The social dimension of practical assent

Abstract

Ordinarily, we consider ourselves agents and authors of our own thinking. We experience and conceive of thinking as an activity rather than a process occurring to us that happens to be located in our mind. Furthermore, we consider such activity as autonomous, generated by our own intellectual powers, rather than occasioned by the external environment or hetero-directed. What grounds can we offer to support the claim that we are “practical selves”—that is, the origin and authors of our thinking? Our claims cannot rest solely on agential awareness because there can be activity without an agent, “a deed without a doer” (Gomes 2024, 66). This is Lichtenberg's problem, which Anil Gomes sets out to solve in The Practical Self.

The problem arises for epistemological theories such as Descartes's and Kant's, insofar as they endorse what Gomes calls the “isolationist methodology,” recommending that one start by characterizing thinking as a self-conscious activity to understand what thinking really is (Gomes 2024, 2, 68). As Gomes remarks, the isolationist methodology should not be seen as a sign of confidence but as a way to highlight the centrality of self-reflecting capacities in grounding knowledge of the world. By centering on agential awareness, this methodology brings to the fore the deliberative and first-personal aspect of thinking (Gomes 2024, 73). Thinking is doing—something more akin to action than to an event happening in our mind. Further tightly connected claims follow from this characterization. Firstly, to the extent that thinking is an activity in which we engage, thoughts are deliberations held first-personally: we can assess them in a variety of ways and exercise first-person authority over them.¹ Because we stand in a relation of first-person authority with our thoughts, our assessment of them directly impacts the way we keep, revise, or discard them.² If we see no reason to believe that it is going to rain tonight, we should also think that recurrent thoughts that it is going to rain ought to be discounted and discarded. If we have previously asserted that it was about to rain, then we should stand corrected, take back the assertion, and acknowledge it as false. It is a sign of irrationality to resist one's own authority—and a kind of irrationality more akin to self-alienation than to incoherence. Failing to follow up our own thoughts that some belief should be discarded or revised radically differs from (reasonably or unreasonably) failing to concur with somebody else's view that we should do so. In contrast, we have no such authority over others' states of mind. While we can exercise some (epistemic, moral) authority on others, for instance, by correcting them, showing them that they are mistaken, or corroborating their thoughts with proof, such authority is indirect and may not have any effect on them. Correspondingly, we can resist the authority of others and reject what we provide as evidence: sometimes we do so reasonably, with good reason, and other times unreasonably (e.g., out of arrogance). Nonetheless, rejecting others' (epistemic or moral) authority is never straightforwardly a matter of irrationality.

Secondly, and relatedly, insofar as thinking is an activity produced by our mind rather than an impersonal process that takes place in it, we are in charge of our own thinking. We can claim and take responsibility and be held accountable for what we think. Expressing thoughts in the form of an assertion or declaration makes us responsive to the normative standards and expectations associated with asserting and declaring: if asserting and declaring constitutively aim at truth, in asserting and declaring falsities, we fail to reach this target. However, asserting and declaring falsities is also something we do to others, and there are normative expectations relative to truth-telling as a moral duty.³ First-person authority also means that as agents of our thoughts, we have the responsibility to check that they are attuned to reality. This is a responsibility we bear toward others and is directly connected to the ways we make ourselves accountable to them.

The salience of this communicative, expressive, and even performative aspect of thinking, with its normative and evaluative entailments, introduces a third claim that to be conscious of oneself, one must also recognize oneself as situated in a world of distinct objects. It is a condition of self-consciousness that we are related to an objective world.

All these claims are at least implicit—if not reflectively endorsed—in ordinary epistemic practices. We ordinarily consider ourselves in charge of our thoughts and assess such thoughts in various ways. In discussing the weather as much as political opinions, we claim responsibility for what we think; correspondingly, we attribute to others the responsibility for thinking what they think. We also ordinarily think that what we think matters not only to us but also to others in that they are affected by our thoughts—that is, by their contents and the way they are communicated and expressed. These problems are vivid to agents who represent themselves as situated in an objective world, and engage with other subjects who are similarly positioned.

