Trust and Psychedelic Moral Enhancement

Moral enhancement proposals struggle to be both plausible and ethically defensible while nevertheless interestingly distinct from both cognitive enhancement as well as (mere) moral education. Brian Earp (Royal Institute of Philosophy Supplement 83:415–439, 12) suggests that a promising middle ground lies in focusing on the (suitably qualified) use of psychedelics as adjuncts to moral development. But what would such an adjunctive use of psychedelics look like in practice? In this paper, I draw on literature from three areas where techniques for moral development have been discussed: psychotherapy (e.g., Overholser 2010; Burns 1980) education (e.g., Uhl and Lütge, 2018), and AI-assisted enhancement (e.g., Lara and Deckers, Neuroethics 13(3):275–287, 17) in order to propose more concrete ways in which to use psychedelics as adjuncts to moral development. It is shown that in each of these areas, we can see that trusting relationships (e.g., Baier 1986; Hawley 2019) between the facilitator and the agent will very plausibly maximize the success of this type of moral enhancement. Finally, I appeal to literature on informed consent for use of psychedelics (e.g., Smith and Sisti, Journal of Medical Ethics, 22; Johnson et al., The Journal of Psychopharmacology 22(6):603–20, 23) and on the therapeutic relationship in psychotherapy (e.g., Dryden and Reeves 2013; Horvath et al. 2011) to outline concrete suggestions for facilitating dimensions of trust most likely to maximize the benefits of (adjunctive) psychedelic moral enhancement. The result is a newly detailed practical proposal for how we might best facilitate moral enhancement by using drugs as adjuncts to moral development


Introduction
A central strand of research in the contemporary literature on moral enhancement takes as a reference point Persson and Savulescu's controversial [1,2] argument that we have an urgent imperative to pursue moral (rather than merely cognitive) bioenhancement to protect ourselves from a "minority which is morally corrupt" (2008, p.163) and which could plausibly enough exploit our cognitive advances to design and deploy weapons of mass destruction. Much of the literature that followed focused on criticizing some aspect of Persson and Savulescu's proposal-for example, some believed that the reasoning problematically generalised to a rejection of scientific progress, 1

others took
Abstract Moral enhancement proposals struggle to be both plausible and ethically defensible while nevertheless interestingly distinct from both cognitive enhancement as well as (mere) moral education. Brian Earp (Royal Institute of Philosophy Supplement 83:415-439, 12) suggests that a promising middle ground lies in focusing on the (suitably qualified) use of psychedelics as adjuncts to moral development. But what would such an adjunctive use of psychedelics look like in practice? In this paper, I draw on literature from three areas where techniques for moral development have been discussed: psychotherapy (e.g., Overholser 2010; Burns 1980) education (e.g., Uhl and Lütge, 2018), and AI-assisted enhancement (e.g., Lara and Deckers, Neuroethics 13(3):275-287, 17) in order to propose more concrete ways in which to use psychedelics as adjuncts to moral development. It is shown that in each of these areas, we can see that trusting relationships (e.g., Baier 1986; Hawley 2019) between the facilitator and the agent will very plausibly maximize the success of this type of moral enhancement. Finally, I appeal to literature on informed consent for use of psychedelics (e.g., Smith  issue with the notion that cognitive and moral enhancement could be separated as distinct goals, 2 and many more argued that there would be negative consequences associated with pharmacologically enhancing altruism and our sense of justice in the way that Persson and Savulescu suggest. 3 Meanwhile, other bioethicists responded to the explosion of work on moral enhancement with more fundamental criticisms of moral enhancement as a project-type, suggesting that its objectives are unclear and that existing concrete proposals are unsatisfying. For example, de Melo-Martín [6] calls the debate "confusing and confused", citing a range of inconsistencies and shaky empirical support, 4 and after reviewing 85 distinct pieces of work on the topic Specker et al. [11] identify what they describe as a need for "focused debate on realistic options" in the moral enhancement debate. (Ibid., p.15). Earp [12]-in noting that he finds it tricky to parse what "moral enhancement" is supposed to refer to-offers perhaps the clearest articulation of a dilemma that contributes to this unclarity and appears to be at the heart of much existing work on moral enhancement. The first horn of the dilemma would have bioethicists focus on the development of radical proposals that are novel and philosophically interesting, but which are unlikely to be neurologically feasible 5 (at least not without having other, dangerous impacts on our functionality). Meanwhile, the second horn of the dilemma would have us shift focus to proposals that are more practically achievable and less ethically concerning but which are not clearly very interesting-or, at least, not interestingly different from traditional moral enhancement programs (e.g., moral education) or merely cognitive enhancements (e.g., nootropic drugs.) 6 In the face of this dilemma, Earp suggests that we might be able to have best of both words-in other words, a moral enhancement program that is within reach and ethically justifiable, but which is also interesting and distinct enough to "merit the flood of ink that continues to be spilled in this area" (Ibid., 421). Such a program, he suggests, would involve using psychedelic drugs in certain carefully specified and regulated ways.
