Marcus on forms of judgment and the theoretical orientation of the mind

In Belief, Inference, and the Self-Conscious Mind (BISCM),1 Eric Marcus analyses and responds to a core set of puzzles concerning belief and inference (Marcus, 2021). Most centrally on offer are: an explanation of non-evidential/non-observational doxastic self-knowledge (ch. 2), an explanation of the unintelligibility of Moore-Paradoxical statements (ch. 3), an account of inference (built on an interpretation of the 'Taking Condition') fit to solve the Lewis Carroll regress (chs. 4–5), and an account of what constitutes the unity of the rational mind (ch. 6). Marcus' answers depend on viewing beliefs as essentially self-known and expressible (through asserting their contents ("P"), or avowing them in a self-ascription ("I believe that P")). Beliefs are also held to (metaphysically) necessarily exclude their contradictories (27), and (metaphysically) necessitate their implications (135). Or at least they do in their paradigm form; when they are 'in mind' (an important theoretical concept Marcus makes much use of throughout the book (see p. 28 et passim)). In this paradigmatic form, Marcus terms them judgments. A significant achievement of BISCM is in how, despite the strength of these and other core commitments (theses variously acknowledged to be "very strong" (39), "wildly unpopular" (22), and "singularly anachronistic" (27)), Marcus remains utterly clear-eyed about the need to pass the plausibility test. Nowhere, for example, does he lose sight of the fact that we are "distinctively liable to irrationality" (27), including to self-deception (which is analysed in terms of both belief and the doxastic self-knowledge which essentially accompanies it on Marcus' view, being 'out of mind'). A second, and related strength is in the connections Marcus traces between both expressive and rational mental capacities. This is relatively rare in many of the literatures Marcus is engaging with, especially in the first half of the book. Marcus (building on earlier work) shows how such a synthesis can bear serious fruit, not least in his excellent analysis of Moore-Paradoxical unintelligibility. There is a lot one could say about what is exciting and valuable about each element of Marcus' discussion. And inevitably for a book so bold and full of riches, there is also a lot one might question about these elements. But as Marcus emphasises, what is on offer is "… not a collection of discrete solutions to putatively separate problems but a unified conception of the mind in its theoretical orientation" (8). And what I want to press him on is a big-picture question: what are the boundaries, and what is the internal structure, of the mind's 'theoretical' orientation? For, for all the important and detailed work throughout the book, in which a general conception of believing is suggested, tested, refined, and applied, I don't yet have a stable sense of Marcus' answers to these questions. As noted, the main job of BISCM is to develop and defend an account of belief (this is true of the book as a whole, since the account of inference offered in the second half is an "implication" (130) of the explicit account of belief in the first). But one thing that is refreshing about Marcus' discussion is that he gives belief much less of a totalitarian status than it is often given in philosophy of mind. (Not uncommonly, human rationality gets boiled down to interactions between just two kinds of attitude: belief and desire. Sometimes, belief is thought to do the whole job on its own.) The general idea … is that we have different cognitive abilities, the exercise of each of which culminates in correspondingly different mental states and whose perfect exercises correspond to distinctive forms of knowledge. What one 'thinks' might be a matter of what one learned from the evidence, or on the basis of aesthetic experience, or philosophical consideration, or … Just as a table is never simply coloured, but always a specific colour, so a thinking/knowing that p is never a thinking/knowing that p, but always a thinking/knowing that p of some specific sort. It may be a theoretical judgment … or it may be something else. (150) An idea which has been taken increasingly seriously in the past couple of decades is that some judgment is not theoretical but 'practical' – not (as I will put it) truth-seeking, but goodness-seeking. This is not Marcus' claim in the passage above (although he does defend it in other work (Marcus, 2018)). His thought here is more radical. It is that there is more than one way to judge that such-and-such is so – to recognise "the world as being a certain way" (145). Marcus' claim is that not all truth-seeking judgment is theoretical (belief) judgment; not all truth-seeking capacities are theoretical capacities. Call this 'judgment pluralism'. For Marcus, judgment pluralism promises an avenue of "escape from long-standing confusions about the mind" (150). Cards-on-the-table time: I am very sympathetic to a general outlook which embraces a plurality of forms of cognitive contact with the world, making for a plurality of forms of factual knowledge. I agree that an unwillingness to acknowledge such a plurality has contributed to "long-standing confusions". Assuming that all propositional knowledge of contingent truths takes the shape of belief-based-on-evidence, for example, gives rise to all sorts of difficulties. To take a particularly obvious example: it forces us to choose between viewing those forms of knowledge which we ordinary conceive as evidentially ungrounded (e.g. 