1 Introduction

What are the aims and norms of our interrogative attitudes? An interrogative attitude, say, being curious about why the sky is blue, is directed at answering a question; in this case, the question ‘Why is the sky blue?’ But when exactly does this attitude achieve its aim?Footnote 1

According to one popular view, the aim of our interrogative attitudes is knowledge. That is, what we aim at when we are in an interrogative attitude is knowledge. Knowledge can be naturally seen as the state that closes an inquiry, a state that satiates an attitude of, say, curiosity. When we inquire, we seek to know the answer to the relevant question. And once we come to know the answer to the question we were curious about, our interrogative attitude appears to have achieved its aim. However, a number of counter-examples and theory-based considerations have moved some philosophers to question the knowledge aim of inquiry. Epistemologists have suggested that sometimes we seek certainty or additional confirmation of an answer when we open up an inquiry (assuming that one can know without being fully certain). Furthermore some theorists think that sometimes interrogative attitudes aim at understanding rather than mere knowledge (assuming that understanding goes beyond knowledge).

In the light of such a disagreement about the possible aims of inquiry, one might be tempted to consider a pluralist option. That is, given the apparent plausibility of each of the options, one might be tempted to conclude that all of the above mentioned are genuine aims of our interrogative attitudes. On such a pluralist view about the aims of inquiry, sometimes we aim at mere knowledge when we inquire, sometimes we aim at something more demanding, e.g., certainty, at other times still, we aim at understanding (assuming it is not mere knowing).

Despite its intuitive appeal, however, the simple pluralist view about the aims of inquiry faces a substantial problem. It faces what I will call the problem of the standard pluralist dilemma. On the one hand, it has to explain what makes these various aims, aims of inquiry specifically. That is, a proponent of pluralism has to explain what is the common element in all these possible aims that makes them aims of inquiry (of interrogative attitudes) and prevents them from being a totally unconnected plurality of aims of, say, unrelated attitudes we could have. On the other hand, the proponent of pluralism has to make sure that the proposed explanation of the common element in these various aims that our interrogative attitudes might have is not unificatory enough. That is, the explanation has to remain pluralist (if one wants to defend pure pluralism) and not collapse into a monist view according to which there is one fundamental aim of interrogative attitudes after all. For a monist can insist that the fundamental aim is captured specifically by the common element that we have to provide if we think that the aims of inquiry are aims of one sort of phenomenon and not of totally disconnected attitudes.

As far as I can see, the problem of the standard pluralist dilemma has not yet received an in-depth treatment in the recent debates on inquiry. It is a serious lacuna that risks undermining the whole pluralist approach to the aims and norms of inquiry. A theorist that is moved by the appeal of the pluralist intuition has to provide an answer to the standard pluralist dilemma. Failure to do this presents the risk that one’s pluralism remains ad hoc and lacks theoretical motivation beyond the need to respond to counter-examples to the knowledge aim of inquiry.

The aim of the present article is to provide a solution to the standard pluralist dilemma as it concerns the aims of inquiry. The solution that I will propose in what follows can be called a version of a moderate pluralism. In short, it will consist in embracing one horn of the standard pluralist dilemma and accepting that we have to identify a common element in all the possible aims of inquiry if we want to theorize about aims of inquiry specifically, and not of, say, farming or digesting. I will elaborate on the common element in all aims of inquiry. I will bite the bullet in response to the other horn of the dilemma, however. I will accept the conclusion that the view I will develop is not a pure pluralism, since it does appeal to a common element in all aims of inquiry. At the end of the day, the view is a monist view at the most fundamental or abstract level. However, I will resist the conclusion that the view I develop is not a pluralist view at an intermediate level of abstraction. Indeed, I will show that inquiry has a plurality of aims, e.g., knowledge, certainty, understanding, that are unified at a more abstract level. That’s why the resulting view can be best characterized as a moderate pluralism about aims of inquiry.

Before moving on, I should note that the question about the aims of inquiry is closely related to the question about the fundamental norms of inquiry. And this also leads us to the closely related question: should we accept the existence of a plurality of norms of inquiry? The background idea here is that the (substantive) aim of inquiry specifies a fundamental norm of interrogative attitudes: the norm that specifies the internal standard that can be used to assess interrogative attitudes. One popular idea that goes hand in hand with the knowledge aim of inquiry is that ignorance is the fundamental norm of interrogative attitudes. Roughly, according to the ignorance norm, one should not be at the same time in an interrogative attitude about a given question and know the answer to this question. Now, if we accept that there might be other aims of inquiry (and here we assume that the aims are supposed to be constitutive of the sort of attitude that interrogative attitudes are), then it seems we should also accept that there is a plurality of fundamental norms of interrogative attitudes. In short, the standard pluralist dilemma appears to concern not only the question of the aims of inquiry, but also the topic of the fundamental norms of inquiry. Thus the solution to it has to apply to both, the aims and norms of inquiry.

To put my cards on the table from the outset, the more concrete solution that I will elaborate on in what follows is to suggest that the most general aim of inquiry is to find reasons to F (e.g., to believe), where reasons are appropriate answers. And the most fundamental norm of inquiry is the norm that proscribes being in an interrogative attitude while already having the relevant reasons. When we ask questions, we are sometimes asking for arguments and premises for good patterns of reasoning, and we are sometimes asking for explanations. The aim of good arguments is to lead us to knowledge in most cases, or to certain knowledge in some cases. The aim of a good explanation is to lead us to a better understanding. We can explain how knowledge, certainty, and understanding can all be more specific aims of the general aim of finding the relevant reasons. That is, if we also assume that reasons are appropriate answers to the relevant questions (cf. Hieronymi, 2005; Logins, 2022).

In what follows, I will develop all of the above claims and assumptions in more detail so that the distinct parts of the whole theory are articulated to fit together in an overarching picture. More concretely, section two elaborates on the appeal of pluralism and the dilemma that it faces. I consider there in more detail the view according to which knowledge is the aim of inquiry and ignorance is its norm. I also present two recent worries about the view and how these worries might motivate one to move towards a pluralist picture about the aims and norms of inquiry. Section 2 concludes by elaborating on how the standard pluralist dilemma is inevitable for a simple pluralist view and how a desideratum of reasons-responsiveness in inquiry should motivate our way out of the dilemma. Section 3 introduces a theory of reasons and elaborates on how it can help us to specify the aims and norms of inquiry by appeal to reasons in a way that could solve the dilemma. Section 4 works out the details of our new proposal and explains the relevant cases and considerations. Section 5 concludes.

2 Knowledge, pluralism of aim and norms, and a dilemma

This section elaborates on the intuitive appeal of pluralism about the aims and norms of inquiry and shows how a simple pluralist view leads inevitably to a problematic dilemma. I first introduce and detail the motivation for the view according to which knowledge is the aim of inquiry and ignorance is its norm. I then show that this view faces some serious worries and counter-examples that have recently motivated philosophers to look for alternatives. I then introduce more detail about the standard pluralist dilemma for the pluralist view of the aims and norms of inquiry. Finally, I present a desideratum that should help provide a solution to the dilemma.

2.1 The appeal of pluralism: knowledge and other aims

According to a popular view, knowledge is the aim of inquiry and ignorance is its norm. The idea that knowledge is the intrinsic aim of inquiry can be summed up by the following principle:

  • The knowledge aim of inquiry (KAI): for any inquiry E concerning question Q, the goal of E is to get to know the answer to Q (see Williamson, 2000; Whitcomb, 2010; Kappel, 2010; Kelp, 2011, 2014, 2021a, b, c; Millar, 2011; Rysiew, 2012; Friedman, 2017; van Elswyk & Sapir, 2021, among others for endorsements of KAI).

