Wednesday, January 30, 2019

Episode #52 - Devlin on Sex Robots and Moral Panics

Kate Devlin.001.jpeg  

In this episode I talk to Kate Devlin. Kate is a Senior Lecturer in the Department of Digital Humanities at King's College London. Kate's research is in the fields of Human Computer Interaction (HCI) and Artificial Intelligence (AI), investigating how people interact with and react to technology in order to understand how emerging and future technologies will affect us and the society in which we live. Kate has become a driving force in the field of intimacy and technology, running the UK's first sex tech hackathon in 2016. She has also become the face of sex robots – quite literally in the case of one mis-captioned tabloid photograph. We talk about her recent, excellent book Turned On: Science, Sex and Robots, which covers the past, present and future of sex technology.

You download the episode here or listen below. You can also subscribe on iTunes or Stitcher (the RSS feed is here).

Show Notes

  • 0:00 - Introduction
  • 2:08 - Why did Kate talk about sex robots in the House of Lords?
  • 3:01 - How did Kate become the face of sex robots?
  • 5:34 - Are sex robots really a thing? Should academics be researching them?
  • 11:10 - The important link between archaeology and sex technology
  • 15:00 - The myth of hysteria and the origin of the vibrator
  • 17:36 - What was the most interesting thing Kate learned while researching this book? (Ans: owners of sex dolls are not creepy isolationists)
  • 23:03 - Is there are moral panic about sex robots? And are we talking about robots or dolls?
  • 30:41 - What are the arguments made by defenders of the 'moral panic' view?
  • 38:05 - What could be the social benefits of sex robots? Do men and women want different things from sex tech?
  • 47:57 - Why is Kate so interested in 'non-anthropomorphic' sex robots?
  • 55:15 - Is the media fascination with this topic destructive or helpful?
  • 57:32 - What question does Kate get asked most often and what does she say in response?

Relevant Links

  • Laodamia - Owner of the world's first sex doll?

Thursday, January 24, 2019

New Paper - On the Wisdom of Algorithmic Markets: Governance by Algorithmic Price

Another day another new paper. This is one I co-authored with Pip Thornton. It applies Hayek's famous 'wisdom of the market' argument to algorithmically constructed marketplaces. This is a working paper. You access the full draft by clicking on the paper title below or on the other links.

Title: On The Wisdom of Algorithmic Markets: Governance by Algorithmic Price
Authors: Pip Thornton & John Danaher
Links: SSRN; Philpapers
Abstract: Leading digital platform providers such as Google and Uber construct marketplaces in which algorithms set prices. The efficiency-maximising free market credentials of this approach are touted by the companies involved and by legislators, policy makers and marketers. They have also taken root in the public imagination. In this article we challenge this understanding of algorithmically constructed marketplaces. We do so by returning to Hayek’s (1945) classic defence of the price mechanism, and by arguing that algorithmically-mediated price mechanisms do not, and probably cannot, accumulate and signal that same kinds of knowledge that Hayek felt were essential to effective market governance. Indeed, we argue that algorithmically-constructed marketplaces are closer to, though distinct from, the central planning model that Hayek critiqued. Regardless of how you feel about Hayek’s argument, this has important consequences for how we respond to the rise of algorithmic governance tools in both markets and elsewhere.

Tuesday, January 22, 2019

New Paper - Regulating Child Sex Robots: Restriction or Experimentation?

I have a new paper about what is for many people an unthinkable topic: child sex dolls/robots. But, as I say in the introduction to the paper itself, you sometimes have to think about the unthinkable. The paper is set to appear in a special edition of the Medical Law Review later this year. You can access a preprint version of it below.

Title: Regulating Child Sex Robots: Restriction or Experimentation?
Journal: Medical Law Review
Preprint Links: Philpapers; Academia; Researchgate 
Abstract: In July 2014, the roboticist Ronald Arkin suggested that child sex robots could be used to treat those with paedophilic predilections in the same way that methadone is used to treat heroin addicts. Taking this onboard, it would seem that there is reason to experiment with the regulation of this technology. But most people seem to disagree with this idea, with legal authorities in both the UK and US taking steps to outlaw such devices. In this paper, I subject these different regulatory attitudes to critical scrutiny. In doing so, I make three main contributions to the debate. First, I present a framework for thinking about the regulatory options that we confront when dealing with child sex robots. Second, I argue that there is a prima facie case for restrictive regulation, but that this is contingent on whether Arkin's hypothesis has a reasonable prospect of being successfully tested. Third, I argue that Arkin's hypothesis probably does not have a reasonable prospect of being successfully tested. Consequently, we should proceed with utmost caution when it comes to this technology.

Friday, January 18, 2019

Why You Should Love/Hate Moralists

You know the type. Always quick to blame you for your moral complacency. Always righteously indignant at your moral failings. Always keen to highlight their virtue and your vice. I am talking about moralists, of course. The self-appointed guardians of moral order. The people who pour scorn on the rest of us.

If you are anything like me, you won’t like moralists. You will find them intrusive, overbearing and annoying. You’ll tend to ignore them or ridicule them. Maybe not to their faces. Maybe you will wait until they are out of earshot. But once they are you’ll whisper to your friends “Who does she think she is?” “She should mind her own business!”.

I know this attitude intimately. Sometimes I feel like I spend my whole life surrounded by moralists. It is a professional hazard — university departments are full of them. But why do I dislike them so much? Why do I find them so off-putting? Philosophers have tried to answer these questions, and it turns out that many of them share my discomfort. They dress it up in the form of reasoned criticism. They argue that moralists degrade and undermine the system of morality: that they foster a climate of moral intolerance and help to kick-start a moral arms race.

I find these arguments reassuring. They convince me that I am right to be morally sceptical and reticent. But this also makes me suspicious. When I probe the depths of my disdain for moralists I worry that its true origin lies in the fact that I think moralists are often in the right — that oftentimes their moral critique is on point. I want to explain why I think this in what follows.

1. What is moralism? What is a moralist?
Let’s start by clarifying what a moralist actually is. Philosophers have a particular take on this. They don’t see moralists as people who are just uniquely sensitive to moral norms and who eagerly identify and police our moral failings. They also don’t see moralists as people who use regressive or incorrect moral beliefs to criticise others. Catholic priests might be very keen to police the boundaries of our private sexual lives, but the problem with their doing so is not that they are moralists (though they may be that too). The problem is that they have the wrong moral beliefs and use those to criticise others (if you are a staunch traditional Catholic you won’t accept this — more on this problem later).

The vice of moralism is something different. Alfred Archer writes about this in his article ‘The Problem with Moralism’. As Archer sees it, the really unique thing about the moralist is not that they get morality completely wrong but, rather, that they almost get it right. Their moral compass is just slightly miscalibrated and it is this miscalibration that does all the damage. The moralist uses their ‘almost correct’ morality to apply excessive and unwarranted moral criticism to others. As Archer points out, there are four distinct errors that the moralist typically makes:

Type 1: The moralist demands that others engage in supererogatory acts, not just obligatory acts. Obligatory acts are those that you are morally bound to do. It is perfectly appropriate for someone to demand that you do them. Supererogatory acts are above and beyond the call of moral duty. They are morally good (possibly very good) but they are not obligatory. The moralist tends to conflate the two and demand that we all go above and beyond the call of duty. An example might be the moralist demanding that you engage in some act of heroic self-sacrifice to save the lives of others. While most people would agree that heroic self sacrifice is morally commendable, few would argue that it is obligatory.

Type 2: The moralist demands that others engage in particular acts that are morally indifferent or neutral. In other words, the moralist tells us that we have to pick and choose between different acts that have no positive or negative moral value. An example might be the moralist claiming that you have a moral obligation to eat a particular flavour of ice-cream or see a particular film, when actually it doesn’t matter which flavour you eat or film you see.

Type 3: The moralist blames others for engaging in immoral acts for which they have a legitimate moral excuse. An example might be a moralist blaming someone for breaking a promise to meet them at a particular location when the person had some serious illness that made it impossible for them to make it.

Type 4: The moralist applies excessive blame to others for minor moral infractions. An example might be a moralist arguing that someone should be given a long jail sentence (or other severe punishment) for jay-walking, or that someone should be ostracised and ignored for being five minutes late to an appointment.

Again, the interesting thing in each of these cases is not that the moralist gets morality completely wrong. They have, roughly, the correct conception of what is good and what is bad. The problem is that they overreact and over-demand.

