Not Evidential has recently been defended by a number of philosophers:¹

These philosophers present various arguments in favour of Not Evidential, and the merits and significance of their position and arguments have been debated at length in the literature.²

This paper raises and considers an objection to Not Evidential that hasn’t, to my knowledge, received attention so far in the literature but which highlights a real shortcoming of a descriptive account of philosophical methodology that endorses Not Evidential. The objection in question centres on the fact that there are various aspects of philosophical practice that are perfectly explicable if philosophers do use intuitions as evidence but are rather more puzzling if Not Evidential is true. I’ll call these aspects of philosophical practice “I-Practices” and introduce them properly in the next section.³

The objection to Not Evidential considered and rebutted in this paper can be presented as follows:

The key task of this paper is to show that P2 is false by providing an explanation of I-Practices that is compatible with Not Evidential; whatever else might be said for or against Not Evidential, there is a satisfactory alternative explanation of I-Practices that proponents of Not Evidential can give.

This paper thus presents a very limited defence of Not Evidential. The aim is not to argue in favour of Not Evidential by showing that it can give the best all-things-considered account of I-Practices. The aim is merely to show that one specific charge won’t stick. The charge that won’t stick is the charge that the proponent of Not Evidential can’t satisfactorily account for I-Practices. I’ll show they can if they want to.⁴

The structure of the paper is as follows: § 1 outlines the relevant I-Practices, i.e., practices that seem difficult to explain given Not Evidential; § 2 generates a possible alternative explanation for these practices using a device involving genealogical reflection, i.e., I-Practices serve a purely explanatory role; § 3 argues that this alternative explanation is not only compatible with Not Evidential but has some baseline plausibility, being able to account for the use of I-Practices within the context of contemporary philosophy, including concrete examples from recent highly cited articles; § 4 wraps things up and addresses some likely objections.

1 I-Practices

What are I-Practices? What do philosophers do that seems to betray a use of intuitions as evidence? What would be difficult to explain given Not Evidential? The following is intended as an illustrative, although not exhaustive, list.⁵

Case Wielding

Philosophers appeal to cases a lot. Infamously, Gettier (1963) did so. But focusing on one paper, or some small collection of supposedly paradigmatic uses, fails to capture the extent to which philosophers wield cases. Enter any philosophical seminar room or lecture hall in the world, and you’ll be lucky to avoid a helter-skelter of cases in discussion. Cases are often imaginary or outlandish, but sometimes they are based on real cases or are otherwise realistic in some sense. Whether or not philosophers should do this, they do it a lot. Some aspects/kinds of philosophers’ use of cases are often described as “appealing to intuitions about cases as evidence.” Indeed, that’s a natural way to describe the practice. If we think of the relevant cases eliciting or involving intuitions (whatever we mean by that), then the way that use of such cases helps to move philosophical debate forward, shapes which accounts we are willing to give credence, and can serve to change our minds seems naturally to be described in terms of those intuitions playing an evidential role. Our question will be whether an alternative explanation can be given of ways in which philosophers appeal to cases that might typically be interpreted as appealing to intuitions as evidence. Can the defender of Not Evidential account for this phenomenon?

Idea Motivating

Philosophy can be a bit obscure. Why would anyone be interested in this? That’s a question the person in the street might ask if told what a lot of us worked on. Indeed, that’s likely a question many of us would ask when considering our colleagues’ work if they didn’t provide some motivation for thinking about their idea, topic, or argument. It is important to give your audience some sense of why anyone would care about what you are talking about. Consequently, we motivate our theorizing by pointing out how the problem we are grappling with arises from our ordinary ways of thinking about \(X\), the things we naturally want to say about \(X\), and so on. This is not the only way to motivate a view. But it is an effective one. Moreover, many cases of motivating an idea are naturally described in terms of intuitions, e.g., “To provide some motivation for this view, just consider a tension between two very intuitive claims about the moral status of future generations.” Moreover, it is sometimes very natural to describe the use of these “intuitions” as establishing an evidence base—or a partial one—for the philosophical discussion to follow. Again, our question will be whether an alternative explanation can be given in such cases. Can the defender of Not Evidential account for this phenomenon?

Assumption Footstamping

Making progress in philosophy could be difficult if interlocutors decided to be like an annoying child and ask, “But why?… Back up that claim… prove it…” after every claim adduced. In any debate, a lot must be taken for granted. Now, simply making assumptions isn’t an I-Practice. That alone isn’t difficult to explain for the proponent of Not Evidential as one can make assumptions just for the sake of exploring what would happen if we held certain things fixed. “Assumption footstamping” is a more specific manoeuvre within philosophical discourse. Philosophers often put certain claims beyond question, at least for the duration of particular debates, talks, question sessions, papers, etc., in a way that goes beyond simply making an assumption. Sometimes this practice is described in terms of intuitions: “I’m just treating this as a basic intuition about domain \(X\). It could be denied, but I can’t really make sense of why anyone would want to do that, and, in the kind of debate I’m interested in, we’re just not interested in engaging with that kind of position, so let’s just assume it in what follows.” The move is treating certain claims as enjoying a particular status: they are taken for granted in the debate; we are not interested in arguing for them; we simply take ourselves to be justified in assuming them. Insofar as the relevant claims are intuitive, this kind of manoeuvre seems naturally described in terms of philosophers taking intuitions to provide evidence and in some way appealing to them as enjoying a certain evidential status. Again, our question will be whether an alternative explanation can be given of ways in which philosophers set out the assumptions made in their projects, which might typically be interpreted as treating intuitions as evidence. Can the defender of Not Evidential account for this phenomenon?

Intuition Respecting

Say you’ve defended your favourite philosophical position. Now, you give your audience some bonus material. You point to various ideas that your audience might endorse, which might appear to conflict with the theory you have advocated or the argument you have given for it, and demonstrate how there is in fact no conflict. You acknowledge various ways in which your approach connects up with other ideas in philosophy, and show that there are no problems that result, and point out various nice things that your account allows one to say about these other ideas. You can do something similar in advance. You can lay out at the beginning various beliefs that you think your audience might endorse. You focus in on beliefs that your audience is probably not going to want to give up on, i.e., if your accounts were to conflict with the relevant beliefs, your audience would probably not be on board with your arguments. You also focus on beliefs that, for example, your account’s competitors end up having to reject and that you think your audience would rather not reject. You can generate a list of desiderata that any satisfactory account must satisfy. Such practices are often going to be naturally described in terms of “intuitions” being taken to count as evidence in favour of the relevant theories, as the reason that many of the relevant claims are on the table in the first place is that they are intuitive/counter-intuitive. “There are various intuitions that this sort of account allows us to respect…” or “There are various intuitions that are at stake in this debate…” or “There are various intuitive claims that you might want to respect.” Again, our question will be whether an alternative explanation can be given in such cases where philosophers might naturally be interpreted as treating intuitions as evidence. Can the defender of Not Evidential account for this phenomenon?