One major motivation for (KAI) comes from a general observation that inquiry aims to close the relevant question under discussion. Specifying this general aim of closing the question (answering the question) can be seen as characterizing the aim of inquiry in a light-weight way (cf. Kelp, 2021c). The most plausible way to specify this aim in a more substantive way, according to this line of thought, is to appeal to knowledge. In short, nothing short of knowledge of the relevant answer is satisfactory when we aim to close/answer a given question.

To give further support to the idea that knowledge is the (substantive) aim of inquiry, Chris Kelp has recently proposed a number of cases that seem to favor knowledge over alternative accounts of the aim of inquiry (see Kelp, 2021). Among these are the cases of reward for a successful inquiry, commitment release (when hired to answer a question), and progress of inquiry. Without going into details of the examples, we can observe the common element in all of these: having a true or even a true and justified belief doesn’t seem to be enough in these situations; knowledge is required for one to be entitled to a reward, to be released from a commitment and so on. The best explanation of these results seems to be that knowledge is the constitutive aim of inquiry.

If one is attracted to the knowledge aim, then one might be naturally led to the ignorance norm of inquiry:

  • The ignorance norm of inquiry (INI): where p/not-p is a complete answer to a question Q, one ought not: have an inquiring attitude towards Q at time t and know that p/not-p at t (cf. Whitcomb, 2017; Friedman, 2017; van Elswyk & Sapir, 2021; see also Palmira, 2020, for discussion and for a permissibility version of the norm).

A popular line of defense of (INI) is to appeal to our ordinary language use and common-sense judgments about inquiry and to claim that (INI) is the best explanation for these. Consider, for instance, the following assertion: “I know there’s oat milk in the store, but I’m curious/wonder if there’s oat milk in the store” (cf. van Elswyk & Sapir, 2021). This assertion sounds very odd. (INI) explains why. The state of affairs that it seems to express is prohibited by (INI): the norm tells us that one should not at the same time inquire (be curious/wonder) and already have the knowledge of the answer to the relevant question. Indeed, (INI) seems to vindicate our common-sense judgments about cases that would correspond to the above assertion. There seems to be something untoward in a case where, say, I see the oat milk section in front of me in the grocery store and thus know that the store has oat milk, and yet, at the same time, I wonder or am curious whether the store has oat milk. This untowardness is predicted and explained by (INI): one ought not to be in an interrogative attitude and, at the same time, know the answer to the relevant question.

The knowledge aim and the ignorance norm seem clearly to capture something important about interrogative attitudes. However, recently, a number of authors have observed that they cannot be the whole story. Indeed, there seem to be cases of inquiry where we aim at something beyond mere knowledge. Moreover, theory-based considerations about our fallibility indicate that it can be appropriate to continue inquiry even when one already knows the relevant propositions, since one might want to be certain about it (assuming that we can know that p without being certain that p). The remainder of this subsection elaborates on two worries in particular, and shows how these considerations seem to motivate a pluralist picture of the aims and norms of inquiry.

The first worry for the knowledge-centered approach is a theory-driven concern that arises from considerations about our fallibility and the need for certainty in certain contexts [a version of this line of objection appears in Woodard (2022), Falbo (2021, 2022), Beddor (forthcoming)].Footnote 2 It is theory-driven insofar as it relies on a very popular and plausible principle, namely, fallibilism about knowledge. We can specify fallibilism roughly as follows.

  • Fallibilism: S can know that p, without p being guaranteed to be true (or certain for S) based on the evidence S has for p (cf. Brown, 2018, for one recent defense of fallibilism).

Fallibilism enjoys great popularity as a response to radical skepticism (cf. Unger, 1975). In situations where being right about things is particularly important, say, from a practical point of view, it seems that double-checking and continuing or reopening an inquiry is not only permissible but the appropriate thing to do. Consider the following well-known case provided by Jessica Brown:

SURGEON

A student is spending the day shadowing a surgeon. In the morning he observes her in clinic examining patient A who has a diseased left kidney. The decision is taken to remove it that afternoon. Later, the student observes the surgeon in theatre where patient A is lying anaesthetised on the operating table. The operation hasn’t started as the surgeon is consulting the patient’s notes. The student is puzzled and asks one of the nurses what’s going on:

Student: I don’t understand. Why is she looking at the patient’s records? She was in clinic with the patient this morning. Doesn’t she even know which kidney it is?

Nurse: Of course, she knows which kidney it is. But, imagine what it would be like if she removed the wrong kidney. She shouldn’t operate before checking the patient’s records.

(Brown, 2008, p. 176).

The Surgeon case was brought up initially in the debate on pragmatic encroachment (the idea that whether one knows that p depends in part on what is practically at stake on one being wrong about p), and in particular against the view that knowledge is the norm of practical reasoning/action (see Hawthorne, 2004; Stanley, 2005; Hawthorne & Stanley, 2008, among others). Assuming fallibilism about knowledge, we can accept that the surgeon knows the relevant facts. Crucially, would there be anything wrong with the surgeon reopening the inquiry? That there is nothing wrong with the surgeon reopening an inquiry about Q while knowing the answer to Q constitutes a counter-example to the ignorance norm of inquiry. It also seems to imply that knowledge is not always the aim of inquiry. That is, it implies either that there can be several aims, and both knowledge and certainty are the aims, or that knowledge is not the aim, if there is no radical plurality of aims. What exactly is the aim of inquiry in the Surgeon’s case? A natural answer would be to say that the surgeon aims to be certain that p, where p is the answer to the relevant question Q.Footnote 3

The second problem case for the knowledge-centered approach comes from considerations involving another much-discussed epistemic state, namely, the state of understanding.Footnote 4 It appears that in many cases, we continue inquiring because we want to gain a better understanding of what we already know. Here is one recent example of such a case from Elise Woodard:

MATHEMATICAL PROOF

Indeed, such cases sometimes occur in proof-checking in mathematics. For example, the great mathematician Michael Atiyah once reported having proven a theorem-thus knowing that it held-while simultaneously seeking to understand why it held (Woodard, 2022, p. 3).

One might, however, wonder whether this example constitutes a genuine counter-example to the knowledge-centered approach. For one thing, one might be worried that when we look closely enough into the details of the case, we realize that the question that the mathematician is supposed to investigate here (‘Why does the theorem hold?’) is not the same as the question to which the mathematician is reported to know the answer (‘Does the theorem hold?’).Footnote 5

However, we can avoid the above worry altogether by specifying a case where the question that one is investigating in trying to get a better understanding is not different (at least at the surface level) from the question to which one already knows the answer. Here is one such attempt:

THE BLOB

I know that the blob, Physarum polycephalum, a single-celled species of slime mold, without a brain or central nervous system, has the ability to learn (and has the ability to transmit information), but I don’t understand how this is possible. I continue to be curious. I learn that the blob has the ability to learn because it has repeatedly shown that it can modify its path upon coming across an obstacle in the next attempt to get to the destination. I know why the blob can learn. Roughly, that the blob can learn follows from two plausible claims, namely, that it can modify its path and that if an organism modifies its path, then it can learn. But I am still curious; I still continue to read and follow the research on why the blob can learn. I aim to understand why the blob can learn—How come it has this ability?

It would seem that in this case, the question is the same—‘Why can the blob learn?’Footnote 6 I know the answer to this question - it can learn because it has shown (repeatedly) that it can integrate information about obstacles and modify its behavior accordingly. But I don’t yet fully understand why the blob can learn.Footnote 7 I am looking for a more fundamental explanation—an explanation that could make sense given what we know about the role of brains and the central nervous system and the typical capacities of organisms without these. I know that the blob can learn because it avoids obstacles, but I am curious why the blob can learn. Crucially, there doesn’t seem to be anything inappropriate with my continuous curiosity. My inquiry doesn’t seem problematic in any way. If these observations are on the right track, then we have here a case where one knows an answer, and yet it can be appropriate for one to continue the inquiry, contrary to what the ignorance norm of inquiry suggests. And this would show again that there might be additional aims of inquiry beyond mere knowledge.