It should also be noted that this typology of moralism focuses on the negative, i.e. on moral criticism and moral demands. It is possible — and Archer acknowledges this — for a moralist to miscalibrated on the positive side too. So, for example, a moralist might be too quick to assign moral praise to an action. But even though this positive form of moralism is conceivable, it is the negative form that is probably more common and more irksome. We now need to figure out why this is the case.

2. How Moralists Undermine Morality
The main argument against moralism is that it corrodes and undermines morality. In other words, moralists have a perverse effect on the very thing they seem to cherish the most. Again, Archer presents a good version of this argument in his article ‘The Problem with Moralism’. The argument starts by focusing on the value of legitimate moral criticism and then concludes that moralists tend to undermine this with their excessive moral criticism. In basic outline, the argument works like this:

  • (1) Legitimate moral criticism is socially valuable.

  • (2) Moralists tend to undermine the force and salience of legitimate moral crticism.

  • (3) Therefore, moralists tend to undermine something that is socially valuable.

Premise (1) should be common sense. Moral norms are important. If we all follow the moral code (and if that moral code is correct), society benefits. If some people (or some institutions etc) fail to do so, it is good if we can single them out and criticise them for their shortcomings. Legitimate moral criticism of this sort allows us to identify moral injustices and motivate moral reform. Many of the great social justice movements of the past (anti-slavery; civil rights reform; anti-colonialism; women’s rights etc.) have been premised on legitimate moral criticism. It is what gives them their energy and urgency. If people didn’t pay attention to legitimate moral criticism, or tended to ignore or overlook it, these movements would be deprived of a vital resource. The same is true on a much smaller scale. Moral criticism, when done right, has real bite: it can encourage someone to rethink what it is they are doing and change their conduct for the better. But if it loses this bite, this personal reform becomes more difficult (I won’t say ‘impossible’ because there are, presumably, other levers that could be pulled to change someone’s conduct for the better).

Premise (2) is the core of the case against the moralist. The support for this is, in effect, a classic ‘boy who cries wolf’ argument. If moralists constantly overreact and over-demand, then they denude the force of legitimate moral criticism. If everything becomes morally obligatory, then you may as well say that nothing is morally obligatory; if minor infractions result in excessive blame, then it is quite likely that people will become desensitised to proportionate blame. The signal gets lost in the noise, and the people to whom the criticism is directed build up resentment and eventually start to ignore all forms of moral criticism.

In their article on ‘Moral Grandstanding’, Justin Tosi and Brandon Warmke go a bit further and argue that moralism (they don’t use the word but I think they are talking about the same basic phenomenon) has a poisonous effect on our public moral discourse. If one individual or group engages in excessive moral criticism of another, the other tends to respond in kind. The opposing party hasn’t just made a mistake or got things wrong, they have made a completely, obviously corrupt and unjust mistake. They are a moral cancer that must be stamped out. The end result is, according to Tosi and Warmke, increased polarisation in the public sphere and increased cynicism about moral talk. As they put it:

Ramping up [of moral criticism] contributes to group polarisation, where individuals come to hold more extreme views after deliberating with others, rather than moving toward a moderate consensus. The result of a moral arms race is that people will tend to adopt extreme and implausible views, and refuse to listen to the other side…Another consequence of grandstanding is that many people stop taking moral conversations seriously. They become cynical about the moral claims they hear in public discourse because they suspect that the speaker is simply trying to show that his heart is in the right place, rather than trying to help others figure out what we should do or believe.

(Tosi and Warmke 2017)

I suspect this is one reason why I often feel irked by moralists. I question their sincerity. I doubt whether they apply the same moral standards to their own behaviour and I suspect they are just engaging in a kind of virtue signalling (note: Tosi and Warmke also resist using this term in their article but I think it is apt).

If premise (1) and (2) are persuasive, then the conclusion follows. Moralists are a problem because they undermine the value of legitimate moral criticism. Archer goes on, in his analysis, to argue that this criticism still applies if you think that morality is quite demanding (as some moral theorists do) because even if morality is demanding there will still be some distinction between superogatory acts and obligatory ones, some distinction between obligatory acts and morally neutral or indifferent acts, some legitimate excuses for moral failings and some sense of proportionality for moral blame. So, for him at least, moralists are always a problem no matter what conception of morality you hold.

3. In Defence of Moralists
That’s the case against moralists. As I say, I find it reassuring because it confirms that my irksome attitude towards them is correct. And, to be fair, I think that some of the fears raised by the critics — undermining legitimate moral discourse and fostering resentment and polarisation — are valid and worth taking seriously. But I also worry about any argument that confirms my prejudices and preconceptions. It prompts me to probe deeper and consider alternative explanations for my attitude. Perhaps the problem is not that the moralists are miscalibrated but that I, myself, am miscalibrated? Maybe the case against moralists is unfair?

Let me start with a minor criticism of the argument. I worry that there is something faintly paradoxical in criticising moralists in the above terms. Essentially, what the preceding argument does is to highlight a moral failing (Archer calls it a ‘vice’) among moralists. To criticise them for that moral failing is at least somewhat ironic: you are morally criticising those who engage in moral criticism. Admittedly, there is an easy way out of this ‘faint’ paradox. You can just say that your moral criticism of the moralist is legitimate. But that’s arguably too easy and is itself a problem if moralised discourse can, in and of itself, be poisonous. This is something that Tosi and Warmke point out in their article and is one reason why they discourage people from using their argument to criticise others. Instead, they encourage people to use the argument to reflect on their own behaviour.

This minor criticism points the way to a deeper criticism — one that I think is the source of my ambivalence about the argument as a whole. The argument presumes that we are able to easily identify moralists, i.e. that we can tell who it is that is engaging in excessive moral criticism. But can we really? I certainly worry about my own ability to do this. Unless I assume my own moral infallibility, it’s quite possible that I am the one getting things wrong. Instead of the moralist being overreactive, maybe it is I who is being underreactive?

This worry becomes more compelling when we reflect on how moral reform usually takes place. Consider, once more, Archer’s four examples of moralistic excess (the conflation of the superogatory with the obligatory etc). And consider all the great moral reform movements of the past (the anti-slavery movement, the women’s rights movement etc). Isn’t it the case that most of those reform movements were initially perceived to be engaging in excessive moral criticism? And didn’t they ultimately work because they shocked the majority out of their moral complacency? These movements pointed out that what the majority thought to be morally indifferent or neutral was, in fact, morally obligatory; that what they thought to be a legitimate excuse, was in fact anything but; and that what was thought to be a minor infraction or inconvenience was, in fact, a serious moral error.

So I worry that this is why I find moralists irksome. It’s not that I think they are wrong or excessive. It’s that I think they might be right and that I am simply resisting uncomfortable change. I find this to be particularly true when I encounter the so-called ‘moral saints’ of this world, i.e. people who dedicate their lives to doing the most good they can possibly do. Larissa MacFarquhar’s book Stranger’s Drowning is filled with compelling portraits of such people, all selflessly dedicating themselves to improving the lives of others. In a famous philosophical essay, Susan Wolf once argued that such moral saints must live impoverished lives because by dedicating themselves to doing good they have no time for the usual humanistic pleasures. But I think moral saints also show us what is possible for beings like us. They show us that what we perceive to be excessively demanding (maybe even impossible) is in fact achievable.

Tuesday, January 15, 2019

Episode #51 - Moen on the Unabomber's Ethics

ole martin moen

In this episode I talk to Ole Martin Moen. Ole Martin is a Research Fellow in Philosophy at the University of Oslo. He works on how to think straight about thorny issues in applied ethics. He is the Principal Investigator of “What should not be bought and sold?”, a $1 million research project funded by the Research Council of Norway. In the past, he has written articles about the ethics of prostitution, the desirability of cryonics, the problem of wild animal suffering and the case for philosophical hedonism. Along with his collaborator, Aksel Braanen Sterri, he runs a podcast, Moralistene (in Norwegian), and he regularly discusses moral issues behind the news on Norwegian national radio. We talk about a potentially controversial topic: the anti-tech philosophy of the Unabomber, Ted Kaczysnki, and what's wrong with it.

You can download the episode here or listen below. You can also subscribe via iTunes or Stitcher (the RSS feed is here).