Weights and Balances Philosophy

Sometimes glorified under labels such as “reflective equilibrium,” there is a notable style of philosophizing that aims to evaluate the relative merits of theories by working out how they stack up against all the various considerations that might favour or count against them.⁶ For example, in the philosophy of time, presentism fairs well with respect to accounting for various basic ideas about time flowing and about the privileged present, but fairs less well with respect to the truth-maker problem and the modal asymmetry of past and future. Conducting weights and balances philosophy is somewhat less rigid than having a list of desiderata. But the idea is similar: there are various things it would be nice if a theory did, and the best theory will be the one that does the most. In this style of philosophy, being able to accommodate or account for “intuitive” ideas is frequently counted among the good features a theory might have, and having counter-intuitive implications among the theoretical costs of a theory. As the last sentence demonstrates, the language of intuitions is very naturally adopted to describe certain aspects of weights and balances philosophy. The active inclusion of “intuitions” among the weights and balances is naturally taken to indicate that intuitions are serving as evidence in philosophy. Again, our question will be whether an alternative explanation can be given of cases of weights and balances philosophy where the standard interpretation might be that intuitions are being treated as evidence. Can the defender of Not Evidential account for this phenomenon?

Negotiating Dilemmas, Inconsistent Triads and So On

Here’s a frequent feature in the philosophical terrain: We face a dilemma. We are pulled strongly in two directions at once and can’t immediately see how to resolve the two pulls. At other times, we come to recognize that there are three or more claims we are pretty attached to that are not consistent. Some claims may be on the table because of established dogma. Others may be empirical facts. Often, however, the reason these claims are on the table is that they are attractive in a way naturally described as being “intuitive.” The fact that we pause when we come to these features in the landscape and allow the shape of philosophical discussion to be guided by them might be taken as a sign that “intuitions” are playing a guiding role in the debate. These features of the philosophical landscape exert no small influence. The entire trajectory of debates in certain subfields is structured in relation to certain dilemmas or paradoxes. Thus, it is natural to see many instances of this as indications of philosophers treating intuitions as evidence. Again, our question will be whether an alternative explanation can be given of such cases. Can the defender of Not Evidential account for this phenomenon?

2 Genealogical Reflection

The fact that philosophers engage in I-Practices is easily explicable given the falsity of Not Evidential, and so makes denying Not Evidential attractive. With respect to each I-Practice, there is an element that we might provisionally label “intuition” that seems to play something like a supporting role. These “intuitions” seem to play an important role in what views philosophers come to believe are true. In many cases, even if these “intuitions” wouldn’t be the first port of call for a philosopher who wanted to convince another of their view, the “intuitions” are nonetheless there in the background being treated as claims to which the relevant philosophers are epistemically entitled: they seem to guide theorizing, speak in favour of views, and generally seem to be part of the implicitly assumed story about why some theories are better than others.

It is not immediately obvious how one might explain the fact that philosophers engage in I-Practices, given the truth of Not Evidential. So, the defender of Not Evidential faces a theoretical burden. How can the defender of Not Evidential account for the fact that philosophers engage in I-Practices? What function might I-Practices be playing within philosophical practice, if not justificatory/evidential? This is, at least at first glance, a significant burden for the proponent of Not Evidential.

To think about possible alternative explanations, I have found it helpful to employ the device of genealogical reflection: to consider a specific kind of imaginary “philosophical state of nature.” The kind of “state of nature” it is helpful to reflect upon is a very artificial construct. The idea is to isolate in our imagination an environment in which (in a sense) issues of evidence and of justification are simply not in play and yet in which something that (in other respects) resembles philosophy (or “protophilosophy”) exists.⁷ The idea is to try to imagine an environment where agents are not involved in any activities within which practices of adducing evidence or attempting to provide justification would arise (but which would of course still be a world in which evidence and justification exist). We can then ask whether there would be any clear purpose to employing I-Practices in such an environment, any function they might serve.

I should clarify the aims of the genealogical reflection in this section. The genealogical reflection in this section is only in the business of generating a hypothesis.⁸ I use the genealogical reflection to search for a possible alternative explanation for I-Practices that would be compatible with Not Evidential. What the genealogical reflection helps us to see is that I-Practices could serve, at least in principle, a purely explanatory function rather than a justificatory/evidential function.⁹ It is worth repeating that the genealogical reflection alone is not intended to provide any support for this hypothesis. It will be important to demonstrate—given the aims of this paper—that this hypothesis has some baseline plausibility, but that is a task taken up in the next section (§ 3).

I should also clarify the nature of this genealogical device. Let me clearly set aside a number of distinct kinds of genealogical reflection that are not part of the exercise below.¹⁰ The aim isn’t to speculate about the actual history or provenance of I-Practices or philosophy. I am not interested in reflecting on the nature or role of intuitions themselves (the mental state), the word “intuition,” or the use of the word “intuition” in philosophy. Those projects might be interesting in their own right, but they are not relevant to the task in hand. The object of the exercise is simply to provide some illumination as to what functions the above I-Practices might serve in philosophy—specifically, what functions they might play that are compatible with Not Evidential. To that end, the genealogical reflection pursued is pure fantasy to help generate ideas.

First Stage: Solitary Thinking

We need to try to imagine an environment in which people do not engage in justifying philosophical claims to others or attempt to provide evidence for philosophical claims, and yet in which something that (in other respects) resembles philosophy or “protophilosophy” exists. A natural place to try and start is an environment in which someone is engaged in (something like) philosophical thinking but isn’t engaged in trying to justify her position or adduce evidence for it simply because she does it without engaging with anyone else.

So, picture a protophilosopher wandering about the streets, interacting with her surroundings, making knowledge attributions, referring to things, approving and disapproving of various things in various ways, thinking about stuff, making judgements and beliefs, being fairly confident about some stuff and super unsure of other things, being au fait with various culturally peculiar practices, beliefs, myths, and so on. She might reflect on these and other ordinary aspects of life, ask questions about them, and eventually come to have some interesting thoughts about them. That is to say, she generally tries to make sense of the world, including her own thoughts and experiences.

Does this solitary protophilosopher have any use for I-Practices? Perhaps some. For instance, she may purposefully structure her thinking around what to say about various cases. However, I think it is plausible that the only real need for most I-Practices emerges within interactions with others. So we need to consider a slightly different imaginary environment.

Second Stage: Interactions

Is there a way we might change our solitary protophilosopher’s environment—such that a need for I-Practices might emerge—without introducing practices involved in adducing evidence and justification? We can simply imagine an environment in which our protophilosopher engages with others but doesn’t try to persuade them of anything.

So, suppose our protophilosopher has for the first time reflected philosophically on her world, asked some new questions, and hesitantly considered some answers. Now she wants to share her enquiry with others. This is tricky. How might she proceed? It would be natural to attempt to retreat to familiar ground: “So, you know \(X\).” “Sure.” “Well, if you think about it, there’s something odd about…” Because a large part of philosophy starts from reflecting on our ordinary ways of thinking about the world, a large part of the relevant common ground will consist in ways in which her interlocutors are themselves disposed to think. Our protophilosopher might start with cases with which the audience is familiar or which are not too difficult to grasp, and with judgements that the audience itself tends to endorse.

Does our protophilosopher now have any use for I-Practices? I think she clearly does. In particular, the activities described in the previous paragraph are instances of Case Wielding and Idea Motivating as described above. It is important to note that these I-Practices can clearly be of use to her, despite the fact that she is nowhere near a context of justification. The nature of her communicative act is purely explanatory or pedagogical. It takes place in what I will call an explanatory context: a context in which the aim is not to provide evidence for theories; a context in which the primary aim is simply to communicate unfamiliar thoughts; the aim is that someone else comes to understand what you are on about.

Calling this an “explanatory context” might invite some misunderstanding, so I should take a moment to explain what I mean by “explanation” in this paper.¹¹ An explanation has the following features: (a) All things being equal, it is a good making feature of an explanation that one’s audience comes to understand whatever the hell you are on about; and (b) there is a sense in which an explanation fails in the event that it just doesn’t help one’s audience “get it.” It is this sense of “explanation” I use.¹²

Although we have seen how some I-Practices might have a use outside of a context of justification, we haven’t yet seen that all of them might have such a use. In particular, it is unclear whether our interacting protophilosopher would have much use for anything like Assumption Footstamping or Weights and Balances Philosophy. So we need to consider a slightly different imaginary environment.

