The core of this book revolves around establishing the soundness of the following argument:

The Philosophical Personality Argument (PPA)

  1. 1.

    “Philosophically relevant intuitions are used as some evidence for the truth of some philosophical claims.

  2. 2.

    Some differences in philosophically relevant intuitions used as evidence for the truth of some philosophical claims are systematically related to some differences in personality.

  3. 3.

    If philosophically relevant intuitions are used as some evidence for the truth of some philosophical claims and those intuitions are systematically related to some differences in personality, then one’s endorsement of some philosophical claims is at least partially a function of one’s personality.

  4. 4.

    Therefore, one’s endorsement of some philosophical claims is at least partially a function of one’s personality.” (A. Feltz & Cokely, 2012a)

This chapter attempts to establish the truth of premises 1 and 3. Chapters 2–4 were devoted to establishing the truth of premise 2. For those who are more interested in the empirical results concerning philosophical intuitions, refer to those chapters. For those who are only interested in the more sustained argument about what practical implications the PPA has, please skip to Chap. 7. In this chapter, we argue that the PPA has important implications for some philosophical practices. In particular, our goal is to not only connect the findings from earlier chapters to provide evidence for 1 and 3, we will go on to argue that the conclusion represented in statement 4 means that we should not use (or at least significantly discount) intuitions in some major philosophical projects. This view has been characterized as a “restrictionist” view (Alexander & Weinberg, 2007). Along the way, we will consider and answer some objections to the PPA.

Philosophical Projects

To be clear from the beginning, we do not think the results or our arguments from previous chapters have restrictionist implications for all philosophical projects. Our restrictionist target is much more, well, restricted. We do not think that the empirical results that we have reported in Chaps. 2–4 call into question all of philosophy or all philosophical projects. Rather, we think that our results support a limited form of restrictionism only for some philosophical projects. To illustrate, many have noted and discussed the implications of empirical data for a particular project in philosophy—conceptual analysis (see below for a fuller discussion of conceptual analysis) (Kauppinen, 2007; Ludwig, 2007, 2010). But that is not the only general philosophical project one could engage in. As Sosa notes,

[T]he use of intuitions in analytic philosophy, and in philosophy more generally, should not be tied to conceptual analysis. Consider some of the main subjects of prominent debates in analytic philosophy: utilitarian versus deontological theories in ethics, for example, or Rawls’s theory of justice in social and political philosophy, or the externalism/internalism debates in epistemology, and many others could be cited to the same effect. These are not controversies about the conceptual analysis of some concept. They seem moreover to be disputes about something more objective than just our individual or shared concepts of the relevant phenomena. Yet they have been properly conducted in terms of hypothetical examples, and intuitions about these examples. The objective questions involved are about rightness, or justice, or epistemic justification. ((Sosa, 2007b, p. 59), see also (Sosa, 2007a, 2009))

We agree with the general idea that Sosa presents. Conceptual analysis is not the only general philosophical project philosophers engage in. For example, other projects could be normative projects that may be primarily concerned with the identification of how one ought to behave or what one ought to believe.

Along these lines, Stich and Tobia (2016) argue that empirical results concerning intuitions have different implications for different philosophical projects. They point out three types of projects in contemporary philosophy (that do not exhaust the projects in contemporary philosophy): (1) projects in conceptual analysis, (2) Neo-Platonic projects, and (3) normative projects. Conceptual analysis aims to provide an analysis of philosophically relevant concepts. For example, the Justified True Belief (JTB) account of knowledge could be understood as an analysis of the concept of “knowledge” (see for some related examples Audi (2011)). On this account, one knows some proposition p, if and only if (1) p is true, (2) one believes that p, and (3) one has good evidence for p. Neo-Platonic projects attempt to discover the truth about some non-conceptual or non-linguistic philosophically relevant phenomenon. For example, one attempts to discover what knowledge is or what intentional action is and not what some people’s concepts of knowledge or intentional action are. To put the distinction in other (somewhat misleading) terms, conceptual analysis deals with things in the head and Neo-Platonic projects deal with things in the world and universe (Stich and Tobia (2016) refer to this as the distinction between mental and extra-mental projects). Finally, normative projects address how we ought to be. Two of the most prominent fields in this project are epistemology and ethics. Epistemology largely deals with what we ought to believe and ethics largely deals with how we ought to act.

Few restrictionists want to call into question all philosophical uses of intuitions. Rather, restrictionists are often worried about certain classes of intuitions used in the service of certain projects (J. Weinberg, 2006). In this chapter, we will be mostly concerned with the impact the PPA has for Neo-Platonic (i.e., extra-mental) projects. In short, if the PPA is sound, then many Neo-Platonic projects run the risk of not being able to be reliably conducted given the current, dominate approaches to Neo-Platonic projects. But first, we begin with a sketch of a defense of each of the premises in the PPA.

The Truth of the PPA’s Premises

Premise 1

The PPA is valid but are all the premises true? We think they are. Premise 1 describes what has been called the “practice of philosophy” (Alexander & Weinberg, 2007). As Hilary Kornblith puts it, “Most philosophers do it openly and unapologetically, and the rest arguably do it too, although some of them would deny it. What they all do is appeal to intuition in constructing, shaping, and refining their philosophical views” (1998, p. 129). That is, intuitions are often used as evidence for some philosophical claim (Bealer, 1998; Pust, 2000, 2001). For example, “intuitions are supposed to function like observations” in empirical sciences (Sosa, 2007a, p. 106) (see also Sosa (2009)).

As these quotes indicate, it is commonly thought by many contemporary philosophers that intuitions are crucial pieces of evidence for some philosophical claims. However, unpacking exactly what it means for intuitions to be used as evidence for some philosophical claims is tricky. On the face of it, one might think that there are at least three issues that need to be addressed in order to have confidence that Premise 1 is true: (a) one must understand what an intuition is, (b) one must understand how those intuitions are used as evidence, and (c) it must be true as an empirical claim that intuitions are used as evidence. Currently, debates exist about each of (a)–(c). Here, we attempt to address some of these issues.

What is an intuition? Perusing the philosophical literature on intuition does not help much. Consistent with the general theme of this book, there is a plurality of notions of what intuitions are (for a fuller discussion of how intuitions have been characterized, see Feltz and Bishop (2010)). According to David Lewis, “‘intuitions’ are simply opinions” where “some are commonsensical, some are sophisticated; some are particular, some are general, some are firmly held, some less. But they are all opinions” (Lewis, 1983, p. x). According to Peter van Inwagen, for example, “‘intuitions’ are simply beliefs—or perhaps, in some cases, the tendencies that make certain beliefs attractive to us, that ‘move’ us in the direction of accepting certain propositions without taking us all the way to acceptance” (Van Inwagen, 1997, p. 309). According to Ernest Sosa, an intuition is “a representationally contentful conscious state that can serve as a justifying basis for belief while distinct from belief, not derived from certain sources, and possibly false” (2007b, p. 57). George Bealer argues that intuition is a “sui generis, irreducible, natural (i.e., non-Cambridge-like) propositional attitude that occurs episodically” and are distinct from “physical intuitions, thought experiments, beliefs, guesses, hunches, judgments, common sense, and memory…not reducible to inclinations, raisings-to-consciousness of non-conscious background beliefs, linguistic mastery, reports of consistency; and so forth” (1998, p. 213).

Many philosophers who write about intuitions claim that they are non-inferential. For example, Lisa Osbeck claims that “the salient feature common to various accounts of intuition is its non-inferential status” (2001, p. 119). Alvin Goldman and Joel Pust “assume, at a minimum, that intuitions are some sort of spontaneous mental judgments. Each intuition, then, is a judgment ‘that p’, for some suitable class of propositions p” (1998, p. 179). But some naturalistically inclined philosophers take intuitions to be the result of some inferential process. Michael Devitt claims that “intuitive judgments are empirical theory-laden central-processor responses to phenomena, differing from many other such responses only in being fairly immediate and unreflective, based on little if any conscious reasoning” (2006, p. 491). Hilary Kornblith argues that intuitions “are corrigible and theory-mediated” (2002, p. 13).

Philosophers disagree about whether intuitions are commonsense, untutored judgments or whether they can arise (non-inferentially) after considerable learning and reflection. So L.J. Cohen contends that “an intuition that p is…just an immediate and untutored inclination, without evidence or inference, to judge that p” (1981, p. 318). On the other hand, Laurence BonJour takes intuitions to be “judgments and convictions that, though considered and reflective, are not arrived at via an explicit discursive process” (1998, p. 102).

