In his fascinating book, The Practical Self (Gomes, Reference Gomes2024), Anil Gomes takes up the perennial question of how self-consciousness connects to objectivity. Descartes and Kant each tried to argue from reflection on the mind to the existence of a mind-independent world. Gomes finds both approaches wanting and proposes a two-part alternative:
-
(i) Agency: We should have practical faith that we are doxastic agents responsible for judging and doubting.
-
(ii) Objectivity: Faith in agency is sustained by social practices of mutual accountability, through which we connect to an objective world.
Part (ii) beautifully portrays how our nature as rational animals is reinforced by our nature as social animals, cashing out the cliché that criticism is the highest form of praise—praise Gomes generously extends to readers like me. By challenging my assumptions, he affirms, and has me affirm, my agency as a thinker. I will try to return the favor.
I contend that Gomes cannot make do with faith alone. On both planks of his view—agency and objectivity—he needs what he denies himself: evidence. In §1, I argue that without evidence he is vulnerable to charges of bad faith and circularity, and pose a dilemma for his practical faith. In §2 and §3, I show how Descartes’s notion of clarity provides, ironically, what Gomes needs: evidence for both objectivity and agency. Finally, in §4, I return to the image from Lichtenberg that Gomes invokes as his launching pad—of thoughts striking us like “lightning”—to show that Lichtenberg’s concerns reach beyond our agency in judging and doubting to our passivity in flashes of creative insight. The general lesson: we need evidence for passivity no less than agency and objectivity—and, in all cases, Cartesian clarity provides it.
1. Gomesian Faith
Gomes’s entry point is this famous passage where Georg Lichtenberg questions Descartes’s cogito argument, “I am thinking, therefore I am”:
We become conscious of certain representations that do not depend on us; others, at least we believe, depend on us; where is the boundary? We know only the existence of our sensations, representations and thoughts. One should say it is thinking, just as one says, it is lightning. To say cogito is already too much as soon as one translates it as I am thinking. (K 76; quoted p. 72)
Lichtenberg is often misread as worrying about thoughts without a thinker—thoughts that do not belong to a subject called “I.” But as Boris Hennig (Reference Hennig2018) emphasizes—and Gomes agrees—Lichtenberg assumes the thoughts we are conscious of are our own. Some simply happen to us, as if we were struck by lightning, while others “at least we believe, depend on us.” Gomes notes that judging and doubting do not merely happen to us: “giving and withdrawing assent is something we do” (p. 9). We are doxastic agents—or so we believe.
Yet this self-conception faces an immediate challenge: hard determinism. The problem arises from the idea that being part of nature is incompatible with being free. Gomes assumes this incompatibilism, channeling Kant:
If thinking is a process subject to natural laws—a process like that of digestion or growth, to use Lichtenberg’s comparison—then we are not really thinking for ourselves even though thinking is going on within us. Kant’s philosophy bequeaths us an “incalculable gulf fixed between the domain of the concept of nature … and the domain of the concept of freedom,” as Kant himself notes in his third Critique ([Kant Reference Kant, Guyer and Matthews2000] CPJ 5:176).
If human nature belongs wholly to nature, the hard determinist concludes, we are not agents. The challenge extends beyond bodily movement: as Lichtenberg observes, it encompasses thinking itself:
Just as we believe that things occur outside of us without any contribution on our part, the representations of these things might also arise within us without any contribution on our part. Indeed, it is without any contribution on our part that we have become what we are. (H 150)
When we judge or doubt in response to reasons, we seem to make up our own minds. Yet hard determinism treats this experience as illusory. “How reason moves us,” Gomes writes, may be “like the beating of our hearts, something which happens to us, but which we do not do” (p. 79).
Incompatibilism cuts both ways. The hard determinist affirms determinism and denies freedom; the libertarian affirms freedom and denies determinism. Libertarianism, however, carries heavy metaphysical baggage: it requires a dualism—Kant’s “incalculable gulf”—between nature and freedom. Every physical event has a prior cause, but for traditional libertarians, the agential self or will transcends the physical world.Footnote 1 Because the will belongs to the immaterial soul (Descartes) or noumenal self (Kant), we possess a godlike power to be the uncaused cause—the first cause—of our judgments and choices. As Kant puts it in a passage Gomes cites approvingly, when you exercise your will, you are “the originator of action,” “its first beginnings” (Kant, Reference Kant, Heath and Schneewind1997, EV 27:559, cited on p. 82). This is the libertarian power of agent causation. Given Gomes’s incompatibilism, his faith in agency amounts to faith that we have this godlike power.
Compatibilists, by contrast, eschew the supernatural in defending freedom. For them, our choices are as much part of nature as the beating of our hearts—yet we choose freely. However, we defend freedom—as libertarians or compatibilists—hard determinism purports to refute us. It presents a reason to concede we aren’t really free, and we need a reason to affirm that we are. Gomes calls this the “epistemic problem” (p. 74):
The Epistemic Problem
What reason do we have to affirm that we are doxastic agents?
An epistemic reason—or “theoretical reason”—for believing a proposition is evidence, empirical or a priori, that it’s true. Hard determinism creates an epistemic problem for affirming agency precisely because it provides an epistemic reason for denying it.
Gomes does not counter with an epistemic reason for affirming agency. Indeed, he insists we cannot have one—and this, I’ll argue, undermines his account.