Third Stage: Differences of Opinion

Can we imagine our protophilosopher’s circumstances evolving in such a way that a use for all I-Practices would emerge and yet there would still not be a context in which their function was to justify or persuade? I think we can.

Consider that her explanation attempts can fail: an explanation that strives to begin from common ground may fail if it turns out what she supposed to be in the common ground wasn’t; she might draw attention to some feature of “the ordinary way of thinking about things” that turns out simply to be a quirk of her personal experience. If our protophilosopher reaches for some common judgements, patterns of thought, or ideas from which to move forward and convey her thinking, it is important that she finds some. Otherwise, the audience just isn’t going to be brought up to speed, they’re not going to understand, and they are not going to be able to share the protophilosopher’s ponderings, questions, or tentative answers.

How might our protophilosopher discover that she has failed in this way? An interlocutor might simply look baffled and walk away. But if they are willing to try to engage, they might ask questions of their own or explain exactly where they get off the bus. “I can see why you would say that about \(X\) kind of case, but I’m inclined to think differently about the sort of case you offered” or “That doesn’t quite match how I tend to think about \(X\)” or “I’m struggling to see how \(Y\) fits in to this… What would you say about that?” Within an interaction of this kind, there could be a clear use for all the I-Practices. In explaining her position, it can be helpful to situate it within the slightly broader landscape of her or their thinking—to help her interlocutor understand the topic within which the interesting question she’s encountered crops up. Structuring the topic around the “fixed points” of the background to her question in terms of (perhaps intuitive) assumptions-to-be-held-fixed or certain dilemmas, etc., could be a helpful way to do this, but doing so needn’t indicate that those fixed points are adduced as providing or enjoying any particular epistemic support. Devices such as these—or laying out intuitions that are respected or claims to be balanced against each other—can serve to articulate how the ideas and questions she is thinking about are related to the worldview of her interlocutor to help them understand it.¹³

So, once a protophilosopher is engaged in conversation with someone else—even if both of them are merely trying to understand where each other is coming from and not attempting to argue for a view or to change each other’s mind—you have a conversational context in which all the I-Practices serve a function. Let me emphasise that, in such a context, it doesn’t make sense to talk of intuitions (or indeed anything) being used as evidence. The nature of the interaction is (by stipulation) one in which the sole exercise is attempting to explain/understand. I-Practices in such a context are not playing a part in arguments for positions or being used to persuade other parties of any position. Whatever is being used, it is being used in the elucidation of ideas and in the sharing of one’s ideas with others, and that includes anything that you might be tempted to call an “intuition.”

3 Explanatory Practices

The genealogical reflection in § 2 has generated a hypothesis, a possible alternative explanation of I-Practices in philosophy. What function might I-Practices play in philosophy, if not justificatory/evidential? The hypothesis: they serve an explanatory role. The next step is to assess how plausible it is that this is the role I-Practices actually play within contemporary philosophy. I will make the case that it is somewhat plausible.

My case for its plausibility has four parts. First, I make the case that it is still plausible to regard an I-Practice as serving an explanatory role even when deployed by philosophers who are presenting arguments and attempting to persuade each other of positions. Second, I take each I-Practice in turn and demonstrate that we can, in the abstract, make sense of each of them serving a purely explanatory role within contemporary philosophy. Third, I take a series of concrete examples of I-Practices from recent journal articles and demonstrate that we can give a plausible reading of each of them such that the function of the I-Practice is merely to explain and not to justify or adduce evidence. Fourth, I address the concern that while case-by-case one can give a plausible reading in which I-Practices play a purely explanatory role, what the proponent of Not Evidential needs to be plausible is an alternative reading of the whole philosophical corpus in which I-Practices are not understood as typically functioning to adduce evidence or to justify.

3.1 Plausibility within a Context of Persuading

Is it plausible to regard an I-Practice as serving an explanatory role when it is used within the context of contemporary philosophy, in which philosophers present arguments and attempt to persuade each other of positions? The context in which philosophers use I-Practices is very different from the imaginary environment we invoked in our genealogical reflection. Philosophers are not just a bunch of people explaining their ideas to each other. Philosophy, as we know it, is a project of joint enquiry into the nature of reality, including ourselves and our place in it. We are looking for the truth, or something like it. We defend positions. We object to others. We offer reasons in favour and reasons against. The nature of our conversations is one in which persuasion plays a big role. Philosophers offer evidence for and against positions. There are elements of the picture of philosophy I just presented that you might want to deny.¹⁴ But, for sake of argument, suppose this picture is right.

Our context and the imaginary context in which we placed our protophilosopher are undeniably very different. Indeed, were we to imagine our protophilosopher’s environment evolving to more closely resemble our context, many things in their lives would change. However, it is important to recognise that the function of I-Practices need not change. There is no in principle reason why I-Practices couldn’t serve an explanatory function in a context such as ours.

It might seem implausible that I-Practices serve an explanatory function in philosophy because they appear to play a justificatory/evidential role. But this appearance would be expected, even if their role is explanatory. It would be easy to mistake the function of explanatory I-Practices in philosophy. One reason for this is that understanding an idea, at least in the sense of it making sense to one, is something like a precondition for accepting the idea. There will be, if you are good at persuading, a decent correlation between those to whom you successfully explain your position and those who are persuaded of your position. However, that doesn’t mean the I-Practices (playing a role in the relevant process of explanation) themselves serve the function of persuasion. It is at least a tenable view that in a context such as ours—a context of arguments, persuasion, and joint enquiry into the nature of reality—the function I-Practices themselves are playing is really still only an explanatory one. And that whatever persuasion we manage is achieved by other means. And that whatever evidence we adduce is adduced in distinct practices.

The idea that some of the things philosophers do—in the process of defending positions and attempting to ascertain the truth—are not themselves playing an evidential function is essential to the argument of this paper. While a simple idea—and I hope commonsense—it is worth taking a little time to clarify, as I have found the idea invites misinterpretation. It would be implausible, I take it, that everything a philosopher does in the process of defending a position is best understood in evidential terms. When I press the letter “T” on my keyboard as I type this, it seems unnecessary to insist that the function of that keystroke was to adduce evidence or anything similar. Likewise, when I say, “Here is the view I am attacking… and the reason I think it is flawed is…” we can make sense of the idea that the function of my activities during the first ellipsis is not best understood as involved in providing evidence against the relevant view. So long as we can make sense of that, we can make sense of the idea that philosophers’ use of I-Practices, though occurring within a discourse in which philosophers defend positions, etc., nonetheless doesn’t itself function to provide evidence or justify positions.

Note something else about the ellipses case. You will perhaps have observed that, in this case, there are possible (and perhaps prima facie plausible) revisionary readings of the case in which the activities during the first ellipsis do serve an evidential function despite the way they are explicitly framed. This exemplifies a more general truth that is also important to this paper: we should be open to the possibility that revisionary readings of philosophical methods might be the best; it is reasonable to think that our best descriptive account of the methods in any domain might produce some surprises, such that some of the aspects of the relevant approach that were thought to be serving such-and-such a function are revealed by the best descriptive methodology to be doing something else. The proponent of Not Evidential who adopts the idea that I-Practices serve a purely explanatory role is going to be doing something similar. I don’t pretend that their account wouldn’t be a surprising one.