One might think then that a natural reaction to the observations introduced above is to accept that there is a plurality of aims of inquiry and a corresponding plurality of norms of inquiry. After all, we do find it plausible that in many cases knowledge is the aim of our interrogative attitudes (and ignorance is their norm). However, we also find it plausible that on some occasions we aim for something above knowledge. Maybe sometimes we aim for certainty or additional confirmation. And maybe sometimes we aim for understanding. To posit a plurality of aims and norms of inquiry then is a natural move in light of the above intuitions and otherwise plausible theoretical frameworks that seem to pull in different directions (e.g., fallibilism vs. the fundamentality of knowledge). And there does seem to be an emerging tendency among zetetic epistemologists to move in the pluralist direction (cf. Millson, 2021; Falbo, 2021, 2022; Woodard, 2022).

Before moving on to discussing a worry for pluralist accounts, let me insist on why pluralism might seem such an attractive view. Pluralism about the aims and norms of inquiry might appear as a good option, given that assuming that there is only one aim of inquiry and that there is only one norm of inquiry leads to the conclusion that the three views that we just observed are mutually incompatible and yet all seem very plausible. The three views are: knowledge is the aim of inquiry, certainty is the aim of inquiry, and understanding is the aim of inquiry. One might wonder why these views are incompatible. Here is one way of elaborating the incompatibility. If monism is right, then there is only one aim of inquiry. If knowledge is the aim of inquiry and fallibilism about knowledge is true, then there are situations where we are not certain that p, while knowing that p. And in some such situations it seems absolutely appropriate to (continue to) inquire into the relevant question in order to obtain certainty. But if knowledge is the aim, it is not appropriate to inquire into the question. The aim has already been achieved. Moreover, if it is appropriate to inquire into the question when one seeks certainty, then it is plausible to think that certainty is the aim. But if monism is right it cannot be the case that knowledge is the only aim and certainty is also the aim. A contradiction can be derived, it seems. Furthermore, if, however, certainty is the fundamental aim, then there should be situations where one has achieved knowledge but is still not 100 percent certain about the answer. In some such cases, it seems that one has completed the inquiry and achieved its aim (say, one is 99 percent certain and knows that p) and yet, one has not achieved the aim if the aim is certainty (see Beddor, 2013). The two claims stand in a tension. Finally, assuming that understanding doesn’t reduce to knowing a single proposition, or being certain about a single proposition, there will be situations where one has one of these states, but not others. One putative view of aims and norms will predict that one state is the aim, whereas other views will predict that the other state is the aim. Thus, it seems that, assuming monism, the three views of aims and norms are mutually incompatible.Footnote 8

2.2 The standard pluralist dilemma

The pluralist picture might appear as the most plausible one, in the light of the above cases and considerations. However, anyone who is tempted by a pluralist approach has to be sure to avoid what I call the standard pluralist dilemma. The need to avoid the pluralist dilemma applies in the case of the aims and norms of inquiry as well. What I call the standard pluralist dilemma is the following challenge: if there is a plurality of clearly distinct things of one sort, then what is their unificatory principle? (the need to justify the unity in plurality: what’s the common element? The difficulty here is that by attempting to identify the common element, we might end up with an overarching monist view after all); but if we give up the attempt to find a common element in the plurality of the relevant phenomena, then how do we explain that the relevant phenomena do appear to have something in common? (the challenge here is to provide a plausible error theory that would explain why we are so massively wrong in thinking that the relevant phenomena have something in common). This subsection elaborates the dilemma in more detail.

To see the general structure of the dilemma, consider a case where it can be easily met. It is common knowledge that democracy comes in various forms. Yet we might imagine someone challenging this by raising a standard pluralist dilemma. One might say that well, either all forms of democracy have a unifying element, in which case, a pluralist about democracy has to explain why this common element is not enough for endorsing a monist view of democracy. On the other horn, if the proponent of pluralism about democracy thinks that there is no fundamental common element in forms of democracy, then they have to explain why it appears to us that there is something unifying in some forms of government, namely, why it makes sense for us to talk about forms of democracy rather than forms of governance in general. In short, on this horn, one has to explain why we are wrong in putting various forms of democracy under one general label of democracy. In other words, if one takes the radical pluralist horn, then one has to provide an error theory that could explain why we are so massively wrong in thinking that there are forms of democracy.

It seems that in the case of democracy the challenge can be met by insisting that there is indeed a common element in forms of democracy (taking the first horn of the dilemma) and then by insisting that the apparent plurality can be maintained at the more fine-grained level of a theory of democracy. In short, the solution seems to amount to endorsing a moderate pluralism where we have an independently plausible theory of both the unity (democracy) at a more general level and of the plurality at a more fine-grained level, by maintaining, say, that democracy just is the power of a majority in a group and then explaining that this power can be implemented in a number of distinct ways (parliamentary representation, direct democracy, etc.).

It can also be useful to see a case where the dilemma cannot be easily met. Here is one tentative proposal. Consider a putative theoretical move in contemporary epistemology where, in light of persistent Gettier counter-examples, the theorist postulates a plurality of propositional knowledge.Footnote 9 Crucially, it is postulated that one sort of propositional knowledge avoids Gettier counter-examples. Clearly, the motivation for such a putative pluralism is based in the mere need to avoid Gettier cases and, if nothing more substantial is proposed, one’s pluralist theory of propositional knowledge appears purely ad hoc. Suppose that the theorist postulates a common element in propositional knowledge and then postulates the plurality at a more fine-grained level of theorizing. However, the worry doesn’t go away. Without a further theoretically independent motivation for such a moderate pluralism, the theory just appears arbitrary. After all, what explains this alleged plurality at a more fine-grained level? It would seem that such a simple pluralist view cannot respond appropriately to the standard pluralist dilemma.

Now, if there is one point I would like readers to take away from the present article, then it is that merely postulating a pluralism about the aims and norms of inquiry in the face of counter-examples to the knowledge-centered approach is not enough. One needs to motivate a theoretically satisfactory response to the standard pluralist dilemma.

Applying the standard pluralist dilemma to the case of the aims and norms of inquiry, we get the following challenge: (i) if there are distinct (constitutive) aims and norms of inquiry, then what unifies them—What makes them all aims and norms of inquiry? In other terms, we have to explain what is the common element in all of these aims and norms. Yet, if we can find a common element, then why can’t we use it in an overarching monist definition of the fundamental aim and the most fundamental norm of inquiry?; (ii) on the other hand, if we give up the idea that there is a unificatory common element in the aims and norms of inquiry, then we need to propose a non-ad-hoc theoretically plausible account of why we are so massively wrong in thinking that there is a common element in all instances of inquiry. Why does it seem that the cases we consider, that is, cases of seeking knowledge, cases of seeking certainty, and cases of seeking understanding, are all instances of the same general phenomenon: inquiry?

An in-depth engagement with the pluralist dilemma about the aims and norms of inquiry still seems to be lacking. Yet, if we embrace pluralism, we owe an explanation of why there is a plurality of aims and norms of inquiry. We have to respond to the challenge of the pluralist dilemma for the aims and norms of inquiry. Without explaining why there is such a plurality, a pluralist approach risks being theoretically unsatisfying and runs the risk of appearing as a mere ad hoc move in the light of an observed conflict of intuitions about cases. In short, we should be satisfied with a pluralism about the aims and norms of inquiry only when we have provided some non-arbitrary theoretical grounds that explain it, that is, only when we can provide an explanation of why there is a plurality of aims and norms of inquiry.

Let me insist that the fundamental problem with the standard pluralist dilemma is not so much the dilemma itself, as the need to deal with it in a theoretically satisfactory and well-motivated way. One has to make sure that the view one is endorsing in the light of the dilemma, say, a pluralist view, is not a mere ad hoc move in the light of conflicting intuitions, but that there is a more fundamental explanation of why there is a plurality even if there is also a common element.