Some philosophers believe that intuitions come with a characteristic feeling or conviction that what is intuited is true. Guy Claxton thinks that intuition “comes to mind with a certain aura (or even conviction) of ‘rightness’” (1998, p. 217). Stephen Hales thinks that “to have an intuition that A is for it to seem necessarily true that A” (2000, p. 137). But as we have already seen, those philosophers who take intuitions to be beliefs do not suppose that intuitions must come with these sorts of seemings, although they can include an inclination to accept a belief.

There is a cacophony of views about intuitions. Feltz and Bishop (2010) suggest that we can capture some of the variation in views about intuitions in terms of the following menu:

Menu A: Choose one each from the As, the Bs, and the Cs.

A1. Intuitions are beliefs or inclinations to believe.

A2. Intuitions are sui generis propositional attitudes.

B1. Intuitions are inferential judgments.

B2. Intuitions are non-inferential judgments.

C1. Intuitions include only untutored judgments.

C2. Intuitions include tutored and untutored judgments.

This menu defines eight different views about intuitions (in terms of the various possible combinations of As, Bs, and Cs).

So, what do we mean by intuition? The answer may be somewhat anti-climatic. We take no substantive position on which combinations of A-C intuitions are because we don’t take it as that important what the psychological states of the intuitions are—e.g., if intuitions are different from judgments, if intuitions can be inferential or not, etc. This is a debate we think we can safely avoid because the PPA does not need to assume anything substantive about the nature of intuitions. As we hope will become clear, what we take as important about intuitions is their contents and not their psychological properties. For example, when one has a Gettier intuition, it is not the fact that one has an intuition that is important. Rather, what is important is that one has the intuition that has as its content that the person does not know.

Another crucial notion is what it means for intuitions to be used as evidence. There are several philosophical conceptions of evidence (Achinstein, 2000). For our purposes, not much hangs on the correct philosophical account of evidence. Premise 1 is consistent with many philosophical conceptions of evidence. To illustrate, take the following popular analysis of evidence: X is evidence for (against) Y if and only if X makes the truth of Y more (less) probable (Achinstein, 1994; Maher, 1996). This analysis of evidence can be easily modified so that intuitions are evidence for some philosophical claims: an intuition (or cluster of intuitions) I is evidence for (against) some philosophical claim C if and only if I increases (decreases) the probability that C is true.Footnote 1 For example, intuitions about Gettier cases typically are thought to decrease the probability (perhaps to 0) that the JTB account of knowledge is true. Hence, intuitions about Gettier cases are evidence against the JTB account.

However, having an I for or against C is not sufficient for I to be used as evidence. It is widely accepted that intuitions are defeasible evidence for the truth of some philosophical claims (Bealer, 1998; Sosa, 2007a, 2007b, 2009). Just as one could have visual intuitions that the moon is larger than the sun without using that intuition in one’s astronomy, one could have a philosophically relevant intuition without using that intuition as evidence in one’s philosophical theory. One could discount intuitions in a number of ways. To illustrate, one favored method of generating philosophical theories is wide reflective equilibrium (Bealer, 1998; Daniels, 1979; Goodman, 1955). Wide reflective equilibrium counsels revising a theory if it conflicts with a deeply held intuition shared by many (along with appropriate background theories). By a process of mutual adjustment between intuitions and background theories, one settles on a theory. In creating equilibrium between intuitions and principles, some intuitions could be discarded. For example, the intuitions about the moon could be discarded given background theories and other evidence. Those discarded intuitions do not increase or decrease the probability that some philosophical claim is true. In such situations, those discarded intuitions are no longer used as evidence for the truth of a philosophical claim. As a result, we understand I used as evidence when I enters into a justificatory process where I figures into the probability that a philosophical claim is true.

Of course, whether reflective equilibrium is the right method for philosophical theorizing (Stich, 1998) or if intuitions should play an evidential role is not uncontroversial (Cappelen, 2012; Deutsch, 2010, 2015; Timothy Williamson, 2007). The actual correctness of either is not our main concern here. What is important is that this account captures how many philosophers in fact treat the contents of intuitions. A perusal of the philosophical literature reveals the contents of intuitions used in ways consistent with the above analysis of evidence where the contents of intuitions are used as evidence to decrease or increase the probability of some philosophical claim. We find Chinese rooms (Searle, 1980), Swamp Men (Davidson, 1987), counterfactual interveners (Frankfurt, 1969), and strangely wired video games (Bratman, 1984) that are meant to generate an intuition in the reader. The content of these intuitions is then used as evidence either for or against philosophical claims. Many philosophers take the contents of these types of intuitions to be valuable or even irreplaceable parts of philosophical practice (Bealer, 1998; Daniels, 1979; Jackson, 1998; Ludwig, 2007; Pust, 2000, 2001; Sosa, 2007b). It is this practice that many empirically minded philosophers have been interested in (Alexander & Weinberg, 2007; E. T. Cokely & Feltz, 2009a, 2009b, 2011; J. S. Miller & Feltz, 2011; Stich, 1990, 1998; J. Weinberg, 2006; J. Weinberg et al., 2001). So, while not all philosophers take intuitions to be central evidence for philosophical claims, many philosophers think that intuitions provide some important evidence for many philosophical projects.

To summarize, it is largely irrelevant to our argument what intuitions are. However, we think that it is also clear that the contents of intuitions are often thought to be, and are often actually used as, evidence for some philosophical claims. If the contents of intuitions are used as evidence for or against some philosophical claims, then Premise 1 is true.

Premise 2

We have presented evidence in Chaps. 2–4 that personality predicts some philosophically relevant intuitions. For example, some intuitions about free will and moral responsibility are predicted by the heritable personality trait extraversion (see Chap. 2) (Andow & Cova, 2016; Cokely & Feltz, 2009a; Feltz, 2013; A. Feltz & Cokely, 2008, 2009; Feltz & Millan, 2015; Feltz, Perez, et al., 2012b; Nadelhoffer et al., 2009), as are some intuitions about intentional action (see Chap. 3) (Cokely & Feltz, 2009b; Feltz, Harris, et al., 2012a). Similarly, some intuitions about moral objectivism are predicted by the heritable personality trait openness to experience and some intuitions about virtue are predicted by the personality trait emotional stability (see Chap. 4) (Cokely & Feltz, 2011; Feltz & Cokely, 2008, 2012b, 2013b). This research indicates that personality traits can, at least in part, predict a variety of philosophically relevant intuitions in a variety of philosophically relevant domains.

At this point, one may wonder about the strength of the empirical evidence that personality predicts philosophical disagreement. While sustained defenses occurred in the previous three chapters, here we will briefly consider the recent “replication crises” that has pre-occupied researchers in experimental philosophy and psychology more generally. This pre-occupation is justified in the light of recent findings of fraud and high-profile (if disputed) attempts at replications (Open Science, 2015). Indeed, even in experimental philosophy there have been results that have been difficult to replicate (Seyedsayamodst, 2015). One might think that personality’s relation to philosophically relevant judgments could be the same—namely, that there is in fact no relation between personality and one’s philosophical judgments. Indeed, some have argued that some of the paradigmatic examples of personality and philosophical judgments (e.g., between extraversion and free will) have not been replicated (Mortensen & Nagel, 2016). Thus, either given the failure of replication or the possibility of failure, we should have little confidence that Premise 2 is true.

Science involves risk, so premise 2 could turn out to be false, of course. However, the mere possibility that the relations aren’t real is nothing all that new or threatening to Premise 2. Rather, the question is whether the on-balance evidence suggests that Premise 2 is true. Failure to replicate a few or even a substantial number of times does not in itself call into question the reliability of the relation between personality and philosophically relevant intuitions (see, e.g., the meta-analysis in Chap. 2). There are any number of explanations why experiments could fail to replicate—explanations too numerous to enumerate here. What is important is that estimating the reality of the underlying relations that is difficult to determine with one or two experiments (Cumming & Calin-Jageman, 2017). That said, we think the body of work speaks for the reality of the relations.Footnote 2 While being certain of Premise 2 is not to be had, there is growing evidence that the relations are in fact robust and reliable, persisting independent of wide differences in abilities, presentation formats, ages, cultures, education levels, expertise, and general demographics. The weight and consistency of the evidence suggests we’d be foolish to expect that the evidence will not continue to mount. In other words, it is theoretically possible there is no relationship between personality and philosophical intuitions, but at this point that possibility appears extremely unlikely.