Yet for Gomes, Lichtenberg both identifies the problem and hints at its solution. Continuing the passage challenging Descartes’s cogito, Lichtenberg writes:
To assume the I, to postulate it, is a practical requirement. (K 76; quoted on p. 72)
Building on this, Gomes argues we have a practical reason for faith in our own agency:
Therefore,
I will raise problems for the first premise in §1.1 and the second in §1.2. Then, I’ll question the second half of Gomes’s view, his treatment of objectivity, in §1.3.
1.1. The bad faith problem
In good Cartesian fashion, I begin by striking at the foundation. Premise (1) is bold. Almost every philosopher who has weighed in on agency/freedom has offered epistemic reasons on one side or the other. Premise (1) entails that they are all necessarily wrong; there cannot be evidence of the sort they offer.
Gomes does not consider arguments supporting agency (libertarian or compatibilist), which suggests he is not entitled to premise (1). Meanwhile, he countenances an argument against agency (hard determinism), which contradicts premise (1). Remember, Gomes presents hard determinism as the central challenge to our self-conception as agents. If every event in nature is law-governed, every thought, judgment, and action is the necessary product of prior causes. On this picture, our sense of “making up our own minds” is no more an act of agency than the beating of our hearts. Lichtenberg underscores this, and Gomes invokes it to frame the Epistemic Problem: what reason do we have for affirming that we are doxastic agents?
Yet Gomes does surprisingly little with the challenge. He does not attempt to rebut hard determinism, nor does he allow that arguments for agency could succeed. Instead, he claims the question is “theoretically undecidable.” We cannot, he insists, have epistemic reasons either way. But this leaves him in an unstable position. The determinist argument is precisely what he uses to motivate the problem—it functions as an epistemic reason against agency. By denying the possibility of countervailing reasons for agency, Gomes leaves the only acknowledged evidence pointing against agency. To describe the matter as “undecidable” is therefore misleading. What remains is not a balance of opposing considerations but a one-sided case against agency. By refusing to rebut hard determinism or permit counter-evidence, Gomes undercuts his own premise of undecidability.
This gap leads to a deeper worry. If the only recognized evidence is on the side of hard determinism, then the appeal to faith in our agency is not faith in the face of undecidability. It is faith in the face of undefeated contrary evidence—a form of self-deception, or bad faith.
When Gomes considers a version of the bad faith objection, he explains that faith is a kind of “acceptance” and that, absent evidence for belief, we may accept a claim out of faith “in service of certain ends.” For example, “A doctor might accept for the sake of treatment that a patient has a viral infection, even when the symptoms underdetermine the diagnosis” (p. 111). I grant that my doctor may accept an unconfirmed diagnosis to pursue a cure. However, if my doctor has unopposed evidence refuting the diagnosis but accepts it anyway to subject me to a procedure, my doctor would be treating me in bad faith.
Kant’s treatment of freedom and God in the Critique of Pure Reason offers a revealing contrast. In the Third and Fourth Antinomies, he presents arguments both for and against freedom and God. The thesis and antithesis in each case are in direct conflict. Transcendental Idealism then splits the difference, explaining how each side can hold in different respects: determinism reigns in the phenomenal world, while freedom (and God) remains possible in the noumenal realm. On this basis, theoretical reason shows the possibility of freedom and God without establishing actuality, leaving room for practical faith.
Indeed, Kant wrote the first Critique to show that the topics of rationalist metaphysics (God, freedom, and the immortality of the soul) lie beyond the scope of theoretical reason, and so—in what could have been Gomes’s slogan too—Kant declares, “I had to deny [theoretical] knowledge in order to make room for faith” (Kant, Reference Kant, Guyer and Wood1998, Bxxx). In later works, he argues that to submit ourselves to the moral law, we are practically required to have faith in God (as Gomes notes, p. 118) and in our own freedom. Thus, Kant contends that freedom is “the fact of [practical] reason” in the second Critique (Kant, Reference Kant and Gregor1996, I.I.1.7, 5:31) and that we have “practical knowledge” of our freedom in the Groundwork (Kant, Reference Kant and Gregor1996, III, 4:448).
Kant’s faith in freedom is not bad faith because it emerges against the backdrop of the Antinomies, where he models exactly what Gomes’s view lacks: epistemic reasons on both sides. Gomes’s account would be strengthened by supplying parallel evidence in favor of agency, but he explicitly denies that such evidence is possible.
I will later offer a Cartesian account of the required evidence for agency. If Gomes could use that evidence to counterbalance his evidence against agency, he could salvage a weaker version of premise (1): that the question of agency is theoretically undecided. Although this falls short of undecidability, it would vindicate his claim that “so far as the evidence is concerned, we must suspend judgement” (p. 110).
But even so, does he establish that a “practical requirement” should break the tie?
1.2. The dilemma for the practical requirement
Recall Gomes’s next step:
Gomes is no Pragmatist. He is not saying, as William James would, that since evidence does not settle the matter, we should have faith in agency because prudential benefits will ensue. Instead, Gomes’s view is that when we engage in self-conscious evaluation, “the demands of rational coherence require us to assent to the claim that we are the agents of our thinking” (p. 126).
This claim invites a dilemma. Self-conscious evaluation either presupposes agency or it does not. If it does, the following holds:
If I am reflectively evaluating my perspective, then I am a doxastic agent.