3.2 Plausibility for Each I-Practice

To make the case that it is plausible to view I-Practices as serving explanatory roles when used in contemporary philosophy, we can now look at each of the I-Practices in turn. In each case, our question is the same. Is it plausible to view this I-Practice as playing a fundamentally explanatory role—as involving the use of intuitions to explain rather than persuade, or in the business of making sense of things rather than in discerning and establishing the truth—in the context of contemporary philosophy, even if the I-Practice might at first seem to betray the use of intuitions as evidence?

Can We Make Sense of Case Wielding Being Explanatory?

Suppose you are thinking about society-level decisions and their impacts on the next generation and generations to come. There are various tensions that arise. One such tension is revealed by the so-called repugnant conclusion (see Arrhenius, Ryberg and Tännsjö 2017). If you are going to explain tension to someone, you need to start appealing to some pretty wacky cases—not as evidence but just to help them understand your viewpoint. One is not explaining that such-and-such is true of a case or offering the truth of such-and-such as evidence in favour of a view; one is simply explaining the view in relation to familiar examples or examples whose structure is well-suited for the task. For example—and here, fittingly, the purpose of my example is simply to illustrate the nature of my claim in this section and not to provide evidence for any particular view—the idea might be that when a philosopher appeals to a pair of cases, those of the “trolley problem,” in the context of discussion of deontological and consequentialist ethics, the function of the appeal is simply to illustrate the content of one or other view, or indeed both views—to explain what would be true were \(X\) true.

We needn’t deny that the nature of other philosophers’ interest in your viewpoint is ultimately whether it is true/defensible. The idea is simply that such a practice of Case Wielding itself needn’t be in the business of adducing evidence that one’s viewpoint is true. Your interlocutor can challenge the spin that you put on cases, they can fail to see where you are coming from, the conversation can be one in which persuasion is the main aim and in which your interlocutors reach for new cases during their attempt to persuade you. But we can still make sense of the idea that, in reaching for the cases in attempts to persuade, the real work being done is to explain a position to you in the hope that you judge it to be true, rather than to adduce evidence.

Can We Make Sense of Idea Motivating Being Explanatory?

In § 1, the example I used to introduce the I-Practice of Idea Motivating involved providing some motivation for some debate: noting an apparent tension between two claims that we might ordinarily endorse, showing that the problem we are grappling with arises out of our ordinary ways of thinking about something. Consider utterances such as, “There’s something to be said in favour of \(X\), and there’s something to be said in favour of \(Y\).” Such utterances are fairly familiar in philosophy. And while we might interpret them in terms of articulating pro tanto evidence or prima facie justification, there is another way to understand what is going on. The key thought would be as follows: ultimately, the reason that we motivate the debate in these terms is that it helps to explain to people the shape of the philosophical debate.

You might object that such an interpretation only makes sense in case the speaker is offering the intuitiveness of these claims as a reason to accept them—why else would there be an interesting puzzle?—but that’s not right. There is a philosophically interesting puzzle to be found simply in noting a tension in our ways of thinking and exploring that tension: we ordinarily think both \(X\) and \(Y\), which seems inconsistent when you think about it; I wonder if we can resolve that. That puzzle can be presented without the starting position being “we have evidence for \(X\) and evidence for \(Y\)…” Now, of course, there would be no tension to be explored unless truth was in the mix somewhere—the tension is precisely that we tend to think both, but they can’t both be true—but that doesn’t mean that when philosophers motivate ideas in this way, they are adducing evidence or employing a device whose function is to persuade. My point is simply that we can make sense of the relevant I-Practice—Idea Motivating—playing an important role in philosophical conversations where, even if the conversation is concerned with truth and arguments, the function of this I-Practice is not to provide support, adduce evidence, etc., but purely explanation.

Can We Make Sense of Assumption Footstamping Being Explanatory?

Take an example of a philosophical debate that revolves around a particular knotty problem, perhaps the free will debate. Although our interest is in philosophical methodology (i.e., research) rather than pedagogy, it will be helpful to consider for a moment how one might introduce the topic for the first time in an undergraduate class. You need to find a common ground from which everyone in the conversation is willing to work. The conversation doesn’t yet have a set of assumptions. And, even though the game is still exploratory and explanatory, recognizing and drawing attention to relevant common assumptions can play an important role in ensuring that everyone in the discussion is able to understand the ideas being bandied about. When you are trying communally to make sense of the world in a way that involves sharing and explaining ideas, there needs to be some common ground that helps one articulate one’s thoughts to those in the conversation. Moreover, to ensure that everyone is making sense to each other, it can be helpful to police that conversation—making clear what everyone in the conversation accepts without argument and even excluding those potential interlocutors who don’t accept these things from the conversation.

Now consider the philosophical debate about free will outside the context of teaching introductory classes. It is quite plausible to view the role of footstamping in the debate between philosophers as continuing to play the exact same role. The nature of the conversation is different, for sure, as participants in academic debate are interested in persuading each other of various claims. Moreover, these assumption footstamping practices might well seem to take on a different role since the positions considered in the relevant conversation may well stand or fall on some set of assumptions. It may look like, especially from the outside, these footstamping practices involve presenting certain “intuitions” as being taken as evidence in the debate. But if we can still make sense of these practices being essentially explanatory, then rather than the articulation of assumptions being in the business of noting what is taken to be evidence in the debate, the practices are simply involved in the business of explaining the nature of the positions, arguments, and debates concerned.

We are now in the position of being able to deal with the last few I-Practices relatively quickly. It should be rather clear how the general pattern of thinking might be applied to each of them, given the discussion so far.

Can We Make Sense of Intuition Respecting & Weights and Balances Philosophy Being Explanatory?

Why respect intuitions? Why draw attention to the fact that something you want to say is compatible with some things that others want to say and is not compatible with some things that they want not to say? Doing this can help to make sense of ideas for an audience by helping to locate the idea within the audience’s worldview. This can still be their function in the truth-oriented project of an enquiry that is contemporary philosophy. Admittedly, the nature of the conversation is different, and the relevant parties are interested in truth and in discerning and evaluating favouring and disfavouring considerations. But, still, the inclusion of something intuitive or even described as an intuition on a list of desiderata or favouring considerations can still be interpreted as a communicative act, to communicate that this is a claim that, in the current debate, is taken to have something to be said for it, and importantly, an act that doesn’t adduce the claim as evidence or as enjoying any particular evidential status.

Can We Make Sense of Negotiating Dilemmas, Inconsistent Triads, and So on Being Explanatory?

Why structure philosophical debates in these ways? On the current hypothesis, their main function is communicative. Crystallizing issues around sets of inconsistent propositions brings a tight focus on the precise issue in question. They break an issue down into discrete chunks, which can be explained separately. Structuring debates in this way ensures that everyone is on the same page and that they do not talk past each other. Within the philosophical community, this might naturally be interpreted as laying out the evidence, but we can also make sense of the idea that structuring discourse around dilemmas, etc., is to employ an explanatory device. Intuitive claims feature, where they do, in part because they help to smooth the process of communication: I can more easily understand what you are telling me about a particular theory if you can relate it to some intuitive claims I find very easy and familiar.