The aim of the remainder of this article is to elaborate and motivate a moderate pluralist approach to the aims and norms of inquiry. To do so, I will endeavor to provide a theory-driven account of why there is a variety of aims and norms of inquiry. The explanation will postulate a common element at a more abstract or general level of explanation of aims and norms of inquiry and will also elaborate on why at a more fine-grained level of explanation we still need to postulate a variety of more specific aims and norms of inquiry.

2.3 Reasons-responsiveness desideratum

In order to move on and attempt to solve the standard pluralist dilemma for pluralists about the aims and norms of inquiry, it would be helpful to fix a criterion that would help us settle the matter. This section introduces one possible attempt to fix such a criterion. More concretely, it introduces a desideratum that any plausible theory of inquiry should aim to satisfy. Making this desideratum explicit may help us to move the debate forward. The capacity to satisfy this desideratum should be a criterion that guides our search for a more fundamental theory of the aims and norms of inquiry. This desideratum appears to be present in the form of an implicit assumption in debates. One place where the desideratum is discussed explicitly is a recent publication by Chris Kelp, who writes:

“One prima facie plausible candidate [of a constitutive norm of inquiry] is that in inquiry one must form or work toward positioning oneself to form beliefs via epistemic abilities.[...] That’s why going to the local brainwashing service and having a certain belief installed will not count as inquiring into the corresponding question” (Kelp, 2021c, p. 368).

One way to understand Kelp’s claim here is that an account that predicts that installing a belief through a procedure of brainwashing or similar can count as an inquiry should not be accepted. Inquiry cannot be a mere acquisition of beliefs, even if true and justified. According to Kelp, this no-brainwashing (not mere belief acquisition) desideratum is best satisfied by an account that focuses on epistemic abilities. Yet he recognizes that alternative accounts could also satisfy it. For instance, he writes:

“While I think this is roughly right, alternatives are conceivable. Evidentialists, for instance, might say that inquiry essentially involves gathering evidence (and that’s why using the brainwashing service in the case below doesn’t qualify as inquiring)” (Kelp, 2021c, p. 368, fn. 6, my emphasis).

But we might wonder how exactly we should understand the no-brainwashing desideratum that Kelp refers to here. If the desideratum in its most general form is not tied specifically to the virtue-theoretic notion of epistemic abilities, nor to the evidentialist requirement that involves an appeal to having evidence, then how could we describe this desideratum in general?

I would suggest that the more general way of understanding the no-brainwashing desideratum can be made more precise by an appeal to reasons. More precisely, the suggestion is that genuinely inquiring agents are responsive to (normative) reasons. The difference between someone who just instals a belief in their head by visiting a brainwashing service and someone who arrives at the belief through an episode of genuine inquiry is that the latter and not the former is responsive to reasons there are to believe the relevant proposition.

A person who is genuinely inquiring has to be able to be moved by reasons in the sense of basing their attitudes/actions on considerations that genuinely speak in favor of the relevant response (where basing might be understood causally with no need for extra demanding meta-cognitive states). Whether we want to further explain this responsiveness to reasons in terms of abilities or evidence gathering is a further question, a question upon which more fine-grained accounts might disagree. For instance, a proponent of a virtue-theoretic account might insist that this responsiveness to reasons is nothing above and beyond having an epistemic ability, whereas an evidentialist might assert that all reasons to believe are evidence and that by definition being responsive to reasons to believe amounts to having or gathering evidence. In what follows, I will appeal to the most general characterization of the no-brainwashing desideratum. According to this characterization of the desideratum, then, a plausible account of the aims and norms of inquiry has to predict that inquiring agents are responsive to reasons, where being responsive means at least that the agent can, in a sense, base their response on the reasons that speak in favor of the response in question.

3 Reasons and inquiry

3.1 Question-centered view of reasons

The no-brainwashing desideratum, understood as a desideratum that a theory of the aims and norms of inquiry should explain why only subjects who are responsive to reasons can count as genuinely inquiring, can help us to move the debate forward. The idea that I will develop in the rest of this paper is that focusing on normative reasons can help us to solve the dilemma that arises for pluralism about the aims and norms of inquiry. But to see how exactly the solution could work, we need first to introduce a theory of reasons that is fit for the job. The remainder of this section does just this. Namely, I will introduce the Erotetic, that is, question-centered, theory of normative reasons. In order to keep things manageable, I will not be able to defend the theory in detail here [see Logins (2022) for a fuller defense]. We can also think of the theory being a useful tool for solving the pluralist puzzle as constituting an argument in its favor. The fact (if it is one) that the proposal is theoretically fruitful can be taken on its own as a point that speaks in its favor.

Before we move on, let me stress that by reasons we understand (a rough characterization) considerations, and more specifically, facts, that speak in favor of (or against) a response, where a response can be an action, an intention, a belief, or another attitude (cf. Scanlon, 1998; Parfit, 2011, among many others). So, for instance, the fact that the building is on fire is a reason for us to evacuate the building; the fire fact speaks in favor of the evacuation response. Normative reasons are commonly distinguished from motivating reasons (also known as operative reasons, cf. Scanlon, 1998), that is, our actual basis for actions and attitudes, and they are sometimes also distinguished from mere explanatory reasons (cf. Alvarez, 2010; Engel, 2015), that is, from considerations that explain our actions and attitudes.

According to the Erotetic view, normative reasons are answers to normative questions that can be interpreted either as quests for arguments/reasoning or as quests for normative explanations. The idea that reasons are connected to questions has been suggested by Pamela Hieronymi (2005). I have recently developed this insight of focusing on questions in Logins (2022). According to this more recent version of the Erotetic (that is, question-centered) approach, normative reasons are appropriate answers to normative ‘Why F?’ questions. If that there is a fire is a reason for me to run, it is because that there is a fire is an appropriate answer (that I may possess or not at the moment) to the question (addressed to me): Why run out of the building? (why should/ought I run out of the building?). More precisely:

  • The Erotetic view of reasons] “For that p to be a reason to F for S is for that p to be (a part of) the content of an appropriate answer to a (S directed) question ‘Why F?”’ (Logins, 2022, p. 168).

A crucial point that the Erotetic view is in a position to explore fruitfully is the fact that ‘Why?’ questions in general, and thus also normative ‘Why F?’ questions, can have two fundamentally distinct readings. On one of these, when we ask why something is the case, we are seeking a (deeper) explanation of why the thing is the case. Typically when we ask for an explanation with a ‘Why?’ question, we are ultimately aiming at a better understanding of the relevant phenomenon. When I ask ‘Why is that person a foreign spy?’, I might well know that the person is a foreign spy, but by asking this question, I want to understand better why he is. I am looking for an explanation when I ask the question in this reading. That he was promised good money and thought he would not be caught might suffice as an answer to the question in this explanation-seeking reading. But in a different context, my question “Why is that person a foreign spy?” asks for an argument or a premise in reasoning towards the conclusion that the person is indeed a spy. Perhaps I don’t know that the person is a spy; maybe I just overheard someone saying that he is. Moved by curiosity and shock at the accusation, I utter, ‘What? Why is that? Why is that person a foreign spy?’ When I ask this question in this state of mind, I expect my interlocutors to answer something like, ‘Well, the police have found this evidence; they have phone calls, bank transfers, and intercepted messages; and he was also at that secret meeting.’ Now, if ‘Why?’ questions, in general, can have two different readings, then it is only expected that normative ‘Why F? (Why should I F?)’ questions can also have this double reading.