Premise 3

Premise 3 also appears to be true. An intuition is used as evidence when it enters into one’s justificatory process (e.g., wide reflective equilibrium), and as a result, the content of the intuition increases or decreases the probability that a philosophical claim is true. If the contents of intuitions are used as evidence, then the view that one ends up endorsing is a function of the contents of those intuitions. In addition, gathering evidence suggests that some philosophically relevant intuitions often used as evidence are systematically related to global personality traits. Intuitions in the domains we have documented (e.g., free will, intentional action, ethics) are influenced by personality and continue to be used as evidence for some philosophical claims. Hence, if a widely endorsed method of philosophical inquiry is used and philosophically relevant intuitions vary as a function of personality, then the philosophical view one ends up holding will be at least partially related to one’s personality.

Implications of the PPA

We have tried to be careful in arguing that the implications of the PPA are dependent on what projects one is engaging in. In this section, we distinguish three types of projects one could engage in: Neo-Platonic projects, conceptual analysis, and normative projects (Stich & Tobia, 2016). The implications of the PPA for each of these projects is different, as we detail in what follows.

Neo-Platonic Projects

Neo-Platonic (or “extra-mental”) projects attempt to discover the non-conceptual, non-linguistic truth of the relevant philosophical phenomenon by using rational reflection along with relevant intuitions. On this project, through rational discourse, we can come closer to achieving or approximating the truth. The PPA suggests that some agreement or disagreement in Neo-Platonic projects is not solely a function of purely rational arguments aimed at a progression toward the truth. Rather, some features irrelevant to the truth of the content of the intuition (e.g., personality traits) may be driving mechanisms of philosophical agreement or disagreement.

We think that there is wide agreement that personality is not related to truth for almost all Neo-Platonic projects. As we mention below, that would be like claiming that one’s personality is related to the truth about how many electrons copper has or how many moons Jupiter has. For these reasons, the PPA is another argument for restrictionism. Restrictionism holds that “the results of experimental philosophy should figure into a radical restriction of the deployment of intuitions as evidence” (Alexander & Weinberg, 2007, p. 61) because “it involves deploying a source of putative evidence that is sensitive to non-truth-tracking factors” (J. M. Weinberg et al., 2010, p. 332). There are many results from psychology and related disciplines that suggest that many intuitions that could be or have been used in Neo-Platonic projects are associated with extraneous features. These extraneous features thereby call into question whether those intuitions are reliable guides to the Neo-Platonic truth.Footnote 3 For example, some philosophically relevant intuitions in some paradigmatic cases vary with respect to socio-economic status (Haidt et al., 1993), culture (Huebner, Bruno, & Sarkissian, 2010; Machery et al., 2004; J. Weinberg et al., 2001), the presentation order of scenarios (Feltz & E.T. Cokely, 2011; Swain, Alexander, & Weinberg, 2008),Footnote 4 and one’s perspective (Alexander, Betz, Gonnerman, & Waterman, 2018; A. Feltz, Harris, et al., 2012a; Nadelhoffer & Feltz, 2008). Several theorists contend that associations such as these call into question the truth of the content of those intuitions (Alexander & Weinberg, 2007; Horvarth, 2010; Sinnott-Armstrong, 2008; Stich & Tobia, 2016; J. Weinberg, 2006; J. Weinberg et al., 2001). The actual mind-independent truth about determinism’s relation to free will and moral responsibility should not depend on the order in which questions about that relation are presented (just as, e.g., the weight of a 10 pound bar does not vary just because one was previously holding a feather, although people may tend to judge that weight differently after holding a feather). Personality is like these factors. If one has the intuition that free will and moral responsibility are compatible with determinism, it would at least be odd to defend the truth of the content of the intuition by appealing to the fact that one is extraverted. It would be like saying that one’s personality is relevant to whether the atomic number of gold is 79.5. Yet some of those free will intuitions are predictably related to personality. As such, those intuitions may be related to extraneous factors. Given that there is currently no argument that successfully allows us to always prefer one set of intuitions to another set (introversion and extraversion seem to be equally irrelevant to the truth of whether determinism is compatible with free will and moral responsibility. See below for a fuller discussion of objections), one should not dismiss one set of intuitions. To the extent that these intuitions vary with irrelevant factors such as personality, it does not appear that some philosophical disagreements about some Neo-Platonic projects are solely a function of rational disagreement (see (Sommers, 2012)). The use of any source of evidence in Neo-Platonic projects that does not only track the truth should be restricted. Hence, the use of intuitions that vary with personality should be restricted in Neo-Platonic projects.

But some care needs to be taken in interpreting restrictionist implications of the PPA. To illustrate, take Horvarth’s (2010) formulation of the restrictionist’s Master Argument:

“(A) Intuitions about hypothetical cases very with irrelevant factors.

(B) If intuitions about hypothetical cases vary with irrelevant factors, then they are not epistemically trustworthy.Footnote 5

(C) Intuitions about hypothetical cases are not epistemically trustworthy.” (p. 448)

It seems like the PPA nicely complements the Master Argument. But, one worry is that the conclusion of the PPA does not support (A). One might think that personality could be relevant to the truth of particular philosophical claims. To take just one example, Prinz (2007) argues that the truth of some moral claims is essentially related to one’s emotions. For example, when one says “Eating your dead relatives is wrong,” the truth of that claim essentially involves a sentiment (i.e., disposition to experience disapprobation) in the speaker. Without the sentiment toward eating a dead relative, what the speaker says is false on Prinz’s account. On the assumption that personality can influence what sentiments one is likely to have, it seems that it is possible that personality is reliably related and relevant to the truth of some philosophical claims (e.g., “Eating your dead relatives is wrong”).

We do not dispute that personality could be related to the truth of some philosophical claims (in fact, some of our main claims require it). However, our main point is about whether personality is related to the truth of the content of intuitions for Neo-Platonic projects. Here, it is illustrative to see two different claims that might be made on Prinz’s view:

(M) An action has the property of being morally wrong (right) just in case there is an observer who has a sentiment of disapprobation (approbation) toward it. (2007, p. 92)

(N) Eating your dead relatives is wrong.

Presumably, personality can be importantly and reliably related to the truth of (N) but not the truth of (M). One plausible explanation is that (M) is an example of a Neo-Platonic claim whereas (N) is a normative claim. (M) is a “Metaphysical Thesis” about the nature of a philosophically relevant phenomenon indicating a requirement for something to be morally wrong or right (Prinz, 2007, p. 92). The truth of Neo-Platonic claims like (M) is not supposed to be related to one’s personality. However, (N) is a normative claim. The truth of (N) could be importantly related to one’s personality (see Chaps. 4 and 5). Somebody who is extremely low in emotional stability might think that (N) is true whereas people who are high in emotional stability may be more likely to think (N) is not true. Even if one argues that (N)-like statements support the generalization to (M), there could be another who does not think that the (N)-like statements support (M) as a mind-independent fact about the world. If people who are high and low in emotional stability have qualitatively different intuitions about (M), at least one of them is wrong. As a result, personality could be critically important for the truth of some normative claims, but it does not appear that personality is important or relevant for the truth of many (if not all) Neo-Platonic claims.

Defenders of traditionally conducted Neo-Platonic projects may ask “is the PPA problematic for Neo-Platonic projects that use intuitions as evidence?” The defender may argue that on the conception of evidence we have been using, having an I for (or against) a C is not sufficient for using I as evidence for (or against) C. Once philosophers learn that some of their intuitions are systematically related to some personality traits, they may appropriately discount those intuitions and not use them as evidence. And, the argument goes, excluding or discounting some intuitions is a natural part of philosophical practice. As a result, the PPA should be of no concern for those using intuitions as evidence because almost all philosophers incorporate relevant empirical evidence into their philosophizing already. Philosophers can admit that learning some intuitions are systematically related to personality is important information to incorporate into theorizing, yet that information does not call into question Neo-Platonic theorizing based on intuitions as the PPA suggests.

