There’s a battle in the field of ethics between three approaches—Consequentialism, Virtue Ethics and Deontology, but this framing is all wrong, because they’re all on the same side. By treating ethics as an adversarial all-or-nothing (zero-sum) debate, we are throwing out great deal of baby for the sake of very little bathwater.
First of all some (very basic) definitions.
- Consequentialism: holds that the morality of an action is determined by its outcomes (or more specifically its expected or intended outcomes) in terms of what we value. Utilitarianism, a prominent form of consequentialism, explicitly formulates this in terms of the increase in utility (happiness) and the avoidance of harm (suffering).
- Virtue Ethics: holds that the morality of an action is derived from the motivation for that action, is it virtuous or vicious? Moral virtues like courage, compassion, humility, and vices such as cowardice, cruelty and arrogance, have been codified by philosophers from all cultures back to Confucius and Aristotle and continue to be updated with virtues today, generally focused on greater inclusivity.
- Deontology: holds that the morality of an action is derived by its adherence to rules, rights or duties, whether religious or social.
The way I see it, there is no conflict between these three approaches, they simply represent different levels of focus, which are applicable to different sorts of situations. They are not only compatible, but also complimentary and interdependent.
How Do We Derive Virtues?
… and what makes something a vice rather than a virtue? Take compassion and cruelty—opposites on the virtuous spectrum. Would it make sense to swap these and make compassion a vice and cruelty a virtue? Is it arbitrary? There have been times in history where these may have been reversed, but why?
An Unfortunate Situation
Imagine a society where children are taught to be cruel, perhaps to hurt and hunt animals, to fight and hurt each other. Why would this happen?
Well, what if they needed to be prepared for a harsh world, full of actual threats coming from violent neighbouring tribes? In this society compassion might be a liability, trusting your neighbour might lead to the death of you and your family, your capacity for cruelty might be necessary in order to avoid conflict with others (via intimidation). This is an extreme example, but it shows that to reverse a norm, you need extreme opposing consequences .
Today
The example is not meant to suggest that cultures in the past were all like this, it’s just to illustrate that, given certain conditions, we can see how such social values could arise. It also isn’t meant to entirely exonerate modern societies from spreading some cruel messages to our progeny. But in today’s society, thankfully, compassionate behaviour is generally not a liability, and leads to much greater well-being for each other, and most of the time cruelty is also generally abhorred. Is it a coincidence then that cruelty in our society generates suffering and is abhorred and compassion creates greater well-being and is celebrated? I would suggest not. This is because the moral foundation of virtues is essentially consequentialist.
What we are doing here is actually using consequentialism to inform virtue ethics—saying “if people act with these virtues and avoid these vices, everyone will be better off”. This not only holds for us, but also our unfortunate hypothetical tribe. But if virtues are derived this way, why not simply default to consequentialism? Not so fast. For the next section I’ll use Utilitarianism as the Consequentialist position of choice.
Utilitarian Calculus
For the strident Utilitarian, analysis is complex and time-consuming, not always practical for informing actions in-the-moment. Virtues are essentially rules of thumb or general principles that can be quickly applied to situations. We can act confidently on virtues secure that we have a good (consequentialist) foundation for them, without having to run a complex calculus each time we face a new situation.
Another key quality of virtue ethics is that it makes our actions more predictable. If we were to run an individual utilitarian calculus on each situation, we might find ourselves making self-serving decisions in some situations and not in others, meaning we might not be able to be trusted to do the virtuous thing, if there isn’t a threat of consequences in a particular situation. For instance, feeling a pang of hunger as one leaves a store when the attendant has left, might lead us to swipe a chocolate bar on the way out, because the cost-benefit analysis in this particular situation might reveal that eating the chocolate bar would bring us far greater pleasure than the shop owner’s pain. Whereas a virtue of honesty would make this a no-brainer—don’t steal.
Does this make consequentialist ethics wrong?
Not at all, because we can understand that such “vice”-ious behaviour would lead to a low-trust environment over all, where we lose the benefits of a high-trust society. Consistency, integrity and trust are features of a healthy society that mean people don’t have to constantly be looking over their shoulder. From a utilitarian perspective a high-trust environment has far greater utility than a low-trust one, and so, once again we see that virtues are themselves justifiable using consequentialist logic.