3.3 Plausibility in Recent Articles

Perhaps the above isn’t enough. To make the case that the alternative hypothesis is plausible, perhaps we also need to see it give a plausible reading of real examples from contemporary literature.¹⁵ So let’s turn to some examples from contemporary literature.¹⁶ Looking at these examples is particularly helpful for addressing any residual concern that the alternative explanation can’t make sense of the way that philosophers engage with the I-Practices of other philosophers.¹⁷ For example, the worry might be that the idea that the function of these practices is explanatory can’t make sense of the way that philosophers respond to and engage with each other’s appeals to cases. Some of the examples are helpful for addressing such worries as they involve examples of engagement with (rather than use of) I-Practices.

A note on methodology in this section: what is needed is to demonstrate that the alternative explanation under consideration is a somewhat plausible explanation of what is going on, to which the proponent of Not Evidential might appeal. The alternative explanation under consideration is that the function of I-Practices is to explain (rather than justify or persuade). It isn’t necessary that the alternative explanation is the most plausible explanation, the only possible explanation, or something similar. The most important thing is that it isn’t completely implausible for the proponent of Not Evidential to hold that in the cases considered, the I-Practices serve an explanatory function.¹⁸

Example #1: Pryor (2000)

Pryor begins as follows:

Consider the skeptic about the external world. Let’s straightaway concede to such a skeptic that perception gives us no conclusive or certain knowledge about our surroundings. […] Let’s also concede to the skeptic that it’s metaphysically possible for us to have all the experiences we’re now having while all those experiences are false. Some philosophers dispute this, but I do not. (2000, 517)

Later on, he makes a (in some respects) similar move, this time more explicitly framed in terms of “intuitions”:

For a large class of propositions, like the proposition that there are hands, it’s intuitively very natural to think that having an experience as of that proposition justifies one in believing that proposition to be true. What’s more, one’s justification here doesn’t seem to depend on any complicated justifying argument. […] One might be wrong […] But it seems like the mere fact that one has a visual experience of that phenomenological sort is enough to make it reasonable for one to believe that there are hands. No premises about the character of one’s experience—or any other sophisticated assumptions—seem to be needed.

I say, let’s take these intuitive appearances at face value. (2000, 536)

How might we interpret these moves? There is a straightforward interpretation according to which the moves serve to set aside certain claims in a debate as basic intuitions that enjoy a certain epistemic privilege, such that it doesn’t make sense to question them. This seems to be a species of what I called “Assumption Footstamping” above.

But can these moves plausibly be explained—as per our alternative strategy—as serving a purely explanatory function within a context in which Pryor is certainly interested in justifying a conclusion? Yes, we can. Here, the alternative explanation for these practices is that they (particularly in the first instance) serve to help Pryor isolate the topic of interest, the locus for further discussion, and (particularly in the second instance) to communicate the nature of the claims involved.

Example #2: Elga (2007)

Elga appeals to some cases in setting up the issues for discussion:

When it comes to the weather, I completely defer to the opinions of my local weather forecaster. My probability for rain, given that her probability for rain is \(60\)%, is also \(60\)%. […] I treat her as an expert about the weather. That means: conditional on her having probability \(x\) in any weather-proposition, my probability in that proposition is also \(x\).

In treating my forecaster this way, I defer to her in two respects. First, I defer to her information: “As far as the weather goes,” I think to myself, “she’s got all the information that I have—and more.” Second, I defer to her judgment: I defer to the manner in which she forms opinions on the basis of her information.

In the above case, we may suppose, I am right to treat my forecaster as an expert. But advisors don’t always deserve such respect. For example, suppose that the forecaster has plenty of meteorological information, but I can see that she is dead drunk and so isn’t responding properly to that information. In that case, I shouldn’t treat her as an expert. Or suppose that the forecaster responds perfectly well to her information, but I can see that I have information that she lacks. In that case too, I shouldn’t treat her as an expert.

Even in such cases, I shouldn’t just ignore her opinion. […] If my forecaster is drunk or otherwise addled, then I should only partially defer to her judgment. I postpone discussion of such cases. For now, suppose that I do completely defer to my forecaster’s judgment. Nevertheless, I think that she lacks relevant information that I possess. What then?

An example will suggest the answer. Suppose that my forecaster lacks one highly relevant tidbit: that I have been secretly seeding the clouds for rain. Suppose that I’m sure her probability for rain is low—\(5\)%, say. In this case, I shouldn’t set my probability for rain to that same low value, since my cloud-seeding activities make rain much more likely. But I should be guided by the forecaster’s opinions. Roughly: my probability for rain should be what hers would be if she were informed that I’d been seeding the clouds. (Elga 2007, 479, emphasis in original)

It is tempting to read such a use of cases as involving Elga justifying his opinion to the reader by adducing evidence in the form of intuitive verdicts about these cases that Elga assumes the reader will share.

But we don’t have to think of the device of appealing to cases as playing that kind of role here. Yes, Elga is, in general in this article, interested in defending a view and not just communicating it. But we can still understand an appeal to cases—indeed, the above appeal to cases—within that context as playing a purely communicative/explanatory role: the appeal to illustrative cases serves simply to articulate a certain view about appropriate norms for deference. Of course, it is key to the illustration that his readers don’t radically depart from his understanding of the cases. But we don’t have to think of the specific manoeuvre, viz., appealing to these cases, as serving to provide support for any position or offer reasons for accepting it. We can think of this manoeuvre as merely serving to communicate the position in terms likely to be easily intelligible to the expected audience.

Example #3: Hohwy (2016)

Hohwy’s paper is involved in making an indirect case for a particular model of brain functioning: PEM (prediction error minimization). Part of that indirect case involves engaging with the idea of extended cognition—the idea that cognitive processing is not skull-bound or body-bound but extends into the world—and engaging with part of the standard case for thinking of cognition as being extended. Part of the standard case for thinking of cognition as being extended is that extended cognition models respect the idea that there can be a functional equivalence between neurally-based processes and processes partially realised in external resources such as a notebook or smartphone. Hohwy’s engagement with this involves pointing out that PEM may be able to avoid any pressure to accept that cognitive processes extend into the world since, according to PEM’s model of what the brain is doing, there isn’t obviously a relevant functional equivalence between what the brain is doing and processes involving notepads and smartphones.

[I]t is far from clear that notebooks and smartphones actually play any part of the functional role set out by PEM. There does not seem to be the right kind of hierarchical message passing between the notebook or phone and the rest of the neural system to implement variational Bayes with respect to hidden causes. The challenge is to specify the role of notebooks or smartphones, or any other thing, such that it clearly plays an appropriate prediction error minimization role. (2016, 270)

Hohwy isn’t obviously directly involved in any of the I-Practices itemized above. Instead, he is engaging with others’ use of I-Practices. It is natural to read the proponents of extended cognition models as presenting a case that rests on intuitive responses to certain kinds of cases—intuitively, there is no relevant difference between certain cases—and to read Hohwy as trying to undermine that case by undermining the intuitive evidence, or perhaps as disputing the relevance of the intuitive evidence depending on how we parse the case.

But we don’t have to read the role of cases in this way. It is plausible instead to read the cases, as used by proponents of extended cognition, as serving simply to explain what it means to say that cognition is extended and to articulate the sorts of features that need to be present for it to be appropriate to treat a case as one of extended cognition. We can then read Hohwy, specifically regarding his engagement with these cases, as explaining how PEM isn’t committed to thinking that those features are present in the relevant cases. Of course, in the paper as a whole, Hohwy isn’t just pointing out who is committed to what. He ultimately has views and intends to defend them. But we don’t have to think of either (a) the use of cases by the proponent of extended cognition models or (b) Hohwy’s engagement with their use of cases as having to do with adducing or undermining evidence. It’s plausible to view these specific aspects of the discourse as serving purely explanatory roles.