'self-knowledge' of our own states of mind) as either not really knowledge, or as not really like what we think they're like.2 I think neither option is attractive. So, I think some kind of pluralism about our epistemic capacities is attractive and absolutely worth exploring. But what exactly are the non-theoretical (and non-practical) forms of judgment that Marcus has in mind? And what is the principle of individuation by which they are to be distinguished? I think that Marcus lacks a univocal answer to the second question. So there is more than one possible Marcusian answer to the first. But I am not sure whether either of the answers his discussion suggests clearly draw the boundaries of the mind's theoretical orientation in precisely the places he wants them to be. Doxastic / 'theoretical' judgments: e.g. that one's brother is a loser. Judgments of doxastic self-knowledge: e.g. that one believes that one's brother is a loser.4 Aesthetic judgment: e.g. that Carlos' concerto is beautiful. Philosophical judgment: e.g. that physicalism is false. Although some of these forms of judgment are tied to specific domains (one's own beliefs; aesthetic phenomena; philosophical issues), distinctions between different judgment-types are not merely a matter of topic or content.5 Different forms of judgment are the exercises of differently-structured cognitive capacities. These different capacities, (for doxastic judgment, for judgments of self-knowledge, etc.) have different proprietary contexts for their rational exercise, often involving responsiveness to different sorts of rational grounds. More concretely: well-formed doxastic judgments rest on epistemic reasons; judgments of doxastic self-knowledge are precisely not evidentially grounded, but are a concomitant of holding the first-order belief. Aesthetic judgments depend essentially on one's first-person appreciation of an artwork, and philosophical judgments are grounded in one's "philosophical reasoning and sensibility" (147). Part of Marcus' case for acknowledging various forms of non-theoretical judgment depends on the idea that although it is possible to theoretically judge propositions about one's own beliefs, or about aesthetic or philosophical matters, to be true, the canonical kind of judgment proper to these spheres takes a distinctive form. Believing on the basis of (e.g.) a careful analysis of one's own behaviour, that one believes that one's brother is a loser is quite possible. But it shows one to be alienated from one's first-order belief. It is not the ordinary kind of doxastic self-knowledge that we typically have – indeed which on Marcus' view is constitutive – of our own beliefs (see e.g. 52, 56, 58, 60, 81–2, 133). Equally, thinks Marcus, he can judge that Carlos' concerto is beautiful purely on the basis of Lydia's testimony. But this will not be properly aesthetic judgment (144), which would require him to have experienced the concerto for himself. Finally, a student might believe that physicalism is false out of deference and respect for their teacher, who is vocal on the matter. But this belief will not reflect their philosophical judgment, which would have to express their own grasp of the philosophical issues (147). So, the picture is of different kinds of cognitive capacity, which have different conditions for their rational exercise, whose manifestations are categorially different kinds of mental state, which in their paradigmatic forms constitute different forms of judgment, these ideally amounting to different forms of (propositional) knowledge. Marcus' own taxonomy explicitly identifies four forms of judgment: theoretical (doxastic) judgment, and three forms of non-theoretical judgment (judgments of self-knowledge, aesthetic judgment, and philosophical judgment). Marcus never claims that these four forms exhaust the genus of truth-seeking judgment. And there would seem to be a good case for distinguishing further forms, at least by the 'differently structured capacity: differently structured judgment' criterion introduced above. Consider first judgments grounded in testimony. According to a reductionist Humean conception of testimony, testimonial judgment is a matter of belief-on-the-basis-of-evidence, roughly: "so-and-so said that p, people tend to speak the truth, so p". But the Humean approach has been argued to face several difficulties (Coady 1992; Moran 2018), and in light of these, various non-reductionist pictures have been suggested, according to which testimonial grounds are sui generis. Richard