While such a position is possible, we think it would be a major concession to restrictionists for at least two reasons. In one sense, consulting closely with empirical psychology may not be odd or new for those working on some Neo-Platonic projects (Sosa, 2007a). Some have argued that the conditions under which one is free or morally responsible may be helpfully informed by empirical science (Nahmias, 2007). Empirical psychology may tell us under what conditions one lacks freedom-relevant control over a behavior (e.g., one’s glucose is too high or low, cf. Baumeister (2008)). However, in another sense, consulting with the empirical sciences would be important and new for Neo-Platonic projects. In this sense, it is not merely that empirical evidence has some role to play in assessing some philosophical claims. Empirical evidence has a role to play in evaluating the intuitions deployed in Neo-Platonic projects. For example, is one’s glucose being too high or too low relevant to the control required for free will and moral responsibility? The answer to this question may depend on what intuitions one has about the particulars of the case involving glucose, and those intuitions may be systematically influenced by personality. It isn’t odd to think that glucose’s effect on behavior is relevant. It is odd to think that one’s personality influences one’s thinking about the importance of glucose’s effect on behavior.

We know of very few philosophers (if any) who discard intuitions because the intuitions are likely influenced by the type of personality they have. For example, compatibilists and incompatibilists neither reject their own intuitions nor the intuitions of others because of personality’s relation to those intuitions. But the PPA suggests that these intuitions sometimes need to be dramatically discounted as evidence in some Neo-Platonic projects. More generally, the PPA suggests that in order to have confidence in intuitions used in Neo-Platonic projects, we need to have a deeper psychology of philosophical intuition (see Stich and Tobia (2016)) and philosophical expertise. Hence, the PPA would constitute a significant change to the way some philosophers go about doing Neo-Platonic projects.

Second, philosophers who think that intuitions related to personality can be discounted find themselves in a precarious philosophical position: The viability of their approach depends critically on empirical science. That is, at the end of the day, for their position to be viable there had better not be large and systematic relations of one’s personality with the intuitions that one has. There is already evidence that personality is related to intuitions in problematic ways for Neo-Platonic projects. We suspect that the evidence indicating that philosophically relevant intuitions are systematically related to personality (and other stable individual differences) will continue to grow. If our suspicion turns out to be right, then these philosophers are hostage to empirical results. What if philosophically relevant intuitions in a substantial number of fields are systematically related to personality traits? If that is the case, then the defender of intuitions in Neo-Platonic projects would be committed to barring those intuitions as evidence for Neo-Platonic projects. But if a sufficient number of intuitions are thereby barred, there will not be a sufficient amount of evidence to theorize upon. These considerations lead to a rather striking suggestion: In their current form, many intuition-driven Neo-Platonic projects simply should not, and possibly cannot, be done.

It is important to note that the worry posed by the PPA is not merely a skeptical hypothesis. We aren’t just positing the possibility that personality is systematically related to intuitions used in Neo-Platonic projects. Rather, the PPA (and the evidence reviewed in the previous chapters) indicates that this possibility is likely actual. What we currently know is sufficient to justify the worry that many Neo-Platonic projects based on intuitions are in trouble. What we need to do, then, is to investigate the extent to which intuitions in the relevant Neo-Platonic project are associated with personality traits (and other similar psychological variables). And that means that almost all philosophers engaged in Neo-Platonic projects using intuitions as evidence need to consult empirical evidence about their own (and others’) dispositions and intuitions more closely (i.e., via scientific methods).

Given these considerations, we favor the following principle:

(E) Empirical evidence should play a substantial role in many philosophical projects.

We take (E) to be the general attitude of experimental and empirically minded philosophers. The empirical evidence is both evidence relevant to the Neo-Platonic project and evidence about intuitions deployed in those projects. Many philosophers engaged in Neo-Platonic projects reject (E) in some way. One could reject (E) by simply ignoring (E). However, we think following principle is true:

(R) Most fully adequate philosophical views should take into account all available, relevant evidence.Footnote 6

Indeed, we think that (R) is entailed by reflective equilibrium—one of the core philosophical techniques that use intuitions as evidence. Ceteris paribus, if a philosophical theory does not account for available and relevant evidence, then that theory will at least in some cases (and perhaps in most cases) be inferior to one that does take into account that evidence. Personality’s influence on intuitions is available and relevant. Therefore, ignoring personality’s influence on philosophically relevant intuitions is in most cases not tenable.

Conceptual Analysis

Generally, whether intuitions ought to be trusted depends on what intuitions are to be trusted for. Our main target in this chapter is some Neo-Platonic projects. However, we will briefly comment on two other projects in which intuitions have been used as evidence—conceptual analysis and normative projects (Stich & Tobia, 2016). Indeed, we want to stress that the arguments that we have provided do not have restrictionists implications for intuitions in all philosophical projects. Philosophy is a diverse field with many different projects, some of those projects can use intuitions and perhaps even should use them as evidence. One is in conceptual analysis. One prominent approach to conceptual analysis attempts to provide a set of individually necessary and jointly sufficient conditions for something to be classified under that concept. For example, an analysis of “knowledge” attempts to give the individually necessary conditions that if a person lacks, then the person does not know, and the jointly sufficient conditions specifying that if a person satisfies all those conditions, then one knows. As has already been mentioned, the Justified True Belief account of knowledge conceived of as an analysis of the concept provides the following individually necessary and jointly sufficient conditions for knowing: (1) a person believes that P, (2) P is true, and (3) the person has justification for believing that P.

If people have different intuitions about knowledge, does that mean conceptual analysis about knowledge cannot be done? Not necessarily. Our view is consistent with the view expressed by Machery (2017) (who also does not think that all intuitions ought to be restricted). He holds a view of “naturalized conceptual analysis.” On this view, conceptual analysis could still be done but it should pay close attention to empirical results indicating that different people have different intuitions. There are many ways that using these empirical results about intuitions still could be used in a valuable evidential role. To illustrate one way, perhaps the diversity of intuitions simply means that different (groups of) people have different concepts. There is some evidence that at least for some philosophical concepts the multi-concept view is true. Recall some of the discussion about intentional action intuitions from Chap. 3. There, we reviewed some evidence that suggests that there could be two or three different concepts of intentional action that are responsible for the pattern of results observed in the side-effect effect (Cushman & Mele, 2008; Mele & Cushman, 2006; Nichols & Ulatowski, 2007). Additionally, our evidence suggests that we can at least partially predict who is likely to have those concepts with extraversion. It could be, then, that extraverts and introverts have different intentional action concepts. Maybe something similar is happening in at least some other philosophical domains. While we don’t currently have specific evidence to support the following claim, it could be that some people have different concepts (and corresponding intuitions) in all of the areas we have discussed (free will, some areas of ethics, intentional action) (Sosa, 2007a, 2009). If all of this is correct, then conceptual analysis can still be done with the important caveat (or restriction) that the conceptual analysis is not about the only concept (e.g., the concept of intentional action).Footnote 7

We do not take a stand whether the multiple concepts approach is tenable. There do seem to be at least some potential costs if one adopts a view that there are potentially many concepts in a specific philosophical area. Some may find it objectionable and costly to concede that the analysis they are providing is not about the only concept in that philosophical domain (Stich, 1998). Many philosophers think they are offering an analysis of the philosophically relevant concept such as knowledge, freedom, or intentional action. To give up the notion that we are analyzing the relevant concept may thereby seem to diminish the importance of some conceptual analyses. Concentrating on our own concept that may be shared by a few others may often be of relatively less interest than investigating a single conceptual truth (Bishop & Trout, 2005). However, one benefit of the multiple concepts approach is that the variation of intuitions with personality does not pose a significant challenge to conceptual analysis and in some ways could help illuminate conceptual disagreement. Some people, because of their natural tendencies associated with personality (or other experiences and individual differences), may have different philosophically relevant intuitions. Indeed, as will become clearer in responses to objections to the PPA, it is important that the variation does not undermine some aspects of conceptual analysis. Consequently, the kinds of systematic differences in philosophical intuitions could be problematic for some approaches to conceptual analysis but not others.