Show Notes

  • 0:00 - Introduction
  • 2:05 - Should we even be talking about Ted Kaczynski's ethics? Does it not lend legitimacy to his views?
  • 6:32 - Are we unnecessarily anti-rational when it comes to discussing dangerous ideas?
  • 8:32 - The Evolutionary Mismatch Argument
  • 12:43 - The Surrogate Activities Argument
  • 20:20 - The Helplessness/Complexity Argument
  • 23:08 - The Unstoppability Argument
  • 26:45 - The Domesticated Animals Argument
  • 30:45 - Why does Ole Martin overlook Kaczynski's criticisms of 'leftists' in his analysis?
  • 34:03 - What's original in Kaczynski's arguments?
  • 36:31 - Are philosophers who write about Kaczynski engaging in a motte and bailey fallacy?
  • 38:36 - Ole Martin's main critique of Kaczynski: the evaluative double standard
  • 42:20 - How this double standard works in practice
  • 47:27 - Why not just drop out of industrial society instead of trying to overthrow it?
  • 55:04 - Is Kaczynski a revolutionary nihilist?
  • 58:59 - Similarities and differences between Kaczynski's argument and the work of Nick Bostrom, Ingmar Persson and Julian Savulescu
  • 1:04:21 - Where should we go from here? Should there be more papers on this topic?


Relevant Links

  • Ted Kaczynski on Wikipedia (includes links to relevant writings)
  • "The Unabomber's Penpal" - article about the philosopher David Skrbina who has corresponded with Kaczynski for some time


Monday, January 7, 2019

The Politics of Sexual Exclusion: Notes on Srinivasan's 'Does Anyone Have a Right to Sex?'

[These are expanded reading notes. They are part of my effort to learn more about the ethics of sexual exclusion.]

Does anyone have the right to sex? That’s the question posed in the title of a prize-winning essay by Amia Srinivasan. The essay addresses a contentious and politically charged issue. To put it bluntly, some people are sexually excluded. That is to say: they are not seen as potential/desirable sexual partners. Since sexual intimacy is typically seen as a ‘good’ (something that, all else being equal,* makes life better), this sexual exclusion could be viewed as an injustice. This is particularly true if certain groups of people are systematically sexually excluded. Could they have a ‘right’ to sexual inclusion?

This question has been actively debated for quite some time, particularly in the disability rights community. But it is, to many, a deeply unpalatable idea. People worry that it feeds upon patriarchal and misogynistic ideas of sexual entitlement. Srinivasan’s essay addresses these concerns head-on by engaging directly with the ‘incel’ movement and the way in which its members talk about sexual exclusion and sexual entitlement in violent and misogynistic terms. And of course it is not just talk: at least two members of the online ‘incel’ community have been responsible for mass killings. It’s hard to debate the philosophical niceties of sexual inclusion and exclusion in the shadow of such atrocities.

But Srinivasan’s essay does so in an engaging and unsettling (in the best sense) way. I highly recommend reading it. In this post, I want to try to excavate the logic of the main arguments and ideas in Srinivasan’s essay and offer some critical reflections of my own. Despite the title and the initial set-up, I think the essay turns out to be less about the right to sex and more about offering a corrective to a long-standing debate within feminist theory. It then uses this corrective to raise a dilemma for those who want to argue about the politics of sexual inclusion/exclusion. That’s my reading of it anyway. I’ll try to explain why I take that perspective in what follows.

1. The Politics of Sexual Desire
Srinivasan’s essay starts with a discussion of the Elliot Rodger case. Rodger was a member of an online incel discussion group. On May 23rd 2014, Rodger killed six people and injured fourteen others in a mass shooting/killing spree that took place in and around the University of California, Santa Barbara. Rodger released a manifesto that explained his actions. The manifesto fixated on his sexual exclusion, specifically on his frustration at not being seen as an object of sexual desire by ‘beautiful blonde’ ‘white’ girls (Rodger’s manifesto and online commentary was replete with racist statements). He saw them as cruel, stuck-up and heartless. They had to be punished for his exclusion.

Rodger’s horizon of sexual desire seems to have been very narrow. He longed for a certain type — the beautiful blonde white girls — not for other types. Why was this? Srinivasan and others see Rodger’s limited sexual horizons as a product of patriarchal conditioning. The ideological climate in which he lived and breathed taught him that only certain kinds of women were sexually ideal and so being ignored or overlooked by these women was perceived, by him, to be a grave injustice. Now, to be clear, Srinivasan doesn’t think that patriarchal conditioning is the only thing that explains Rodger’s behaviour — other factors were clearly at play — but she does use his case a springboard for asking the broader question: should we take a more critical perspective on the ways in which ideologies shape and construct our sexual desires?

This is where a long-standing debate within feminist theory rears its head. According to the radicalist school of thought, we should pay a lot of attention to the social conditioning of sexual desire. In particular, we should pay attention to the way in which male and female sexual desires are constructed around relationships of subordination and domination. Men are seen as sexual aggressors: they must be controlling and dominating. Women are seen as sexually subordinate: they must be submissive, coy, and sexually compliant. This conditioning is all part of the patriarchy: the ideology that prevents ensures male primacy and prevents true equality between the sexes. Under the pervasive influence of the patriarchy, all sexual desires and experiences must be questioned. It is very difficult (perhaps even impossible) for them to be truly authentic or autonomous. Thus, women who claim to enjoy heterosexual sex (and possibly even lesbian sex) are more than likely victims of a ‘false consciousness’. This highly politicised approach to sexual desire is associated, in particular, with the work of one of the leading radical feminists, Catharine MacKinnon.

The radicalist school of thought has been criticised by the sex positive school of thought. Sex positive feminists think that women can experience genuine and authentic sexual pleasure, and that not all sexual desires and experiences need to be questioned. In a famous article called ‘Lust Horizons: Is the Women’s Movement Pro-Sex?’ — which Srinivasan returns to several times in her essay — Ellen Willis sets out the case for the sex positive view, arguing that MacKinnon and her ilk are unnecessarily inducing shame in women for their sexual desires, reinforcing a negative view of women as victims, and denying women any possibility of genuine sexual agency. Willis is much less inclined to be sceptical about expressions of sexual desire and pleasure than MacKinnon: if a woman testifies, under appropriate conditions, that she enjoys certain forms of sex, and that she engages in those forms of sex authentically and of her own volition, then who are we to question her?

In offering this critique, Willis is supported by the move towards ‘intersectional’ analysis in feminist theory (roughly: the idea that individuals don’t fall neatly into a single identity category, such as ‘male’ or ‘female’ but, instead, exist at the intersection of multiple overlapping identity categories). The intersectional mode of analysis casts doubt on the notion that all women experience the negative effects of patriarchal conditioning in the same way. It favours trusting the testimony of individuals over general theories (following the strictures of standpoint epistemology). If individuals say they desire or are empowered by certain forms of sex, then so be it. This is true even if the sex they want is commercialised (i.e. prostitution/sex work), or includes elements of dominance and subordination, or other superficially concerning features.

As Srinivasan puts it:

The case made by Willis in ‘Lust Horizons’ has so far proved the enduring one. Since the 1980s, the wind has been behind a feminism which takes desire for the most part as given — your desire takes the shape that it takes — and which insists that acting on that desire is morally constrained only by the boundaries of consent. Sex is no longer morally problematic or unproblematic; it is instead merely wanted or unwanted. 
(Srinivasan 2018)

Should this victory for sex positive feminism be celebrated? Srinivasan has her doubts.

2. Srinivasan’s Argument: The Danger of Depoliticising Sexual Desire
The sex positive approach reinforces a privatised model of sex. Your sexual desires are your own business. All’s fair in love and sex, so long as you respect the consent and autonomy of your sexual partners. Consent becomes the only currency of any significance in debates about the permissibility of sex.

The problem, for Srinivasan, is that this overlooks and ignores the obvious ways in which our sexual desires are politically and socially conditioned to be discriminatory. Srinivasan does not set out her critique of sex positive feminism in formal terms in her essay. I’m going to try to do this by offering a semi-formal reconstruction of the argument. Some of the language in this reconstruction might seem a little strange but I will explain it after I have set it out. Here it is:

  • (1) Sex positive feminism encourages us to view consent/wanting as the sole constraint on permissible sex (i.e. to adopt a privatised/laissez faire model of sexual desire).

  • (2) If we treat consent/wanting as the sole constraint on permissible sex, then we end up treating sexual desire as primordial/axiomatic, not political.