The appeal to two readings of ‘Why?’ questions helps us to reconcile two intuitively central functions of reasons: the function in reasoning and the function in explanation. The Erotetic view explains that in some contexts, normative reasons are premises in good patterns of reasoning. Reasons have this function in contexts where they are responses to ‘Why F?’ questions in their argument/premise-in-reasoning-requiring reading (when an appropriate answer to the relevant ‘Why F?’ question requires a premise in a good argument/pattern of reasoning towards the conclusion F/one ought to F). And the Erotetic view also explains that in other contexts, normative reasons are elements of the explanation of the relevant normative facts (e.g., that one ought to F). Reasons have this function in contexts where they are responses to ‘Why F?’ questions in their explanation-requiring reading (when an appropriate answer to the relevant ‘Why F?’ question requires providing some elements of explanation of why one ought to/should F). Normative reasons have these two central functions, according to the Erotetic view, because a fundamental dualism of normative reasons holds. Some normative reasons are tied to good patterns of reasoning; other normative reasons are tied to an explanation of normative facts. A given context - which reading of a possible “Why-F?” question is the relevant one in a context—determines the relevant sort of reasons. This dualism of reasons is well grounded in and thus theoretically well motivated by a more fundamental point of the Erotetic view, according to which normative reasons are answers to normative ‘Why F?’ questions, which can have two possible readings: the reasoning reading and the explanation reading, as introduced above (see Logins, 2022, pp. 160–191, for more details on the view). In what follows, we adopt the Erotetic view of normative reasons, according to which normative reasons are appropriate answers to normative ‘Why F?’ questions, and explore its fruitfulness in theorizing the aims and norms of inquiry.

3.2 Reasons-centered view of inquiry

We specified above that a criterion that should guide our search for a theory of the aims and norms of inquiry is the desideratum according to which our theory should not predict that brainwashing or similar activity of mere acquisition of beliefs or mere gathering of information can be an instance of genuine inquiry. We elaborated on this desideratum by specifying that a good and inclusive way of thinking about it is to interpret it as a desideratum according to which any plausible theory of the aims and norms of inquiry should predict that an inquiring subject is responsive in her interrogative attitude to (normative) reasons. That is, genuine cases of inquiry should come out on any plausible view of inquiry as episodes where, if successful, an inquiring attitude results in an attitude or action that is responsive to reasons, reasons that genuinely speak in its favor. A genuine inquiry appears to be connected to the inquirer being responsive or sensitive to reasons in the sense that the inquirer’s attitude/action that is an output of the inquiry is based on the relevant considerations, i.e., reasons that speak in favor of the attitude/action (yet these reasons need not be sufficient for the attitude/action).Footnote 10

A reasons-centered account of the aims and norms of inquiry can easily account for the reasons-responsiveness desideratum. Our detour through the theory of reasons now enables us to specify such an account in a more elaborated way, that will show how reasons and inquiry are connected. In the remainder of this section, we elaborate on the details of the Reasons view of the aims and norms of inquiry.

Consider the following proposal:

  • The reasons aim of inquiry: general version When S’s (response) \(\Phi \) is question Q dependent, A is the aim of the S’s inquiry (of type t) concerning Q if and only if (and because) A is to find sufficient reasons for S to \(\Phi \) (response of type t), for any relevant \(\Phi \).

The general idea here, simply put, is that the aim of our inquiries (understood in the sense of interrogative attitudes) is to find sufficient reasons for the relevant responses on our part, where the relevant responses are connected to the relevant questions. So, for instance, the aim of S’s inquiry concerning the question ‘What caused the Notre Dame fire?’ is for S to find sufficient reasons to believe that p (alternatively to believe that p1-pn, or, perhaps, to grasp/entertain p1-pn, more generally, to have the relevant attitude about p), where p is a complete answer to the question ‘What caused the Notre Dame fire?’ (And for S to find sufficient reasons to believe that p (or to have another theoretical attitude about p, where p is a complete answer to the question ‘What caused the Notre Dame fire?’) is to be the aim of S’s theoretical inquiry concerning the question ‘What caused the Notre Dame fire?’)

Note that the account is restricted to apply specifically to cases where the relevant responses are question-Q dependent. This is merely needed to ensure that we don’t get trivial results that appear counter-intuitive. More concretely, a response \(\Phi \), i.e., a choice in a very broad sense between (possible) beliefs/attitudes \(\phi _1\), ..., \(\phi _n\) is Q dependent if and only if the most appropriate option between \(\phi _1\), ..., \(\phi _n\) conditional on having an appropriate response to Q is not the same option as the most appropriate option between \(\phi _1\), ..., \(\phi _n\) conditional on not having an appropriate answer to Q [compare to Hawthorne and Stanley (2008, p. 578), on the p-dependency of choice].

Given the Reasons aim of inquiry, a naturally connected view is that reasons also set the standard, or specify the norm of inquiry. Here is then a general version of a reasons-centered norm of inquiry:

  • The Reasons norm of inquiry: general version When \(\Phi \) (a response) is question Q dependent, one ought not: to have an interrogative attitude concerning the relevant question Q at a time t and possess sufficient reasons to \(\Phi \) at a time t.

The Reasons norm of inquiry prohibits the combination of an interrogative attitude about a question while having at the same time sufficient reasons for the response (belief, attitude) that is properly connected to the relevant question under investigation. So, for instance, the Reasons norm prohibits continuing research on the question ‘What caused the Notre Dame fire?’ when one already knows, say, that the fire was caused by an electrical malfunction.

An account that specifies the aim and the corresponding norm of inquiry in terms of possessing sufficient reasons to believe that p/not-p perfectly respects the reasons-responsiveness desideratum. However, one may still wonder how exactly this view helps to deal with the standard pluralist dilemma. The next section elaborates on exactly this. The key to the proposal is to make use of the Erotetic view of reasons as appropriate answers to questions that can come in different sorts.

4 The Reasons view of inquiry and modest pluralism

The aim of this section is to show how the reasons-centered approach (of the aims and norms) of inquiry can solve our initial tension and respond to the challenge that we began with. The success of the reasons-centered approach in this respect will be taken as an argument in its favor.

4.1 The reasons view and its precisifications

Our initial problem was the following one. We observed that at least three distinct and apparently incompatible views about the aims and the corresponding norms of inquiry all enjoyed a high degree of plausibility. When thinking about certain cases, knowledge seems to be the aim of inquiry. Yet, in different cases, it would seem that agents are aiming for certainty. Yet, in still other cases, we seem to aim for understanding when we inquire. The worry then is: how do we account for this apparent plurality of aims (and the corresponding norms) without endorsing a merely ad hoc pluralism? In short, we need to face the pluralist dilemma applied to the aims and norms of inquiry; either we need to explain in a theoretically satisfactory way what is the common element in all the aims and norms, or we need to provide a satisfactory error theory that would explain why we are massively wrong in thinking that there is a common element in inquiry in general.

Here is then how the Reasons view of the aims and norms of inquiry solves our worries. It allows us to explain in a theoretically satisfactory way why all three, knowledge, certainty, and understanding, can be genuine aims (and set the corresponding norms) of inquiry while there is a common element that unifies them all. The common element that unifies them all is that inquiry aims at finding sufficient reasons to F. Given that reasons come in a variety of sorts (as our theory of reasons explains), there is also then a variety of aims of inquiry. The idea here is that the general aim can be precisified given that reasons come in different sorts. Each sort of reasons will have a corresponding sort of precisified aim of inquiry.

Recall from the previous section that there are normative reasoning reasons on the one hand and normative explanatory reasons on the other hand. The former are appropriate answers to the normative questions of the form "Why believe that p?" which ask for an argument/premise in good reasoning as an answer. The latter are appropriate answers to the question "Why believe that p?" when this question asks for an explanation of why one ought to or should believe that p.

The Reasons aim and the Reasons norm of inquiry appeal to sufficient reasons to \(\Phi \). The Reasons aim and the norm are then about having sufficient reasons to believe. But since normative reasons on our Erotetic view of reasons come in two sorts, sufficient reasons to believe come in two sorts too. That is, having sufficient normative reasoning reasons to believe is one thing, and having sufficient normative explanation reasons to believe is another. These two have different standards.