Evidence for the antecedent is evidence for the consequent. Since I can know through introspection that I am reflecting in this way, the conditional furnishes a theoretical argument—an epistemic reason—for affirming agency. Modus ponens. Faith is not required.Footnote 2
If, on the other hand, self-conscious evaluation does not presuppose my agency -- if the conditional above is false -- then it’s hard to see why such evaluation would be incoherent if I do not affirm agency. Hard determinists, after all, engage in self-conscious reflection while denying their own agency. They reject the conditional above. Their position may be false, but it seems coherent. Gomes himself appears to grant the coherence of hard determinism as he uses it to motivate the Epistemic Problem. If one can coherently deny one’s agency, one is not practically required to affirm agency for the sake of coherence.
Thus, we have a dilemma for premise (2). On either horn, Gomes fails to establish the practical necessity of affirming agency.
We turn now to the second part of Gomes’s view: his move from agency to objectivity.
1.3. Circularity in the move to objectivity
As Gomes portrays them, Descartes and Kant each proposed a theoretical argument from an aspect of self-consciousness to the claim that an objective (mind-independent, external) world exists. Rejecting those arguments, Gomes posits a practical rather than theoretical link between his chosen aspect of self-consciousness (faith in agency) and objectivity. Granting other people objective status—regarding them as having their own perspectives, independent of ours—we question and challenge each other’s judgments. And as we engage in such practices of mutual accountability, we reinforce our faith that we are indeed responsible agents:
This faith [in our own agency] is sustained by a practice of holding and being held accountable for our judgements which relates us to a world of other people … This faith is sustained by practices which involve a connection to objectivity. (p. 7)
In other words:
But Gomes also runs (5) in the opposite direction, to make an “argument from self-consciousness to objectivity, from isolation back to the world” (p. 7):
This pair of claims forms a circle. If objectivity is to ground and sustain faith in agency, it cannot rest on that same faith. Between the two claims, Gomes seems genuinely committed to (5) rather than (6): his discussion supports the idea that commitment to objectivity sustains faith in agency, not the reverse. What he describes as a “move from self-consciousness to objectivity” actually runs the opposite direction—from objectivity to self-consciousness. To preserve (5), Gomes would need to drop (6) and acknowledge that objectivity must be secured independently, on epistemic rather than purely practical grounds. Unlike agency, he does not claim there cannot be evidence for objectivity. However, he rejects Cartesian and Kantian attempts to provide such evidence, while offering none of his own.
In sum, Gomes’s first plank—faith in agency—requires evidence for agency, and his second plank—faith sustained through objectivity—requires evidence for objectivity. On both counts, he withholds evidence his view needs.
The deeper irony is that one of his opponents, Descartes, provides what Gomes needs—and clarity is the key. We get defeasible but genuine evidence for the objectivity of the world and our own agency by experiencing these things with considerable clarity.
2. Cartesian Clarity for Objectivity
Though I have argued for this interpretation of Descartes in other work culminating in a book, Clarity First: Re-envisioning Descartes’s Epistemology (Paul, Reference Paulforthcoming; cf. Paul Reference Paul2018; Reference Paul2020; Reference Paul2023; Reference Paul2024), I lack space to rehearse the textual evidence. My goal is not to convince you that Descartes held these views, but to show they are worth considering as an alternative to Gomes’s picture.
In this section, I argue, against Gomes, that Descartes is a Direct Realist: when you veridically perceive the sun, for example, what appears to you is numerically identical with the real sun (§2.1).
This leaves open the epistemological question: what gives you reason to assent to your perceptions? Descartes’s answer is: clarity—to the extent that it’s clear to you that the sun is real, you thereby have reason to judge that the sun is real (§2.2).
How good a reason must you have—and so how clear must your perception be—to rationally permit assent? Descartes’s answer is: it depends on context. In ordinary life, your sensory experience of the sun and countless other things is clear enough to license assent, and it would be irrational to doubt them just because it seems logically possible an evil deceiver might be tricking you—thus the senses themselves give us sufficient reason to affirm the existence of external things without any “argument from self-consciousness.” This is Descartes’s unappreciated Commonsense Realism (§2.3).
However, in the context of the Meditations, where the bar for assent is maximally high, the apparent possibility of a deceiver gives you sufficient reason to doubt even your clearest sensory experience and thus the existence of the whole external world. Descartes therefore needs a different strategy for proving the external world exists—a strategy based not on sensory experiences per se but rather on your perfectly clear introspection of your sensory experiences—and this is where he does attempt an argument from self-consciousness to objectivity (§2.4).
I agree with Gomes that this argument fails, but the remedy is not an alternative appeal to self-consciousness based on faith in agency. Instead, we should stop hankering for perfect certainty where it is neither available nor needed—we should simply come back to our senses.
2.1. Direct realism
Imagine walking through the desert, where there appears to be a lake glimmering ahead. In the good case, the lake is real; in the bad case, the shimmering is a mirage. What, in each case, immediately appears or is present to your awareness? In other words, what do you perceive directly, not by perceiving something else which serves as a proxy or representative?
Since in both cases it appears there’s a lake, many philosophers assume you directly perceive the same thing in both cases. Because no real lake exists in the bad case, they conclude that even in the good case you directly perceive not the lake itself but an idea of the lake—a mental image existing only while perceived. Gomes accepts this view: “The immediate objects of perception are dependent for their existence on the mind” (p. 24). He thus endorses what is known as:
The Veil of Ideas
We directly perceive only our own ideas, never external bodies themselves.
On this picture, the senses are introverted. You try to look outward at the world, but you are inevitably looking inward at ideas in your mind.
If bodies exist in the external world (Realism), we can perceive them only indirectly, by perceiving our own mental images which serve as proxies or representatives of external things (Indirect Realism). In the good case, your mental image of the lake is caused by, and resembles, the actual lake. But the image and the lake are two distinct things. According to Indirect Realism, veridical perception of the lake is like watching a live video of the lake on your TV, except your “screen” is an ideational veil inside your mind.