Example #4: Nichols and Knobe (2007)

Nichols and Knobe’s paper involves experimental philosophy. The way they present their project makes it clear that they (like Hohwy 2016) are engaging with other philosophers’ use of I-Practices rather than directly using them themselves:

The debate between [compatibilism and incompatibilism about moral responsibility] has invoked many different resources […] But recent discussions have relied heavily on arguments that draw on people’s intuitions about particular cases. Some philosophers have claimed that people have incompatibilist intuitions (e.g., Kane 1999, […]); others have challenged this claim and suggested that people’s intuitions actually fit with compatibilism […]. But […] relatively little has been said about the specific psychological processes that generate or sustain people’s intuitions. And yet, it seems clear that questions about the sources of people’s intuitions could have a major impact on debates about the compatibility of responsibility and determinism. There is an obvious sense in which it is important to figure out whether people’s intuitions are being produced by a process that is generally reliable. (Nichols and Knobe 2007, 663–664)

It is natural to read Nichols and Knobe as being concerned with investigating whether intuitions about cases that they take to be used as evidence by others should be used as evidence.¹⁹

To assess whether we can give a plausible alternative explanation of the practice in this case, it is helpful to go back to the literature they are engaging with. For example, here is Kane (1999):²⁰

The second stage of this two-stage argument in support of compatibilism will concern me here, the one that goes through [the principle that if an action is undetermined at a time \(t\), then its happening rather than not happening at \(t\) would be a matter of chance or luck, and so it could not be a free and responsible action] […] to show that indeterminism would […] undermine, freedom and responsibility. What is at stake here is not merely the clever arguments of philosophers; for it happens that the case for [such principles] is a powerful one. It is difficult to see how indeterminism and chance can be reconciled with freedom and responsibility. Philosophers have tried to bring this out in a number of ways […] We may think of these as the varied intuition pumps that support [such principles]. […]

2. […] suppose a choice occurred as the result of an undetermined event (say, a quantum jump) in one’s brain. Would that be a free choice? Being undetermined, it would appear to be more of a fluke or accident than a free and responsible action. […]

3. […] Imagine that you are intending to make a delicate cut in a fine piece of cloth, but because of an undetermined twitching in your arm, you make the wrong cut. Here, indeterminism is no enhancement of your freedom, but a hindrance or obstacle […].

4. Even more absurd consequences follow if we suppose that indeterminism or chance is involved in the initiation of overt actions. Arthur Schopenhauer imagined the case of a man who suddenly found his legs start to move by chance, carrying him across the room against his wishes. (1999, 219–220)

Although Kane talks of such “intuition pumps” supporting principles, we don’t have to think of this being their true function. Suppose the relevant philosophers are not, by appealing to these cases, providing evidence against the idea that indeterminism is the friend of freedom, is there any way to make sense of what function these appeals might be playing? We can see these manoeuvres as being concerned with explaining a position (i.e., that which the relevant philosopher thinks is false). It is plausible to view the function of the “intuition pumps” being simply to help us see what it means to say that indeterminism is the friend of freedom.

Example #5: Haslanger (2000)

Haslanger’s paper offers revisionary accounts of gender and race. As she puts it, “the task is to develop accounts of gender and race that will be effective tools in the fight against injustice” (2000, 36).²¹ She makes a couple of moves in the vicinity of either intuition-respecting (of the retrospective variety) or weights and balances philosophy. After offering an account of gender, she says, “One advantage of this account of gender is the parallel it offers for race” (2000, 43). Then, following the main discussion, Haslanger opens the closing sections, saying,

Let me now turn to summarize some of the advantages of the proposed definitions. […] [An] interesting question is whether it is useful to think of these groups in these terms: Does it serve both the goal of understanding racial and sexual oppression, and of achieving sexual and racial equality to think of ourselves as men or women, or raced in the ways proposed?

By appropriating the everyday terminology of race and gender, the analyses I’ve offered invite us to acknowledge the force of oppressive systems in framing our personal and political identities. Each of us has some investment in our race and gender: I am a White woman. On my accounts, this claim locates me within social systems that in some respects privilege and in others subordinate me. Because gender and racial inequality are not simply a matter of public policy but implicate each of us at the heart of our self-understandings, the terminological shift calls us to reconsider who we think we are. […]

[T]he appropriation under consideration intentionally invokes what many find to be positive self-understandings […] and offers analyses of them which emphasize the broader context of injustice. Thus there is an invitation not only to revise one’s understanding of these categories […], but to revise one’s relationship to their prescriptive force. By offering these analyses of our ordinary terms, I call upon us to reject what seemed to be positive social identities. I’m suggesting that we should work to undermine those forces that make being a man, a woman, or a member of a racialized group possible […]. This, I hope, will contribute to empowering critical social agents. (2000, 46–48, emphasis in the original)

One might interpret this kind of move as resting on the premise that—at least within critical anti-racist/feminist theory—we should adopt concepts that are effective tools in the fight against injustice; a premise that might be regarded as having the status of a basic intuition (of some kind) at least within the relevant domain of discourse.²² Suppose we run with an intuition-based interpretation of such a move. Is it plausible to interpret the use of the intuition in a purely explanatory way? I think we can. We can somewhat plausibly read the function of such a use of intuitions to be to explain the view proposed and its implications, and the relation between that view and various other positions, albeit within the context of an argument for that view.²³

3.4 Plausibility across the Whole Corpus

One might have some residual doubts even if one accepts that there is a somewhat plausible reading of any individual case along the lines of the alternative explanation I suggest. The thought might be something like the following: I am willing to entertain the possibility that one or two of my friends lie about whether they dye their hair, but I am not really willing to entertain the possibility that they all do so. And what the proponent of Not Evidential needs is more analogous to the latter. The proponent of Not Evidential needs to be able to look at the entire philosophical corpus of I-Practices—those which would otherwise invite an explanation in terms of intuitions being used as evidence—and give a plausible reading of that whole body of practice that is compatible with their position. Nonetheless, it is worth just quickly clarifying what exactly the proponent of Not Evidential needs to be a plausible thing to say about the entire philosophical corpus (and what they do not need to be plausible).

The proponent of Not Evidential doesn’t need it to be plausible that all the instances of I-Practices in the corpus serve a purely explanatory function (although if it is, that’s no problem). Why not? First, because the alternative interpretation of I-Practices I’ve provided here needn’t be the only plausible alternative interpretation. Indeed, by making the case that one alternative interpretation is available, this paper might well provide some indirect evidence that further alternative interpretations may be available. So this particular alternative interpretation ultimately need only be a part of an overall explanation. Moreover, while it is implausible to interpret, for example, all instances of I-Practices as mere functionless fluff or elaborate verbal ticks, it isn’t implausible to think that a small proportion of I-Practices might be accounted for in this way.