What then are the standards for sufficient normative reasoning and explanation reasons? Let us begin with normative reasoning reasons. Here is one proposal for how to make these standards more precise. For a consideration (a fact) r to be a sufficient reasoning reason for S to believe that p is for r to be (a content of) a premise in a good and not defeated reasoning from a fitting premise-response (e.g., a fitting belief) together with some further considerations to a belief that p.

To support this way of understanding the standards for sufficient normative reasoning reasons, we can appeal to the fact that a number of theorists have recently proposed to conceptualize normative reasons by appeal to their role in good patterns of reasoning. According to the Reasoning view, reasons are appropriate contents of premises of good patterns of reasoning towards the relevant F-ing (see Williams, 1989; Setiya, 2014; McHugh & Way, 2016; Way, 2017; and Asarnow, 2017, for versions of this idea).Footnote 11 On the present view, proponents of the Reasoning view of reasons are right but only about one sort of normative reasons, normative reasoning reasons. The conceptualization of sufficient normative reasoning reasons by appeal to patterns of good reasoning inherits all the advantages of the Reasoning view and none of its disadvantages.

Note that my characterization of sufficient normative reasoning reasons has a clause that specifies that having sufficient reasons requires that the relevant pattern of reasoning is not defeated. This clause ensures that the agent effectively arrives at a fitting conclusion-response, in this case, to a fitting belief that p, given that the agent starts with the relevant premise-responses.

Now, sufficient normative explanation reasons can be specified as follows: for a consideration (a fact) r to be a sufficient explanation reason for S to believe that p is for r to be (a content of) an explanans (together with other elements) of a genuine explanation from appropriate explanantia-responses (attitudes whose contents are proper explanantia) to a belief that p (explananda-response—attitude whose content is an explanandum-response) that is not defeated. The idea here is to focus on the corresponding responses that are proper to the explanation (explanantia-response, explanandum-response). I am working with the assumption that attitudes that fill the role of explanantia-responses and explanandum-responses are beliefs. When one’s relevant (sets of) beliefs stand in the relation that properly captures an explanatory relation, one has an instantiated pattern of explanation. Patterns of explanation can be genuine and good and yet defeasible. A pattern of genuine, good explanations can be defeated, for instance, when a less surprising or more economical explanation is provided. For instance, that it rains might genuinely explain why it is wet inside my room. Yet this explanation is defeated by the explanation that it is wet inside my room due to the fact that the roof has a hole in it and it rains (see also Broome, 2013, pp. 47–49, for a similar view of explanation). A genuine explanation that is not defeated is one where there are no less surprising or economical explanations easily available.

To support this way of understanding the standards for sufficient normative explanation reasons, we can also appeal to the fact that another recently very popular view of reasons has proposed to conceptualize normative reasons by appeal to their role in explanations. According to the Explanation view, normative reasons are defined by their role or function in the explanation of the relevant normative facts: ought facts (see Broome, 2004, 2013; Nebel, 2019) or facts about values (see Finlay, 2014, 2020; Maguire, 2016).Footnote 12 According to the present interpretation the proponents of the Explanation view are absolutely right, but only about one sort of normative reasons, namely the normative explanation reasons (see also Brunero, 2013, 2018; Schmidt, 2021; Logins, 2020, for recent objections to and overviews of the Reasoning and Explanation views of reasons).

It is no surprise, I think, where this is going. The final step for our solution then is to suggest that what constitutes sufficient reasoning reasons to believe that p will be only either fitting premises in good patterns of reasoning that are not defeated and lead to knowledge in general of the relevant conclusions, or fitting premises in good patterns of reasoning that are not defeated and lead to the relevant conclusions being certain for one (a special kind of knowledge, a certain knowledge). And that sufficient explanation reasons are then elements that constitute explanantia in a good genuine explanation that is not defeated, which means that understanding of considerations that constitute an explanandum is secured through this explanation. A natural thought here is that explanations aim at providing us with understanding.

We are now in a position to solve our initial puzzle. Our proposal is that the Reasons-centered approach allows us to recover all three of the initially plausible-sounding proposals: the aim (and the corresponding norm) of the inquiry being set by knowledge, by certainty, or by understanding. There are several ways to precisify (interpret in more fine-grained ways) the general Reasons aim/norm of inquiry. We can say that in some cases, the Reasons aim/norm amounts (can be precisified) to the knowledge aim/ignorance norm, but in other cases, the Reasons aim/norm amounts to the certainty aim/norm; yet in still further cases the Reasons aim/norm can be precisified as the understanding aim/norm of inquiry. The unity in this plurality is guaranteed to be not arbitrary, given the underlying general appeal to reasons. But given the ambiguity in reasons, which is inherited from the fundamental ambiguity of ‘Why?’ questions in general, we have a theoretically motivated account of the plurality of aims and norms of inquiry. I develop this idea in more detail below, where I consider a potential worry for the present view. In the next few paragraphs, I present how exactly all the more fine-grained aims/norms (knowledge, certainty, understanding) are derived from the more general Reasons account of inquiry.

First, inquiry may well have an aim of knowledge (and the corresponding norm of ignorance might well apply to inquiry too). Here is a way to make this idea more precise:

  • The Reasons aim of inquiry: the knowledge version When S’s belief that p/not-p is question Q dependent, and when Q is ‘Why believe that p/not-p?’ in its general premises/argument-seeking sense, A is the aim of S’s inquiry concerning Q if and only if (and because) A is to find sufficient reasoning reasons (in general) for S to believe that p/not-p, that is, reasons that guarantee knowing p/not-p.

This precisified version of the Reasons aim of inquiry then leads naturally to the following version of the norm of inquiry:

  • The Reasons norm of inquiry: the ignorance version When one’s belief that p/not-p is question Q dependent, and Q is “Why believe that p/not-p?” in its general premises/argument-seeking sense, one ought not to: have an interrogative attitude concerning the relevant question Q at a time t and possess sufficient reasoning reasons (in general) to believe p/not-p, that is, know p/not-p at a time t.

Note that we assume here that any question Q, which is a content of the inquiry, can be translated into a question of the type ‘Why believe that p/not-p?’ where p/not-p are possible to complete answers to Q. The idea here is to rely on a standard view of the meaning of questions (cf. Hamblin, 1973) on which questions are sets of propositions, namely, it is a partition of all the relevant possibilities. So, for instance, the meaning of the question ‘Who killed the victim?’ just is a partition with all the relevant individuals as distinct cells of the partition (relative to a context). Taking seriously the no-brainwashing desideratum of inquiry suggests that the aim of closing the question ‘Who killed the victim?’ by inquiry cannot be successfully accomplished by merely picking out one of the cells of the question (that is, one individual) even if it turns out to be correct. One needs to do something more to answer the question ‘Who killed the victim?’ through a genuine inquiry. This ‘something more’ is what corresponds to finding a proper answer to the question ‘Why believe that p/not-p?’ where ‘p’ picks out one of the cells of the question ‘Who killed the victim?’ So, it would seem that the relevant ‘Why?’ questions can be translated into ‘Why believe p?’ questions.Footnote 13

The focus on sufficient reasoning reasons to believe that p, of the general sort (and assuming that sufficient reasoning reasons in general guarantee knowledge of the relevant proposition), allows us to recover a sense in which ignorance is indeed the norm of inquiry and knowledge is its aim. The important point is that knowledge is the aim, and ignorance is the norm, specifically when the content of an interrogative attitude can be interpreted as the question ‘Why believe p/not-p?’ in its general premises/argument-seeking sense.