The Veil of Ideas is a metaphysical thesis about the structure of sense-perception, but it has epistemological ramifications. The veil cuts us off from direct perceptual evidence of external things. To get evidence of external things, our only recourse would be inferring their existence through an argument beginning with self-conscious reflection on our own ideas—the kind of argument Gomes rejects.
Descartes is often said to be an Indirect Realist, but he is not. He would reject the Veil. Those who say he’s committed to the Veil point to how he redefines the term “idea”: “I am taking the word ‘idea’ to refer to whatever is immediately perceived by the mind” (3O/R, 7:181), and indeed to “any object of thought” (5O/R, AT 7:366)—“in general everything which is in our mind when we conceive something, no matter how we conceive it” (To Mersenne, July 1641, AT 3:393). The fact that something is being perceived (thought of) is what makes it count as an “idea.” Thus, he stipulates this:
Definition of “idea”
x counts as an “idea” in S’s mind =def. S is perceiving (or thinking about) x
An idea is, by definition, an intentional object. Whatever I perceive is thereby an idea, but that does not mean it’s merely an idea. Descartes distinguishes two non-exclusive modes of being: something may exist “in reality” (independently of perception), or “in the mind” (as an idea)—or both at once. Thus, one and the same thing can exist both in reality and in thought; a real object can also be an intentional object. The sun, for instance, exists in the external world; when I perceive or think of it, the sun also exists in my mind as an idea: “the idea of the sun is the sun itself existing in the intellect,” Descartes says. He explains that the term “idea” is “an extraneous label” which does not describe the sun’s intrinsic nature; it merely expresses an external relation between me and the sun while I am perceiving it (1O/R, AT 7:102). The relation is temporary. Compare another extraneous label, like “boyfriend”: Sam Smith and Brandon Flynn were boyfriends only while seeing each other, but—as they will assure you—they did not cease to exist once they stopped. Just as Sam and Brandon exist independently of each other but count as boyfriends only while they are seeing each other, the sun exists independently of your perceiving it but counts as your idea only while you are perceiving it. Descartes is thus committed to:
Direct Realism
In veridical sense-perception, we directly perceive objects in the external world (which simultaneously count as our “ideas,” but only while we perceive them).Footnote 3
How does this view handle misperception? Perception -- whether veridical or illusory -- is always of something: an intentional object, an idea. In both cases, you may have a visual experience as of a lake being there before you. Yet the two cases differ in kind. In the good case, the intentional object is also a real object. In the bad case, the object is merely intentional, not real. In Descartes’s language, you may be said to perceive your “idea” of the lake in both cases, but the idea is not the same kind of thing in each. When the perception is veridical, your idea is the real lake, which is literally wet, and so forth. When the perception is illusory, there is no real lake, nothing literally wet—there only appears to be. Thus, Descartes can accommodate illusion without trapping the senses behind an inner veil.
Now let us look closer at the content of appearances. Whatever I perceive is an idea in my mind, but that does not mean it appears to be an idea in my mind. As Descartes recognizes, normal sense-perception does not present its objects as ideas in the mind but as things existing in reality. His skeptical arguments in Meditation One presuppose this about the senses: the possibility of deception arises precisely because the senses appear to present an external world. If the senses presented only internal ideas, their falsity would pose no threat—the evil demon could not deceive us about the external world by making things not as they appear. In the deceiver scenario, Descartes writes, “the sky, the air, the earth, colours, shapes, sounds and all external things are merely the delusions of dreams which he has devised to ensnare my judgement.” This shows the senses purport to reveal an external world. When I assent to what they tell me, I believe—“falsely believe” in that scenario—that bodies exist outside my mind (M1, AT 7:24). Hence, normal sense-perception is outward-facing: it presents its objects as bodies existing in reality, not as ideas in the mind.
In fact, since normal sense-perception has only first-order content, it cannot present anything as an idea. The sun is among my ideas (= I have a perception as of the sun) is a higher-order claim. It would be the content not of vision but of a higher-order perception, namely introspection/reflection. The difference is crucial:
My vision presents the sun as existing in reality.
(Visual content: The sun is real.)
My introspection of vision presents the sun as existing in my mind.
(Introspective content: It visually appears to me that the sun is real.)
The Veil of Ideas treats vision as if it were introspection of vision. The senses aren’t introverted. Introspection is.
In sum, sense-perception presents its objects as existing in reality—and in veridical cases, they truly do (Direct Realism). But Direct Realism alone does not answer the skeptic’s challenge. As the skeptic reminds us, some perceptions mislead: what appears real is not always so. This raises the epistemological question: what gives us reason to assent to our perceptions?
2.2. Presentationalism: Reasons are due to clarity
A “reason” for assent is normative: it suggests you should assent (unless it’s defeated by a reason for doubt). But a “reason” is also explanatory because if you assent for that reason, it provides a “rational explanation” for your assent (Anscombe, Reference Anscombe1957; Davidson, Reference Davidson1963)—or as Gomes puts it, your assent “makes sense” or is “intelligible” to you from your own point of view.
Descartes holds that, in good conditions (when objects are relatively near, large, well-lit, etc.) sense-perception:
-
• is very clear,
-
• is very reliable, and
-
• provides very good (though defeasible) reason for assent.