Nonetheless, to provide a philosophical foundation for these ordinary attitudes and practices proves difficult. The connection between agential awareness and objectivity seems especially elusive: Descartes and Kant offer different a priori arguments for linking self-consciousness to the objective world. Descartes's argument appeals to God's beneficence; Kant's argument is that recognizing something as an object requires understanding that it exists independently of one's subjective perceptions, which, in turn, allows the subject to understand oneself as one among many other subjects in the same world. Self-consciousness requires a unified awareness of oneself over time. The process of unification required for self-conscious judgment is carried out by the subject, but for such unity to exist, there must be a stable framework of external objects that are perceived as existing independently of one's mind. Lacking an objective world, it would be impossible to form a coherent sense of self out of a bundle of disconnected perceptions.

Gomes aligns with most contemporary philosophers in finding these strategies untenable (Gomes 2024, 28). The pars destruens of his essay is devoted to showing that both the experiential and conceptual foundations of our understanding of ourselves as agents of thinking fail to explain the practical self: “our status as agents does not show up in our experience of the world. And there is nothing conceptually incoherent about the idea that we might be the mere passive recipients of all our thoughts” (Gomes 2024, 80). Nonetheless, Gomes thinks that Lichtenberg's challenge can be successfully addressed by seeking “an alternative route from self-consciousness to objectivity, from isolation back to the world” (Gomes 2024, 79). His pars construens combines two strategies. The first builds upon Kant's account of rational assent and aims to demonstrate that the grounds of the agential awareness of thinking are practical rather than epistemic (Gomes 2024, 80–81). Like the existence of God, being the agent of one's thinking admits of no empirical or conceptual proof and is thus theoretically uncertain. However, there are practical reasons for according our assent to both of these claims. The philosophical task is to show that having faith in ourselves as agents of our thinking is a requirement of practical rationality. The second strategy is invoked to secure the connection between this faith and objectivity. To this purpose, Gomes turns to practices and institutions, drawing selectively from Kant's account of religion and Strawson's view of reactive attitudes.

In a way, this is an odd pair. On Kant's account, self-consciousness is the route to responsibility as well as to the objective world. For Gomes, one step is missing. Kant's transcendental argument rests on the unwarranted claim that the awareness of intellectual activity suffices to ensure awareness of oneself engaging in that activity (Gomes 2024, 83). Gomes proposes that the gap be bridged by refocusing on practices; in this context, he relies on Kant's account of religious institutions and practices. However, differently than for Strawson, for Kant, social practices do not provide objectivity of any interesting kind, they help sustain moral commitment.

The emphasis on institutions and practices serves to highlight the social dimension of practical assent, thereby placing the centrality of social modes of negotiating the boundaries between the self and the world, and the self and others, into sharp focus. This represents a promising avenue for overcoming the limitations of the isolationist approach to thinking. I concur with Gomes that the social dimension of practical assent is crucial to establishing the connection between agential self-awareness and objectivity, but I have some reservations about the viability of his proposal. Ultimately, my worry concerns the role that practices and institutions play in his argument.

Gomes's first strategy builds on an argument that is central to Kant's practical philosophy and is receiving growing attention in epistemology. Kant's practical assent argument responds to a problem regarding the capacity to assent to the possibility of some ends. It exploits the distinction between choice and wish: to will an end, one has to think that the end is attainable. We are morally required to set the highest good as an end, and the argument for the existence of God is based on the presumption that rational agents are rationally committed to the pursuit of moral ends. Gomes's strategy consists in arguing that the claim about agential awareness is analogous to claims such as the existence of God. Although they lack epistemic support, these claims are not groundless. The absence of empirical or conceptual grounds prevents us from believing in the existence of God as a matter of epistemic certainty but there are—or, more precisely, we have—reasons to believe in it on practical grounds.