Second, proponents of Not Evidential don’t need to be committed to the idea that no philosopher has ever used intuitions as evidence in order to have an important point. Cappelen (2012, 1) may claim that philosophers don’t rely on intuitions as evidence extensively or “even a little bit,” but he conceives of his main point going through even if some philosophers do rely on intuitions sometimes (in a way that isn’t characteristic of philosophy, 2012, 16; or as the result of being misled by a false metaphilosophical idea, 2012, 1). And, indeed, Cappelen is explicit that the claim he targets—“Contemporary analytic philosophers rely on intuitions as evidence for philosophical theories”—is to be understood as “a claim about what is characteristic of philosophy,” which “allows for some exceptions, but is true only if it applies to a wide range of paradigms of contemporary philosophical practice” (2012, 15–16, my emphasis). Similarly, Deutsch—even though he states his personal opinion is that “intuitiveness does not play even a small evidential role in any philosophical argument” (2015, 78), and says things like it is “simply a myth that philosophers employ a method, in reasoning about thought experiments and cases, whereby they make evidential appeals to intuitions” (2015, xvi)—is clear that the claim he really takes issue with in his book is the position that philosophers “treat intuitions as the only, or at least as essential, evidence, when it comes to the truth or falsity of judgments about thought experiments” (2015, 77, my emphasis), and says that his point is that “[appealing to the intuitiveness of a judgment about the thought experiment in order to provide (what he or she regards as) essential evidence for the truth of the judgment] is not at all common in analytic philosophy” (2015, 174, my emphasis).

So, for my argument to go through, it doesn’t need to be plausible that all I-Practices are purely explanatory. Nonetheless, it does need to be plausible that large proportions of I-Practices in philosophy are purely explanatory. How plausible? Well, there is no precise threshold. That’s not how theoretical costs and benefits work. I have said that a significant cost to Not Evidential would be that it rendered I-Practices inexplicable. But it is difficult to put a figure on how significant a cost this would be—and readers will likely vary in their assessments. It is also just one cost among many relevant costs and benefits relevant to theory choice in this area. Various considerations have been raised and debated for and against Not Evidential in the literature—see footnote 2—and so it is difficult to give a definitive idea of how this one consideration stacks up against other relevant considerations.²⁴ Perhaps the best I can say is simply that the more plausible an alternative explanation of I-Practices in philosophy the proponent of Not Evidential can give, the better.

How plausible is it that large proportions of I-Practices in philosophy are purely explanatory? I can only report that for myself, it is sufficiently plausible to take Not Evidential more seriously than I did before, i.e., when the position appeared to render many aspects of philosophical practice inexplicable. Proponents of this position can say something sensible about the kinds of aspects of philosophical practice that might have looked anomalous given their position: they serve a function in communication between philosophers, and (for most of them) that function is purely communicative rather than supportive—to explain key ideas and dialectics. I have made this case in general terms (in § 3.1 and § 3.2) and only appealed to specific cases (in § 3.3) in order to demonstrate that the general interpretative strategy doesn’t come unstuck when encountering the details of individual cases.²⁵

4 Concluding Remarks

There is a recent trend in metaphilosophy to defend Not Evidential. I-Practices present a prima facie challenge to Not Evidential. A natural explanation of the fact that philosophers widely engage in I-Practices is that philosophers use intuitions as evidence. But it is not obvious how to explain I-Practices given the truth of Not Evidential. It is not compelling to think that I-Practices are empty and contribute nothing to philosophy. The purpose of this paper is simply to explore whether there is any alternative story—a story that is compatible with Not Evidential—to be told about the function of I-Practices within philosophical practice.

The conclusion is that there is a somewhat plausible story that one might tell. The brief genealogical reflection above provided the seed of an alternative to viewing I-Practices as involving an evidential use of intuitions: the contribution of I-Practices is that they help philosophers to understand what each others’ positions are; they are part of philosophers’ attempts to explain ideas to each other; I-Practices play a purely explanatory role in philosophy. To accept this explanation is not to accept a view that says philosophers are always only in the business of explaining positions to each other, but rather, specifically, that the function of our I-Practices within our wider practice is an explanatory one. I have not defended the claim that this is in fact the function of I-Practices. Merely that it is a somewhat plausible story that is available to the proponent of Not Evidential. The availability of this alternative explanation of I-Practices provides indirect support for Not Evidential.

Before closing, let me see off a few potential lines of resistance and acknowledge some limitations to my argument.

Objection #1

If the role of I-Practices is purely explanatory, then none of our positions are justified in philosophy, and that’s false!

The idea here might be that all the time philosophers have taken themselves to be providing evidence for their positions, they have merely been explaining their positions, and thus their positions lack justification. However, note that just because I-Practices don’t involve adducing evidence or providing justification, this doesn’t mean that philosophers don’t adduce evidence for their positions or justify their positions, nor does it mean that philosophers don’t have evidence for their positions or have justified positions.²⁶

A subtler, but related, objection is that in many cases—such as those case studies selected above—once we interpret the relevant I-Practices within a paper or literature as serving purely explanatory functions, it becomes somewhat untenable to even interpret the relevant philosophers as providing arguments for their views. The thought might be: that the only bits of those papers or literature that are plausibly interpreted as involved in justifying positions or evidence adducing are the bits involving the employment of I-Practices; so, once we interpret I-Practices as serving purely explanatory roles, there’s no plausible reading available on which these philosophers are arguing for their positions.²⁷

In response, I think it is important to make two points. The first is to concede that there may be some instances in which the proponent of Not Evidential, who interprets I-Practices as I suggest, is forced into interpreting a philosophical text that we typically think of as providing an argument for a view as doing nothing of the kind. I’m not inclined to think this is a big problem, as I see no reason to think they’ll be forced into such an interpretation for vast swathes of the philosophical literature—only the occasional instance. But I acknowledge that it is a cost for the proponent of Not Evidential, who interprets I-Practices as I suggest, that they may sometimes interpret a philosopher as not presenting an argument (when the philosopher themselves thinks they are). However, the alternative explanation I argue is available to the proponent of Not Evidential isn’t typically going to say, “\(X\) thinks they are giving an argument, but they are not.” It says, “\(X\) is giving an argument, but the function of device \(Y\) within the discourse is purely explanatory.” And that only involves going against \(X\)’s self-conception of their practice insofar as they had any commitment specifically to the function of \(Y\) being something more than purely explanatory—and I’m not sure how common that is.

The second is to note that a single passage of philosophy may contain both an I-Practice that serves solely to explain the claim that \(p\) as well as present the claim that \(p\) as evidence. This is the interpretation I envisage the proponent of Not Evidential, who interprets I-Practices—as I suggest—as playing a purely explanatory role, taking in relation to any cases in which a passage containing an I-Practice seems to be the only plausible candidate for a passage in which support is offered for a position. Consider the following imaginary excerpt from a text in which the I-Practice of appealing to cases plays a role:

Suppose that no intuitions are adduced as evidence in the presentation of this text. Instead, taking a line typical of proponents of Not Evidential, let’s say that what is adduced as evidence is something like the claim that some berries are not cherries or else something like dewberries are berries, dewberries are not cherries, or similar. If that’s right, that deals with the question of, “How is the argument supposed to work if it doesn’t involve the use of intuitions as evidence?” But it would still leave open the question that motivates this paper: “What role is the appeal to cases playing there?—can it just be empty fluff?” And the answer proffered in this paper is that the appeals to intuitive judgements about cases are not serving to adduce evidence or present reasons; rather they just explain some relevant claims—claims that might themselves be being adduced as evidence.²⁸

Objection #2

If the role of I-Practices is explanatory, then I-Practices (often) involve the use of intuitions as evidence (which is inconsistent with Not Evidential and the fact that philosophers do use I-Practices).

There are various reasons one might hold this conditional. There are two I have encountered.