Which of the first two conditions explains the last? When I look at my hands, I have very good reason to believe I have hands. Is that because my sense-perception is very clear? Or because it’s very reliable? Following (his interpretation of) Hume, Gomes thinks Descartes endorses the latter answer:
Reliabilism
A kind of perception provides reason for assent in virtue of reliably representing truths.
Hume and Gomes reject Reliabilism because a reliable connection to truth does not by itself make the truth “manifest” to us since it falls outside our point of view, and so could not make our beliefs “intelligible” to us.Footnote 4
But Descartes rejects Reliabilism too! For him, reason for assent comes from something internal to the mind, namely clarity.
Clarity is the appearance of truth. I also refer to it as “presentationality.” To say you perceive something clearly is to say it appears true to you—that it is presented to you, manifest to you, or strikes you as true. These are synonyms, not a conceptual analysis. Clarity is a phenomenal quality, a matter of what it’s like, or how it feels, to have a certain kind of perception. So clarity is primitive: it cannot be defined by analyzing it into more basic constituents. The difference between clarity and its opposite, obscurity, is to be “learned by examples rather than by rules” or analytic definitions (AT 7:164). So let us consider examples.
Start with visual experience. If you are looking at a deer from far away, it may be obscure to you that it’s a deer. When you get closer, it may become clear to you that it’s a deer. There is something it’s like to see the deer clearly, a quality absent when you see it obscurely. The more clearly you see that it’s a deer, the more strongly that proposition is presented to you, or strikes you as true.
Now for intellection. This proposition may be obscure to you at first:
The sum of the numbers 1, 2, and 3 is equal to their product.
It becomes clear through a simple deduction:
1+2+3 = 6
1×2×3 = 6
Therefore, 1+2+3 = 1×2×3
When the proposition becomes clear to you, it goes from being a proposition you are merely considering to one that strikes you as true. Notice what it’s like when that happens. It’s like the truth gets illuminated. Clarity is “a light in the intellect” (AT 7:192).
So, clarity is a primitive (indefinable) phenomenal quality, whereby what you perceive is presented to you, or strikes you, as true. Further, other epistemically significant perceptual qualities are defined in terms of clarity. A perception is:
-
• Obscure to the extent that it lacks clarity.
-
• Confused to the extent that it is “fused with” other perceptions in a way that lessens clarity.
-
• Distinct to the extent that it is “sharply separated” from other perceptions in a way that increases clarity.
This gives us the two-fold structure of Clarity First: clarity is primitive and prior to other epistemically significant qualities or states.
In addition to being definitionally prior to the aforementioned perceptual qualities, clarity is explanatorily prior to epistemic goods, beginning with reason/justification for assent:
We must distinguish between the subject-matter, or the thing itself which we assent to, and the formal reason which moves the will to give its assent: it is only in respect of the reason that transparent clarity is required. (2O/R, AT 7:147–8).
Descartes contrasts the content or “subject-matter” of a perception with the form of perception—namely clear perception—which provides reason to assent to the content. Indeed, he adds, “this formal reason consists in a certain inner light” (Ibid.), and that light, as we have seen, is clarity. Mental illumination—clarity—is something over and above the content it shines upon, and what gives you reason to assent is not the content itself but the clarity with which you perceive it, the quality you experience whereby it is presented to you as true. More precisely:
Descartes’s Presentationalism
Because of the (presentational) nature of clarity, to the extent that you are perceiving p clearly (i.e., to the extent that p is presented to you as true) you thereby have a reason to assent to p, and so you should assent to p—unless this reason is defeated by a reason for doubt.Footnote 5
Presentationalism makes the intuitive claim that to the extent you clearly see or feel that you have hands, you thereby have reason to judge that you have hands. Since the clarity of your experience is sufficient to give you this reason, nothing further is necessary. In particular, higher-order reflection—what Gomes calls “self-consciousness”—is not necessary. This is a same-order theory of reasons. In order to have reason to assent to p, your perception as of p must be clear—but you do not need to reflect on your perception so as to judge, or even consider, whether your perception is clear, or whether clarity is a guide to truth, or any other higher-order proposition.
2.3. In ordinary life: Commonsense realism
Since clarity comes in degrees, so too does the quality of reason it provides. When you are looking at something from afar and it’s only somewhat clear to you that it’s a deer, you have some reason but not very good reason to judge that it’s a deer. When you get close and it becomes clear to you that it’s a deer, you now have much better reason to judge that it’s a deer.
Even so, vision does not afford perfect clarity, so the kind of reason it gives you is accordingly imperfect. Imperfect reasons for assent can coexist with reasons for doubt. This clash is illustrated at the end of Meditation One:
I shall never get out of the habit of confidently assenting to these opinions, so long as I suppose them to be what in fact they are, namely very plausible opinions—opinions which, despite the fact that they are in a sense doubtful, as has just been shown, it is still much more reasonable to believe than to deny. (AT 7:22*)
The meditator’s visual experience that she has hands, for example, is very clear—very “plausible” (it appears true)—and so provides very good reason for assent, making it “much more reasonable” for her to judge that she has hands than to judge that she does not.
However, it strikes her as possible that a supreme deceiver is deluding her with such perceptions, so she also has a reason for doubt. Thus, she experiences a conflict between her reason for assent and her reason for doubt.
Imperfect Reasons
Imperfectly clear perception provides an imperfect reason for assent—one which can coexist with a reason for doubt.
Which reason determines what one should do? In Descartes’s view, that depends on which context one is in—ordinary life versus pure inquiry—as set by one’s aim.