The problem happens when a grumbling tummy in-the-moment or a fit of rage can curtail our ability to play out the entire utilitarian calculus. We are not moral robots after all, we are primates.
Evolutionary Psychology
This brings us to the question—why are we like this? Why don’t we act “right” automatically? We have evolution to thank for many positive qualities, from our intelligence and ability to communicate, our dexterity and ability to build things, and moral qualities like our capacity for empathy, courage and love. But for the million or so years that nature was determining what it was to be human, it was doing so in an entirely different environment than the one in which we now live (the one we have made for ourselves). Our modern environment caters to many of our needs for security, shelter, food and health—our evolutionary environment did not, it was full of many immediate threats to security, constant scarcity and where a scratch against the wrong branch could lead to a painful death. Like our hypothetical tribe who, out of necessity, raise their children to be cruel, there are many instincts we evolved that prepared us for a much more dangerous environment, and in turn make us much more dangerous than necessary.
Ethics Rules
If we had not had such a history, and were moral robots, we could easily just trust our programming to logic our way to pro-social behaviour, but for better or worse that is not our lot. Our neat little consequentialist-virtue ethics synthesis may help us understand what constitutes “good” behaviour, but it is not enough to guarantee we will actually act in accordance with it. So, we need (some) rules for when our instincts lead us to behaviour that is destructive to the way of life we share today, and that’s where Deontology comes in.
Deontology is in the business of building a list of the specific rules, duties and rights that are essential for order, where transgression is an absolute no-no. And we have a list of such rules, they’re called laws.
Now, we can see that the legal system is going to be dependent on consequentialist logic (generally) to derive these rules, after all disorder—that which laws seek to avoid, is a consequence of lawlessness, and disorder is assumably bad because it leads to a range of other negative consequences for individuals: getting murdered etc. So, could we do away with virtue ethics in favour of a much more specific consequentialist-deontologicalparadigm?
Well, once again we run into an issue of capacity, in the same way as it is impractical to run a full utilitarian calculus on a situation in-the-moment, it is practically impossible to list every possible transgression (no matter how minor), let alone monitor those transgressions. When I said that we need “some” rules, this is because rules are potentially infinite. They are also context specific—punching someone in the face in a boxing match, for example, is very different from doing the same thing to a stranger in the street. One you get paid for, the other gets you arrested. So, every rule needs further caveats and conditions to be taken into account, making complete ethical accounting unfeasible.
Laws are therefore necessarily limited in scope to crimes—actions that cannot be tolerated in a functioning society. It’s simply impractical (and over-kill) to apply this approach to anti-social behaviour in general, and then there’s good behaviour to which laws don’t have any relevance (unless you’re already in prison, and can reduce your sentence “based on good behaviour”). Virtue ethics applies to all good and bad behaviour, including behaviour that doesn’t reach the legal threshold.
So…
The way I see it, these three approaches to ethics are all necessary and interdependent. We need a consequentialist foundation like Utilitarianism to determine Virtues and Laws, we need Virtues to create a society that can act cohesively and with predictable outcomes, and we need Laws to guard against our anti-social evolutionary instincts. Meanwhile the practical benefits of virtues and laws in their respective domains fit into a larger consequentialist calculus, one that we can adjust based on a changing social environment. In short, debates about the merits of different ethical foundations are all well and good, but by looking at the issue from a non-zero-sum perspective, where the whole can be greater than the sum of its parts, we can appreciate a mutually beneficial synthesis that incorporates all three approaches.
Originally published at https://nonzerosum.games.
I object that we need laws to control antisocial instincts.
Antisocial instincts come in all forms, for example, cannibalism, child sexual abuse, or the public practice of torture. Although we have laws for these, such antisocial acts are so rare that in Western societies they fall under the purview of psychiatric pathology rather than criminal law. However, all these behaviors existed in human cultures of the past, and it wasn't criminal laws that made them disappear, but rather cultural change (moral evolution).