The fact that no appeal to experience can be grounding can be seen as a point of convergence with Kant and Lichtenberg; the idea is that the receptive nature of experience, which undermines the autonomy of awareness, undercuts the explanation of thinking as a spontaneous activity.⁴ Gomes deploys an analogous strategy to respond to Lichtenberg's challenge: making the postulate of the self as an agent of thinking a practical requirement (Gomes 2024, 86–88). Practical postulates are claims to which we are rationally entitled to assent (Gomes 2024, 87). What kind of practical attitudes does practical assent license? Gomes focuses on faith and takes the existence of God as the paradigmatic example of an object of practical assent, on which to model the argument for the practical self. However, Kant also identifies hope as a key practical attitude for practical assent. These attitudes are often used interchangeably, and scholars have put forward different ways of understanding their offices.⁵ I propose that we consider faith and hope as distinct concepts in relation to different intentional objects to clearly identify the two different kinds of agency that they promote.⁶ When directed at God, faith allows individuals to rationally commit to moral principles under a specific kind of uncertainty, that is, in the absence of empirical proof of the ultimate compensation of their striving for the moral ideal and the attainment of the highest good. Its object is thus an ultimate order of reality, in which happiness and virtue are harmonized. Rational faith does not entail confidence in God's existence or assistance (R 6:68) but is directed to and ensures the coherence of moral and natural goods, virtue, and happiness. By contrast, hope is directed to and serves moral perfection (R 6:62) and points to the real possibility rather than the mere conceivability of human moral achievement independently of divine external assistance. To effectively engage in the moral struggle requires hope in humanity (R 6: 57). The human being “must be able to hope that, by exertion of their own power, they will attain to the road that leads in that direction […] For they ought to become good human beings yet cannot be judged morally except on the basis of what can be imputed to them as done by them” (R 6: 51). Hope is directed to the realization of an ethical community on earth, through human perseverance in the pursuit of the moral ideal (R 6: 100). In that sense, hope can be kept separate from faith directed at God.

Attention to both attitudes of practical assent is crucial to appreciate that while humans suffer from the cognitive limitations peculiar to their embodiment and social embeddedness, they can also use the powers of practical rationality as the key resource to address their distinctive predicament (CPrR 5: 19,25, CJ 5:257–260). Kant often takes for granted the complementarity of hope and faith, but for present purposes, there is a methodological gain in distinguishing them based on their intentional objects. By using this stratagem, I intend to bring into sharp focus that both knowledge and the moral struggle are collective problems, to be solved by engaging in a variety of shared rational actions. Some of these are articulated by practices and institutions whose structures demand complex normative and legal frameworks.⁷ This approach provides a straightforward route to the second strategy for solving Lichtenberg's problem.

To recap, Gomes's first strategy accounts for the claim that we, as self-conscious subjects, must have faith in ourselves as thinking agents. The second strategy aims to establish that this faith is sustained by engaging in social practices. Gomes endorses Strawson's characterization of the participatory stance, from which agents assess actions, make claims, and take responsibility for them. The practice of holding people accountable provides a framework supporting the contention that we are the agents of our thinking. This framework is not necessary to sustain this faith, and, to this extent, the argument is less ambitious than Kant's; however, it is a way of underpinning the assent that we have reached on purely practical grounds. Gomes thinks that this fills a gap overlooked by the isolationists and is sufficient to establish the relevant notion of objectivity (Gomes 2024, 4).

In elaborating the second strategy, Gomes avails himself of the Kantian repertoire of considerations regarding the role of religious practices and institutions in reinforcing the agent's commitment to morality. Kant considers religious practices “intrinsically contingent” (R 6:101, cf. 6: 103) and irrelevant to the grounding of morality. Nonetheless, he does not think that religions are useless or disposable; on the contrary, he urges that “some historical ecclesiastical faith or other, usually already at hand, must be used” (R 6:109). Insofar as morality is fully grounded in reason, the appeal to religious symbols and rituals is not meant to strengthen the grounds of morality but to make them vivid in the feeble minds of humans.⁸ Rational faith fits the “peculiar weakness of human nature” (R 6: 103). For instance, Christ is to be taken as a prototype of moral goodness (R 6:61), but this adds nothing to its rational justification, and it cannot be deemed a vehicle of moral ideas. The function of exemplars is symbolic as opposed to representational. The emphasis on symbols brings to light the notion that religion does not afford any kind of knowledge. Thus, in Kant's view, symbols and analogies do not convey similarities in content but establish a structural resemblance that always require interpretation.⁹ The insistence on the symbolic nature of religious language points to the importance of a rational interpretation as a potentially disruptive activity. As a rational activity, interpretation is the locus of freedom, a barrier against idolatry and anthropocentrism that proves indispensable for enlightenment (Wood 2020, 123). Understood as a rational activity, hermeneutics stands in contrast to the “literal interpretation [of the Scriptures] that contains absolutely nothing for morality, or even works counter to its incentives” (R 6: 110).