In “ordinary life,” we aim to survive and attain various practical ends, so we need to rely on our senses. The senses cannot be perfectly clear, but they are typically “clear enough” for “moral certainty,” which is “sufficient for action” and “for the conduct of life.” If you don’t assent to any appearances, you cannot act (this is the ancient “argument from inaction” against skepticism). If it clearly appears you’re in the path of a barreling truck, you have “moral certainty”—this became the legal standard of certainty “beyond reasonable doubt” (Whitman, Reference Whitman2008)—and “no sane person” would stand there suspending judgment just because it seems logically possible that an evil demon is at play (5O/R, AT 7:350-1).
Practical considerations are relevant on this picture, but they do not form an alternative basis for assent. Descartes is not claiming that assent is “practically required” in the sense of Gomesian faith. Nor is he a Pragmatist like William James who claims you may be rationally permitted or even obliged to assent to a claim on the basis of prudential reasons absent sufficient evidence. Descartes is an Evidentialist: he thinks you are permitted to assent to a claim only when you have sufficient evidence for it, provided by sufficient clarity. But there is a further question as to how much evidence is needed to justify assent in a given case, and the practical considerations of ordinary life simply lower the bar. The fact that you aim to survive and pursue other practical ends is not non-evidential justification for assent. Instead, it lowers the threshold for the amount of evidence you need to justify assent.Footnote 6
In ordinary life, then, Descartes endorses a clarity-based form of:
Commonsense Realism for Ordinary Life
When we have very clear sensory perceptions as of the external world, we thereby have very good reason to make judgments about the external world—and in ordinary life, it would be irrational to suspend judgment just because the deceiver hypothesis raises a “slight and metaphysical” reason for doubt.
This is not a move “from self-consciousness to objectivity.” Our belief in the external world is justified at the first-order level, by the clarity of our senses.
Descartes’s Commonsense Realism prefigures that of G.E. Moore (Reference Moore1939, Reference Moore, Ambrose and Lazerowitz1959). But while Moore is often dismissed for being stubbornly dogmatic, Descartes is more subtle: he acknowledges there’s a time and place—“once in life”—to take skepticism seriously.
2.4. In pure inquiry: An argument from self-consciousness
“Once in life,” Descartes recommends, you should go on a Meditations retreat where you suspend all practical concerns and aim for perfect certainty beyond the slightest scintilla of doubt. Given this aim, any reason for doubt means you should doubt. So although the senses are typically clear enough for practical purposes, they do not meet the meditator’s exacting bar. Perfect certainty comes only from perfect clarity, which comes only from the intellect, not the senses.
Introspection is always intellectual, even when directed at one’s senses, so it can be perfectly clear. Typically, it is not, however, because it’s confused with, and obscured by, the senses (Paul, Reference Paul2018). So you need to make introspection perfectly clear—clear and distinct—by “sharply separating” it from the senses. The way to do that is by doubting all your senses. This is achieved through the method of doubt in the First and Second Meditations, so that in the Sixth Meditation, Descartes’s proof of the external world begins with an introspective claim:
Therefore,
Unlike Commonsense Realism, this is an argument from self-consciousness to objectivity. Premise (1) is about the senses, but not based on the senses. It’s based on perfectly clear introspection of the senses. So for Descartes, every premise and hence the conclusion can be grasped through perfectly clear intellection and thus with perfect certainty. However, I agree with Gomes that the argument fails, though for different reasons: Descartes does not succeed in establishing premises (2) and (3), so we do not really get perfect certainty of the external world.
But thankfully, we do not need perfection. Nor do we need an argument from self-consciousness aiming at perfection. Without the perfectionism of pure inquiry, Commonsense Realism remains intact: in ordinary life—when we actually need our senses to survive and navigate the world—they are clear enough. So instead of letting perfection be the enemy of survival, we should satisfice. To adapt a clever analogy Gomes uses for a different point (p. 162): perfect certainty is like the ability to fly; it would be nice to have it, but the fact that we do not is no reason to mourn.
3. Cartesian Clarity for Agency
Gomes asserts, “Our status as agents does not show up in our experience” (p. 102), inferring this from the premise that experience or perception is necessarily passive or receptive. The only way he considers rejecting the conclusion is by denying that premise—by positing, despite the passivity of most perception, an exception for “agential awareness”:
And someone who thinks there is such a thing as agential awareness is likely to insist on the differences between our receptive knowledge of the rest of nature and our agential awareness of our own activity. Both are forms of experiential awareness, but the latter is somehow distinctively active or non-observational in a way that the former is not. (p. 94, my bold)
Descartes agrees with Gomes’s premise: we are active in willing, passive in perceiving. But it doesn’t follow that we don’t experience our own agency, as Gomes holds; nor that our experience of agency must itself be agential. Both views conflate the state of perceiving with the object perceived. Gomes assumes that if perceiving is passive, what is perceived must also be passive; the alternative assumes that if we perceive agency, perceiving itself must be agential. The solution lies in distinguishing the state of perceiving from the object perceived: perception is passive even when what we perceive is active, as when we introspect our own acts of will.
We do experience ourselves as agents. That’s why hard determinism feels so unsettling. If you showed me evidence that I cannot fly, I’d shrug: I never thought I could. But if you offered reasons to think I do not make my own choices and judgments, I’d be deeply troubled, because my inner sense of agency is so persistent and profound—and I presume yours is too.
Consider the contrast:
Case 1: A sudden gust of wind hurls you off the dock into the lake.