  • (3) If we treat sexual desire as primordial/axiomatic, then we provide a cover for (or legitimise or unduly ignore) the way in which sexual desire manifests discriminatory ideologies such as racism, ableism, transphobia (etc).

  • (4) Sexual desire is political, not primordial/axiomatic.

  • (5) We should try to combat/undermine/neutralise discriminatory political ideologies.

  • (6) Therefore, we need to push back against (or reconsider) the laissez faire approach of sex positive feminism.

This reconstruction is messy, and it is a little bit difficult to string together the premises into a logically valid form, but I think the premises do fairly capture the key propositions in Srinivasan’s critique of the sex positive approach. Hopefully, premise (1) will make sense as is: it is based on the description of Willis’s work from the previous section. The only thing that needs to be said here is that Srinivasan does point out that Willis’s original essay was a bit more nuanced and ambivalent than is suggested by premise (1). Thus, the focus of the critique is not on Willis’s work per se but rather on an idealised/popularised form of sex positive feminism. The other premises require a bit more unpacking. I’ll go through each of them now, quoting relevant passages from Srinivasan’s essay to show how I derived them.

I’ll start with premise (2). This should be relatively straightforward. The idea being expressed here is that if we adopt the sex positive view, we end up treating sexual desires as pre-political ‘fixed points’ (i.e. axioms) in our ethical thinking. They are placed among the things that cannot be challenged or questioned. The assumption — and it is a big assumption in Srinivasan’s essay, one that I will return to later — is that if we saw them as being politically conditioned/shaped we would not approach them with the same fixed mindset. We would instead see them as things to be challenged and questioned.

Premise (3) allows us to appreciate why we might want to question or challenge sexual desires. This is really the keystone of Srinivasan’s argument and the hingepoint of her entire essay. The claim being made in that premise is twofold: (i) that sexual desire is, in fact, infected with racist and other discriminatory ideologies and (ii) that treating desire as axiomatic and pre-political gives those discriminatory ideologies a free pass. I’m just going to quote an entire paragraph from Srinivasan’s essay because I think it summarises the point better than I ever could:

When we see consent as the sole constraint on OK sex, we are pushed towards a naturalisation of sexual preference in which the rape fantasy becomes a primordial rather than a political fact. But not only the rape fantasy. Consider the supreme fuckability of ‘hot blonde sluts’ and East Asian women, the comparative unfuckability of black women and Asian men, the fetishisation and fear of black male sexuality, the sexual disgust expressed towards disabled, trans and fat bodies. These too are political facts, which a truly intersectional feminism should demand that we take seriously. But the sex-positive gaze, unmoored from Willis’s call to ambivalence, threatens to neutralise these facts, treating them as pre-political givens. In other words, the sex-positive gaze risks covering not only for misogyny, but for racism, ableism, transphobia, and every other oppressive system that makes its way into the bedroom through the seemingly innocuous mechanism of ‘personal preference’. 
(Srinivasan 2018)

Srinivasan goes on to give more detailed examples of how oppressive and discriminatory attitudes make their way ‘into the bedroom’. She provides an illuminating story about the way in which Grindr allows its users to sort potential partners along ‘discriminatory grooves’: Asian men, for example, are often ignored/overlooked, or only attract the attention of ‘Rice Queens’ who think they are ‘good at bottoming’. And it’s not just Grindr. All online dating platforms allow users to do this, to at least some extent. The sexually excluded and ‘unfuckable’ (to adopt Srinivasan’s dysphemism) constantly remain on the fringes of sexual acceptability thanks to algorithmic preference management. This is a political problem, not just a personal one. As she puts it later in her essay:

... simply to say to a trans woman, or a disabled woman, or an Asian man, ‘No one is required to have sex with you’ is to skate over something crucial. There is no entitlement to sex, and everyone is entitled to want what they want, but personal preferences — NO DICKS, NO FEMS, NO FATS, NO BLACKS, NO ARABS, NO RICE NO SPICE, MASC-FOR-MASC - are never just personal. 
(Srinivasan 2018)

From the recognition of this problem, the rest of the argument unfolds in a predictable way. Premise (4) may not be logically necessary for the argument but it is something that is repeatedly emphasised by Srinivasan. Our sexual desires are politically shaped and conditioned; they are not — contra the sex positive view — primordial pre-political axioms. The significance of this, I take it, is that if they are political then it is right and proper to assess them in them in same terms as other political institutions and facts. This could include seeing them as tools of oppression and injustice and taking steps to dismantling them (much as the radical feminists asked us to do in the 1970s and 1980s). This, then, is where premise (5) comes in. It suggests a direction or purpose to any such reconstructive exercise: we should work to reshape sexual desire so as to prevent systematic discrimination from reasserting itself in the bedroom.

The conclusion — that we need to reconsider the wisdom of sex positive feminism — then seems to follow.

Or does it? Here, we need to be cautious. Even though I think my reconstruction of her critique of sex positive feminism is plausible, it’s important to realise that Srinivasan is much more ambivalent about it than I have let on to this point.

3. The Need for Caution: The Dilemma of Politicisation
Srinivasan’s ambivalence stems from her starting point: the case of Elliot Rodger. If we take his manifesto at face value, then Rodger was sexually excluded. He was not seen as sexually desirable because he was shy, nerdy, introverted (or whatever it was that he perceived to be the cause of his sexual exclusion). Is his sexual exclusion also to be couched in political terms? Should we see him as a victim of a grave injustice in how our sexual psyches get shaped and conditioned?

There is a serious danger here. Srinivasan doesn’t want us to shame people for their sexual desires, or to deny them any possibility of authentic sexual agency (like the radicals seemed to do). But she also wants us to ‘repoliticise’ sexual desire so that we are more cognisant of the systematic sexual exclusion of certain groups of people. In making the case for the repoliticisation of sexual desire she focuses in particular on the sexual exclusion of disabled people, black/brown women, Asian men, transgender women and so on — all people who belong to one or more recognised minority class. If you are generally progressive and left-leaning in your political beliefs, you’ll find it relatively easy to be sympathetic to their situation. Speaking from my own perspective, I’ve read several pieces on sexual inclusion and disability — in particular the work of Tracy de Boer — and I find it easy to be swayed by her reasoning into thinking that there is an injustice that arises from the sexual exclusion of disabled persons, and that this injustice ought to be addressed.

But this logic then cuts both ways. If sexual exclusion is an injustice, then surely the sexual exclusion of nerdy, shy white dudes is also an injustice? And once we start seeing it in those terms, don’t we head down a very slippery slope? Don’t we start thinking about sex as a 'good' that needs to be fairly distributed among all people? And don’t we then start seeing it as a good to which everyone has some entitlement?

…there is a risk too that repoliticising desire will encourage a discourse of sexual entitlement. Talk of people who are unjustly sexually marginalised or excluded can pave the way to the thought that these people have a right to sex, a right that is being violated by those who refuse to have sex with them. That view is galling: no one is under an obligation to have sex with anyone else. 
(Srinivasan 2018)

To avoid that galling view, Srinivasan embarks on an analysis of how sex may or may not be similar to other distributive goods, arguing that we should resist analogies with things like food or education, and in the end, that ‘for better or worse, we must find a way to take sex on its own terms’. But this isn’t a particularly satisfying conclusion since it doesn’t provide a resolution to the problems raised by Srinivasan’s analysis. It doesn’t tell us how we should proceed with the repoliticisation of sexual desire, if at all. In essence, she diagnoses the problem without resolving it:

The question, then, is how to dwell in the ambivalent place where we acknowledge that no one is obligated to desire anyone else, that no one has a right to be desired, but also that who is desired and who isn’t is a political question, a question usually answered by more general patterns of domination and exclusion. 
(Srinivasan 2018)

This gives us what I will call ’Srinivasan’s Dilemma’, which I think is the real conclusion/message of her essay.

Srinivasan’s Dilemma: If we depoliticise sexual desire (and thus accept sexual desire as axiomatic/pre-political), then we risk ignoring/legitimising/perpetuating, systematic forms of discrimination in the manifestation of sexual desire. But if we politicise sexual desire, and recognise its role in systematic forms of discrimination, we risk legitimising a discourse of sexual entitlement and rights.

The real challenge then is whether there is some way to slide between the horns of this dilemma, or whether we have to impale ourselves on one of them (perhaps an unfortunate metaphor in the circumstances).