The first line of reasoning exploits a putative link between understanding and modal knowledge. The thought might be that philosophers often use conceivability as a test of possibility. Insofar as that is true, a device employed for helping someone merely understand/conceive of something in philosophy is often ipso facto in the business of evidencing a philosophical theory. Thus, it might be argued, there is no such thing as a purely explanatory practice in philosophy, since a successful explanation provides evidence for a modal claim. However, the types of explanations involved in I-Practices do not seem to fit this picture. For instance, explaining questions and articulating the basic structure of debates: these are not explanations that fit into this picture whereby explanations provide understanding, which is in turn linked to conceivability; the type of understanding they provide doesn’t seem to have any straightforward link to which worlds are possible. The types of explanations that I-Practices are concerned with is not in explaining scenarios or thought experiments in the hope that one’s audience will recognize that the described scenario is conceivable. For just one example, in “Case Wielding,” a case is used to explain a theory. And it doesn’t seem right to me to say that explaining a theory is automatically a matter of providing evidence that the theory is possibly true or necessarily false. Moreover, just because an explanation helped make certain evidence available wouldn’t mean that the explanatory practices used were ipso facto practices whose function is to provide evidence.

The second line of reasoning exploits a model of philosophical enquiry as being largely “abductive,” i.e., seeking the best explanations of the available evidence. Insofar as philosophers are always or very often doing that, might be the thought, there is no useful distinction between justificatory and explanatory practices. There are various concerns one might have about this thought. The most important point to make, however, in response to the thought as an objection to my argument in this paper, is that it exploits an equivocation. Suppose we are in the context of a general collaborative effort to determine the best explanation of the available evidence about some philosophically interesting phenomena. Within such a context, there will be some practices I perform whose function is to provide support for one or another explanation being the best explanation, and there will be some practices I perform whose function is to communicate the precise nature of some candidate explanation under discussion. Given what I mean by “explanatory practices,” there is plenty of room for practices performed within the context of abductive enquiry to not be explanatory practices, and explanatory practices within such a context are not automatically involved in supporting/evidencing/justifying.

Let’s quickly elaborate on why one might think the distinction was not empirically tractable in the right way. First, we can’t test this by simply asking philosophers what methods they use. It is a cornerstone of the whole debate in this area that philosophers may well be largely mistaken about what they are doing when they philosophize. It is on the table, for instance, that despite thinking that they have been using intuitions as evidence all along, many philosophers are mistaken about this—that’s what proponents of Not Evidential tend to hold. Second, we can’t seem to test this by examining philosophical texts, talks, and discussions. Why? Since philosophers are generally trying to make the case for or against some position, it is unclear what would enable one to discern, from the surface features of our texts or speech, whether some particular device really has a purely explanatory function—indeed, the cases studied above demonstrate one can interpret cases in various ways. So, the thought might be, this is not a plausible hypothesis because it is amorphous and not empirically tractable.

The most important thing to do in response is to note that there are virtues to an account of philosophers’ use of I-Practices beyond being able to give an empirically adequate account of I-Practices. There may be non-empirical considerations that speak in favour of an account that focuses on an explanatory function rather than a justificatory one. There may also be empirical considerations relating to distinct but related matters, philosophers’ use of intuition-related language (see Andow 2015a), what else philosophers do in close proximity to I-Practices (see Cappelen 2012), the extent to which philosophers are disposed to update their views in relation to I-Practices (when encountering certain framings of debates or cases), the extent to which philosophers use I-Practices in contexts that are explicitly oriented around explanation (teaching, textbooks). Ultimately, deciding the best view of philosophical practice is going to involve the synthesis of a whole host of considerations. And so, while I agree that a plausible alternative explanation of philosophers’ use of I-Practices ought to be “testable” in the sense that there ought to be some considerations that would support it rather than the standard explanation, the potentially relevant considerations are not limited to empirical observations directly of philosophers’ I-Practices. One place to start looking for such relevant considerations would be, of course, the various considerations that proponents of Not Evidential adduce in favour of their view in the first place (see the opening section for references and Nado 2016 for a critical overview).

Objection #4

When talking to each other, philosophers don’t need to explain basic concepts. It is plausible that when talking to new students and to members of the public, philosophers need to explain their ideas. It is not plausible that, when addressing each other at conferences and in print, philosophers have any great need to explain concepts, positions, or ideas.

I have encountered this objection, but it rests on either a misunderstanding of my analysis or on an implausible worldview. Take a look back over this paper and estimate the proportion of the wordcount that is involved purely in articulating and clarifying the view under consideration. My estimate is that it is the vast majority. That is my experience of writing philosophy, reading philosophy, giving talks, and sitting in talks. When addressing philosophers, maybe we don’t need to explain exactly the same things that we need to explain when addressing nonphilosophers, but surely there is still a lot of explanation that needs doing. I don’t claim that I-Practices are used among philosophers to explain exactly the same ideas that one has a need to explain when addressing nonphilosophers; the claim is that I-Practices can be interpreted as playing the same general explanatory function. I think it would be very implausible to think that communication between philosophers didn’t require a lot of explanation.

Objection #5

The alternative explanation given for philosophers’ use of I-Practices is problematic because there is nothing that unifies I-Practices. There is no assurance that this alternative explanation will extend to all phenomena—including other I-Practices that may exist but which haven’t been explicitly addressed in this paper—that make attractive the view that philosophers use intuitions as evidence.

This is a genuine limitation of the current project. I have only demonstrated how the alternative explanation can handle the specific set of practices considered above. These are the most salient examples of practices that would otherwise seem difficult for the defender of Not Evidential to explain. And I’ll admit that there are not any other salient practices of this kind that I can think of. Indeed, the list invoked in this paper is already much longer than that with which most of the methodological literature around intuitions has concerned itself (the fixation tending to be on appeals to cases). Nonetheless, it may well be that there are some such practices I haven’t thought of, and it would be important to ask whether the suggested alternative explanation can be extended to them. If it didn’t, the defender of Not Evidential would have to look for a different alternative explanation. However, this is only a slight possible limitation to the main contribution of this paper, which is to give the defender of Not Evidential the resources to discharge the theoretical burden of not being able to account for I-Practices—they can at least account for the most salient ones.

Objection #6

I agree that I-Practices, as described in the proto-philosophical context, plausibly play only an explanatory role. However, I don’t accept that the very same practices exist within contemporary professional philosophical discourse. For example, any “Weights and Balances” philosophy conducted by proto-philosophers would clearly be in a merely suppositional/exploratory/speculative/explanatory spirit, but that means that what proto-philosophers do is far less committal than what actual philosophers do, sufficiently different to mean it is a distinct practice (or at least that the plausibility of the explanatory reading doesn’t carry over).

I have a few things to offer in response. First, there are lots of things that look different once one enters an academic context; I’m willing to bet you could train an AI to identify academic contexts based solely on data about blinking, gesticulation, or turn-taking in conversation. I wouldn’t want to confuse the performative elements of academic discourse with anything deep about the practice. Second, I don’t buy that the distinction is as sharp as the objection suggests. One thing that the work of folks like Cappelen (2012) and Deutsch (2015) makes clear is that the “stereotype” putative instance of “using an intuition as evidence” is pretty elusive in the literature. My impression is that philosophers’ actual use of I-Practices is far less committal than a stereotype might suggest. But that’s an empirical claim that would be interesting to pursue: how committal are philosophers in their actual use and engagement with I-Practices? Third, and finally, even if how things look in philosophy is very different in this respect than how we imagined things in the philosophical state of nature—and so different that we don’t want to say they involve the same practices—that only really detracts from the point of this paper if, given how things actually look in philosophy, the alternative explanation I have claimed has some baseline plausibility doesn’t. Above, I demonstrated how the alternative explanation can give a somewhat plausible reading of a number of contemporary cases, and I’m confident I could do so for a wider range of cases (although that will have to wait for future work).