Case 2: It appears to you that it would be fun to dive into the lake, and—for that reason—you choose to dive, and you do so.
In Case 1, you experience being hurled as something done to you, not by you; you feel passive, coerced. In Case 2, you experience both choosing to dive and diving as things you are doing—you feel active and free.
To say that agency “does not show up in our experience” is to claim that diving by choice feels no different from being hurled by the wind. But clearly it does. And on Descartes’s Presentationalism, these different experiences provide evidence for different judgments: in Case 1, it’s quite clear to you that you are not the agent of your motion, so you have good reason to believe you are not; in Case 2, it’s quite clear to you that you are the agent of your choice and action, so you have good reason to believe you are.
Even though perceptual appearances are entirely passive—things simply strike us a certain way—we retain an active power to respond to them. In Case 2, the appearance that diving would be enjoyable is not something you generate; it presents itself to you, giving you a reason to dive. Yet your assent to that appearance—choosing to dive—is a voluntary mental act. The physical dive is voluntary only because it enacts this prior mental action. In this sense, judging and choosing share the same basic structure: each is an act of assent to an appearance, differing only in whether the content is descriptive or prescriptive. When you assent to a descriptive appearance, you judge; when you assent to a prescriptive appearance, you choose.
But some philosophers who readily affirm our practical agency nevertheless deny that we possess doxastic agency. They hold that although we act freely—and are therefore responsible for what we do—we lack comparable freedom or responsibility with respect to our judgments. Some claim that introspection shows a sharp contrast: they feel agential when choosing what to do but passive when forming judgments about how things are. And they presume you experience the same. As you look at the page before you, it may seem that you cannot help but judge that a page is there, whereas it still feels as though you retain control over whether you keep reading it.
Descartes would diagnose this failure to notice your own doxastic agency as a confusion. Because the appearance is passive, if you fail to distinguish it from your assent, you will naturally mistake your assent for something passive as well. In introspection, the active element of assent is thus obscured by the passive perception with which it is confused. To introspect your assent clearly and distinctly, you must distinguish it from your perception. And the way to do that is to withdraw assent from the appearance, i.e., to doubt. Let us see this in action.
Even with practical choices, it is easy to overlook our agency. Much of our choosing and judging occurs automatically or habitually, with little reflection. When we move fluidly through our surroundings—responding to what they afford without conscious deliberation—we notice the objects that appear, not the agency with which we respond. Clarity requires attention: you cannot perceive something clearly unless you focus on it, so you cannot notice your agency clearly unless you introspect it. When absorbed in your environment, your own acts of assent recede from view.
Deliberation brings them into sharp relief. When you pause to weigh competing reasons, you enter a state of doubt in which assent (choice) is temporarily suspended. In this metacognitive posture, you attend not just to how things appear but to your own power to endorse or withhold endorsement. This experience makes the distinction between appearance and assent vivid. Two reasons pull in opposite directions—diving looks fun, diving looks dangerous—and in holding them together, you feel that it is up to you which way to go. This phenomenology of two-way power is a direct experiential manifestation of agency. Your freedom and responsibility for giving and withholding assent now become introspectively clear.
Doubt illustrates the same point in the domain of judgment. When you discover that a red sphere is lit by red light, the sphere still looks red, but you withhold assent from the judgment that it is red. The appearance stays the same; the judgment changes. As we’ll see in a moment, Descartes emphasizes this in Principles I.6: even if an evil deceiver manipulates all your sensory appearances, you remain free to avoid error by withholding assent. Here too, doubt forces you to distinguish the passivity of appearance from the activity of assent, revealing your agential power in thinking. Again, your doxastic agency becomes evident to you through the self-conscious activity of doubt.
This suggests a new way of seeing the grain of truth in the second part of Gomes’s view. Gomes is right: when I engage with an interlocutor who challenges my beliefs, this does reinforce within me the sense that I’m responsible for my beliefs. But pace Gomes, this is not a matter of faith. It’s a matter of evidence, provided by the arrival of introspective clarity. When other people challenge our beliefs, they prompt us to deliberate and doubt—and in doubting, our freedom and responsibility for giving and withholding assent become introspectively clear.
This is why it is precisely at the climax of his method of doubt—when he has withdrawn assent from all his sensory perceptions—that Descartes remarks on his manifest agency as a thinker. Right before asserting “I am thinking, therefore I am” in Principles I, article 7, which Gomes cites as the target of Lichtenberg’s critique, Descartes says this in article 6:
We have free will, enabling us to withhold our assent in doubtful matters and hence avoid error. But whoever turns out to have created us, and however powerful and however deceitful he may be, in the meantime we nonetheless experience within us the kind of freedom which enables us always to refrain from believing things which are not completely certain and thoroughly examined. Hence we are able to take precautions against going wrong on any occasion. (AT 8A:6, italics in original, my bold)
An evil deceiver might inflict me with illusions, but he cannot force me to assent. There’s a feat of Cartesian jujitsu here: by hypothesizing the deceiver I find reason to doubt all appearances and—through doubting—I manifest my freedom, which he cannot touch. By contemplating the deceiver’s supreme power, I gain evidence for my own.
How does this evidence fare against Lichtenberg’s hard determinist challenge? Descartes himself thinks he introspects his freedom with perfect clarity, which precludes any reason for doubt and shows there must be a flaw in the hard determinist’s argument. He would also claim to identify the flaw. He thinks he has decisive evidence for dualism according to which our minds, including our wills, are not part of the deterministic order of nature.