Furthermore, today we engage in such antisocial acts as spending money in fancy restaurants, voting for politicians who preach hate, or having a talented doctor make money performing cosmetic surgery. These types of behaviors mean that many unfortunate people die prematurely from hunger or disease in poor countries. And no law prohibits them.
So there is only one valid ethic: virtue ethics... as long as it is practiced in a morally evolved cultural environment where antisocial instincts are controlled by behavioral habits and not by penal laws. There is no doubt that cultural evolution (moral evolution) has meant that we no longer need laws to prevent the practice of cannibalism, nor to schedule spectacles of human torture (although in Spain some animals are publicly tortured), nor for many other horrible things that were once part of our customs.
What is the limit of moral evolution? How can we promote it as efficiently as possible to the point of eradicating a non-altruistic human culture in the future? That should be a question that should concern us more.
In any case, we will make no progress if we assume that we will always need laws to control antisocial instincts, when we know they can be culturally controlled.
I don't see that you have any support for this claim. Laws against cannibalism and child sexual abuse have undoubtedly made those practices less prevalent, that's how laws work. Laws are informed of course by moral evolution, and I advocate for updating rules based on a utilitarian calculus (unlike traditional deontology that gets these values out of a religious text or other such arbitrary source).
How do you determine what are virtues though? If there is not some consideration of consequences?
Do you think that we can culturally evolve to not need any behavioural guard-rails, given that we are not robots, but rather evolved primates whose instincts were shaped for a much more dangerous environment? I mean, it would be nice to believe that, but I don't see much evidence for it. Child abusers still exist in societies where cultural norms assume it to be wrong, child abusers largely know that what they are doing is wrong, and understand why, but are never the less compelled to do it—Don't you think it's a little naive to think that this can be changed simply by continued cultural evolution?
Thank you very much for agreeing to a bit of discussion about the possibilities of moral evolution as a guide for ethical evaluations.
Of course, I admit that, in today's world (whether in Western European society or in Pakistan), judges and legislators need the most precise ethical criteria possible to elucidate moral dilemmas and priorities of social interest. I have no academic training in ethics, but I recently came across the expression "functional morality," which somewhat sums up the situation we find ourselves in when evaluating events with moral content. I know that some consider that in today's society, the evolution of morals is dictated by jurisprudence.
There is no law prohibiting cannibalism in the United Kingdom (nor bullfighting... something we do need in Spain—I am Spanish). Yes, there are laws against child abuse (I don't know if they exist in Afghanistan), but generally speaking, we all accept that customs change first (moral evolution, among other things), and then these changes (logically initially driven by influential minorities) become entrenched in society... and also in the laws.
When I consider that only "virtue ethics" is valid, I mean that, since human nature is immutable, a certain conception of moral perfection is viable (through future cultural changes). Kant, with all his deontology, was unable to overcome his prejudices against women or "inferior races." In the 21st century, I watch the terrible scenes from Gaza on television at lunchtime, and although they disgust me, they don't take away my appetite... as would stumbling upon an Auto-da-Fé in my town square, similar to those performed in the 17th century.
I'm afraid what's wrong with you is that, like any scholar of the 19th or 18th century, you believe that moral evolution has stopped right at the customs of your time, and you lack the perspective to imagine that there will be a 23rd and a 24th century. And that is truly naive.
There are effectively laws protecting against cannibalism, as there are laws against murder, and against abuse or desecration of a corpse.
Where do you get this idea from? I explicitly stated in my comment:
I certainly don't think we've arrived at an end-point to moral development, and I also recognise that morals will change depending on what suits our civilisation (if we live in a system that effectively incentivises pro-social behaviour then this will hopefully enable increased freedoms) I'm just saying there's likely a limit in terms of our instincts, and addressing mental hardware on a biological level has its own ethical questions.
To be clear, I'm not a fan of deontology, and I don't think it's useful for deriving values, I think moral actions are largely a function of virtue ethics, but I think that we need some metric by which we measure virtues, and that lies in some form of consequentialism.