In a similarly constructive spirit, this paper aims to flesh out the details of what an effective psychedelic moral enhancement proposal might look like by expanding on aspects that are as yet under described, and by demonstrating how certain kinds of trust will plausibly be vital in maximizing the effective use of psychedelic drugs as adjuncts to moral development. To that end, I will also suggest ways we might facilitate such trust in the setting of moral enhancement.
Here is the plan for what follows: §2 highlights the areas of Earp's adjunctive psychedelic enhancement proposal that require further exploration and elaboration. Next, in "Approaches to Facilitating Psychedelic Moral Enhancement" section, I flesh out the core idea of psychedelics adjuncts to moral development with reference to literature on moral development in psychotherapy (e.g., [15], education (e.g., [16]), AI-assisted enhancement (e.g., [17]), and I propose several ways in which they might play this role in these areas. For each of the relevant techniques and 3 See e.g., de Melo-Martín [6] for a convincing argument that someone "enhanced" in this way might be more likely to commit terrorist attacks in the name of responding to unjust oppression. For more of the many responses to Persson and Savulescu, see Jotterand and Levin [7], Hardcastle [8], Azevado [9] and Beck [10]. 4 For example, she reads Persson and Savulescu as saying moral enhancement is both necessary and merely advisable and as being optimistic about moral enhancement's development as well as claiming it is in its infancy (see her p.20 for both criticisms). Regarding empirical support, she highlights that most studies cited in support of moral enhancement proposals focus on WEIRD populations-i.e., those that are Western, educated, industrialised, rich and democratic), and notes the influence of artificial lab conditions on the results (p. 32-33). 5 Specker et al.'s [11] systematic review of the moral enhancement literature also raises this worry, suggesting that we ought to pursue a more focused debate on realistic options of biomedical moral enhancement and concrete moral questions these treatments raise. 6 For example, nootropics such as Adderall or Ritalin might be thought to improve us morally by contributing to more focused moral reasoning, by way of improving cognitive performance more generally. See e.g., Bostrom and Sandberg [13] and Maslen et al. [14] for overviews of what cognitive enhancement might involve and the ethical issues it raises. approaches suggested, we will see that trust between the facilitator and the agent will maximize the success of this type of moral enhancement, and I will use work from Baier [18] and Hawley [19] to outline the nature of the relevant efficacious aspects of trust, including especially perceived commitment (Hawley) and goodwill (Baier). Finally, I will appeal to literature on the therapeutic relationship in psychotherapy (e.g., [20,21]) and on informed consent for use of psychedelics (e.g., [22], Johnson [23] to outline concrete suggestions for facilitating the kind of trust that is likely to maximize the benefits of using the aforementioned techniques during psychedelic moral enhancement. The result, I hope, is a newly detailed-if still incomplete-practical proposal for how we might best facilitate moral enhancement when using psychedelic drugs as adjuncts to moral development.

Psychedelics as Indirect and Adjunctive Moral Enhancers
As noted in §1, Earp proposes that the (appropriately qualified and voluntary) use of psychedelic drugs could be at the heart of a feasible, defensible, and interesting moral enhancement proposal -one that threads the needle between the two horns of the dilemma outlined.
Before we turn to considering suggestions for specific techniques that would be compatible with such an approach to moral enhancement, let's lay some groundwork by outlining roughly how Earp envisions such a proposal working and highlighting several undescribed areas that invite further development.
Earp ([12], p. 423-27) devotes a lengthy section to the discussion of the connection between drugs and spirituality in order to establish that psychedelic substances have long been viewed as providing significant insights into "the nature of reality and human existence" (p. 424)-and, crucially, that these insights frequently transform our understanding of how we ought to live and act. The most prominent drugs considered for this kind of moral development include lysergic acid, diethylamide (LSD), psilocybin and psilocin from "magic" mushrooms, mescaline from cacti, and MDMA (which has a different mechanism of action but is nevertheless classed by some as a psychedelic). 7 Changes in attitude and perspective facilitated by such drugs are potentially morally transformative-for example, according to a double-blind study by Griffiths et al. [26], psilocybin users often experience increased patience, cognitive flexibility, positive thinking, compassion, social concern, and perceptiveness with respect to others' thoughts and feelings. Similarly, Smith and Sisti [22] report findings indicating a positive impact of psilocybin on personality and worldview and find that where people benefit from taking psilocybin, they often describe a diminished sense of their own importance-sometimes called "ego dissolution". 8 Again, it is easy to see how we might think these sorts of experiences would facilitate moral behaviour; such experiences can help to foster pro-social behaviour (e.g., compassion, social concern), and they either preclude or might serve to diminish (e.g., in the case of egodissolution) certain kinds of behaviours (e.g., rage, aggression, selfishness) that undermine cooperation and well-being. Further, there is a clear overlap here between the experience-and attitude-types facilitated by psychedelics and the broad aims of moral enhancement advocates like Persson and Savulescu [1,2], Protopapadakis [30] and Crutchfield [31] -including and especially the promotion of compassion and the minimisation of selfish (non-cooperative) behaviour. For present purposes, I will assume that psychedelic moral enhancement would reliably have the outcomes described.