But you needn’t go that far. You might satisfice. Giving up on perfect certainty, you might embrace commonsense realism about your freedom parallel to Commonsense Realism about the external world. You might say that your inner experience of freedom is decidedly clearer than any metaphysical argument to the contrary (cf. Van Inwagen, Reference Inwagen1983). It’s not that you have no reason for doubt, but that freedom wins on the balance of evidence.
Finally, you might be even more modest and call it a draw: the evidence for and against agency is equally clear. Remember that this antinomy is the inoculation against bad faith that Kant has and Gomes lacks. It would shore up Gomes’s claim that “as far as the evidence is concerned, we have to suspend judgement,” which is the core of premise (1) of his argument for faith in agency. For faith to justifiably break the tie, Gomes would still have to resolve the dilemma for his “practical requirement” in premise (2). But that step is moot unless he has evidence for agency to counterbalance his evidence against it—evidence provided by his opponent, Descartes.
So much for agency. What about passivity?
4. Cartesian Clarity for Passivity
While Gomes duly worries about whether we are active in judging, he takes it for granted that we are passive in perceiving. Why should we believe the latter? Here again, Descartes’s Presentationalism answers: because through introspection we clearly experience our passivity in perceiving.
You experience yourself as being able to influence your perceptions indirectly—for example, by manipulating objects in view (moving your hands before you), shifting your gaze, redirecting attention, dimming the lights, putting on glasses, or taking a psychedelic drug. But when those factors are held fixed, it does not feel like you can directly control how the world appears. In Case 2, where it appears that there is a lake and that it would be fun to dive into it, you cannot alter that appearance at will—nor does it feel as though you can. By contrast, you experience sensory objects as “adventitious,” given from without. Indeed, you must feel receptive in this way for the objects of your perception to appear real (recall §2.1), as existing, to borrow Gomes’s lovely phrase, in “a world that is not of your making” (p. 1). Contrast imagination. When you fantasize about your ideal vacation, adding palm trees and piña coladas at will, the scene both is and appears to be “invented” by you, as Descartes would say (AT 7:38, 312). You have good reason to judge that you are active in fantasizing and passive in sensing because you introspect these things clearly.
Now, while Lichtenberg alludes to the argument for hard determinism to challenge our agency in judging, you might think there’s no corresponding challenge to our passivity in perceiving. But—as Lichtenberg recognizes—there is, specifically concerning our creative ideas. He is not merely concerned with how we can be the “origin” of our judgments, which are typically mundane. As a writer and scientist with creative ambitions, he wonders more deeply how we can be original in coming up with our own ideas:
When I write about something, the best things always occur to me in such a way that I cannot say from where. [K 183, quoted in Hennig, “Lichtenberg’s Point,” 271.] …
There is an element of chance, it seems, in all discoveries, even those we seem to arrive at through effort. Putting things already discovered into the best order, even the leaps of discovery seem just as little the work of our free will as is the beating of our heart—What I have said elsewhere belongs here: we should not say “I am thinking” but “it is thinking,” just as one says “it is lightning.” (L 806, translation amended)
Lichtenberg’s “lightning” is the flash of original insight, the leap of discovery, the idea that pops up “I cannot say from where.” These passages are not about doxastic agency, which we seem to exercise in giving or withholding assent from ideas we already have. They’re about creative agency in originating new ideas.
Lichtenberg stands in a long tradition of thinkers who have wondered how creative ideas occur to people unbidden: “Eureka!” “Aha!” Like being struck by lightning. Or thrown by the wind.
For Plato, the “wind” comes from the Muse inspiring—literally “breathing into”—the poet, who is merely a passive vessel. With Longinus and many others including Kant, the Muse gets naturalized: inspiration comes from innate talent within, but still bypasses the will (Paul, 2026). Being inspired with new ideas accordingly feels like something that happens to you, not something you do.
And yet, the origination of ideas is something we credit people for and regard as self-expression. These considerations suggest that being inspired with novel thoughts is something you do. So how can origination be active and passive at once? This is what I call “The Paradox of Creative Agency” (Paul, Reference Paul, Kind and Langkau2026), though I am highlighting the passive side of it here.
This problem for creative passivity inverts the problem for doxastic (and moral) agency:
-
• Doxastic agency: When judging and doubting, we feel like agents, but there are reasons to believe we are passive (i.e., the argument for hard determinism).
-
• Creative passivity: When new ideas pop up, we feel passive, but there are reasons to believe we are agents (i.e., creativity is praiseworthy and self-expressive).
Lichtenberg worried about both. Just as hard determinism raises the Epistemic Problem for agency, the fact that creativity is praiseworthy and self-expressive raises an Epistemic Problem for passivity. Descartes’s Clarity First epistemology gives us resources to address both, to account for at least prima facie evidence for our agency and our passivity—and for the objectivity of the world—insofar as we experience these things with considerable clarity. There are counter-arguments in each case, but their rational weight, too, will be a matter of how clearly we grasp them.
And we should welcome the doubt. More broadly, we should be grateful for the Socratic gadflies who stir us out of complacency and credulity into searching, self-conscious deliberation, as Gomes does so creatively in his inspired book. For in that state, in the vertigo of suspension, our agency as thinkers becomes evident to us, and this evidence is vital—if only as a prerequisite to faith.Footnote 7
Elliot Samuel Paul is an Associate Professor of Philosophy at Queen’s University in Canada. His book, Clarity First: Re-envisioning Descartes’s Epistemology is forthcoming with Oxford University Press, and he is co-authoring a new book on creative agency.
Open access