Normative Projects

Normative projects are aimed at, among other things, telling us what we ought to do (ethics) and what we ought to believe (epistemology). What implications might the systematic differences in philosophical intuitions associated with personality have for normative projects? Somewhat of an aside, we want to note a striking lack of evidence about personality and epistemic intuitions. To our knowledge, there is no evidence suggesting that epistemic intuitions of the kind of interest to traditional epistemologists (e.g., about knowledge, justification, warrant) are robustly associated with personality. Of course, personality is associated with a host of different kinds of beliefs such as religious belief (Saroglou, 2002) and prejudice (Sibley & Duckitt, 2008), and personality disorders can be associated with distorted beliefs. But to our knowledge, research has not yet identified any data suggesting that personality is systematically associated with epistemic intuitions that are of interest to philosophers. Other researchers have reported individual differences in epistemic intuitions including gender (Buckwalter & Stich, 2014), culture (J. Weinberg et al., 2001), and socio-economic status (J. Weinberg et al., 2001), but none of these studies reported personality differences (however, see Kim & Yuan (2015); Seyedsayamodst (2015) for failed attempts at replication).

One explanation for this lack of evidence is that there simply hasn’t been enough research conducted on epistemic intuitions and personality. We take it that this is the most likely explanation since the research on personality and philosophical intuitions is relatively new. However, another explanation is that there simply is no interesting variation of epistemic intuitions and personality. One reason why there might be no interesting relation is that there might be something special about the nature of epistemic intuitions. For example, Jennifer Nagel (2012) has argued that many epistemic intuitions find their source in basic human capacities for mind reading. That is, most normally functioning humans have the basic ability to understand others’ emotions, thoughts, intentions, and beliefs. Attributions of knowledge are like these other mind-reading abilities. Since we all have these basic abilities, then we all have the same epistemic intuitions, more or less. Consequently, we shouldn’t expect there to be much variation with respect to epistemic intuitions and personality, gender, or socio-economic status.

Given the current state of the science, we are agnostic whether there are personality differences with respect to epistemic intuitions. However, we would like to note that if the reason why we shouldn’t expect differences in epistemic intuitions is because of people’s basic mind-reading abilities, then we should also not expect there to be differences in judgments of intentions among people with different personality traits. Attributing intentions to others is one of the paradigmatic instances of mind reading. However, we do see those differences in attributions of intentions between people with different personality traits (see Chap. 3). Whether similar differences can be found in epistemic intuitions remains to be seen.

Regardless of whether there are differences with respect to epistemic intuitions, there is growing and substantial evidence that there are systematic differences in moral intuitions associated with personality (see Chap. 4). There are already some well-developed theories concerning what implications systematic variation in moral intuitions could have for philosophical theories. As we have already discussed, Jess Prinz (2007) has argued that in order for somebody to make a moral judgment at all, one must have or have had an emotional reaction to the actions. If differences in personality cause differences in emotional reactions, then those differences could provide important pieces of evidence for normative views like Prinz’s. Or, alternatively, Tamler Sommers (2012) has argued that the systematic diversity in intuitions about moral responsibility puts pressure on any universal, objective set of conditions for freedom and moral responsibility. If Sommers’s is right, then those intuitions associated with personality could provide evidence for when somebody ought to be held morally responsible which could be different for different people with different psychological frameworks.

Not only does diversity of moral (and perhaps epistemic) intuitions help provide some evidence for some theories, the diversity can also help illuminate other, possibly surprising, debates (see Chap. 7 for a fuller discussion). For example, the diversity can help illuminate some contemporary debates about paternalism (e.g., Nudging (Sunstein & Thaler, 2003; R. Thaler & Sunstein, 2003; R. Thaler & C. Sunstein, 2008)). Given the systemic variation in some people’s philosophical intuitions, it stands to reason that these differences sometimes are expressed and are related to people’s values. However, paternalistic strategies, by their very nature, often aim to promote some specific set of values or goals. This means that if there is variation in people’s values, then some people’s values will not be served by the paternalistic strategy. Thwarting some people’s values is an important cost associated with paternalistic strategies. If there are alternative ways to paternalistic policies to help people make better decisions, then all else equal those alternative ways should be evaluated to determine whether the paternalistic policy is more costly or if the alternative way of helping people is most costly. For example, informing people could allow for the benefits of paternalism (encourage better choices consistent with one’s values) without commensurate costs (e.g., thwarting some sets of values).

Objections to the PPA

We now turn to objections that one might have to the PPA and its implications. While it would be impossible and inefficient to respond to every possible kind of defense one could mount against the implications of the PPA for intuitions in Neo-Platonic projects, in this section we focus on a cluster of common defenses. In evaluating the following objections, recall two plausible principles we have argued for earlier in the chapter:

(E) Empirical evidence has a substantial role to play in many philosophical theories.

And

(R) Most fully adequate philosophical theories must take into account all available and relevant evidence

There are many ways one could reject (E) and respect (R). As we already indicated, whether using intuitions as evidence is appropriate depends a great deal on the goals one has using those intuitions. We think the strongest (and perhaps only) case for restrictionism is with respect to Neo-Platonic project, so we will use Neo-Platonic projects as our test case for objections.

The Self-Defeating Argument

Is the kind of view we are expressing even conceptually or theoretically possible? Some theorists think that it might not be. One potential way to show that our view is mistaken is by arguing that some of our assumptions call into question some other claims that we make. This general approach has been come to be known as the Self-Defeating Argument (Bealer, 1998; Bonjour, 1998; Goldman & Pust, 1998; Horvarth, 2010; Pust, 2001). In general terms, the Self-Defeating Argument holds that the only way to justify some of the claims we have made is to, well, use intuitions. But, if all those intuitions are associated with extraneous factors, then they should not be used for evidence (i.e., restrictionism about intuitions). So, on that argument, given our view, we might end up with nothing that could be used to support some of the key premises in the PPA. As Horvarth states, “the experimental challenge might easily lead to epistemic self-defeat because some of the relevant intuitions are themselves needed in order to justify the epistemic principles that form the very basis of the experimental restrictionists’ own methodological criticism” (2010, p. 459).

In particular, restrictionists need evidence that the following principle is true:

(EU+) If intuitions about hypothetical cases are unreliable and their unreliability is explicitly noted and explicitly noting their unreliability does not give rise to self-defeat, then they are not epistemically trustworthy. (Horvarth, 2010, p. 461)

Horvath observes that (EU+) is not “directly intuitive.” That is, the intuitions are not directly generated by considering cases (similarly to the intuition A1, B2, and C2 combination above). Rather, the principle (EU+) involves testing out the principle on cases or other kinds of reflection. But, testing out intuitions in responses to cases or other kinds of reflection are exactly what the restrictionist aims to restrict—those intuitions are associated with extraneous factors that would call into question their contents. As a consequence, the restrictionist’s view is self-defeating because the very evidence that is needed to support restrictionism is not available based on restrictionist principles.

Here it is important to mark again a distinction we (and others) have made. The PPA does not suggest that all philosophically relevant intuitions are suspect. We advocate “not the root and branch removal of all intuitions, but just the pruning away of some of the more poisoned branches” (Alexander & Weinberg, 2007, p. 71). Intuitions may be indispensable in projects in conceptual analysis and normative projects. Indeed, as we have seen, variability in intuitions related to personality can be theoretically important in those projects (cf., Prinz (2007)). The PPA supports restricting intuitions in some Neo-Platonic projects and also recommends scientifically documenting and tracking of non-trivial influences on intuitions for conceptual analysis and normative claims. As a result, there could be a wide swath of philosophically relevant intuitions that could be used to justify the premises of the PPA, (E), and (R). To illustrate, all the premises in the PPA are either descriptive or conceptual. (E) is a natural consequence of the PPA, and (R) seems to capture a straightforward conception of a fully adequate theory. We don’t see any reason that we would be required to use Neo-Platonic intuitions to criticize Neo-Platonic intuitions (that would seem to be self-defeating). If we are not required to use Neo-Platonic intuitions and we do not use Neo-Platonic intuitions to justify the premises of the PPA, (E), or (R), then our argument supporting restrictionism is not obviously or necessarily self-defeating. Hence, it is not obvious or necessary that the PPA is self-defeating. Of course, one or more premises of the PPA could turn out to be false. But that criticism is very different from the criticism that somehow the PPA leads to epistemic self-defeat (and would require a different argument and evidence).

