p if and only if S would come to believe that p were she to consider and reflect (perhaps extensively) on p, drawing on her adequate understanding” (840).

    I think we may assume that consequentialists have reflected extensively in forming their view, and if they are professional moral philosophers, it would be presumptuous to assume that they do not have an adequate understanding of the Rossian view they reject, or do not draw on that adequate understanding just because they reject it. But if the consequentialist has an adequate understanding, has reflected extensively on the Rossian view and its alternatives, yet still rejects the Rossian view, then we cannot ascribe a disposition to believe the Rossian view in the specified sense. Furthermore, we cannot insist that such a disposition is a necessary condition of adequate understanding, since we are assuming these philosophers have not only a merely basic grasp of the propositions they reject, but have an adequate understanding.

    Having said that, I think that many non-Rossians are disposed to believe the Rossian principles they reject, but that is because we can be inclined to believe intuitions we think or even know are false. But that inclination often tends to fade with time such that one loses that disposition, just as one may gradually loses the disposition to say that the lines in a Muller-Lyer illusion are different even though they seem to be so. So one cannot assume that everyone who has an adequate understanding of Rossian principles is disposed to believe them even in the inclination sense. Indeed I am not sure why we should think that non-Rossians must be disposed to believe the view they reject, though it would be nice if they were.

  5. Philip Stratton-Lake says:

    Sorry that last comment was unintentionally posted anonymously.

  6. Robert Cowan says:

    Thanks very much for your comments, Philip!

    Regarding seemings: I’m quite happy to allow that seemings can justify beliefs, and that they may do so even in the absence of presenting truth-makers. I’m also wiling to concede that adequate understanding might not be playing a crucial epistemic role when it comes to justification. However, when it comes to the question of how intellectual seemings ground non-inferential knowledge I’m inclined to think that appealing to adequate understanding may be epistemically crucial, i.e., it ensures the reliability of intellectual seemings. The question then is: what does adequate understanding involve? That’s where the distinction between ‘perceptualist’ and ‘conceptualist’ views comes in. The former, but not the latter, will think that adequate understanding involves some kind of perceptual-like awareness of abstracta.

    Regarding dispositions: You are right to point out that the comments I made in the reply to the precis don’t sit very comfortably with the simple view of dispositions that I discuss on p.840. However, later in the paper (p. 841) I suggest that we should jettison that simple view in order to accommodate cases of masked dispositions. I don’t go into details on what sort of account one might hold, but there are a few that we could appeal to, e.g., ceteris paribus accounts, ideal conditions accounts. These would, I think, cohere better with the comments I made in the reply. But there still remains the important question which you raise: do we have a reason to ascribe a disposition (whose manifestation is masked) to non-Rossians? I think that we might garner some limited evidence for this from the non-Rossian’s judgments and inclinations to judge. It would, however, be good to have more evidence on the judgments of non-philosophers concerning the Rossian principles.

    I should say that I’m actually quite sympathetic to the idea that non-Rossians don’t have a disposition to believe the Rossian principles. But that’s partly because I think that we lack sufficient reason to believe that the Rossian principles are self-evident (a point which I argue for later in the paper). So I think that we may ultimately be in agreement on the point about dispositions, although perhaps for different reasons.

  7. Philip Stratton-Lake says:

    It may be that the relevant disposition is not to endorse Rossian principles, but some other normative theory, and those dispositions may be understood in various ways. But if it is not the Rossian theory would it be something like it, perhaps with an extended list? (Despite the fact that many philosophers think that Ross’s ‘theory’ is just a list, it is quite a short list, and does cover the territory well enough.) I really don’t think the dispoition could be to believe a act consequentialist theory. Indirect and rule-consequentialist theories are also implausible candidates. So if it is not the Rossian view (or something close to it) that those with adequate understanding are disposed to believe then we need to know what the theory is that is built into adequate understanding. Otherwise it is very hard to assess the conceptualist view. The same applies to different accounts of how a disposition is to be understood. It is very hard to assess the view till we know how this key term is understood.

  8. Robert Cowan says:

    Thanks again for you comments, Philip!

    Regarding dispositions: I completely agree that dispositional claims are difficult to assess, particularly when there isn’t agreement on what exactly a disposition is. So I think that the response I give to the disposition-to-believe objection against Rossian Conceptualism isn’t clear-cut. But proponents of the objection face the same problem – that’s partly why I think we are left with an impasse.

    Regarding what content might underlie adequate grasp of moral concepts: I agree that it isn’t very plausible that the consequentialist principles that you mention are ‘built in’ to the possession conditions (constitute the content of the implicit conception) for the concept MORAL REASON. I also agree that there are some difficulties in assessing the Rossian Conceptualist view, mainly because we lack sufficient data on people’s judgments about the Rossian Principles/particular cases where the Principles are relevant. In the paper, however, I suggest that, given the extant data that we do possess, we lack sufficient reason for thinking that the Rossian Principles are built in to the possession conditions for the concept MORAL REASON. What could the alternative be? One possibility that I discuss in the paper – and which I think fits better with the disagreement we observe among moral philosophers – is that built in to the possession conditions for MORAL REASON is a principle which specifies nonexhaustive sufficient conditions for cases of, for example, promise keeping to ground an overridable but ineradicable moral reason (mutatis mutandis for other act-types on the Rossian list). How extensive the list is, and what the content precisely is, may be difficult to specify (that wouldn’t be unique to moral concepts, e.g., the content of the implicit conception underlying the concept NUMBER may be not be something that we can easily specify). One suggestion which I find quite attractive is that the content identifies relevant features of prototypical cases of promise keeping, etc.