4. Concluding Thoughts
That’s my take on Srinivasan’s argument. Perhaps I have misinterpreted it; perhaps I have overlooked some crucial detail that would give me a different perspective. I’m happy to be corrected. I want to close, however, with three comments on the reconstruction I have provided.

The first comment is purely interpretive. I see Srinivasan’s article as fitting into a particular sub-genre of feminist analyses of sexual ethics. Specifically, I see it as fitting into a sub-genre whose paradigmatic instance is Ann Cahill’s essay on ‘Unjust sex vs Rape’, which I covered last year. The main claim defended in this sub-genre is that ‘consent’ is not the sole criterion of sexual permissibility and that we need to be more nuanced in our ethical assessment of sexual relations. In particular, we need to recognise forms of sexual activity that are ‘unjust’ without amounting to ‘rape’ or ‘sexual assault’. Part of the mistake of the radicalist school — at least in the eyes of its critics — was a tendency to conflate these two things, leading them to be ridiculed for holding the view that ‘all sex is rape’ (which, as I understand it, is not something they ever actually stated). The Cahill-approach is to avoid such extremism while still being sensitive to ethical shortcomings in our sexual practices. Cahill herself does this by focusing on ways in which patriarchal conditioning hijacks women’s sexual agency; Srinivasan does it by focusing on how discriminatory ideologies hijack everybody’s sexual agency.

The second comment is also interpretive but has more substance to it. It has to do with Srinivasan’s claim that sexual desire is ‘political’. This is central to her entire analysis, but it is potentially problematic. This is because there are two possible interpretations of that claim. The first interpretation is that it is a claim about the origins/explanations of sexual desire. It is saying that if our sexual desires are the product of political conditioning then they are not (or at least not solely) the product of something else like evolutionary programming or autonomous self-authorship. The second interpretation is that it is a claim about how we should ethically assess sexual desires, i.e. that we shouldn’t give them a ‘free pass’ but should instead assess them according to the criteria we use to assess all political institutions and facts. As you might have noticed, I favoured the second interpretation above. Let me explain why.

I admit that it is tempting to favour the first interpretation. Doing so would make Srinivasan’s argument slot into familiar territory (people love to debate the origins of sexual preferences after all). It would also be consistent with some of the things she says (and may, for all I know, actually be her view). But I think favouring this interpretation would be a mistake. Although many people think that the origins of a sexual preference makes an ethical difference, I think this is unfortunate because it can get you into endless and unproductive debates about what best explains a sexual preference. What often matters most is not where a sexual preference comes from but whether it expresses itself in a problematic way. If our sexual desires are unjustly discriminatory, then they are unjustly discriminatory and we have reason (though maybe not an overwhelming or decisive reason) to change them. It doesn’t matter whether this is the result of evolutionary programming or social conditioning or autonomous self-authorship.

A controversial analogy might help. Some adults have a consistent sexual preference for children. There are many attempts to explain why this is the case. Maybe it is due to genetic factors, maybe it is due to early environmental/developmental conditioning, maybe it is something that people acquire, gradually, over their adolescence, maybe it is due to some combination of all these factors. It would be nice to know, but, importantly, it wouldn’t change our ethical assessment of those sexual preferences if we did know where they came from: we would still view them as ethically problematic and would still have reason to change them or block their expression. The same reasoning applies to discriminatory sexual preferences, at least when they have systematic and oppressive effects.

You might disagree with this. You might think the origins do make an ethical difference. I'll admit that they might, in two circumstances neither of which is really relevant here. First, they could make a difference if what we are interested in is assigning punishment or blame for the expression of that desire. If we think the desire was not chosen or endorsed by the individual who expressed it we might be less inclined to punish or blame them. But this seems irrelevant to Srinivasan’s argument. It doesn’t seem to me that she is interested in assigning punishment or blame for our discriminatory sexual preferences. They could also make a difference when it comes to assessing how easy it is to reprogramme or change our sexual desires. If our desires are hardwired into us through genetic programming, then the assumption often is that they are less malleable than if they are socially conditioned. This genetic fixity could undermine Srinivasan’s argument. For example, a proponent of evolutionary psychology could take issue with her claim about the ‘political’ nature of sexual desire by arguing that there are difficult-to-change evolutionary constraints on what people find sexually desirable. We cannot easily reprogramme or change those sexual desires — not without causing significant hardship or creating an authoritarian system of sexual reprogramming, and even then it might be doomed to failure.

But, of course, this isn’t the right way to think about it. Just because something is genetic or evolutionarily programmed doesn’t mean that it is hard to change. For example, there is a strong evolutionary connection between sex and procreation but technology allows us to easily de-link sex from procreation if we so desire. Likewise, just because something is socially conditioned doesn’t mean that it is easy to change. For example, there are patterns of wealth inequality that are largely due to legal and institutional facts such as private property and capitalistic modes of production. Nevertheless, these patterns are very hard to change given the political and economic interests at stake. Ease of change is ethically significant, but it is its own thing and needs to be considered on its own terms.

This brings me to the third and final point. This one is purely substantive and has to do with Srinivasan’s dilemma itself. Although I accept that there is something to this dilemma, I do wonder whether she overstates the slipperiness of the slope between recognising the injustice of certain forms of sexual exclusion and fostering a discourse of sexual entitlement. It seems to me that there are things we can do to foster greater sexual inclusion without endorsing a ‘right’ to sex or encouraging the belief that some people have a claim to the sexual use of another person’s body (which is certainly galling and unpalatable). Some such possibilities have been discussed in the literature on disability and sexual inclusion. Advocates and activists in that area argue that basic consciousness-raising around the problem of exclusion and the tendency to see certain people as ‘asexual’ or ‘nonsexual’ can go some way to addressing the problem, as can greater tolerance (and less stigmatisation/legal prohibition) of certain kinds of sexual practice. Technological aids to sexual gratification can also be considered. They might allow for a meaningful kind of sexual inclusion without creating demands over other people (though, to be fair, all of this depends on what exactly people are being excluded from).

Related to this, I don’t see why we couldn’t say that some people have a greater/stronger claim to sexual inclusion than others. People who suffer from multiple forms of discrimination and oppression would obviously seem to warrant greater sympathy when it comes to sexual exclusion than people who are otherwise privileged. Furthermore, I also don’t see why we couldn’t say that some people’s sexual desires have no justifiable claim to gratification because they are harmful. So, for example, someone like Elliot Rodger who harbours violent misogynistic sexual fantasies doesn’t have any ‘right’ to be sexually included because his sexual desires are themselves morally problematic. Indeed, they are part of the problem -- part of the discriminatory ideology -- that we would like to correct.

Still, let me conclude by cautioning that this is a highly contentious and charged issue. I think Srinivasan is right to say that we need to occupy a somewhat ambivalent space.

* If the sex is abusive, coercive and/or non-consensual then obviously this changes the assessment.

Tuesday, January 1, 2019

The Wisdom of Regret and the Fallacy of Regret Minimisation

The Stages of Life by Caspar David Friedrich

I regret many things.

I regret throwing a tantrum on my tenth birthday when I didn’t get the present I wanted. I regret not taking part in those science and math quizzes when I was in school simply because I was annoyed that the teacher had put my name down without asking me first. I regret not thinking about what I wanted to study in college a bit more. I regret not taking that PhD scholarship in London when I was offered it. I regret my hesitation about getting married, and the effect it had on my partner. I regret making my sister cry one Christmas when I pushed my sarcastic sense of humour too far. I regret not applying for that grant when I had a chance. I regret the many cruel things I have said to family and friends over the years — they seem to flow out of me for no apparent reason. I regret not taking the more prestigious job that I was offered just because I worried about the stress and pressures it may have placed on my personal life. I regret not spending more time with my sister before she died.

My regrets are an important part of my life. I don’t think about them all the time, but they occasionally bubble to the surface, and when they do I like to wallow in them for awhile and reflect upon the ‘roads not taken’. Thinking about what might have been gives me a new perspective on what is.

I was recently criticised for taking this approach to regret. A close friend of mine told me that they never think about their regrets. They find regret to be a useless self-indulgence. They argued, reasonably enough, that you cannot change what you have already done. The future is the only thing under your control. So wallowing in regret just makes you feel unnecessarily worse about the fixity of the past, and can also act as an impediment to changing the future for the better. When pushed, this friend eventually conceded that prospective regret (i.e. thinking about what you might regret if you make a particular choice) might have some value. In this regard, they were in good company. I told them that Jeff Bezos — currently the richest man in the world — has long advocated ‘regret minimisation’ as a useful decision-making heuristic.