Gomes does not pick up on the latter point and insists that practices do not sustain assent by means of symbols that express the same content as the assent but by presupposing the claims to which we practically assent: “they scaffold practical assent by providing a framework around which we can organize our lives and through which we can reinforce the commitments of rational religion” (Gomes 2024, 109). The scaffolding view holds that religious practices sustain practical assent by placing it within “a form of life.” They do so in three ways: by making assent intelligible and articulating it in the terms of a specific form of life; by keeping it stable across time, thereby countering the risk of degradation; and by “making it easier to accept that which we are required to accept on practical grounds” (Gomes 2024, 101).

The scaffolding view concurs with Kant's discussion of ritualized actions that reinforce Christian fellowship (R 6:192). Rituals help in “awakening and sustaining our attention to the true service of God” (R 6:193) and sustain faith in God's existence by placing assent within a communal form of life structured around that faith, insofar as they are intelligible solely on the presupposition that God exists (Gomes 2024, 110). While Kant suggests that this is a unique prerogative of Christian Churches (R 6:128–131),¹⁰ the scaffolding view could be legitimately extended to cover other forms of life (Gomes 2024, 109). My question, then, is whether the scaffolding view suffices to support the relevant kind of objectivity.

The force of the argumentation based on the scaffolding view is relative to and bounded by a specific form of life. For instance, the appeal to Christ as a moral exemplar provides no practical reason to non-Christians. To people unfamiliar with the Christian form of life, it means nothing that there are institutions supporting faith in a Christian God. This is not to deny that different forms of life may concur to a significant extent, leading to what John Rawls terms an “overlapping consensus” rather than establishing convergence on core matters (1993, Chapter 4). The result is a general phenomenon of practical assent, but the underlying supporting reasons are issued from separate and mutually independent authoritative sources of normativity. This is a significant limitation for a strategy aiming to warrant objectivity.¹¹

Arguably, the practices that sustain practical assent in the case of thinking about thinking are not limited to a specific form of life as is the case with religious beliefs. Presumably, the conception of oneself as the agent of one's own thinking is far more general than the notion of oneself as say, a Christian. However, this creates a dilemma for the scaffolding view that practices sustain practical assent when they presuppose the assented claim and situate it within a form of life. It is difficult to imagine a human form of life in which subjects do not think of themselves as agents of their own thinking. If so, it is unclear that it is practices that supports the objectivity of the practical assent. The reverse seems to be true.

This is where Gomes and Kant part ways, but their disagreement requires further analysis. Gomes praises Murdoch's insight “that not all contingencies are on a par. Some are central in our lives and must be central, given the creatures that we are” as more suitable (Gomes 2024, 111). It is hard for me to understand the claim that some practices must be central given the beings that we are without concluding that they are rooted in some structural and universal features of being human. The underlying suggestion might be set aside with the following warning: “It is always tempting for the Kantian to hear this as a restricted necessity: they are necessary for us human beings or some such. But this is to miss the point. Something can be central for us even if it is not strictly required” (Gomes 2024, 111).

However, it may be useful to remind the critic of Kant that the kind of necessity in place in this context is subjective (Wood 2020, 32). The assent to the necessary conditions is merely “subjectively sufficient”: it holds only because I have chosen to set an end. When introducing the second strategy that appeals to social practices, Gomes underlines embodiment as a distinctive feature of the Strawsonian framework. The turn to Strawson is motivated by the need to capture the perspective of “human subjects in a world of objects,”¹² in stark opposition to Kant's transcendental idealism while also resisting the contingencies of history (Gomes 2024, 3). The Kantian account of practical assent shows that Kant's framework is designed to capture the condition of the embodied and embedded rational agents we are. To be sure, Kant thought that such attitudes could not stand independently of transcendental idealism; but a more interesting question is whether the general plan could be achieved independently of it.¹³

It is in this context that Gomes introduces the practices of reactive attitudes “as a case study” (Gomes 2024, 113). Reactive attitudes belong in the participant's stance and, in Gomes's view, “presuppose responsibility” (Gomes 2024, 114). Strawson's account determines the conditions for the attribution of responsibility by attending to their disabling conditions, similar to how he approaches perceptual knowledge (Gomes 2024, 115). It is a merit of Gomes's reconstruction that it highlights the cohesiveness of Strawson's philosophy. Nonetheless, it also brings to the fore a tension with the Kantian approach to practical assent, which threatens to pull his two strategies apart. I attempt to articulate this worry by examining the disagreement between Gomes and Kant.