Moran, for example argues that testimonial judgment essentially involves a second-personal appreciation of the testifier's point of view (as itself encompassing one's own). The capacity for testimonial judgment is thus viewed as essentially distinctive in form by contrast with the capacity for ordinary empirical judgment (which can be responsive to entirely third-personal grounds). Now consider judgments grounded in perception. As with testimonial judgment, it has been observed that trying to model an account of perceptual knowledge too on inference-from-evidence leads to seemingly insuperable theoretical problems (often resulting in certain forms of scepticism seeming irresistible). There are a number of ways of trying to avoid these problems (McDowell 2011; Kern 2017; Stroud 2018, esp. chs. 6–8; Millar 2019), but the various approaches are often characterised by a common commitment to viewing perceptual knowledge as a direct upshot of exercising distinctive perceptual-recognitional capacities. If these rough lines of thought are right (and I suspect that Marcus would be sympathetic to something like these ideas), then there are formal differences between the capacities which give rise to inferential, testimonial, and perceptual judgment respectively. The capacities for perceptual or testimonial knowledge require one to be able to recognise and exploit properties and relations which the capacity for inference doesn't itself require one to be able to recognise and exploit (respectively: perceptible properties of objects; the subjectivity of an interlocutor – of which their sincerity will be a mode). Clearly the capacities for testimonial and perceptual knowledge are also formally distinct from one another by this criterion. By the 'differently structured capacity: differently structured judgment' criterion, the judgments we make on the basis of inference, testimony, and perception, are different forms of judgment in Marcus' sense. What should we take from this? In particular, how do these forms of judgment fit into the 'theoretical orientation' of the mind which Marcus offers in the book? Do the judgments involved in perceptual and testimonial knowledge come out as 'theoretical' or 'non-theoretical' by Marcus' lights? Well, we know that for Marcus, inferentially-grounded judgments are theoretical. The whole thrust of the book would seem to tie inference essentially to belief. And with this in mind one might then imagine that – since belief judgment is theoretical judgment, the alternatively-structured judgments constituting testimonial and perceptual knowledge would have to be non-theoretical. Yet in various parts of BISCM, Marcus is explicit in classifying judgments made on testimonial and perceptual grounds as beliefs; as 'theoretical'. For example, in "an ordinary case" of theoretical deliberation, he says, "one considers what one has seen with one's own eyes, what one has been told, relevant memories, any evidence that bears on the question, etc." (13; see also 125). And nor is this an eliminable element of his account. Testimonial judgment in particular is often treated as a paradigm case of 'theoretical' judgment, against which Marcus' categories of non-theoretical judgment are contrasted. Recall that if Eric believes that Carlos' concerto is beautiful simply on Lydia's say-so, then this is doxastic (theoretical) and not aesthetic (non-theoretical) judgment (144). Parallel contrasts are drawn between testimony-based judgment and judgments of the other 'non-theoretical' kinds. If Marcus is to hold that judgments based on inference, on testimony, and on perception, are all equally theoretical, as he would seem to want, then he must either assimilate the latter two to the former – which I suggested above there are good reasons not to do – or to view the category of 'theoretical judgment' as itself internally differentiated into number of species. On this way of looking at things, 'belief' itself (theoretical judgment; doxastic judgment) comes in a variety of forms – inferential, perceptual, testimonial – at least if we are to hang on to the principle, 'differently- structured capacity: differently-structured judgment'. Marcus may well be happy with this. And I don't think there's anything incoherent in the idea. But it leads to a further challenge, which is to provide an account of the genus 'theoretical judgment' – specifically, one which admits what are now three different species (inference-based, testimony- based, perceptually based) as members, whist excluding the three forms of non-theoretical judgment listed above (judgments of self-knowledge, aesthetic judgment, philosophical judgment). We are looking for "a unified account of the mind in its theoretical orientation" (8). Having started by focussing on some ideas from the end of the book, let's scroll right back to the beginning. To believe that p is to represent p as true. Truth is a standard by which beliefs are assessed. A certain sense of the question 'why?' has application to beliefs. Beliefs are governed by the law of