Because you wrote this
Do you think that we can culturally evolve to not need any behavioural guard-rails, given that we are not robots, but rather evolved primates whose instincts were shaped for a much more dangerous environment? I mean, it would be nice to believe that, but I don't see much evidence for it.
The idea that moral evolution might one day lead us to dispense with criminal law (as well as give rise to an economic system based on altruism) is no more improbable than its opposite. And there is nothing in human nature to prevent it, since we now have evidence that human behavior is extremely susceptible to cultural influences.
It is assumed, of course, that there must be limits to this, and above all, we currently do not know the cultural mechanisms that allow us to control aggression and promote altruism. It is also true that, as far as I know, no one is working on it...
What if we seriously reflect on it?
There are enormous differences between the aggressiveness of certain societies, and we do not always understand the factors that make them so different. To give a well-known example, the famous Amish (now numbering 400,000) lack criminal law, judges, police, fines, or prisons. They "govern" themselves by consensus, most likely because, through trial and error, they have developed a preventive control of aggression in individuals based on a "psychological priming" (behavioral patterns). Another well-known phenomenon is that of homicidal offenders who suddenly undergo a religious conversion that transforms their antisocial behavior into prosocial behavior. There is no miracle here, but rather cultural factors that condition moral behavior.
A space dedicated to altruistic work like "Effective Altruism" should not ignore these facts and the possibilities of developing strategies to promote moral evolution. From a consequentialist perspective, there would be no greater advantage from a "cost-benefit" point of view. It's a bit like the difference in the 19th century between individual acts of charity in the conservative tradition and large-scale social policies such as social security, public education, etc. But now it wouldn't be a question of politics, but rather of rational strategies for moral evolution.
Consequentialism has the problem of focusing only on the products of human labor for the common good, and not so much on the productivity of the human factor of production of altruistic deeds ("virtue"). Twenty thousand altruists can save quite a few people from malaria in Africa, but a rational social movement for moral improvement that has an emotional impact on its members in a similar way to how compassionate religions have done so far could reach millions of people.
We shouldn't be ashamed to study some positive aspects of religious movements from a rational point of view. Evolution is "copy plus modification."
There are many other factors that can influence human behavior to promote altruistic, non-aggressive, and enlightened societies. However, it seems to me that legislative (and political) changes, the mass education system, and the technology of the current capitalist economic system have indeed reached their limits.
Let's consider that a hundred years ago, an "effective altruist" might well have thought that the best way to do "the greatest good for the greatest number" would be to make the Marxist revolution triumph on a global scale.
But if one path closes, another may open.
Okay so, writing this:
Does not mean I...
If I said "to not need the guard rails we have at the moment" then you would have a point, but the word "any" means any level of guard rails down to anything above zero.
I'm just going to clarify, I am putting forward an inclusive position that says there can be value in the different perspectives (how much value each has can be anything above zero). You are putting forward an exclusive position that says virtue ethics is the only morally relevant position—therefore you must show that consequentialist calculus or deontology has zero benefit. So, all I have to do is show that any level of guard rails are necessary (anything down to zero) for the deontological part of this inclusive position, and all I have to show is that a consequentialist calculus is in any way useful for deriving virtues.
So, please don't strawman the position by stating that I have claimed more than I am claiming.
And in terms of showing that a consequentialist calculus is useful in validating virtue ethics, you've done it yourself. Comparing a virtue ethics approach to a utilitarian approach:
Why mention the millions of people affected, if consequences aren't important? You are running a utilitarian calculus here to justify the virtue ethics approach, which is exactly what I said utilitarian calculus is useful for.
So, at this point I've questioned if moral evolution could reach the point where there is no need for any guard rails at all (not the current level of guard rails we have at present) and have shown that you yourself use utilitarianism in the way I've suggested we all do.
I do think there is something to learn from Amish and other plain communities, and I support what you say that it might be productive to research what allows for this sort of harmony without the strictures of the legal system. I recognise the benefits of "copy plus modification" in moral evolution. Though I don't think the Amish are the first thing people think of when imagining an open and free utopia—my general impression is that the harmony is born out of an effective brainwashing campaign, but I'm open to research that shows the utility of other mechanisms in the Amish community that might be applied to a free society.