Second, our approach allows us to suggest that sometimes inquiry (our interrogative attitudes) may also have the aim of certainty and the corresponding certainty norm. Here is a way to make this suggestion more precise:

  • The reasons aim of inquiry: the certainty version When S’s belief that p/not-p is question Q dependent, and when Q is ‘Why believe that p/not-p?’ in its more demanding/restricted premises/argument-seeking sense, A is the aim of S’s inquiry concerning Q if and only if (and because) A is to find sufficient reasoning reasons (of the more demanding/restricted sort) for S to believe that p/not-p, that is, reasons that guarantee certain knowledge that p/not-p.

The idea here is that the certainty aim just is a more specific sort of general knowledge aim of inquiry. Sometimes we aim not merely to know but to know with certainty. Certain knowledge is a subset of knowledge. And certain patterns of reasoning/argument might be a better fit for reasoning towards certain knowledge than others. When we seek certainty, the relevant patterns or reasoning/argumentation will be of this more specific sort. Maybe certain sorts of deductive arguments (and the corresponding patterns of reasoning) are relevant here. Or perhaps only patterns of reasoning where one normally moves from premises known for certain to conclusions known for certain will constitute the relevant patterns here. The certainty version of the aim of inquiry leads us naturally to the corresponding norm of inquiry:

  • The reasons norm of inquiry: the uncertainty version When one’s belief that p/not-p is question Q dependent, and Q is “Why believe that p/not-p?” in its more demanding/restricted premises/argument-seeking sense, one ought not to: have an interrogative attitude concerning the relevant question Q at a time t and possess sufficient reasoning reasons (of the demanding sort) to believe p/not-p, that is, know p/not-p for certain at a time t.

What the uncertainty version of the Reasons norm proscribes is to embark an inquiry with the relevant question in this more demanding premise/argument-seeking sense of ‘Why believe p/not-p?’ while having already reasons to believe that p/not-p that guarantee certain knowledge of p/not-p.

Third, our approach also allows us to recover a reading of the general Reasons aim and norm of inquiry that is specifically about understanding and having good explanations. Here is a way to precisify the aim of understanding:

  • The Reasons aim of inquiry: the understanding version When S’s belief that p/not-p is question Q dependent, and when Q is ‘Why believe that p/not-p?’ in its explanation-seeking sense, A is the aim of S’s inquiry concerning Q if and only if (and because) A is to find sufficient explanation reasons for S to believe that p/not-p, that is, reasons that guarantee understanding of why one should believe that p/not-p.

This precisification of the aim allows us then to derive the following norm of inquiry:

  • The Reasons norm of inquiry: the explanation version When one’s belief that p/not-p is question Q dependent, and Q is “Why believe that p/not-p?” in its explanation-seeking sense, one ought not to: have an interrogative attitude concerning the relevant question Q at a time t and possess sufficient explanation reasons to believe p/not-p, that is, reasons that guarantee understanding of why one should believe that p/not-p.

We are assuming that good (normative) explanation reasons for believing that p are undefeated explanatory elements of why one should/ought to believe that p. That is, they are appropriate answers to the ‘Why believe that p?’ questions in their explanation-seeking interpretation.

We are also assuming here that if X satisfactorily explains why p, then X satisfactorily explains why we should/ought to believe that p. This plausible principle will allow us to recover reasons in certain cases where given considerations appear intuitively to be reasons: that you missed a train explains why you will be late; and it also explains why you should believe that you will be late.

One might wonder at this point, however, whether the Reasons view can address an important worry. The Reasons view was advertised above as the solution that a pluralist about the aims and norms of inquiry can endorse in the face of the standard pluralist dilemma applied to the aims and norms of inquiry. However, the Reasons view appears to focus on one kind of thing as the aim of inquiry. The thing in question is said to be finding sufficient reasons to believe that p. Finding sufficient reasons is also said to settle the norm of inquiry. A reader might wonder then whether the appeal to sufficient reasons doesn’t commit the proponent of the Reasons view to monism about the aim and norm of inquiry after all. If so, how can the proposal constitute a genuine defense of pluralism?Footnote 14

To this potential objection, I would like to reply by clarifying the overall dialectic and the sort of pluralism that I aim to defend. The first thing that needs to be noted is that the pluralist dilemma is not a problem per se. As we specify it in the above section on the pluralist dilemma, it amounts to a problem insofar as a proponent of a pluralist view is not in a position to provide a theoretically well-motivated, non-ad-hoc response to the dilemma. A response to a dilemma by definition involves either dissolution (by noting, for instance, that it relies on some ambiguity) or by embracing one of its horns while providing a theoretically satisfactory explanation of why embracing one of the horns is our best bet. The standard pluralist dilemma as applied to the aims and norms of inquiry puts pressure on the pluralist about the aims and norms of inquiry, insofar as it calls for a theoretically satisfactory explanation of why one of the horns has to be accepted. The two horns were that either a pluralist accepts that there is a common element in all aims and norms of inquiry, in which case the pluralist seems to endorse a monist view, or the pluralist accepts radical pluralism and insists that there is nothing in common in the aims and norms of inquiry, in which case the challenge is to explain why we typically think, but are radically wrong, that various phenomena can be unified under the general label of inquiry. The theoretical problem that the dilemma raises for the pluralist lies not so much in the dilemma itself as in the need to provide a theoretically well-motivated answer to the dilemma.

I believe that in what precedes I have provided a theory-driven explanation of what indeed appears to be a common element in all instances of inquiry. Namely, that in inquiry in general we seek to find sufficient reasons to \(\Phi \). Now the challenge is to explain why accepting this view is not incompatible with the view still qualifying as a version of pluralism about the aims and norms of inquiry. In short, to return to the examples above, the challenge is to show that the pluralism about the aims and norms of inquiry is more like the pluralism about democracy than the (putative) pluralism about propositional knowledge in light of Gettier cases. I believe that the pluralism that I have developed here is closer to the pluralism about democracy than to the supposed pluralism about propositional knowledge. The pluralism I have defended is a modest pluralism, meaning that it admits that at the more general level of theoretical explanation there is only one overarching aim (finding sufficient reasons to \(\Phi \)); however, at a more fine-grained level of explanation, a plurality of aims can be recovered. Exactly as in the case of democracy, there seems to be a common element in all instances of inquiry, yet we can individuate differences in various sorts of inquiries at a more fine-grained level. Similarly, democracy might be the power of the majority that comes in various forms (parliamentary, direct, other). In short, the view is indeed giving up the radical pluralist flavor; however, it is theoretically well motivated, given our reasons-based approach, and it can still account for all the relevant data that need to be explained by any theory of the aims and norms of inquiry, as we will see in the next section.

With the more-fine grained precisifications of the general Reasons aim and Reasons norm of inquiry in place, we are in a position to account for all of the initial considerations that seemed to speak in favor of different aims and norms of inquiry. Our account can respect all the relevant considerations for knowledge, certainty, and explanation aims, and the corresponding norms of inquiry. Let us revisit the three main examples that we considered above as paradigmatic cases in favor of the three aims/norms: the oat milk case, the surgeon case, and the blob (and the mathematician) case.

4.2 Explaining all the cases and considerations

Recall that one of the motivations for the knowledge aim and the ignorance norm was that they provide a straightforward explanation of why assertions of the following sort seem absurd: “I know there’s oat milk in the store but I’m curious/wonder if there’s oat milk in the store” (cf. van Elswyk & Sapir, 2021) (and a corresponding explanation of why having mental attitudes expressed by this assertion appears inappropriate). That we should not make this sort of assertion is explained by the ignorance norm, according to which one should not inquire/be curious about a question while knowing the answer to the relevant question. The oddity of the oat milk assertion comes from the fact that it violates flagrantly the ignorance norm (or rather that it expresses states of mind that violate the ignorance norm). In this picture, ignorance is the norm, because inquiry aims at closing a question, and knowing the answer to the relevant question is to close the question, and it is problematic to continue inquiry once the relevant question is already closed.