How, then, ought we to use these drugs safely and effectively in a moral enhancement setting? What policies are the best ones? One key distinction that Earp makes here is between two particular types of roles that substances might play if used to morally enhance ourselves: Determinative role: Enhancement drugs are taken without preparation or direction and cause moral development. 7 See Móró, Levente et al. [24], Roberts [25] and Griffiths et al. [26] for discussion of the impact of such drugs on one's insights and a sense of one's life purpose. See also Nichols [27] for how MDMA differs from these other drugs (indeed, some resist the description of MDMA as a psychedelic). See also Check [28] for more on the relevant psychological impacts of MDMA. 8 See e.g., Griffiths et al. [29] for reports on these effects of psilocybin along with a reduction of depression and anxiety. Adjunctive role: Enhancements are taken after preparation and consultation with experts and facilitate moral development.
Psychedelic moral enhancement, Earp suggests, should best pursued alongside a commitment to their playing an adjunctive as opposed to determinative role in moral development. In a similar vein, Schaefer [32] also offers us a useful distinction that helps us to characterise psychedelic enhancements more narrowly. He contrasts indirect with direct moral enhancement, with the former inducing specific ideas (e.g., being caused to have certain pro-moral beliefs) and the latter merely improving the processes by which moral ideas, motives and behaviours are generated, "without committing to [their] content" (p. 261); psychedelic moral enhancement appears quite clearly to be indirect on Schaefer's taxonomy. Indeed, given that the morally relevant improvements associated with psychedelics mainly appear to be increases in empathy and decreases in selfishness, psychedelics could not typically be direct enhancers (i.e., to simply induce specific ideas) or to be determinative (i.e., to cause moral development without any attempt at facilitation).
This foregoing sketch of a psychedelic enhancement program, construed as both adjunctive (and indirect), seems initially to meet Specker et al.'s [11] call for focused debate on realistic options on moral enhancement. In addition, it appears to do well by de Melo-Martín's lights; there are no (obvious) theoretical inconsistencies in what is being proposed and what it aims to accomplish, and empirical evidence that pharmacological enhancements could play such an adjunctive role is borne out by empirical evidence of the sort discussed. Moreover, using certain drugs as indirect enhancers that work as adjuncts to moral development (alongside moral education) is less clearly susceptible to a cluster of well-known normative objections that bioconservative thinkers have appealed to in order to target much more radical kinds of enhancement proposals -e.g., that enhanced abilities, insights and achievements are worth less because of the role of enhancement. For example, even if the above bioconservative line of argument is compelling when in radical envisaged cases where, e.g., moral beliefs simply arrive fully formed in our heads after popping a pill (or are implanted there via a BCI), it is much less compelling when applied to psychedelics pursued (in an indirect fashion) adjunctively alongside traditional moral education. 9 The same kind of point can be made with respect to other argument-strategies in this cluster. For example, Kass [34] insists that enhancement disconnects performance from effort, resulting in an "easy life" filled with "trivial" successes, and Harris [4] suggests that moral enhancement removes our "freedom to fall" (i.e., our ability to get things wrong, morally speaking), the less meaningful our morally good acts become. Similarly, Sandel [35] discusses that the more enhancements influence our actions, the more credit shifts from us to "[our] pharmacist." (p. 25). 9 One might object that -contrary to what is being suggested here -bioconservative critiques that have been levelled against 'direct' moral enhancement as proposed by Savulescu and Persson would be applicable as well on an Earp-style adjunctive proposal in the case of psychedelics, at least in so far as psychedelics could in some circumstances directly induce certain morally relevant states, as opposed to merely enabling or indirectly facilitating them. The thought, in a bit more detail, would be that if psychedelics directly induced such states, a programme that promotes using them to do so would be subject to at least certain critiques of direct moral bioenhancement (e.g., [33] -e.g., that it might stifle moral diversity, and if employed systematically might suppress divergence and dissent which may manifest as being anti-social. Two points are worth noting here in response: first, on an adjunctive approach, the aim is not direct moral enhancement, even if in some cases it is incidentally the case that psychedelics directly improve one morally. Importantly, there is no commitment on an indirect/adjunctive proposal to countenancing that morality could be so directly induced nor that this would be desirable. In this way, the brunt of bioconservative critiques of direct moral enhancement would not be equally applicable to a indirect/ adjunctive model even if psychedelics in some circumstances were to incidentally directly induce moral improvement, and not merely to facilitate (as is intended on such a programme) moral improvement when combined with, and as an auxiliary to, education/therapy. The second point to note in response here is that, to the extent that the particular critiques of direct moral enhancement -which is that it might 'stifle diversity' or suppress dissent -are thought to be applicable to indirect/ adjunctive moral enhancement, these two objection can be met with reasonable responses. In the case of minimising diversity: a response is that if the objection extends to indirect/adjunctive bioenhancement, it's not clear how it would not also extend to mere moral education of an excellent quality. Regarding the suppressing dissent objection: here the usage of psychedelics in consultation with experts aimed at facilitating moral education will take place in a context in which the political value of dissent can be affirmed. Even more, psychedelics might be useful in helping one better appreciate the value of peaceful political dissent. This kind of reply is not available to a proponent of direct moral enhancement outside an adjunctive context. With psychedelics as adjuncts to moral development, there is deliberate, active learning (as opposed to passive receptivity), and in such a way that makes the foregoing kinds of critiques lose the applicability they might have to enhancements envisioned to be both determinative and direct (in Earp's and Schaefer's senses).
Crucially for what follows, Earp's proposed adjunctive programme for psychedelics-driven moral enhancement is put forward as a solution to what at first looks like an intractable dilemma for moral enhancement advocates. Consequently, he says little about what how preparation, consultation and facilitation would work in practice. The most detailed description of the proposal's aim is that it will "foster states of mind that that [allow] one to engage with the moral domain in a more productive or insightful way, storing away any lessons learned for application in the "real world" once the effects of the drug had worn off" (p. 436, italics mine).
This leaves us with interesting questions about what fruitful psychedelic-facilitated engagement with the moral domain might look like. Here, I would like to sketch some research-informed suggestions for making use of psychedelics as moral enhancers, and in this way, to show how this kind of proposal might be most responsibly implemented in practice.

Approaches to Facilitating Psychedelic Moral Enhancement
To make some progress on fleshing out how psychedelic enhancement might be responsibly implemented, I will look at some example techniques and suggestions from psychotherapy and education. Following this, I will stress the presence of a unifying factor-the importance of trust in making the most of the techniques highlighted in the context of psychedelic-facilitated moral enhancement.
In addition, it is worth emphasising that in order to make the most of the example techniques I suggest below, subsequent reflection (after the influence of psychedelics has diminished) is crucial. This selfreflection component to any kind of implementation of psychedelics in the service of moral improvement should accordingly include the provision of a journal of relevant moral reflection 'prompts' related to what was discussed and explored in the meeting. 10

Questioning and Socratic Dialogues
One case study -which Earp himself notes -which describes the efficaciousness of psychedelics in moral development concerns the use of MDMA in couple therapy sessions (e.g., [39]. However, within psychotherapy, it is plausible that a more effective setup for pursuing aims of moral enhancement would be more narrowly focused and task-oriented-one that is more aligned with approaches in cognitive behavioural therapy, which is rarely used as the central focus for couple therapy, 11 and which typically involves clearly delineated tasks aimed at adapting thinking processes to promote dimensions of individual well-being (including where these have morally relevant effects on other persons). 12 Conveniently, we can adapt recommendations from literature relevant to cognitive behavioural therapy (e.g., a taxonomy from [44]) in order to suit the unique purpose of moral development work. To this end, consider that some types of questions are equally relevant in both settings 13 -such as information-gathering questions like "When did you feel guilty about lying to your partner?", clarifying questions such as "So, you often saw strong emotions expressed at home but you still believed they were a sign of weakness?" and recapping questions like "Do you think it 10 For some representative studies supporting the role of such prompts in facilitating self-reflection and learning, see, e.g., Berthold et al. [36], Veenman et al. [37], and -specifically for strategy for using prompts -see Thillmann et al. [38]. 11 Couple therapists use a range of therapeutic approaches, some of which may involve elements of cognitive behavioural therapy. However, more commonly used modalities include psychodynamic, systemic and emotionally focused therapy. See e.g., Gerson [40] for more about psychodynamic and systemic approaches in this context, and Wiebe and Johnson [41] for more on the evidence base for using emotionally focused therapy with couples. 12 See e.g., Blackwell and Heidenreich [42] and Fenn and Byrne [43] for more detailed discussion of how cognitive behavioural therapy sessions tend to be structured, and the types of dimensions they often focus on (e.g., anxiety, depression and phobias). 13 See e.g., James, Morse and Howarth [45] for more on taxonomies of questions and their different functions. is fair to say that you believe in clearly delineated gender roles?". However, other types of questions should arguably occupy larger amounts of space in specifically moral enhancement sessions. In particular, we might make frequent use of challenging questions like "You have described yourself as forgiving, but can you see that you have given two examples today about ending friendships with people who haven't met your expectations?". Similarly useful might be facilitating questions-questions that encourage deepening of the conservation-like "We've established that you want to understand why you are habitually unfaithful, so what parts of your history do you think we might need to look at?". Additionally beneficial here would be a focus on open questions that can be answered in a broad range of ways rather than closed questions that invite a "yes" or "no" [46].
Socratic dialogue is used in cognitive behavioural therapy as a method for dynamically engaging with the above kinds of questions, and it might likewise help us to see how the moral development facilitator's questions might best most efficaciously structured. Socratic dialogue is a method of asking sequenced, discovery-oriented questions in order to facilitate cognitive change [47]. It has two distinct aims ( [48], p. 151): firstly, to help the person make new links between the way they think and how their actions are influenced by those ways of thinking, and, secondly, to encourage creativity and self-reflection with respect to how the person may start to think differently. James, Morse and Howarth [45] add a third aim-facilitating the development of more adaptive thinking. Socratic dialogue may be especially effective in uncovering inconsistent beliefs, and in helping people to subject their underexplored assumptions to scrutiny [15]. For example, a standard Socratic question in cognitive behavioural therapy might include "What do you think counts in favour of your belief, and what counts against it?". Equally common-and perhaps especially helpful in the setting of moral development-are invitations to consider hypotheticals, such as "What would happen if you broke your promise?".
