From Self-Interest to Morality: How Moral Progress Might Be Possible

One of the most stunning successes I have personally seen in my life is the emergence of the Effective Altruism movement. I remember when Will Crouch (now MacAskill) first presented 80 000 hours to our Graduate Discussion Group and Toby Ord was still a grad student. From their ideas a whole movement has emerged of brilliant young people galvanised into doing good. We are getting the brightest, best people of the current generation coming to Oxford to engage with the Centre for Effective Altruism. Almost every grad student I come across has some connection. Well done Will and Toby, and all those others who have contributed to establishing this movement

So I guess I should not have been surprised when during my visit to Harvard this week, a student contacted me from EA to give an ad hoc talk. I discovered there were cells all over the world and the movement had spread way beyond Oxford.

Anyway, I gave an impromptu talk and predictably there were many questions I could not answer satisfactorily. One the issues I covered was the need to create a new basic (or minimal) secular morality. This is necessary not only to decide what the goals of moral bioenhancement should be (my favourite current pet topic), but indeed how education should be revised and society ordered. Every society has a set of normative commitments. Ours are outdated, archaic and unfit for the challenges of a globalised, interconnected and technologically advanced world.

Since we were talking about Altruism, I said one part of a minimal morality which could be justified on either utilitarian or contractualist grounds is:

Duty of Easy Rescue: when the cost to you of some action is minimal, and the benefit (or prevention of harm) to another is great, you should perform that action.

If everyone agreed to do this, we would never be called upon to make great sacrifices but we might be the beneficiary of large benefits.

I also discussed the importance of dropping familiar and well accepted moral distinctions like the acts-omissions and intention-foresight distinctions. It is a part of folk morality and some deontological codes (like Christianity) that there is a difference between the effects we intend, and those we merely foresee as a consequence of our actions. And many people believe there is a deep distinction between the effects of our actions, and the foreseeable, avoidable effects of our omissions.

These distinctions have dramatically significant effects in the real world. For example, the acts-omissions distinction allows us to avoid taking responsibility for alleviating global poverty or inequality. And the intention-foresight distinction allows people to justify killing innocent civilians as “collateral damage.”

One student, Garrett Lam, very astutely pointed out that if we move beyond any minimal moral requirement (like “don’t kill an innocent person for no good reason”) then we will need to have commitments to some contestible moral framework.

One example is consequentialism. Consequentialists famously deny the acts-omissions and the intention-foresight distinction. Of course, if you are not a consequentialist, you need other grounds to reject these destructive distinctions.

So how can we make moral progress, and create some minimal morality, without commitment to a particular, contestible if not controversial moral theory or set of normative commitments.

Of course, every society has normative commitments – they are inescapable if we are to organise society in any way. And the status quo is pretty bad – we just have to do better than that.

But I suggested a heuristic for how to select what might candidates for components of a revised, minimal secular ethic.

Consider prudence or self-interest. Should you adhere to acts-omissions distinction in your own life? It would be disastrous to use this distinction if you wanted to have the best life. Consider the following two cases.

Eating. Jim is hungry so acts: he looks for a nice meal and eats. He is happy and lives a good life
Not eating. John is hungry but he omits to act. He does not eat and consequently dies of starvation.

It would be absurd to say in the second case: well, all John did was fail to act. He omitted to act so he is not responsible for his own death. Rather, we would say John killed himself. (I have suggested this is one way that patients with severe illnesses can now end their lives legally – voluntary palliated starvation)

Or consider a more realistic case:

Jenny has severe motor neuron disease and is nearly totally paralysed. She wants to die. So she acts. She takes a pill that kills her.

Petra has severe motor neuron disease and is nearly totally paralysed. She wants to die. She is given a drug to relieve her spasms. She has an anaphylactic reaction to the drug and will die unless she is given an adrenalin shot. The adrenalin is in front of her. She omits to take it and dies.

In both cases, Jenny and Petra wanted to die. In one case, it was the result of an act. In another the result of an omission. What matters from the perspective of self interest is not whether one act or omitted to act, but their suffering, their wishes, the effect on their family, whether there was a chance of cure, etc.

In our personal life, what matters is not whether some event is an act or an omission of ours, but the foreseeability of its effects, its costs and benefits, and other features (perhaps whether it is an obligation). But there mere fact it is an act or an omission is totally irrelevant.

Now morality is importantly different from self-interest. It requires some kind of impartiality and sacrifice of self-interest for others. What kind and in what circumstances is hugely contentious. But if acts and omissions are irrelevant to prudence and self-interest, why should they become morally relevant to morality? What matters is foreseeability, avoidability, costs-benefits, nature of obligations, etc

It would be equally crazy to employ an intention-foresight distinction in your own life if your aim was maximizing self-interest. Consider the following pair of examples.

Jack intends to squeeze a pimple to improve his appearance for a party but foresees it will cause a life threatening septicaemia.

This would be crazy and not at all how we reason. We would work out what the effects of our action are likely to be and balance whether the foreseeable benefits outweigh the foreseeable harms, and then intend the action.

Of course the doctrine of double effect requires that the intended good effects be commensurate with the unintended but merely foreseen bad effects. But this still won’t do in the case of self-interest.

Imagine Jane has breast cancer. She intends to cure her breast cancer. She has two options.
Radical mastectomy and chemotherapy. This has a 90% chance of curing her cancer with the foreseeable side effect of removing her breast.

Lumpectomy, chemotherapy and radiotherapy. This has an 80% chance of curing her with the foreseeable side effect of preserving her breast.

In war, the loss of several civilians to eliminate one strategic terrorist target is seen as acceptable “collateral damage.” If this is true, in this case the loss of a breast compared to the saving of a life must be acceptable “collateral damage”. The doctrine of double effect appears to support the permissibility of mastectomy.

But this not how Jane would reason. She would weigh the badness of the loss of a breast against and 10% increase chance of dying. The primary intention – to save life – does not figure dominantly as it does in DDE.

My claim is that if these distinctions – acts/omissions and intention/foresight – do not play a role in our personal lives, in our decisions to act prudently, why should they play a role in morality? To be sure, morality is different to prudence but why introduce pointless distinctions when it comes to morality. Morality is about impartiality, weighing people’s claims and interests, but not introducing specious distinctions.

Here is another way in which thinking about self-interest can assist thinking about morality. Another student asked me how a young person should select a career to do the most good. This is the focus of Will MacAskill’s charity, 80 000 hours which aims to ask how we can spend our working life (80 000) to do the most good.

But forget doing good for others for a moment. Imagine your sole goal is self-interest. You just want to make as much money as possible, or get as much fame? What is the best way to achieve that?

Well, you need to pick some kind of profession that is likely to make money or attract fame. But the choice is very wide. You should then work out what you are talented at or good at. Then decide what you are passionate about or love. That is the best way to succeed: doing what you are good at and love. Success comes as byproduct of that.

This is like the paradox of hedonism. To have a happy life, one should not aim directly at happiness, but rather at good relationships and friendships, achievements, development of talents, etc.

Now the only difference between choosing a career for self-interested reasons and for moral reasons is that the scope of careers open in the moral sphere must include some other-regarding considerations. But that is pretty easy to satisfy – most careers involve engagement with other people.

It is very difficult to predict what will bring about the most good. The inventors of the mobile phone might have been aiming to improve communication, or make a lot of money, but the end result has been a technology that has revolutionised life and finance in the developing world. It has brought about more good than just about any donation of money directly.

My father used to say to me frequently, everything you do, you should it with love. Perhaps that is way to do the most good overall.