Gomes draws on a consonance between Kant and Strawson, whereby both posit that reactive attitudes presuppose a notion of responsibility. Furthermore, neither suggests that the relevant sort of responsibility could be elucidated through a metaphysical investigation into the nature of free will. Both perspectives assume that reactive attitudes emerge within the context of interpersonal relationships, manifesting as a function of our normative expectations regarding how others should treat us and how we should treat them. In order to hold someone accountable, we must view them as having a kind of agency that makes them responsible for their actions. Reactive attitudes and the presupposition of responsibility are “inextricably intertwined” (Strawson 1962). Conversely, for Kant, it is the moral feeling of respect that carries the subjective experience of rational (and moral) agency, which is not relative to a specific form of life but a universal feature of human rational agency—that is, a variety of rational agency that is embodied and socially embedded (R, OQ 84).

The focus on life forms allows Gomes to conclude that “the practice scaffolds our assent within a form of life. This makes it easier for us to accept that which we are required to accept on practical grounds. The practice of holding each other accountable for our thinking cements the formation of group-identities” (Gomes 2024, 117). Further, “there is nothing necessary about this scaffolding. Perhaps there are other practices which could sustain our faith in ourselves as the agents of our thinking or perhaps there are people” (Gomes 2024, 122). By contrast, Kant's appeal to the moral feeling of respect enables him to reject precisely the talk of group identities, not so much to argue for the cosmopolitan ideal but before this—and perhaps more profoundly—for the very possibility of shared rational agency. These are two very different lessons to draw from Kant's Religion. Kant is not a foundationalist, but, in his view, the practice of thinking encompasses humanity at large. To relate thinking to the formation of group identities is to renounce the structural norms of thinking in favor of a “private” use, which is a self-defeating operation (WOT). The divisive issue, then, concerns the boundaries of the relevant normative community.

It seems to me that the very Kantian resources that Gomes interrogates in the Religion may indicate a viable alternative to two opposite models, one seeking convergence and the other striving for overlapping consensus.¹⁴ For Kant, thinking for oneself is a moral duty and an inescapable responsibility. It is also a fragile entitlement, which can be undermined by other individuals and by our own habits, practices, and traditions. To emancipate oneself—the agenda of the Enlightenment—is not a private and individual endeavor. It is a communal problem and a collective pursuit (R6: 5, 97, OQ 79, TPP 307–308) (Wood 2020, 37, 48). The ultimate rational justification for practices and institutions is the individual and shared duty to build an ethical community (R 6:97). The Kantian story thereby connects practices and institutions more directly to objectivity.

First, the practice of attributing responsibility is more fundamental than other practices; it is the very structure of human interactions and the basis for all varieties of social agency. It relies on engaging with one another on an equal normative standing. Second, Kant allows for the experience of morality as a subjective experience of autonomy. Moral agency is experienced, that is, felt under the guise of a distinctive feeling that is not pathological but moral: the feeling of respect. The characterization of moral sensibility is not among the Kantian resources that Gomes exploits, but it could—and, I think, should—be recruited in service of Gomes's project, with two provisos. It could be recruited if and insofar as the moral feeling of respect as a recognition of moral standing coheres with a naturalistic account of psychology, which makes room for feelings occasioned by the activity of thinking rather than solicited by and directed to any particular object.¹⁵ It should be recruited if one wants to secure a route from moral agency to thinking agency. Gomes does not follow Kant on the latter point, but the result is a loose connection between the two argumentative strategies.