The promise of Socratic dialogues to facilitate moral development generally (we'll connect this with psychedelic adjuncts shortly) is further supported by the fact that another recent moral enhancement proposal-that of using artificial intelligence as a Socratic assistant [17]-has Socratic elements at its core. Specifically, Lara and Deckers are envisioning a system that "should work to help us to reach a better decision ourselves, without committing us to any pre-designed ethical perspectives" (p. 281). Examples of the functions Lara and Deckers hope such an AI might provide include pointing out use of ambiguous terminology/encouraging clarity in language use, offering relevant empirical support (or pointing out the lack thereof), raising awareness of factors about human biology and the environment that impact on how we make decisions, and helping the person trace out the consequences of their decisionmaking. While one arguable advantage of this moral enhancement proposal is that it doesn't require drug ingestion, we are still in the relatively early stages of producing such an AI-and the authors note the risk of imbuing it with the creator's moral code or judgements.
For our purposes, the relevant takeaway from Lara and Deckers is that Socratic dialogues are noted as useful in facilitating moral enhancement, and so we should consider such an approach to questioning when we explore how to ask questions in psychedelicassisted moral development sessions. What is more, such an approach -along with being well suited for moral development -is also a promising fit (particularly, through an envisaged positive feedback loop) with a specifically psilocybin-based adjunctive role. This is due to the alignment of the reported effects of psilocybin on patience, cognitive flexibility, and perceptiveness with respect to others' thoughts and feelings. Initially at least, we can see how psilocybin as an adjunctive role might work in Socratic-fuelled moral development in a kind of 'virtuous feedback loop' -e.g., where Socratic questioning and techniques encourage cognitive flexibility, patience, consideration of others thoughts, which are then more readily embraced through psilocybin, when in turn increases the facilitation of the Socratic questioning.

Discovering the Impact of Bias
Uhl and Lütge [16] explore the usefulness of experiments in teaching ethics in the context of business education. They aim to "make students question their own behavior and re-evaluate the implementability of their moral ideals" (p. 203) and encourage them "to transcend their own viewpoint and put themselves in the shoes of others" (p. 206). Both of these objectives 1 3 Vol.: (0123456789) are compatible with a range of ways of spelling out concrete moral aims of moral improvement (including of the sort psychedelic moral enhancement could facilitate adjunctively), and so it will be prima facie promising to consider how such experiments could might be effective for our present purposes in facilitating moral development. Part of what we can draw from Uhl and Lütge is the usefulness of giving different groups of individuals subtly different vignettes that ask them to make hypothetical choices. Perhaps most interestingly for our purposes, Uhl and Lütge used these vignettes to teach students about the Knobe effect (i.,e., [49])-viz., how small changes to wording can "radically change [the] ascription of intentionality to an action." 14 Their thought is that it is one thing for students to learn about the Knobe effect from a textbook and another thing to see it played out in the differences between their responses and the responses of peers given a different vignette. As Uhl and Lütge put it, "not only will students better internalize the effect's relevance after experiencing it, but also will their subsequent discussion of its implications for ethics have a different quality." Further evidence supports their thinking here-for example, Frank [52] showed that students who participating in a "take-some game" got better results on a test of their understanding of the "tragedy of the commons", and Dickie [53] found that controlled studies that took aptitude into account showed students who took part in experimental sections of an economics course achieved better grades. All told, this evidence suggests it might be worth conducting group moral development sessions in order that such lessons might be best internalised.
Again -and as with the case of Socratic dialogue ( §3.1.1) -we can see both that (i) such interactive experiments of 'living out' certain morally relevant lessons can plausibly facilitate moral development, and on this basis would be a practical approach worth considering as a way to communicate moral lessons in psychedelic-assisted moral development sessions. And furthermore, and also like in the case of Socratic dialogue, these kinds of experiments fit conveniently well with a specifically psilocybin-based adjunctive role. Recall here, e.g., the effects reported by Griffiths et al. [26] by psilocybin users, which include (along with cognitive flexibility), positive thinking and social concern. These kinds of experiences promoted by psilocybin might also feature positively in a similar kind of virtuous feedback loop -e.g., where roleplaying experiments whereby one learns (through experience) moral lessons that themselves promote (in the subject's identification with the moral features relevant in the lesson) cognitive flexibility, positive thinking and social concern, which are then more easily and readily experienced through psilocybin, when in turn amplifies or sharpens further such qualities as then promoted through the role-playing experiences.

Trust as Maximizing Benefit from Moral Enhancement Sessions
Suppose then that the kinds of practical strategies for moral development outlined in §3.1 were then combined -in line with an adjunctive moral enhancement proposal like Earp's -with psychedelics (e.g., psilocybin) as kind of multi-faceted moral enhancement strategy, where traditional methods (e.g., Socratic dialogues ( §3.1.1) and role-playing experiments ( §3.1.2) for moral development are combined with psychedelics in a way that will in principle be mutually supportive (for each strategy).
Even on these assumptions, a concrete proposal for implementing something like this adjunctive strategy will need to consider ways to best manage the relationship between the participant and the enhancement facilitator in these sessions.
Firstly -and this is a point we will now explore in some depth -note that it is initially very intuitive that trust in such a relationship would play a key role in improving the effectiveness of moral development sessions-especially given that such sessions (perhaps more so than, e.g., in a business ethics class or in a more traditional education setting) involve discussing personal information about your thoughts and feelings (relevant to moral growth), with the potential to be subjected to moral judgement depending on what is disclosed and how questions are answered. Without a sense that the participant can share information honestly without receiving a punitive response, it is hard to imagine that you would 14 In particular, Knobe's results indicated that perceived goodness or badness of side effects of actions influences people's inclination to attribute intentionality to those side effects. For overviews, see, e.g., Feltz [50] and Nichols and Ulatowski [51].
share very much at all. 15 Not to mention, without trust in the facilitator or the broader organisation, you may be highly aware of the sensitive nature of what you share, and the consequences that its dissemination could have for how others perceive you (both personally and professionally), as well as to feel vulnerable in ways that might lead to a less effective use of psychedelics. 16 An established relationship of trust is important additionally in group enhancement contexts-i.e., where more than one participant is present along with a facilitator. Participants must be willing to be observed as influenced by cognitive biases by the facilitator and group members, and such willingness will likely depend on trusting that there is a shared commitment to respectful conduct. 17 However, we can further our understanding of the relevant trust involved-and begin to get a sense of what might be required to create it-by drawing on some of the recent literature in the philosophy of trust. 18 A commonly drawn distinction in this literature (on interpersonal trust, specifically) is between trust and mere reliance [18], where both involve depending on an another to take care of things (in some way), but where only trust-and not reliancerequires relying on the agent to take care of things (i.e., that on which she is relied) in conjunction with certain kinds of attitudes or beliefs toward the trustor. For example, I might merely rely on a someone by planning in ways that depend on their acting in predictable ways; such reliance requires no particular kind of optimism on my part that the person on whom I am relying will, e.g., take the fact that I am relying on them as reason to prove trustworthy, whereas (presumably) someone I trust will do so. A related distinction here between trust and reliance concerns the place of betrayal -trust as such can be betrayed (and when it is, the trustee is subject to certain appropriate reactive attitudes), in light of whatever normative standing there is between the trustor and trustee, whereas, betrayal is not applicable to mere reliance.
A contested question in this literature is whether in trusting I must believe or hope the trustee will prove trustworthy out of goodwill, or perhaps whether this characterisation is too demanding. For example, Blackburn [66] notes that it seems too strong to require that a person fulfil commitments to an agent in virtue of goodwill, because we place trust in many people we encounter in spite of having no sense that they need have goodwill toward usoften people in professional roles, such as the person who delivers our mail at an agreed time. As Blackburn puts it, we often "[suppose] merely that they have psychological traits sufficient to get the job done." The goodwill account of trust is further called into doubt by cases in which it appears too weak-specifically, in cases where goodwill exists between two people in spite of one not believing the other is trustworthy. For example, consider that we might have goodwill toward people we love deeply even when they have a track record that tells us we cannot rely on them in certain important ways.
In sum, it seems that kind of trust that would seem important for facilitating relationships that facilitate for psychedelic moral enhancement can though perhaps needn't involve a dimension so strong 19 (even if 15 See e.g., Wilkins [54] for exploration of how the communication of unconditional positive regard is a major curative factor in any approach to therapy, and Watson and Steckley [55] for discussion of the evidence in favour of the effectiveness of being unconditionally respecting and valuing of clients in therapy. There is also evidence that this sort of attitude is useful in fostering self-regard (e.g., [56]), and that independence and self-responsibility increase in the absence of external judgement (e.g., [57]. See also Frankel et al. [58] for an argument that unconditional positive regard is especially effective at encouraging change in therapy when combined with empathy. 16 See Bunce [59] for discussion of how negative association with psychedelic use arguably makes one more likely to have a "bad trip", and see Haijen et al. [60] for a study in which negative experiences during psychedelic use were partly predicted by not "feeling comfortable in the environment and with the people that were present during the experience" (p.9). 17 We will explore confidentiality and contracting in more depth in the next section. 18 I am here appealing to two models which give a primary place (respectively) to the normative notions of commitment and goodwill. These dimensions of trust are not distinctive to the accounts Baier and Hawley give, it should be emphasised, and in fact, such notions feature in other accounts. For instance, optimism that the trustee will prove trustworthy through goodwill is a component of Jones' [61] account of trust. Likewise, Hieronymi's doxastic account of trust (e.g., [62]) includes components of commitment, as does Frost-Arnold's [63] account. For an overview of recent work on trust, which includes overviews of alternative accounts that go beyond what I will be discussing here, see, e.g., McLeod [64] and Carter and Simion [65]. 19 For example, it sounds odd to imagine one such person saying "I trust the enhancement facilitator-I'm confident she'll keep my information confidential because she has good will toward me." While such an expectation on the part of the participant towards the facilitator might be a pleasing feature of such work and one we might hope for, it seems rather beside the point in this particular context.

3
Vol.: (0123456789) some rich forms of trust include this goodwill dimension on the part of the trustee).
Katherine Hawley [67], however, offers us an alternative view of trust that doesn't have the same shortcomings, and which is also more practically applicable for present purposes (and in other cases where trust is present when we deal with those with whom we are in a professional rather than personal relationship). Hawley-who thinks that trust is "primarily a three-place relation, involving two people and a task" (2014, p. 2)-focuses not on good will but on the notion of the trusted agent making a commitment to do what she is entrusted to do, where the commitment (rather than goodwill or any other positively valenced affective attitude) is what distinguishes trust normatively from mere reliance. 20 Someone who views themselves as committed to doing something (but not someone who merely can be relied on to do something), will view herself is subject to, e.g., reactive attitudes such as gratitude in blame, and when we trust, according to Hawley, we take it that it is common belief between both parties that such reactive attitudes (in virtue of the commitment present) would be appropriate.
In sum, we've seen that trusting relationships are very plausibly instrumental in facilitating the kinds of adjunctive moral enhancement sessions described in §3.1, and we've seen further -albeit briefly -what some of the key normative dimensions of trusting involve, with transparency of commitment being a dimension of trust with particular applicability in the kinds of cases of interest.
In what follows, let's consider now more concretely how we might best develop that trust. If Hawley is right-then what we should focus on when thinking about how to facilitate trust in moral enhancement facilitators is the commitments they make-and, perhaps, the transparency of those commitments to the participants. Let's now sharpen this idea.

Facilitating Trust to Facilitate Moral Enhancement
With Hawley's commitment-focused account of trust as a working view in hand, let's now consider some ways in which we might encourage the development of trust in the context of psychedelic moral enhancement. What we are looking for here are ways in which we might make the commitments of the facilitator and the sincerity of those commitments obvious to those participating in moral enhancement sessions.

Enhanced Informed Consent and Clear Contracting
When psychedelics are used as part of medical treatment, Johnson et al. [23] suggest preparatory sessions that form the basis for the agreement between facilitator and participant. In such preparatory sessions, as Johnson describes them, participants are introduced to the logistics of treatment sessions, begin to build a therapeutic relationship with facilitators, and take part in "a detailed discussion of the possible range of [psychedelic] experiences" (p. 612). as well as receiving guidance on how to address challenging experiences. Preparatory sessions of the above sort would not only be a useful forum for safeguarding and receiving ethically appropriate informed consent but also for helping to make commitments salient, especially to the participant, in a way that will improve the effectiveness of moral enhancement by encouraging trust.
Relatedly, in recent work, Smith and Sisti [22] have detailed specific issues that should be focused on during the process of gaining informed consent and suggest discussion prompts for enhanced consent in psychedelic psychiatry. One such discussion prompt that Smith and Sisti propose addresses concerns (alluded to briefly in §3.2) that some subjects will experience unwelcome personality changes during psychedelic treatment if their newfound values stand in opposition to their earlier values-most commonly, they are likely to become more open to different experiences and different points of view, may become more or less spiritual, and may be seen as different by loved ones. Secondly, as Smith and Sisti note, facilitators caution that psychedelic experiences are in a certain sense ineffable-that, put simply, we can't tell people exactly what they'll experience, especially not in any fine-grained detail, even when engaging in a careful discussion in the course of informed consent. As they put it: "Those who have experienced this often find it difficult to convey to others exactly what they experienced. Hence, we cannot tell you exactly what this is like, and you may have trouble understanding it before you experience it yourself" (p.3). In addition, more extreme negative reactions-including anxiety and trauma re-exposure-can occur, so these are important to cover responsibly. 21 To promote trust when presenting this information (i.e., including about the ineffability of certain kinds of psychedelic experiences themselves) facilitators may encourage participants to reflect on their own medical and personal history with these risks in mind, while nevertheless noting that there is no foolproof way to predict who will experience these unpleasant effects. Furthermore, given that -at least presently -moral enhancement is not fine-grained enough to rule out the possibility that the participant could change in some new way they didn't specifically endorse, discussion by the facilitator with participants offers good opportunity to probe how they'd feel about such changes. For example: would they be viewed as improvements? Does this warning bring up feelings of resistance in the participant that might be linked to them being under pressure to morally enhance (whether from society or specific people)? 22 Such disclosure and the importance of corresponding discussion stands to clarify perceived commitment on the part of the facilitators, as well as, crucially, the facilitator's own assessment of the limitations of these commitments.
All of the foregoing discussion prompts are direct, honest and communicate a clear commitment to patient care. Moreover, they are thorough in a way that makes transparent a commitment on the facilitator's part to assisting the participant in making a balanced decision about whether to proceed-and, accordingly, have the capacity to build the kind of commitment-transparent trust that will plausibly carry over future moral enhancement sessions are experienced.
Careful informed consent discussions, then, have at least three purposes in my view, and are therefore deserving of careful development: such discussions promote safety, they manage expectations about the process, and they build trust by communicating a sense that the facilitator is committed to transparency and to prioritising participant well-being.

"A Meaningful Human Encounter"
While we have already seen that good will accounts of trust are, at least contestably in the philosophy of trust, both too strong and too weak in different respect, there remains a good reason why Baier's account-type (also embraced influentially by Karen Jones [61] is among the most prominent and widely embraced. 23 Even if we shouldn't strictly define trust in terms of an attitude that implies optimism that there will be good will on the part of the trustee, we will nevertheless in many cases find that the perception of a trustee's goodwill will nevertheless be helpful in facilitating trust. As such, it will be helpful to consider not only ways to illuminate and demonstrate facilitator commitment but also to communicate genuine goodwill. Drawing from literature in psychotherapy on the therapeutic relationship, Mozdzierz et al. ([70] p. 126) observe that "Even though it is a professional engagement, [it] is a human encounter". 24 If we look at research on what determines the outcome of psychotherapy, we repeatedly see studies and literature reviews confirming that "specific techniques contribute much less to outcome than do important 1 3 Vol.: (0123456789) interpersonal factors common to all therapies" ( [71], p. 21). 25 For present purposes, I want to highlight at least three components that Horvath et al.'s [21] work on therapeutic relationships describes as central to that human encounter: (1) a positive affective bond (which is explicitly described by Horvath et al. as involving trust), (2) a mutual understanding of the tasks being worked on, and (3) a consensus on goals.
Firstly, regarding affective bonds, there is extensive work on such interpersonal factors, how to define them and how to facilitate them. Here, I will stress just two relevant and interrelated ingredients-empathy and a non-judgemental stance. Regarding empathy, Lambert and Barley [71] emphasise that the facilitator's manifesting empathy has a powerful impact on how therapeutic encounters are experienced. Meanwhile, as Mearns and Cooper ( [73], p. 126) note, psychotherapists writing on the process of creating trust with psychotherapy clients observe that: "the deepest fears of many [people] is that, once they are seen for who they are, they will be criticised, humiliated and attacked." Relatedly, Dryden and Reeves ( [20], p.86) suggest that "trust emerges [through] being congruent, present, honest, containing, non-judgemental", and by committing to the work "no matter how difficult it gets." This advice bears special relevance when the subject under discussion is morality, and the topics being explored require the participant to reflect on uncomfortable biases, emotional reactions and regrets. Consequently, even in the light of the shared goal being moral development, 26 it will also be valuable for facilitators to avoid moralising, at the risk of the participant shutting down or becoming defensive (even in a psychedelic-enhanced state of increased openness). 27 This is not to suggest that an entirely neutral or value free language is suggested or even feasible when the subject is morality, which is inherently normative. Rather, the point against moralising should be understood as applying to mode of facilitating discussion, rather than a restriction on content. 28 Regarding the second and third ingredients suggested by Horvath et al.-mutual understanding of tasks, and consensus on goals-we should expect that thorough, careful contracting would help to facilitate bringing about and sustaining both. Further specifying tasks and goals could be approached in a range of ways. For example, the moral enhancement work might be split roughly into three potential "types": (1) specific moral struggles the participant has identified in their life experiences (e.g., difficulty with forgiveness or with commitment), (2) ethical questions they find difficult (e.g., the permissibility of abortion, how much to give to charity), and (3) areas that humans in general find difficult (e.g., common cognitive biases and stereotyping). For participants who definitely want to take part but are not sure where to focus, there might also be set of predesigned choices from which to select, which could help to give structure to the moral development work. 29

Concluding Remarks
This paper suggests how an adjunctive 'middle ground' kind of proposal for using psychedelics for purposes of moral enhancement has potential not only in theory (as Earp has demonstrated) but also in practice. In recommending some strategies for responsibly and effectively implementing such a proposal, I've identified -among other things -the important role that trust plays in minimising various kinds of risks while increasing benefits; I've also considered several 25 See also Martin et al. [72] for a meta-analysis of 79 studies of psychotherapy outcomes that indicates an alliance between therapist and client is the most important predicator of a good outcome, with empathy being an especially powerful component. 26 It is worth pointing out that screening/consent process should double-check that this 'match' genuinely exists between facilitator and participant (rather than merely double checking what views the participant holds). specific ways that the kind of trust that might be especially useful in facilitating psychedelic-assisted moral development can be promoted. The result, I hope, is a step in the direction of seeing more clearly what psychedelic moral enhancement would look like in concrete terms in practice, how it might be successful, and what factors should be focal points when considering how such a proposal might be embraced and pursued.

Declarations
Conflicts of Interest This work was conducted as part of the Leverhulmefunded 'A Virtue Epistemology of Trust' (#RPG-2019-302) project, which is hosted by the University of Glasgow's COGITO Epistemology Research Centre, and I'm grateful to the Leverhulme Trust for supporting this research. I have no other financial or non-financial interests to declare.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.