But perhaps the self-defeating argument can be cast in a slightly different way even if we restrict the range of intuitions to those used in Neo-Platonic projects. Regina Rini (2016) has offered a regress argument against those who claim that psychological evidence suggests that some moral judgments do not track the truth. Rini gives a schematic of the regress argument. The first premise states that some psychological processes give rise to some moral judgments. Examples of these kinds of psychological processes have been reviewed above and include order effects, mental correlates of socio-economic status, and personality. The second premise is that psychological processes do not reliably track the truth. Therefore, the moral judgment is unreliable. The problematic part of this schematic is that some moral judgment must be made about whether the psychological processes reliably track moral truth. For example, it is a moral judgment that order is irrelevant to the truth of moral claims. And, the argument goes, if one questions the psychological processes that give rise to some moral judgments, one must also question the psychological processes that give rise to other judgments. And the regress starts.

We think we are very clear that the PPA is problematic for Neo-Platonic projects and not necessarily for normative projects. As such, Rini’s argument as stated does not pose any threat to the PPA. However, perhaps with slight modification, Rini’s argument could generate a problematic regress. Here’s how the modified schematic might look.

  1. 1.

    A set of Neo-Platonic judgments N are caused by psychological processes P.

  2. 2.

    P does not reliably track the truth

  3. 3.

    Therefore, N are not reliable.

But in order for us to determine the truth of (2), we have to know something about Neo-Platonic truths or have a set of Neo-Platonic beliefs on which to base our evaluation of (2). And if we question N, then why shouldn’t we question the Neo-Platonic truths used to evaluate (2)? Hence, the regress starts.

We think a similar reply to Horvarth can be applied to Rini’s argument. We need not appeal to any Neo-Platonic truths in the domain in question but rather to basic truths about what Neo-Platonic projects are conceived to be and some basic logic. Neo-Platonic projects attempt to understand the mind-independent, non-conceptual, non-linguistic understanding of some phenomenon. That means that Neo-Platonic attempt to find the truth about some phenomenon of philosophical importance. At this point, only the concept of Neo-Platonic projects has been employed, and not any Neo-Platonic claim about what Neo-Platonic projects actually are as mind-independent entities (if that even makes any sense). If that is granted, then it only takes some basic logic to show Neo-Platonic projects might not be able to be done if the intuitions are related to irrelevant features because two people could have contradictory intuitions about some philosophical phenomenon, and the content of both intuitions cannot be true. For example, if you have the intuition that P and I have the intuition that not P, they can’t both be correct on a standard Neo-Platonic understanding. That would be like saying it is both true that the moon revolves around the earth and the moon does not revolve around the earth.Footnote 8 As a consequence, no regress is generated.

Defining Intuitions

Some may attempt to circumvent (E) and respect (R) by defining “intuition” so that intuitions cannot be empirically questioned. In this way, it can be claimed that the way restrictionists talk about intuition is “obviously not an intuition in the sense in which philosophers have talked about intuitions” (Ludwig, 2010, p. 437). For example, one may think that one has an intuition only when one has complete (enough) understanding of the relevant concepts involved and only makes judgments based on competencies with those concepts in ideal conditions (Bealer, 1998; Kauppinen, 2007; Ludwig, 2007, 2010). Indeed, some even claim that “it is impossible for intuitions properly understood to be relative” (Ludwig, 2010, p. 427) because there would be “identical judgments in response were the responses made solely on the basis of those competencies and identical understandings of the scenario, task, question, and adequate thought” (Ludwig, 2007, p. 145). As a result, intuitions are “veridical, and it follows that intuitions are not relative to cultures, socio-economic status, times, the ways questions are presented, or anything else, and this is demonstrable a priori” (Ludwig, 2010, p. 442). On this account of intuitions, when people have different responses to a scenario, at least one person does not have an intuition.

The variability in responses associated with personality is easy enough to accommodate while respecting (R) with an exclusive definition of intuition. Intuitions aren’t relative to anything. However, the PPA suggests that intuitions are systematically related to one’s personality. So, when people with different personalities give different responses, at least one of them would not have an intuition. Given that (R) holds that one needs to take into account available and relevant evidence, when there are divergent responses, at least one of those responses is not relevant because the response is not an intuition. As a result, these defenders of intuitions can easily deny (E) while at the same time as respecting (R).

Let’s set aside the fact that there is substantial disagreement about what intuitions are (see above). Several theorists have argued that defining intuitions does not help the defenders of intuitions very much (see also, for a nice discussion, Weinberg and Alexander (2014)). First, stipulating a definition of intuition that makes intuitions invulnerable to empirical challenge is not very satisfying (Horvarth, 2010). But more problematically, intuitions defined in these ways may make it just as likely that philosophers don’t have intuitions (Alexander & Weinberg, 2007). How could we ever tell when philosophers have an intuition? How do we determine (a) who the competent user of a concept is, (b) what ideal conditions are, or (c) if one’s judgment is only influenced by semantic considerations (Kauppinen, 2007)? It seems like it would be very difficult, for example, to determine who are competent users or when ideal conditions obtain (A. Feltz, 2008; Kauppinen, 2007). To illustrate, Sosa writes that the reliability of intuitions “depend[s] on favorable circumstances in all sorts of ways, and these are often relevantly beyond our control. We must depend on a kind of epistemic luck” (2007a, p. 102). Restrictionists would argue that we are often epistemically unlucky when using intuitions in Neo-Platonic projects. There are hundreds if not thousands of studies across disciplines indicating that very minor changes in judgment environments (e.g., framing) can result in large differences in resulting judgments (e.g., Gigerenzer et al. (1999); D. Kahneman (2003)). But more to the point, it is very difficult to determine how lucky one is from the armchair. Horvarth (2010) notes there is quite a bit of evidence that we often aren’t aware of all the causal influences on our responses to particular cases (e.g., Nisbett and Wilson (1977)). Given that we don’t have much evidence that philosophers satisfy a–c, we should not be confident that philosophers even have intuitions. If philosophers don’t have intuitions, then philosophers could not use intuitions in Neo-Platonic projects because they do not have any! We take it as part of the practice of philosophy that philosophers use the contents of their intuitions as evidence to do Neo-Platonic projects. So even if intuitions are philosophical theory mediated (e.g., one knows a lot about free will), the intuitions are used to provide the relevant contents, and it is hard to determine who, from the armchair, has contents reflective of genuine intuitions. Of course, we might be able to determine if philosophers satisfy (a)–(c). But since many factors relevant to determining if philosophers satisfy (a)–(c) are not introspectively discoverable, it won’t be from the armchair. Hence, stipulating definitions of intuition does not insulate defenders of intuitions in Neo-Platonic projects from the PPA and (E).

The No Intuition Defense

Some defenses of philosophical practice attempt to show that the restrictionists argument is somehow misplaced because philosophy doesn’t depend on intuitions at all. So, intuitions aren’t used as evidence in philosophy, making premise 1 of the PPA false.

There are two major proponents of the no intuitions defense and they make slightly different arguments about why intuitions aren’t used in any (substantial) sense in philosophy. First, we’ll look at Hermann Cappelen’s arguments. According to Cappelen, the key question is around centrality of intuitions in philosophical theorizing (i.e., Premise 1 of the PPA), or the claim that “contemporary analytic philosophers rely on intuitions as evidence (or as a source of evidence) for philosophical theories” (2012, p. 3). He goes on to argue that in no way is centrality a feature of philosophical debates.

According to Cappelen, it might look like philosophical debates involve something like centrality. One reason is that philosophers use “intuition talk.” That is, philosophers often use phrases like “It is intuitive that x” that give the impression that intuitions are used as evidence. There are several problems with making the inference from intuition talk to intuitions actually being used as evidence for philosophical claims. First, there is no agreed upon definition of intuition, so inferring what the evidential role of something being intuitive is difficult, possibly highly context sensitive, and not automatically inferable from somebody saying that something is “intuitive.” Additionally, there is almost nothing in philosophy (outside of some specific subdomains like logic) that is universally agreed upon to be intuitive. So not only are there problems with the definition of intuition, there are very few paradigmatic examples that are thought to be intuitive. For these reasons, when one encounters “intuition talk” in philosophical texts, one should be as charitable as one can be when interpreting what is being used as evidence. And, according to Cappelen, when one engages in charitable reconstruction, texts that contain intuition talk can almost always be recast without using intuition talk yet retaining the core of what the authors intend. This, in turn, suggests that intuitions don’t serve as evidence at all because they are removable, can be interpreted as initial judgments subject to revision, or simply refer to pre-theoretical judgments or as initial starting points.

We agree that merely because philosophers use intuitions talk does not mean that they use intuitions as evidence. Premise 1 does not rely on intuition talk as evidence for its truth. Premise 1 requires that philosophers use the content of intuitions as evidence. So, the argument from intuition talk is no threat to Premise 1 of the PPA. Rather, a more important objection from Cappelen concerns whether intuitions are in fact used as evidence. According to Cappelen, there are three ways that would indicate that philosophers use intuitions as evidence.

  1. 1.

    Intuitions with a special phenomenology are used.

  2. 2.

    Intuitions that are rock bottom and in no further need of justification.

  3. 3.

    Intuitions that are the result solely of one’s conceptual competence.

However, when one looks at the actual texts, there is no evidence of 1–3 in paradigmatic examples. For example, in Thomson’s famous violinist case, one might think that the intuitions about the cases play some evidential role in the claim that a right to life does not always outweigh a right to determine what happens to one’s own body. However, 1–3 are not present in the defense of that principle. According to Cappelen, Thomson does not refer to any special phenomenology when making judgments about the Violinist, the intuitions about the violinist are subject to scrutiny and are not taken to be rock bottom, and the intuitions are not solely the result of one’s conceptual competence. In this paradigmatic case, therefore, intuitions do not play a substantial role. To the extent that intuitions do play a role, they play a role of setting a common ground from which to proceed (shared intuitions about the cases). As such, there is no evidence that intuitions play any substantial role in philosophy and if they do, their function is to provide a common framework and background in which to proceed philosophically. Consequently, Centrality is false and entails that Premise 1 of the PPA is also false.

Cappelen notes that Centrality (and by extension Premise 1 of the PPA) is an empirical claim about the actual way that philosophers go about philosophizing. And, as Cappelen notes, it’s ironic that experimental philosophers take a remarkably non-experimental approach to substantiating that intuitions in fact play a role in philosophical debates. How many systematic studies have been done about intuition’s actual role in philosophical theorizing? According to Cappelen, none.

However, it seems uncontroversial that philosophers make judgments about cases, principles, and premises in arguments. Of course, these judgments can be about a variety of things and can take into consideration a variety of different inputs. As already noted, philosophical intuitions are diverse and it is very difficult to identify exactly what an intuition is. Cappelen only identifies three of the many ways intuitions have been identified above. But for Premise 1 to be true, we don’t need any well worked out account of what intuitions are. Rather, all we need is that, as a matter of fact, philosophers use the contents of intuitions as evidence—something that we do not take to be especially controversial or in need of any systematic studies. In fact, it is a virtue of Premise 1 that it does not take any substantive view about the nature of intuitions. All that is needed for the PPA to succeed is that philosophers have judgment (or intuition) biases and that only requires that the contents of intuitions (or judgments) are used as evidence for philosophical claims. Looking again at Thomson’s violinist example, it seems uncontroversial that we have some intuitions about the case and the contents of those intuitions are supposed to carry evidential weight. Of course, we can examine these contents, refine intuitions, and reject the content of some intuitions. However, in each of these instances, the contents of the intuitions are used as evidence for some philosophical claims. Hence, Premise 1 of the PPA is not shown to be false by Cappelen’s argument.

Max Deutsch (2015) has a slightly different approach to the no intuition defense. Deutsch agrees that if intuitions do play an evidential role in philosophy, then the results like those associated with personality would be deeply problematic for philosophy. However, the essential feature of philosophical debates is arguments and not intuitions. If intuitions don’t play an evidential role in philosophy, then Premise 1 of the PPA is false.

Deutsch laudably makes the distinction that we have exploited between the mental state of having an intuition and the content of the intuition. In terms of this distinction, one might think that philosophical evidence comes from something’s intuitiveness (EC1). For example, the intuitiveness of the Gettier examples is evidence against the justified true belief account of knowledge. However, on a very different notion of intuition (EC2), it’s the content of the intuition, not the fact that one has the intuition that is evidence for its philosophical claims. So, it’s not that the intuitiveness of the Gettier cases that is evidence. Rather, it’s the content of the intuition—that the person doesn’t know—that is evidence against the JTB account. Deutsch contends that the EC2 is without question true but not EC1. And many experimental philosophers mistakenly take that EC1 is the way that intuitions are evidence for philosophical claims.

Critical to Deutsch’s claim is that EC1 doesn’t at all play a role in philosophy, which we are happy to grant. Additionally, good use of intuitions (characterized by EC2) are backed up by arguments. We are happy to grant that as well. In this way, it’s not the case that Deutsch simply relocates the problem to a different set of intuitions that are used to justify the premises in arguments. All intuitions could be questioned and arguments could be provided for them (e.g., some form of coherentism). As such, one might think that either Premise 1 of the PPA is false (intuitions aren’t used as evidence) or think that the variability we find is not problematic because arguments could be given for the different positions, and the best arguments will win out.

Here, it becomes useful to be reminded of the distinction that has been drawn between Neo-Platonic projects, projects in conceptual analysis, and normative projects. We agree that the right way to understand intuitions is not simply that something is intuitive (EC 1). Rather, the content of the intuition is what is critically important for philosophical practice (EC 2) (e.g., used as evidence). The problem is that people have intuitions with different contents, or those contents occur with different strengths. Take, for example, the debate about manipulation in the free will debate. Some argue that moment-by-moment manipulation rules out freedom and moral responsibility. The notion is that people have the intuition with the content that one cannot be free and morally responsible if all of one’s actions are ultimately caused and implanted by somebody else. However, hard-liners don’t have this intuition or at least they don’t have that intuition about some cases (e.g., McKenna (2008)). Rather, hard-liners have the intuition that if the person appropriately identifies with the actions, that person is free and morally responsible even if manipulated. Here, we don’t see a clash of intuitiveness (EC1), we see a clash of the contents of intuitions where some people have the intuition that manipulation always rules out freedom and moral responsibility whereas others have the intuition that manipulation does not always rule out freedom and moral responsibility (although we are also likely to see clashes of intuitiveness (EC1) as well). Arguments are given by both sides, but still in many cases, the intuitions are unwavering. Moreover, these intuitions are predicted by personality traits (see Chap. 2). So even though hard-liners and non-hard-liners are possibly justified in having those intuitions because they both have coherent sets of intuitions and arguments, the coherent sets are different. As such, it cannot be the case that both sets accurately capture the relation of manipulation to free will as it exists in a Neo-Platonic sense. To be sure, different people may have accurately captured their concept of freedom, and the differences may have important normative work to do, but they do not help us solve Neo-Platonic questions. So, even if everything that Deutsch says is correct, the PPA still can have problematic implications for Neo-Platonic projects.

At this point, one may object that it is not clear how extensive this kind of philosophical disagreement is. Perhaps debates around manipulation are rare, non-representative cases of philosophical disagreement. In other cases of philosophical disagreement, perhaps extensive philosophical reflection can help achieve convergence. If convergence is not achieved, perhaps this extensive reflection can at least reduce the influence of non-truth tracking features. So, how extensive and robust is the influence of non-truth tracking features? Answering this question is an empirical question and not purely a philosophical or theoretical question that can be handled from the armchair. Here, again, we want to underscore the fact that we think that given the available evidence many Neo-Platonic risk not being able to be done. However, before we can be sure that this actually is the case we need to have more evidence. Given the available evidence, our bet is that philosophical disagreements about Neo-Platonic projects are deep and are not likely to go away soon—and if they do, it’s not going to be because of arguments that justify some intuitions over others.

The Verbal Defense

According to the Verbal Defense, the current evidence based on surveys does not ensure “true disagreement” in people’s intuitions (Horvarth, 2015; Sosa, 2007a). In order for there to be true disagreement, responses gathered by experimental philosophers must be about the same things. But the worry is that different people could interpret scenarios or questions differently and thereby have intuitions in response to different things. There are a number of ways that people could interpret scenarios differently. To illustrate, Sosa (2009) argues that the materials many experimental philosophers use are like stories. Like most stories in fiction, not all details are spelled out in the text. As a result, people often fill in stories in different ways. Participants may do the same thing for the scenarios used in experimental philosophy. People simply fill in the scenarios differently and are thereby representing relevant content of the scenarios differently. These differences may result in different intuitions but not about the same things. Likewise, people may interpret questions asked somewhat differently. For example, when asking whether somebody is morally responsible for an action, people may interpret “morally responsible” in a variety of ways. They may interpret moral responsibility in an attributability sense where judgments are made about the action reflecting the actor’s character. Or, participants may interpret moral responsibility in an accountability sense where one can be held accountable (e.g., punished/rewarded) for acting (Sosa, 2007a). If participants interpret scenarios differently or interpret questions differently, then much of the disagreement in intuitions put forward by experimental philosophers is merely surface or verbal disagreement. In the end, people could be “talking past each other” (Kauppinen, 2007, p. 107). Since such surface variability is not philosophically relevant, we can reject (E) while respecting (R).