But I think both my friend and Jeff Bezos are wrong. Regret is not useless. It is an essential part of the well-lived life and it is important that we appreciate and reflect upon our regrets. Furthermore, the advice that we should ‘minimise regrets’ when making decisions is, oftentimes, practically useless, particularly when it comes to making those momentous choices that end up defining our lives. That said, there is certainly a danger of wallowing unnecessarily in vague existential regrets (and I am guilty of this from time to time) but the solution to that is not to dismiss and ignore our regrets. The solution is to reflect upon our regrets more deeply.

That, at any rate, is what I shall be arguing in this article. I’ll divide my discussion into four main parts. First, I will offer a more precise definition of regret and identify the different forms it can take. Second, I will concede that regret is often unjustified, but argue that this doesn’t mean we shouldn’t think about it more often. Third, I will critically evaluate evidence from psychology that suggests that we can adopt the ‘regret minimisation’ heuristic when making decisions. And fourth, I’ll draw together the strands of the preceding discussion and argue that thinking about regret is an integral part of the well-lived life.

1. What is regret anyway?
Regret is such a common experience that you may think it doesn’t need a definition. We all know what it is anyway. But there are different forms of regret and there are some closely related experiences that overlap with, and are sometimes conflated with, regret. It’s consequently important to have a clear sense of what it actually is. Four major clarifications are in order.

Regret is, first and foremost, a negative comparative concept. It stems from comparing options that arose in some original decision-making context. Suppose you are faced with a decision — A or B? — and you must make a choice. You choose B. Retrospective regret arises when you look back at the choice you made. You compare the life you actually lived (as a result of choosing B) with the life you could have lived (if you had chosen A). You retrospectively regret the choice if you think the live you have lived is worse than the life you could have lived. Anticipative/prospective regret arises when you look forward to the possible futures associated with the choices you face. You compare those possible futures and wonder ‘if I chose A, would I regret not choosing B?’ (or vice versa).

Regret can be associated with trivial decisions or momentous decisions (or anything in between those two extremes). The differences are important. Regretting what you ordered at a restaurant is very different from regretting your decision to decline a big job offer. The former is soon forgotten and repaired; the latter has knock-on implications for nearly all aspects of your life. Regret can also be associated with different timescales. Retrospectively regretting what you did last night is different from regretting what you did 30 years ago. Likewise, anticipating the regret you might feel tomorrow morning if you have another drink is different from regretting how you might feel in 30 years time if you break up with your long-term partner. In addition to this, although philosophers and psychologists often think about regret in relation to binary choices, the reality is that regret can be associated with decisions involving a multiplicity of options, as well as from long sequences of decisions, not just single decision-making moments. The significance of these differences is discussed in more detail below.

Regret is usually understood to be an emotion: you feel bad when you compare options. But regret also has a cognitive component.* You don’t just feel bad about how things have (or might) turn out; you judge one option to be superior to another. I like to think of this in terms of a distinction between first-order and second-order mental states. First-order mental states are immediate, unreflective, largely involuntary reactions to the world (or drivers of behaviour). Second-order mental states are more deliberate, reflective and voluntary assessments of first-order mental states. Regret usually starts off as a first-order mental state: an immediate, unreflective negative reaction to what has been and what might have been. But it then either ends up as a second-order mental state — i.e. a reflective endorsement of the first order feeling — or gets rejected as irrational, unjustified, or unhelpful.

Finally, regret should be, though often is not, distinguished from other closely related mental states. Regret, for example, should be distinguished from remorse. Regret is self-focused and often amoral (or immoral): it is a negative assessment of yourself for the choices you have made regardless of their moral character. Remorse is other-focused and moral: it is a negative assessment of yourself for the choices you have made because of how they have affected others. The precise relationship between regret and remorse is complex. They are definitely closely related: many of the ‘regrets’ I listed in the introduction also provide grounds for remorse. I suspect the best way to think about it is to see remorse as a special, moralised, sub-species regret. Similarly, regret should be distinguished from general sadness or lamentation about the state of the world. Following Christopher Cowley I think it is best to see regret as something that arises from the counterfactual comparison of decisions that we have voluntarily made, not as something that arises from the counterfactual comparison of how history has unfolded. Thus, while I lament the election of Donald Trump as US President, I don’t regret it because it wasn’t my choice. This distinction between regret and lamentation is not always made in the literature. Philosophers, following the work of Bernard Williams, often refer to ‘agent regret’ as a special sub-type of regret that arises from choices we have made, yet still hold out the possibility of regretting things beyond our control. I prefer not to do this.

That’s enough by way of initial clarification. Let’s now consider three arguments about the value of regret.

2. First Argument: Regret is often irrational
The first argument is somewhat unusual, given the set-up in the introduction. It concedes that regret can often be irrational. Many times when we feel profound regret about the choices we have made, we are wrong to do so. Or, to put it in more philosophical terms, many times we should reject our first-order feelings of regret. They are often misleading and unhelpful. The philosopher Paddy McQueen explains why this is the case with something he calls the ‘justified decision perspective’ (JDP) on regret.

McQueen argues that a decision is justified if two conditions are met: (a) it flows from or is grounded in who were are (or were) at the time of the decision and (b) it is based on reasons that are (or were) available to us at that time. If I choose to study law at university because I have a long-standing interest in criminal justice, am committed to the right to a fair trial, and want to dedicate my life to protecting that right, then my decision is justified, given my values and preferences at that time. If I choose to study medicine, despite my passion for justice, and because my father expects it of me, my decision is not justified.

McQueen argues that if a decision is justified at the time that it is made then it should not be regretted. This is true even if the decision turns out badly. Why is this? There are a few reasons. The first is that if your decisions are grounded in who you are and in the information available to you at the time, then, in an important sense, there is no better decision you could have made. You did everything right within your epistemic limits. The second reason is that the badness of any outcome may be (and perhaps often is) a matter of luck or chance. For example, my career in the law might go badly because I don’t get offered the one case that makes my career, or because the government cut back payments to the sector, or because of some general downturn in the economy. These things are beyond my control and could not have been known at the time of the decision: I should not blame myself for them. The third reason is that if we over-ascribe regret to our justified decisions we may learn the wrong lesson from our past failures. If a bad outcome was down to luck, and had nothing to do with the justifiability of our decisions, there is a danger that we avoid making the same kind of decision again, even when it might be the right decision and might turn out well the next time around. In other words, there is a danger that we become irrationally averse to a certain kind of decision just because we now feel badly about the outcome. Professional poker players have a name for this kind of decision-making error. They call it resulting, i.e. overemphasising results in figuring out the best strategy. Finally, and in line with what my friend said to me, if we over-ascribe regret to justified decisions, we just end up making ourselves unnecessarily miserable, i.e. assuming that our lives are worse than they actually are (or were) just because of how things turned out. People who regret getting into a relationship because it ends badly often make this mistake: they ignore all the good stuff that happened before the ending.

Granting that justified decisions should not be regretted, the key move in McQueen’s argument is the claim that regret is often unjustified. This is particularly true, he argues, for momentous decisions and decisions that involve long projections backward or forward in time. It is easiest to understand this argument by considering an example. I mentioned earlier that I often regret my decision not to take a PhD scholarship at a university in London. This was a decision I made just over 11 years ago. I regret it because it was a generous financial offer and it would have been at a good university — much better than the university at which I got my PhD. I sometimes wonder what life might have been like if I took up that offer. I imagine that I would be in a much better place, career-wise, than I am now. The reputational advantage of being at a more prestigious university, combined with the social networks and other opportunities I would have been afforded, could have allowed me to build much more early career capital. I’d surely be a professor at a top tier university by now! On the other hand (and there is usually an ‘other hand’), I did receive an even more generous scholarship at an Irish university, and I haven’t done badly for myself. Furthermore, staying in Ireland helped me to stay together with my long-term partner (now spouse) and at the time I didn’t think we would survive if I moved to the UK (though I tend to discount this more now because we did, ultimately, survive my spending three years in the UK).