Personally I don't think modern society has exhausted its evolutionary path when we look at global trade and the utilitarian benefits (that can be quantified through the gathering of statistics) of social programs and prison reform that we see in Northern Europe. Modern society isn't a capitalist monolith, it's an evolving system with many open paths. If we measure quantifiable results like recidivism in the prison system, the level of extreme poverty in relation to various policies and trade and (meaningful) measures (not GDP) of wellbeing in terms of the welfare state, then we can evolve a system where more people exist in an interdependent relationship, which incentivises harmony in a way that is not coercive but rather mutually beneficial.
We can only do this though by measuring outcomes (consequences) and it might require us to wind back some outdated virtues that keep us following paths that are unproductive: like the value of 'my personal freedom to bear arms', or 'my righteous revenge in the punishment of the wicked', or 'my in-group exists in a zero-sum relationship to a given out-group'. These values are held by some people, and acting on the these values is seen by some as virtuous. The only way to learn that they are not virtuous is for someone to show them that acting on those values has negative outcomes—which again, requires a consequentialist argument.
So, I am in no way discounting the central utility of virtue ethics, it is the guide for 90% of my day-to-day behaviour, it is meaningful and practical, and is the best way to judge a person's character. My daughter actually made a good point the other day that, on an individual level, judging a person by the virtues they exhibit is a much better way of assessing their character than how much good they've done in the world through a utilitarian calculus, because people have very different capacities for effecting good in the world. For instance, 10 years ago judging Elon Musk on his utility would have measured him as one of the best people in the world—which has been completely turned on its head over the past 10 years (depending on your opinion of Elon Musk of course). Whereas if you'd judged him on his character—being abusive to workers, jumping from relationship to relationship, calling a volunteer worker a pedo-guy with no supporting evidence to millions of followers, buying a McLaren F1 and crashing it immediately etc—then with that assessment you'd have been much less surprised at his "turn".
Anyway, just thought I'd add that I agree with you that virtue ethics is important. Please don't (and this may be an unnecessary pre-emtive request) claim in your reply that I think virtue ethics is somehow irrelevant just because I want to also include other perspectives.
I agree with your clarifications and apologize for having misunderstood your consideration regarding the possibilities of moral evolution in the future.
My idea of "virtue ethics" is based on the consideration that the most valuable aspect of moral conduct is the motivating "intention" of the agent. The difference between "altruism" and "prosociality" lies precisely in this. It is possible for a virtuous person, well-intentioned in the sense of charity, benevolence, and altruistic action, to make a consequentialist miscalculation, but they will always be willing to be corrected.
In any of the texts on consequentialist ethics we know, we find examples and dilemmas that question the appropriateness of consequentialism.
Last year, thousands of African immigrants arrived in small boats to the Canary Islands (European Union). It is estimated that around ten thousand perished at sea in the attempt. CONSEQUENTIALIST SOLUTION: From now on, the Spanish navy will sink any immigrant ship that approaches the coast with cannon fire. After having murdered a thousand poor immigrants, terror will prevent any more ships from leaving the African coast: NINE THOUSAND LIVES WILL BE SAVED (the greatest good for the greatest number).
The question would be: does every consequentialist position on altruism lead to the moral improvement of society? I think not.
A "virtue" approach, on the other hand, is unequivocal as long as the model of virtue is consistently altruistic. If I am virtuous, I will also try to do "the greatest good for the greatest number," but I will not overlook the fact that the fundamental objective is to spread a virtuous lifestyle as a cultural model. The main task, therefore, is to define the model of virtue and its viability as an alternative for cultural change.
Of course, I am not claiming that the Amish are a model of virtue. They are superstitious, traditionalist, and even suspected of serious cases of sexual abuse among them. But it is undeniable that their system of controlling aggressive behavior is an example of what can be achieved with cultural changes. Even more surprising when we know that their social system is based on small family farms and that there is economic inequality among them: the most common social environment for neighborly feuds, disputes, and brawls.
Brainwashing? Aren't we conditioned from childhood to be competitive, "assertive," to develop "self-respect," and personal independence?