non-contradiction. A belief is the sort of thing that one can bring to mind. One knows one's own beliefs better and differently than anyone else can. Sometimes people are wrong about what they believe. Recall that our task is to understand how Marcus can distinguish, within the class of what I have called truth-seeking judgment, between judgment-types which are 'theoretical' and those which are 'non-theoretical'. Plausibly, all truth-seeking judgment are likely meet most of these conditions. The exception is principle (iii) (at least on a certain interpretation), so I focus on this.6 Condition (iii) is self-consciously offered by Marcus as an analogue to Anscombe's famous characterisation of intentional action in terms of the agent's giving application to the question why they are doing what they are doing. Anscombe's 'Why?' seeks reasons for action, roughly, "reasons why it would be useful or attractive if the content were true" (Anscombe 1957, 6). By contrast, Marcus' 'Why?' – the 'Why?' in (iii) – seeks reasons for belief, that is, reasons for thinking that a content actually is true. Understood in this way, condition (iii) may seem to be precisely what Marcus needs as a principled way of including judgments grounded in testimony and perception alongside judgments grounded in inference within a coherent genus of theoretical judgment, notwithstanding there being genuine formal differences between them as species. As far as the genus is concerned, it doesn't matter what form one's reasons come in, or whether responding to these reasons requires exercising a distinctive capacity to recognise evidence, or the perceptible properties of objects, or the subjectivity of an interlocutor. All that matters is that in so judging, one exercises a capacity which enables one to respond to a demand to provide reasons which speak to the truth of what one judges to be the case. More good news: judgments of self-knowledge are ruled out of the genus of 'theoretical' judgment by condition (iii). Since these are not made on the basis of reasons, or evidence, but are (for Marcus) simply metaphysically bound up with believing that p, a person will not grant application to the question why she believes that she believes that her brother is a loser (or whatever). Or: insofar as she does grant the question application, she shows herself to be making some form of theoretical judgment after all (so not a judgment of self-knowledge) about her first-order belief (from which she thereby shows herself to be alienated). But the criterion given in condition (iii) does not wholly vindicate Marcus' initial taxonomy, because it does not clearly categorise aesthetic and philosophical judgment as non-theoretical. Both would seem to meet condition (iii). It is true that properly aesthetic judgment requires having experienced a work of art for oneself, and exercised a capacity for appreciation. But that does not mean that one will reject the demand for reasons to think that the concerto is beautiful. Denying this would be denying that aesthetic judgment is rationally responsive. But it is not. On the contrary, the better-developed one's capacity for aesthetic judgment, the more one's experience of Carlos' concerto puts one in a position to say about why it is beautiful – which (at least if one is rational) is at the same time saying why one thinks that it is beautiful. The same goes for philosophical judgment. We can agree with Marcus that a properly philosophical judgment – such as that physicalism is false – must be "an expression of one's own philosophical reasoning and sensibility" (147). But God forbid that this should mean that one will not be able to grant application to a 'Why?' question of the kind described above: one which asks for reasons to think the content of the judgment is true. On the contrary, doing philosophy is absolutely centrally and precisely a matter of recognising the demand to explain why one accepts certain propositions (Marcus himself is wonderfully insistent on this very point (145-152)). So despite the initial promise of the 'Why?'-question criterion for tracing the bounds of the 'theoretical', it is not obvious how this is supposed to work – at least if it is to capture Marcus' initial taxonomy of judgment. What should Marcus say? One option here is to further define the sense of the relevant 'Why?' question so that it cuts the cake in the right place. A 'Why?' which demands only empirical evidence will be too narrow (ruling out a priori theoretical judgments as well as plausibly – as I suggested above – testimonially- and perceptually-grounded judgment); whereas a 'Why?' which seeks any old kind of truth-supporting considerations is, I have just been suggesting, too broad to classify aesthetic and philosophical judgment as 'non-theoretical'. Perhaps Marcus can offer something in between. Another option might be to question the helpfulness of the category of the 'theoretical' in the first place – at least where this is to be defined more narrowly than the category of what above I called 'truth-seeking' judgment.