Is my regret rational? Probably not. Consider the imaginative journey I have to undertake to properly evaluate this regret. I must project myself back in time to my past self (my self of mid-2007), imagine his values and beliefs at the time, and try to strip away what I now believe and value. Then, as if that wasn’t enough, I also have to project myself into a hypothetical present self (my ‘counterfactual self’) and imagine what his life would be like right now. I’m likely to commit three epistemic errors on this imaginative journey. First, I’m unlikely to be able to disentangle my present beliefs and values from my past beliefs and values. My present beliefs and values are undoubtedly going to colour my interpretation of my past decisions. Indeed, you can see me doing this above: I explicitly refer to things I now know about the resiliency of my relationship to long-distance separation that I did not know at the time. Second, the values and beliefs of my past self are probably unknowable and unrecoverable. Unless I wrote down exactly what I thought and believed at the time (and I didn’t) there is just too much epistemic distance between who am I now and who I once was for me to fairly recapture what the decision was like for me at that time. I’ll just end up fabricating a past identity. Third, it is going to be very hard for me to know how my counterfactual self would feel about his predicament. The decision to go to London would probably have been transformative. It would have led to many other changes in my character and outlook. So my counterfactual self would be very different from how I am now. He may well be sitting at his laptop writing an article about how much he regrets going to London, and how he wishes he hadn’t sacrificed his relationship at the altar of his career.

This is to talk purely of retrospective regret. But similar epistemic errors arise in the case of anticipative/prospective regret. It’s just that the imaginative journey is slightly different. In the case of anticipative regret, you don’t have to go back in time, but you do have to project your current self into at least two hypothetical futures and imagine what you would believe and value in those hypothetical futures. This is going to be highly speculative and dubious undertaking and will become more speculative and dubious the further forward in time you go. As I recall, Jeff Bezos likes to imagine what his 80-year-old future self would say about his present choices when trying to ‘minimise regret’. I suspect that in doing this Jeff Bezos is engaging in wishful and biased speculation. If he is anything like me (and to be fair he probably isn’t) he will just use this imaginative exercise as an excuse to project certain presently-preferred values into a fictional future. He’d be better off just focusing on what he currently values and knows.

This is not to say that regret is never justified. As McQueen makes clear in his article, there are times when our decisions are not justified and we know they are not justified given our beliefs and values at the time of the decision. Sometimes we are just not honest with ourselves about what we really care about. We need to be on the look out for those moments. It’s also more likely that our assessment is justified when it involves a ‘short’ imaginative journey into the past or future: then we will actually be able to ground our choices in plausible beliefs about what our past and future selves will value. So regret is sometimes rational and appropriate. We just have to be wary of the epistemic errors we commonly commit when thinking about it. Many times when we feel regretful we should reject and reconsider that feeling.

As I say, I’m willing to concede a lot of ground to this argument. But I think there are some problems with it. First, it imposes quite a high epistemic standard on the assessment of momentous decision-making. While McQueen is undoubtedly correct that there is significant uncertainty involved in comparing the outcomes of different decisions, particularly over long periods of time, we are often called upon to do this. Life throws numerous, momentous choices our way and we have to imagine how things might turn out, to at least some extent. There are better and worse ways of doing this — we can call upon the examples of others for guidance, and analyse data from scientific studies of such choices — but we shouldn’t be afraid of doing this (or of the need to sometimes experiment with our own lives). If we take the JDP too seriously, there is a danger that we think all such efforts are futile.

Second, and perhaps more importantly, we need to remember that McQueen’s argument is only about the second-order assessment of regret. We still have the first order feeling of regret with which to contend. Regret will often float into our minds unbidden and make us feel bad about choices we have made (or are about to make). Granted that these feelings are often irrational, we might like to develop strategies for reducing their presence in our lives. This is where we might like to follow Jeff Bezos’s guidance and minimise the first-order feeling of regret.

Let’s consider this idea in more detail by turning to the second argument.

2. Second Argument: There are ways to minimise regret
The second argument comes in two parts, both relatively simple and straightforward. First, it claims that we can, if we wish, implement certain strategies that help to minimise the feeling of regret. Second, it claims that, even though this is possible, we shouldn’t always strive to minimise regret. I’ve already partly defended this second aspect of the argument in the preceding section by highlighting how speculative and fanciful ‘regret minimisation’ reasoning often is, but I’ll say something more about it below.

Let's start with the claim that there are plausible strategies for regret minimisation. There have been a lot of psychological studies on regret over the years. And I mean a lot. Can we learn anything from these studies? One fairly robust finding, originally set out in a widely-cited paper from Gilovich et al (1995), and since developed in other studies, is that people regret inaction more than action. In other words, if you are faced with a choice between maintaining the status quo (e.g. staying with your current job) and doing something different/new (e.g. taking a new job) you are more likely to regret maintaining the status quo. So one thing you can do if you want to minimise the feeling of regret is to have a bias toward action.

Why do we regret inaction more than action? In a more recent paper, Davidai and Gilovich (2018) argue that there are four reasons as to why inaction is regretted more often than action. The first reason is that it is easier to ‘undo’ actions, i.e. if an action turns out badly you can quickly take steps to reverse its negative effects. The status quo, on the other hand, is harder to undo because there is an accumulated tangle of factors underlying it. The second reason is that it is easier to rationalise actions, i.e. justify them to yourself. Indeed, we humans are very good at inventing explanations for our behaviour; it’s a hardwired psychological trick we perform on ourselves all the time. The third reason is that it is often hard to retrospectively justify inaction. At the time, inaction is usually justified out of a fear of failure or change. This fear might be compelling and visceral in the moment, but we are often much more confident about our capacity to avoid failure when looking back. Finally, we have a tendency to remember unrealised goals more often than realised ones. Once we achieve a desired goal we quickly tire of it and move on to something new. But unrealised goals linger in the memory (‘I coulda been a contender!’). Inaction is usually associated with unrealised goals and is thus a more potent source of regret.

For what it’s worth, I tend to agree with this. I tend to regret inaction more than action in my own life. As mentioned earlier, I often regret not taking up a more prestigious job offer because of my fears about the stress and damage to my personal life that it might entail; but I don’t regret moving to England to take up a new job many years ago, even though that also involved a lot of personal stress and potential damage to my personal life. I see the mechanisms suggested by Davidai and Gilovich at play in these asymmetric attitudes. I can easily see the silver lining in my decision to take the job in England, and I can quickly discount and rationalise the costs associated with it. I can’t do the same with the decision to reject the more prestigious job offer. And yet, when I think rationally and reflectively about what has happened to my life since that rejection, I can’t honestly say that it has been all that bad. On the contrary, it’s been reasonably good. It’s just that it has been more of the same good as I had before. The tantalising possibility that things mightn’t have been all that bad if I took the offer lingers in my mind. It’s hard to shake, no matter how irrational it might be.

Are there any other lessons we can take from the psychology of regret? Maybe. In the same recent paper, Davidai and Gilovich (2018) seem to hit upon a new heuristic for minimising regret. In six experiments, involving mixed cohorts of subjects, they investigated the possibility that our strongest regrets stem from perceived gaps between who we are right now, and who we would like to be and who we think we ought to be. Using work originally done by Edward Higgins, they suggest that we each live with three versions of ourselves: (i) our actual self, i.e. who we really are, or more accurately, who we perceive ourselves to be at the present moment; (ii) our ideal self, i.e. the self we would like to be based on our hopes, goals and aspirations; and (iii) our ‘ought’ self, i.e. the self we think we should be based on moral norms or the expectations of others. Davidai and Gilovich hypothesised that people would experience regret based on perceived discrepancies between their actual selves and their ideal/ought selves, and that discrepancies between the actual self and the ideal self would be stronger sources of regret than discrepancies between the actual self and the ought self. This hypothesis was consistent with previous findings from Bonnie Ware in her interviews with patients in palliative care, and was strongly confirmed by their experiments. Across four of their experiments, approximately 3/4 of people had more regrets about actual-ideal gaps than they did about actual-ought gaps, and they also experienced these regrets more intensely.