All moral development begins with the emotional consideration of mutual relationships. The success of moral virtue consists in the emotional internalization of the abstract principles of morality. Historically, this has been achieved primarily through religious mechanisms (all secular liberalism is inherited from the Reformed Christianity of the nations of Northern Europe), but such mechanisms can be reinterpreted and adapted to a secular approach to altruism. The selection of psychological strategies to reach this goal would be achieved through a process of trial and error. We can draw on something from the Amish, something from the spiritual exercises of ancient times, something from cognitive-behavioral psychotherapy... even from Stanislavsky's "method acting."
In any case, no one is working today to develop a coherent altruistic virtue that could serve as a model for cultural change.
At least the objective idea of "effective altruism" is being worked on, albeit from the limited perspective of consequentialism. And this is a relatively recent movement... one that didn't exist before. Until recently, the only secular approach to social improvement was socialism: establishing social justice through legal coercion (and from there to "the end justifies the means" was only a short step...).
Thank for hearing me. Now that I see your approach is not as hard-line as it seemed to be, I largely agree with your concerns and hopes for the future. I think many of the texts that discuss consequentialism generally do so by straw-manning a very short-term, limited view of consequentialism, and so I'm not surprised that texts you've read on the subject have given you this impression of it.
Playing out those situations for longer or involving more variables generally aligns utilitarianism with our more liberal moral intuitions. For instance the "solution" to the illegal immigration problem omits the entire point of seeking refuge from a country to begin with, that those refugees are hoping for a better life, solving the problem of illegal immigration bars all those people from seeking a better life, intuitively we know this reduces their well-being, and so part of what is so wrong about the solution is this unspoken part of it, which is quantifiable in utilitarian terms.
It's also important to consider why anti-immigration proponents want to keep those people out—out of a fear that those others don't share our values, and so "protecting our values" which is seen as a virtue by many (not me), plays a role in this unsavoury situation.
But as I say in the post, this playing out of the calculus to cover every possibility is not feasible in general, so virtues, or behavioural rules of thumb that generally result in positive outcomes is more practicable. And as I said in the comment, virtue ethics is also a better way to judge character.
I think EA has some major problems especially when it gets into long-termism, and measuring extreme threats that are extremely unlikely against sure threats that are measurably bad.
I'm one of those who think that the problem of the enormous wealth inequality between rich and very poor countries—especially those in sub-Saharan Africa—is the current equivalent of what the fight against slavery was in the 18th and 19th centuries. It's an outrage that there are still people dying of hunger or malnutrition.
That said, many alarmists about "uncontrolled" immigration can also build up consequentialist arguments. I've read that they say that what the inmigrants invest in paying for their dangerous, more or less clandestine journeys to rich countries should be invested in their own countries to prosper economically in their own lands. So if we make it difficult for them to experience dangerous, expensive, and unprofitable "uncontrolled" emigration (it is said that most limit themselves to subsisting precariously in the rich countries where they arrive undocumented), we would be doing them a favor (consequentialism).
If we start from the certain fact that moral evolution has existed, exists, and must continue to exist, the best consequentialist and long-termist calculation is to promote moral evolution. Improving moral behavior within the framework of cultural evolution would lead us to a "virtue ethics" that provides the best results in all areas that can be affected by the intentional, organized, and massive action of a humanity that has evolved morally in the sense of empathy, benevolence, rationality, and altruism (as effective as possible).
Attempts to improve moral behavior in an intelligent and organized way go back a long way: the monasticism invented by Buddhists, the "spiritual exercises" of the Stoics or Epicureans (there is a famous book by Pierre Hadot on this). Today we have many more resources in this regard.
Sigmund Freud, who was by no means a fool, thought that socialism would not improve human life because changing economic structures would not change moral nature. However, he himself admitted that there were psychological resources through which the most pressing human instincts (including the libido) could be manipulated to generate something resembling "holy love."
The best consequentialism is one that rationally analyzes the reality of the factors involved in any issue deemed a priority. If the priority is altruism, moral judgment must take into account the reality of the moral agent, which is human nature (the human being, the "cultural animal").