The PPA may seem to support the verbal defense. As we have documented, people with different personality types have different sensitivities, beliefs, and goals (Costa & Mccrae, 1988; Funder, 2001). These differences may result in people with different personalities resolving ambiguities in scenarios differently.Footnote 9 So, it might be that people with different personality types are not disagreeing about the same contents of the intuitions.

However, if the PPA makes it plausible that apparent disagreement among the folk is only verbal disagreement, it makes it just as likely that disagreement (or agreement) among philosophers is not disagreement (or agreement) about the same things (Alexander & Weinberg, 2007). That is, we cannot be sure that agreement or disagreement among philosophers is about the same content of the intuitions. Not being able to tell when there is true agreement and disagreement results in general skepticism about philosophical uses of intuitions because we could never tell when philosophers are truly agreeing or disagreeing (Alexander & Weinberg, 2007; Machery et al., 2004). Hence, if the verbal defense were to succeed, it would be at the expense of skepticism about intuitions in general. But we take it that philosophers (experimental or otherwise) want to resist general skepticism about intuitions.

Just as in the Expertise Defense, one may argue that the burden of proof may be on the experimentalist to show that philosophers run the risk of only having verbal disagreements. “The appeal to divergence of interpretation is a defensive move, made against those who claim that there is serious disagreement in supposed intuitions. It is only against such a claim of disagreement that we must appeal to verbal divergence. But any such claim need be taken seriously only when adequately backed by evidence” (Sosa, 2007a, p. 103). Fair enough. But there is evidence emerging that adequately backs the worry. As already noted, experts about the free will debate are influenced by specific, heritable facets of their general personality traits: Warm extraverts tend to have more compatibilist friendly intuitions than do introverts (Schulz et al., 2011). As we understand Sosa’s defense, any purported disagreement in intuitions is verbal and that would entail that when there is disagreement between experts, that disagreement would largely be a verbal disagreement. That would mean that the disagreement in expert intuitions between non-warm extraverts and warm extraverts runs the risk of being merely verbal. If that is the case and the empirical results generalize, the verbal defense runs the risk of it being hard to tell when there ever is a substantive disagreement among philosophical experts such as in free will. Consequently, the verbal defense along with relevant empirical evidence risks general skepticism about philosophically relevant intuitions.

Intuition Calibration

The last objection we will consider is what might be called the intuitions calibration objection (for a more extensive discussion of intuition and calibration, see Weinberg, Crowley, Gonnerman, Vandewalker, & Swain, 2012b). Rather than objecting to the relevance of the variance associated with personality (e.g., arguing that philosophers, for whatever reason, won’t display that bias), the intuitions calibration approach starts by accepting the basic empirical science. That is, this objection freely grants that philosophers often display the same (or similarly problematic) biases as non-philosophers. However, philosophers don’t have to display that bias and can in fact incorporate those biases in their theorizing. One way that might happen is by attempting to correct for those biases in intuitions. In this sense, philosophers correcting for a bias in their intuitions are much like other debiasing strategies that are often successful. Take one example: the Planning Fallacy. The planning fallacy is the tendency for people to underestimate the amount of time, effort, or other expenses that projects can require. You may have fallen prey to the planning fallacy yourself when you, for example, thought about how long it would take to write a paper or clean your garage (or write a book!). However, you might also know some of the psychology about the planning fallacy and incorporate that into your estimates of how long it would take to do something. Perhaps you double the amount of time you estimate and that ends up being roughly accurate to the amount of time it actually does take. By incorporating that information, you have removed the bias. Philosophers might be able to do something similar with personality effects—they could use that information to calibrate their intuitions more accurately.

While we agree that debiasing approaches often work very well, we are less convinced that they are going to be of much help to correct for philosophical biases associated with personality. Recall we are arguing a restricted restrictionist’s view that intuitions are not likely to be reliable guides truth for Neo-Platonic projects. These projects attempt to discover the truth about some phenomenon, independent of what anyone thinks of those truths, using intuitions as indispensable sources of evidence (see Cummins (1998) for a discussion related to calibration issues in philosophy). These intuitions are at least in part used to determine those neo-Platonic truths and therein lies the potential problem with the calibration approach. For most (if not all) of the standard cases where debiasing happens, there is some external standard that is used to determine if one’s judgments, decisions, or estimations are accurate. In the Planning Fallacy example above, there is some external, objective measure of how long a project actually takes—time. One’s intuition or estimation about how long a project takes does not factor at all into the measure of how long the project actually takes. But for many philosophical areas that attempt to discover Neo-Platonic truths, the intuitions constitute or are a guide to what those truths are (see Sommers (2010); for a dissenting view, see Timothy Williamson (2007)). Given that intuitions play this role in many Neo-Platonic projects, and given the empirical data that often those intuitions are related to extraneous features like personality, there is no external standard on which to calibrate those intuitions. This variation seems to persist, at least sometimes, in the face of extensive knowledge and training in a subject area. Finally, we know of no argument that allows us to prefer some intuitions over others (should we prefer extraverts intuitions? If so, why, and does that preference require further intuitions?). Consequently, addressing the calibration problem does not, at least at present, insulate Neo-Platonic projects from the risks that we have identified.

Concluding Remarks

If the PPA is sound, some deep-seated ancient debates in some Neo-Platonic projects may not go away without new methods. The reason for this may be surprising. There is some evidence that global personality traits are at least partially genetic in origin (Bouchard & Loehlin, 2001; Jang et al., 2002). And, as noted, intuitions are used extensively in many philosophical projects. Thus, some intuitions used in Neo-Platonic projects are systematically related to personality and personality is at least partly (and often largely) heritable. If both of these two claims are true, then one’s tendency to endorse a particular philosophical view is also at least partially inherited. So, some issues in Neo-Platonic projects are likely to persist through the generations of philosophers with very little resolution. One could take this result as pessimistic, but we think it is actually encouraging. It may help free philosophers from an over-reliance on intuitions and may help encourage philosophers to use other methods and evidence to cover new ground for important, ancient Neo-Platonic projects.

Our view that is a consequence of the PPA might also have implications for inclusivity and diversity in philosophy. Systematic diversity associated with personality is at the core of our argument for the PPA. Perhaps this recognition of systematic diversity may help temper and inform philosophical debates (even if it doesn’t necessarily help us calibrate those intuitions). Philosophical debates are about important issues and many people feel deeply about those issues. Sometimes when there is philosophical disagreement, parties to the disagreement think that the other person is making some sort of mistake. That mistake can be of a variety of forms including having false beliefs or lack of relevant beliefs or not thinking appropriately. However, the evidence we have reviewed here suggests that there may be something else going on. Rather than attribute some epistemic deficit to an interlocuter, that interlocuter may simply have different intuitions (and possibly concepts) involved in the dispute. This recognition might allow different ways to resolve some disagreement and to explore the sources and implications of that diversity (for similar reasoning, see Haidt and Graham (2007)).

At this point, we have reached the end of our empirical defense of the premises in the PPA. Before we turn to some of the more practical implications that the PPA might have, let’s take stock. We think we have established that the PPA is likely to be sound. The soundness of the PPA suggests that at least for some philosophical projects, restrictionism is warranted. If one wants to deny restrictionism, then one would have to provide evidence that somehow the PPA is not valid (it seems to be of a valid form, however) or provide evidence that the PPA has at least one false premise. As an empirical matter, one or more of the premises in the PPA maybe turn out to be false. We think that the evidence reviewed so far makes this possibility very unlikely. Nevertheless, the kind of evidence that would be needed to falsify one of the premises of the PPA is empirical evidence. While we think that our view supports restrictionism, we would be very satisfied if we have achieved the conclusion that at minimum philosophy should be substantially more empirical (Alexander & Weinberg, 2007; Feltz & Cokely, 2012a, 2013a; Feltz & Cokely, 2016).