Davidai and Gilovich tried to figure out why this might be the case. They hypothesised that failures to live up to the ‘ought’ version of yourself inspired more immediate and extensive psychological coping mechanisms. When you breach some perceived moral norm, you feel guilt and anxiety. These are ‘hot’ emotions and they have to be managed immediately in order to reduce psychological anguish. When you fail to live up to some ideal version of yourself, you feel disappointment and dejection. These are 'colder' emotions and so don’t inspire immediate coping mechanisms. Furthermore, the feeling of disappointment tends to build gradually over time and only reaches crisis point relatively late on, when it is harder to pinpoint the mistake and try to repair it. Again, in their final two experiments, Davidai and Gilovich found that this suggested mechanism seemed to explain the discrepancy in feelings of regret.

All of which could be taken to endorse the following tactic for minimising regret: focus predominantly on minimising the gap between your actual self and your ideal self (since you are more likely to ignore this and leave it fester), and not so much on minimising the gap between your actual self and your ought self (since you’ll more naturally deal with that problem when it arises).

There could be some wisdom to this, but I think there are some important caveats. First, as Davidai and Gilovich themselves point out, their studies are relatively preliminary and limited, and may not have been conducted on diverse enough populations (e.g. they didn’t have many older people in their sample who may have learned better how to cope with actual-ideal discrepancies). Given recent controversies about the lack of replicability in psychological science, more work needs to be done to confirm the value of this heuristic.

But even if it a robust finding, I think we should be cautious about trying to minimise regret in this way. One reason for this — which I explored in more detail in an earlier article on hypocrisy — is that ultimately discrepancies between all three versions of yourself may need to be resolved. Indeed, in some ways, gaps between the ideal version of yourself and the ought version of yourself are the most common and the most psychologically disturbing. Making a success of yourself often comes at the expense moral virtue: you frequently have to overlook your duties to others and favour personal ambition. This can lead to a very dissatisfying existence. Speaking from my own perspective, I find this to be the greatest psychological challenge I face. It is also the most intense source of my regrets. It’s all those times that I have favoured personal ambition and success over my duties to others that I feel the worst about. Furthermore, and contrary to Davidai and Gilovich’s suggestion, I don’t find it particularly easy to psychologically manage the ‘hot’ emotions associated with such regrets. My regrets about how I have treated family and friends are the ones that keep me up and night, and don’t go away no matter how much I do to repair the damage I have done. This may just be a personal idiosyncrasy, forged by my intellectual interest in morality, but it’s phenomenologically real to me.

More important than my personal idiosyncrasies, however, is the problem that following this regret minimisation strategy, and indeed the earlier one about inaction, assumes knowledge that we often don’t have. For example, it assumes that we have a good sense of who our ‘ideal’ and ‘ought’ (and, for that matter, ‘actual’) selves are, and ignores the possibility that we have multiple, conflicting, ideal (and ‘ought’) versions of ourselves. It’s this conflict of ideals and duties that, in my experience, is more challenging and more strongly associated with feelings of regret. There are many things I would like to be; there are many duties and virtues I think I should live up to. It’s the fact that I have to choose between them that is the source of most regret.

In addition to this, the regret minimisation heuristic ignores the fact that regret itself is a source of knowledge. If we spent our lives trying to minimise it, we wouldn’t learn and grow as people. In the end, this is the great danger of regret minimisation. It encourages wild speculation about our possible futures, and then breeds complacency and stubbornness when it comes to evaluating the consequences of our choices. If we follow the heuristic, we will probably do an initial cost/benefit analysis of our choices: “I’m more likely to regret option A when I’m 80 years old, so let’s go with option B”. Then, when B doesn’t turn out so well, we’ll more than likely find ourselves pre-committed to rationalising it. We’ll say to ourselves ‘I already determined that A would induce more future regret than B, so I don’t need to reevaluate my choice. I just need to ride it out.’ Trapped in this mindset, we become so afraid of regret that we suppress it; we don’t see it as a source of knowledge that can be used to our advantage.

4. Third Argument: We should embrace regret
This brings me to my final argument. This one draws together the threads that have emerged from the preceding analysis. The argument maintains that regret is not something to be avoided, suppressed or minimised. It is something to be embraced and rationally reflected upon. It is an essential part of life. Three factors bear this out.

First, and following on from what I just said in the previous section, we need to embrace the first-order feeling of regret as a source of knowledge about ourselves. We need to accept that we often don’t know what it is that we really want out of life. We have conflicting visions of ourselves and are notoriously bad at predicting how we will feel about our choices (a problem that psychologists call ‘affective forecasting’). The feeling of regret can reveal what really matters to us.

I’m not the first to make this point. In a nice article about this phenomenon, Justin White argues that regret is often ‘self-revelatory’ and that the information gleaned from regret can used to change our course in life:

We are potentially unique as creatures because the way we think of ourselves can feed back into the way we live. Regret can reveal things about us to ourselves in ways that can then impact how we engage with the world… regret can help us become aware of ourselves, including alerting us to blind spots, to our characteristics we are not fully aware of or to the things we desire or care about without being fully conscious of it…Revelatory regret, then, opens the possibility for a certain kind of self-directed change. 
(White 2017, 238)

You don’t get that advantage if you ignore regret or see regret as a useless emotion. You have to run the risk of regret to learn about yourself.

Second, we need to appreciate that we definitely cannot eliminate regret from our lives, and that sometimes it is impossible to even minimise regret. This is not just because we engage in epistemically unwarranted speculation when trying to do this, though that is a major problem. It is because sometimes we have to choose between different ideals and values, knowing that we will have some regrets no matter how things turn out. Go back to my decision not to take up the more prestigious job offer. I made that choice knowing that there was a strong likelihood that I would regret the outcome, no matter what it was. I decided to stay with my current job because I thought it would be better for my personal psychological well-being and my family life. So far, I have every reason to think that this was correct. Nevertheless, I did this knowing that I would probably regret not taking the chance of the more prestigious job: it was always going to be a tantalising 'what might have been'. But at the same time, I knew that if I took it, I would also more than likely regret the chance of having forgone a more peaceful and contentful personal and family life. There was a ‘tragedy’ to the choice. I had to choose between different visions of the person I think I am (and would like to be). Embracing the inevitability of prospective regret was part and parcel of this. This highlights yet another important function of regret in our lives: it forms our self-identity; it doesn’t just reveal it.

Third, and finally, none of this should be taken to imply that regret is always valuable. Regret can be very destructive. One major reason for this is that regret is always overdetermined. For every choice we make, there are probably dozens of choices we could have made, and thus dozens of possible futures we had to forgo. Some of these futures might have been very good. As these accumulate over the course of a lifetime, it’s very easy to get overwhelmed by what Cowley calls a vague ‘global or nihilistic regret’, i.e. a sense that things could have turned out so much better but you’re not exactly sure how. Cowley argues that we should do our best to avoid that kind of thinking. It doesn't give us any actionable information, it undermines the good things that actually did happen to us, and it just makes us feel miserable.

Cowley suggests that this is the real lesson to be learned from Tolstoy’s famous short story Ivan Ilych. The main character in that story is a successful lawyer and judge, who focuses mainly on his career instead of his family. He develops a mysterious and ultimately fatal disease. While bedridden with this disease he reflects on the choices he has made, and becomes convinced he chose wrongly. The story is commonly interpreted as a cautionary tale about choosing careerism and ambition over family life, but Cowley thinks this is a problematic lesson to draw from it. Ilych’s regrets are vague and non-specific. He’s not sure exactly what he should have done differently. He simply has a strong, possibly unjustified, sense that he ended up in the wrong place. As Cowley puts it:

I find the short story problematic as a cautionary tale about the perils of conventional ambitious bourgeois egotism. Indeed, I take an opposing cautionary tale from it: one should beware of falling into a twilight where one comes to reject all one’s previous life; without any constructive thought for what one should have done differently; without any attempt to distinguish the healthy from the unhealthy values of the highly-placed people in society; without any appreciation for the independent goodness of one’s job performance and job enjoyment. 
(Cowley 2017, 632)

This is the trap I sometimes fall into. I think of all the ‘woulda, coulda, shouldas’ and get the sense that my life is much worse than it could have been. That’s when I wallow in a negative and unproductive form of self-pity. My friend (whose admonitions inspired this article) is right that this emotional stagnation should be avoided. But the best way to do this is not to ignore regret or to think about it less. After all, the feeling of regret often comes unbidden into one’s life. The best thing to do is think about regret more rationally and reflectively. To use something like McQueen’s ‘justified decision perspective’ to properly assess those feelings and learn the right lessons from your past.

This is why I think regret is an important part of the well-lived life.

*Yes, I know that some people think emotions are partly cognitive.