The Reasons norm in its ignorance version respects the initial considerations of the ignorance norm and the knowledge aim entirely. Our suggestion is that the question ‘Is there oat milk in the store?’ can be interpreted as a question: ‘Why believe that there is/isn’t oat milk in the store?’ Now, once we see that the underlying question is a question of the ‘Why believe p/not-p?’ form, we can see that it can have one of the two ‘Why?’ question readings; we can ask either for an argument/a premise in a pattern of reasoning towards p when we ask ‘Why believe p?’ or we can ask for an explanation of p. Which of the two readings is triggered in a given context depends on whether we are in the business of figuring things out (we aim to know the relevant thing) or in the business of trying to understand the relevant thing. The former is then associated with finding good arguments/reasoning toward the relevant p conclusion. The latter will be about finding the relevant explanations. Now, it is reasonable to assume that in the original oat milk case, one is in the business of figuring it out - one wants to know whether, yes or no, there is oat milk in the store. We can think of a case where one doesn’t know or have any clue about there being oat milk in the store. If so, it is natural to hear the question ‘Why believe that there is/isn’t oat milk in the store?’ as a question that asks for a good, that is, sufficient, argument, a premise that one could plug into one’s reasoning with respect to the claim that there is/isn’t oat milk in the store. Note that argument or pattern of reasoning is understood here broadly enough - a pattern starting with an intake of the testimony of an employee to the conclusion that there’s no oat milk in the store might well count as a piece of reasoning. If so, then in the oat milk case, the relevant reading of the ‘Why believe that there is/isn’t oat milk in the store?’ question is a reading that requires obtaining sufficient reasons for reasoning in its general version (the version where sufficient reasoning reasons guarantee knowledge). But—and this is why the oat milk assertion (and the corresponding mental attitudes) appear absurd/inappropriate - the agent, in this case, is said already to possess sufficient reasoning reasons. Actually, I know already that there is oat milk in the store. The inappropriateness of the case is explained then by the case instantiating both a situation where one is still seeking sufficient reasoning reasons (general version) but has them already at the same time (since one knows the relevant propositions).

The Surgeon case above was presented as a case that speaks in favor of the Certainty aim and the corresponding uncertainty norm of inquiry. In short, it is a case where it seems perfectly appropriate for a surgeon to check the patient’s file once more before the operation, indeed, to read the file and to think it through once more, just to be sure or certain which kidney is to be removed. The very fact that there is so much at stake in the operation seems to suggest that the surgeon should be certain about the relevant facts. That there is so much at stake in this situation makes it that to close the question satisfactorily in this case, one has to satisfy quite demanding standards, the standards of certainty. Also, assuming that an anti-skeptical fallibilism is on the right track (i.e., that one can know that p without being certain that p), we may admit that the surgeon knows all the relevant facts already without being totally certain about them.

The Reasons norm in its uncertainty version can explain the Surgeon and similar cases. Indeed, it can explain why sometimes inquiring while knowing is OK when one aims to be certain. There being an uncertainty version of the Reasons norm doesn’t mean that the ignorance version is not a genuine norm. The Reasons norm tells us that one should not inquire and have sufficient reasons to believe the relevant proposition at the same time. But the fact that the relevant question that is the object of the interrogative attitude, in this case, the question ‘Why believe that it is the left kidney/is not the left kidney that has to be removed?’ in its demanding reading (the reading on which the question asks for reasoning reasons that guarantee certain knowledge) makes inappropriate the combination of the interrogative attitude in this case with having certainty (which is guaranteed once one has sufficient reasoning reasons in the demanding sense). In other terms, the ignorance version of the Reasons norm doesn’t apply in this case insofar as the relevant question that is the object of the interrogative attitude has the more demanding reading. This is why in this case combining knowledge and inquiry is not sanctioned. The inquiry in the surgeon’s case is about finding sufficient reasoning reasons that would guarantee certain knowledge of which kidney is to be removed, and not about finding sufficient reasoning reasons (in the general sense), reasons that would guarantee mere knowledge of which kidney is to be removed. The focus on reasons allows us to respect both initial considerations: both ignorance and uncertainty can be norms of inquiry, depending on whether we are looking for reasoning reasons that guarantee mere knowledge or reasoning reasons that guarantee certain knowledge. The unificatory element is that inquiry is about reasons, and reasons come in various sorts.

The blob case above (and the similar mathematician case) was presented as speaking in favor of the understanding aim and the corresponding norm of inquiry. Recall we are considering a situation where a single-celled slime mold, the blob, exhibits certain patterns of behavior that lead scientists to conclude that the blob can learn. After studying these results, I come to know that the blob is capable of learning. Yet I don’t understand how it is possible that the blob is capable of learning. I continue to be curious about this, and my interrogative attitude seems to be appropriate, despite my knowledge that the blob is capable of learning (and a fortiori despite my knowledge that it is possible that the blob is capable of learning).

Once more, the Reasons account can perfectly accommodate this result by accepting that understanding is the aim of inquiry in certain contexts without denying that knowledge can constitute an aim of inquiry. This is so because the question that is the object of interrogative attitudes can have different readings in different contexts. Consider our blob case. In one context, the relevant question (‘Why believe that the blob can learn?’) has a reading that requires an argument/reasoning. This is the context where, if everything goes well, I will obtain knowledge that the blob can learn. I will receive sufficient (normative) reasoning reasons that will constitute appropriate answers to the question in the argument-requiring reading. Yet, in a different context, where I know already that the blob can learn but continue my research, the relevant question that constitutes the object of my interrogative attitude has the explanation-seeking reading. I will receive sufficient (normative) explanatory reasons through receiving an appropriate answer to the question in this reading. And if everything goes well, I will achieve an understanding of why the blob can learn when I receive the appropriate answer to this question in this explanation-seeking reading. So, in this way, we can show that the Reasons approach allows us to maintain a sense in which knowledge is still the aim of inquiry and ignorance is its norm: in contexts where the relevant question (in our case: ‘Why can the blob learn?’) has an argument/premise in the reasoning-requiring reading. Yet we can also maintain that understanding can be the aim (and explanation the norm) of inquiry, when the relevant question has the explanation-requiring reading. When we are in the context where the question has this explanation-requiring reading, a possible case where one has an interrogative attitude and yet continues inquiry is not necessarily a counter-example to the knowledge norm. That is, it is not if the interrogative attitude in this context aims to find sufficient (normative) explanation reasons rather than sufficient (normative) reasoning reasons.

5 Conclusion

What are the aims and norms of inquiry, understood in the sense of interrogative attitudes? Knowledge, certainty, and understanding all seem like plausible candidates. Should we endorse pluralism then about the aims and norms of inquiry? Not if we don’t have a theoretically well-motivated, non-ad-hoc pluralist theory. In this paper, I have elaborated just that: a well-motivated pluralist theory of the aims and norms of inquiry that avoids ad-hocness and yet respects the relevant considerations about the role of knowledge, certainty, and understanding in inquiry. According to the account that I have presented, the aim of inquiry is to find sufficient reasons to F. Since reasons are answers to normative “Why-F?” questions, and “Why?” questions, in general, come in two sorts, we can expect to find distinct and more specific aims of inquiry. These will correspond to distinct sorts of reasons we are looking for in our interrogative attitudes. Some reasons are tied to reasoning and argument. These lead us to the knowledge of the relevant propositions when they are sufficient. A subset of these is tied to certain knowledge and thus leads to certain knowledge. Other sorts of reasons are tied to explanation and lead us to understand when they are sufficient in the relevant context. Focusing on sufficient reasons as the aim of inquiry perfectly respects the no-brainwashing criterion for inquiry. When we get brainwashed into believing things, we typically lack sufficient reasons to believe the relevant propositions. Genuine inquiry and reasons seem to be essentially related.