Third, in Kant's account, respect is the moral feeling of rational agency but also the key moral incentive and a normative constraint on deliberation (G 4:401, CPrR 5:6, 25, 78, 71–87). Respect as a normative constraint helps establish the proper form of the relations between persons: this is a relation of reciprocity based on the mutual recognition of authority. It is this kind of relation that warrants the link to objectivity in Kant's account. To acknowledge oneself as a thinker amounts to acknowledging oneself as a self-originating source of legitimate claims, whose validation requires the recognition of others. Reciprocity is the ground for social practices that organize mutual expectations—that is, not only related to contractual obligations but also to values such as sincerity and civility. Importantly, this dimension cannot be captured by calling attention to the disabling conditions because the participant's stance is one of power and authority based on mutual recognition. And mutual recognition can be withdrawn unilaterally. This is one way in which the analogy with perception breaks down. The withdrawal of recognition can be the result of an illegitimate denial of normative status (Bagnoli 2021).

Gomes's account does not seem to capture this aspect of the social dimension of practical assent. In fact, Gomes openly rejects Tyler Burge's view that the practical self is the locus of power and authority. The thrust of Burge's argument is that critically reasoning about one's attitudes must have an immediate impact on the agent's motivation to keep or revise them. This entails that thinkers have a distinctive sort of power and responsibility over their own thoughts. Against this proposal, Gomes argues that even “if Burge is right that an understanding of critical reasoning requires us to think of ourselves as the locus of power and responsibility, it is a further unargued step that this requires us to think of ourselves as the agents of our thinking” (Gomes 2024, 84). The argument supporting this conclusion rests on the example of bureaucracy, where agents who exercise power and responsibility are distinguished from those who do the work. This distinction is supposed to show that even if critical reasoning requires us “to think of ourselves as the locus of power and responsibility, it is a further unargued step that this requires us to think of ourselves as the agents of our thinking” (Gomes 2024, 86). I do not agree that this example proves the point in question because bureaucracy is one mode of organizing collective action that entails a hierarchical division of work and the corresponding distribution of responsibility. This cannot be generalized to serve as the model for thinking about thinking. The example does establish that it is possible to exercise power and responsibility over actions without being the person undertaking those actions and that one can be the locus of power and responsibility without being the one exercising that power. However, these possibilities are unique to a specific case of collective agency. There is no relevant analogy between the individual case of thinking about one's own thinking and bureaucracy. In the individual case of thinking, there is no one else in charge but the self.

Even in the case of automatic negative thoughts that interfere with our intention and hinder our capacity to act, thereby undermining our self-efficacy, we are not powerless: we can enact techniques that mute or disempower these disturbing thoughts. Paradoxically, such techniques consist in recognizing that these automatic negative thoughts are endogenous, self-engendered, and self-reinforcing, and thus lack any external support. Agents whose action is inhibited by the automatic negative thoughts lose self-efficacy, which is essential to their agential status. Succumbing to automatic forces, they feel self-alienated and unable to take responsibility for action. Resistance to automatic thoughts is a struggle for power and authority. Succeeding in muting automatic thoughts is empowering and ultimately depends on reestablishing a robust relation to the objective world as a world of objects but also, and more importantly, as a world of mutually vulnerable subjects.

While departing from Kant's account of practices, Gomes's argumentative strategy has the merit of uncovering a unity in Strawson's work (across perception and reactive attitudes) by focusing on the disabling conditions of both. But in so doing, he offers a view of objectivity that is flattened into the world of objects rather than a world of subjects—that is, independent sources of legitimate claims on one another.¹⁶ He thus underplays self-consciousness's relation to others as interlocutors and partners in shared rational action and its consequent link to mutual accountability. Practical assent is an act of freedom. The practices of mutual accountability cannot be merely “central” to the human form of life because claiming responsibility for thinking is an indispensable and inescapable act of freedom, and mutual accountability is the way in which such freedom is protected. Social practices can badly mediate, weaken, or distort our relationship to reality and others. They can be alienating and disorienting, thereby undermining our practical selves. Because social practices can displace us as agents of our thinking, they do not provide a robust route from agential awareness to objectivity.¹⁷

All translations are quoted from The Cambridge Edition of the Works of Immanuel Kant (1996), and the quotation rules followed are those established by the Akademie Ausgabe. Kant, Immanuel (1900 ff): Gesammelte Schriften. Hrsg.: Bd. 1–22 Preussische Akademie der Wissenschaften, Bd. 23 Deutsche Akademie der Wissenschaften zu Berlin, ab Bd. 24 Akademie der Wissenschaften zu Göttingen. Berlin. References to specific Cambridge translations of Kant are abbreviated as follows: