I apologize in advance, because I'm trying to get back up to speed in Philosophy, and may butcher terms and possibly, inadvertently for the most part, ask silly questions.

I'm defining rational altruism in the context of Evolutionary Game Theory, where cooperation and sharing has an benefit to the individual in some way, as opposed to pure altruism, as in self-sacrifice.

I'm very much in support of pure altruism, and this is merely to get a gauge on what the experts think at the moment.

  • Is it irrational in which context? Strictly speaking, the only evolutionary aim is to pass on one's own genes, so in that frame of reference it's only rational if it helps achieve that goal. Although, on the other hand, one could also define 'strictly following the biological imperative' as an inherently irrational approach to living, in which case it's entirely rational. It just depends on what you're trying to achieve, which is intrinsic to what game theory is.
    – Cdn_Dev
    Commented Apr 13, 2017 at 19:18
  • @mcraen Thanks for your perspective. (I'm quite heartened to see the number of answers this question has generated, all useful.)
    – DukeZhou
    Commented Apr 14, 2017 at 19:27
  • 1
    You must read The Selfish Gene, I think it'll solve most (all?) your questions regarding to this. We're animals, and you won't understand our behavior unless you understand animal behavior.
    – Rodrigo
    Commented Aug 9, 2017 at 10:03
  • 1
    @Rodrigo Thanks for bringing up Dawkins, with whom I heartily agree on this subject. My point may be partly that the underlying principals derive from Game Theory (her in the context of an evolutionarily stable strategy, and that evolution is merely a physical expression of mathematical principles.
    – DukeZhou
    Commented Aug 9, 2017 at 18:41
  • if we believe that any any altruistic act benefits the actor. then what you call 'rational' altruism and 'pure' altruism are indistinguishable Why would you judge any form altruism irrational? If it is motivated by self-gain then it is not altruism and where it is altruism it may be, as Schopenhauer proposes, the breakthrough of a metaphysical truth. I struggle to see a case for the terms 'rational' and 'irrational;' in respect of altruism.
    – user20253
    Commented Apr 15, 2018 at 12:07

8 Answers 8


I would endorse Schopenhauer's explanation. John Mathews writes:

"In the Foundation of Morality, Schopenhauer asks the question: How is it that a human being can so participate in the pain and danger of another that, forgetting his own self-protection, he moves spontaneously to the other’s rescue? How is it that what we think of as the first law of nature - self-protection - is suddenly dissolved and another law asserts itself spontaneously? Schopenhauer answers: this is the breakthrough of a metaphysical truth - that you and other are one, and that separateness is a secondary effect of the way our minds experience the world in the frame of time and space. At the metaphysical level, we are all manifestations of that consciousness and energy which is the consciousness and energy of life. This is Schopenhauer:

“The experience that dissolves the distinction between the I and the Not I … underlies the mystery of compassion, and stands, in fact, for the reality of which compassion is the prime expression. That experience, therefore, must be the metaphysical ground of ethics and consist simply in this: that one individual should recognise in another, himself in his own true being … Which is the recognition for which the basic formula is the standard Sanskrit expression, ‘Thou art that’, tat tvam asi.”

John Mathews Joseph Campbell and the Grail Myth in At the Table of the Grail, Ed. John Mathews

Note that this does not propose pure altruism. It proposes that is NEVER not in our own interest to help someone else or care for another sentient being.
As ever for the Perennial view when it comes to distinctions, the distinction between altruism and selfishness is transcended. Altruism would be indistinguishable from selfishness for someone who is aware that all sentient being share the same identity.


The key to exploring altruism is that it is a concept that has to do with the "self." If you have a very sharp edged definition of the self, altruism can look irrational. However, philosophers have found that defining a self this way is very difficult. For example, we often see a pattern where the "family" acts as a "self," especially among the rich who inherit a great deal of their wealth and power along hereditary lines.

Arne Naess explored a different definition of "self" which he coined as the "Ecological Self." He defined the Ecological Self to be "that which the Self relates to." One example he gave was of a scientist working with cells under a microscope when a fly landed in the dish under the microscope. The dish had some harsh acids in it for the experiment and it was quickly clear that the fly was never going to fly free. It still took several minutes for the fly to finally succumb, and the scientist watched it die, helpless to prevent its death. Naess argued that if the scientist related to the plight of the fly while it lay dying on the dish, then that fly was part of the scientist's Self.

He then argued that if you choose to define the self in this way, altruism becomes a natural thing that comes from "selfish" desires, which has a really nice effect of pointing out that selfishness is not always immediately bad. He argued that Mother Teresa could be thought of as a very selfish person, she just resonated with such a wide swath of humanity that her acts benefited everyone.

From a game theory perspective, the interesting patterns seem to show up at the point where selves extend across each other. You can reach a point in a game where I am trying to accomplish goal A, and you are trying to accomplish goal B. These goals may be almost conflicting but are almost never perfectly conflicting (even in war, the goals are not perfectly conflicting... both sides don't want to die). In such an environment, both parties may act in unison to accomplish the common part of A+B, even if there was no a-priori reason for them to do so such as common genetic code.

  • Very useful! Thanks also for the Naess reference. Empathy is a central pre-occupation for me, particularly in relation to automata.
    – DukeZhou
    Commented Apr 14, 2017 at 20:35

If you're placing yourself in the context of evolutionary game theory, then what matters is the utility for the genes and the species, not the utility for the individual.

With this in mind, we already see pure altruism in nature: Species, usually insects, where individuals such as worker bees lack reproductive capacity, yet dedicate their lives and even sacrifice themselves for the good of the group. They do so because their genes are 'caring' about the species, not about the individuals themselves.

From this perspective, one might explain acts of human pure altruism in the same matter, maybe that's why people are occasionally willing to sacrifice themselves for women and children, or for nations, with no regards to themselves. It's rational from the group or species perspective.

  • I'm afraid group/species altruism has been debunked in reference to the human race. If you haven't already done so, take a thorough read of 'The Selfish Gene' by Dawkins, who deals with this argument in depth.
    – Cdn_Dev
    Commented Apr 13, 2017 at 19:12
  • 1
    Dawkins is on the more reductive end of where most evolutionary biologists stand on the issue of kin selection. Group selection is actually quite popular amongst many in the past 10-15 years, which makes it hard to state there's been any debunking. Most that I read share quite a bit of flexibility between the poles of the likes of Dawkins and E.O. Wilson actually. . Commented Apr 26, 2017 at 16:08
  • This article by David Sloan Wilson summarizes some of the problems with the extremes in the public debate and where precisely Dawkins is outdated. evolution-institute.org/article/… Commented Apr 26, 2017 at 16:13

The concept of "Altruism", rests on a notion of "Giving".

Now, to give ANYTHING, one MUST first have received SOMETHING, called "Existence" or "Life" through some kind of birth. This is the "Initial Gift", the first breath, this coming "into" reality, and this is spawns the initial thought of "Altruism". Recognizing we are in a living, breathing system, and we are all connected to it and through it, hence there is an underlying "Existence" to life and nature, and we feel gratitude towards this "Existence", relaying it back, through partners and family and society..

Altruism is a function of "Existence" + "A conscious choice". It states the knowledge of having, at some point in time and space, received the "Gift of Existence", and acknowledging this to a degree when you realize, that giving is living, it IS how "existence" works, therefore by giving, we are mimicking the "Existence"

After this is realized, it is merely a matter of HOW much you want to give back to "existence" according to how much you feel "a part" of same said "existence".

So my point being, "A true Altruist" would be an individual, totally connected to "existence", "nature", "The big everything", whatever we want to call "it".

Now, "giving" in a sense, is a perspective, in any case for any individual. We see mom's lifting cars, giving life back to her kid's, bee's humming in hives to sustain the hive, Mother Teresa helping the poor and sick in India etc etc. We DO things for others, realizing we can not function well alone.

Even the seclusive bear realizes this and at certain celestial time of the year, will try to breed. Birds, reptiles, fish, all walks of life instinctively follows this path to a degree..

All forms of life giving their unique "perspective" of "existence" back to other parts of same "existence" and are therefore connected.

Through this uniqueness of "being", we add our own "flavour" back and thus make the cycle continue from the first initial "breath" or "bang" to where we are Now!

So humans, really ponder this "Altruism".

"True Altruism" is the statement of "I realize because I exist", thereby taking a conscious decision to give back in some form, realizing "Existence" is always included in the fellow human.

So "Existence" and "Consciousness" are concepts and simultaneously facts. The assumption that "The Big Bang" came from nothing, rings a little silly. We cannot make things "disappear from existence" or come back from "nothing" INTO "existence". Also we can not reproduce consciousness. Ai is trying to copy that and is going towards the goal like the numbers climb towards an infinity, they'll never reach.

Therefore the assumption correctly would be, "existence" must be eternal and "consciousness" and "Altruism" are functions of that

A game could reflect that fact. The overlapping "selves" in conflicts you mentioned, being a good example.

Setup: 2 opposing armies each get 1 item, then made to make a decision to trade or keep (give or no-give) contemplating the difference between "what i have" and "what can i get". It will reach a fundamental core in most people. play with that :-)

All this is just philosophical, logic-based thoughts.

Thx for a good question!


Personally, I generally feel that defining terms like 'altruism' by casting them in the mould of a mathematical theory (such as game theory) diminishes their essence. It is a sort of reductionism, and an altogether wrong sort if you ask me. For most people learn the term 'altruism' at a relatively young age -- for example as the opposite of 'egoism', as was the case for me. So that then is its true definition, i.e. the definition under which we have acquired the concept, have established its basic properties and relationships with other concepts, have learnt to reason with it, etc. To recast the concept in terms of a mathematical theory denies all this, and moreover denies the non-mathematical and pre-mathematical parts of our faculty of reason. This is what I mean by reductionism: to pretend that all our thinking is either mathematical, or residing in some state of imperfection from which it can only be saved if it is subsequently "mathematized".

Having said all that, let me apologize for being so opinionated on this subject (and me, a mathematician). I just wrote this as a defense for quoting a stoic writer on the subject of altruism. There is a 'real' problem here, in the sense that serious philosophers have felt the need to address it at least from Aristotle onwards, namely the question whether 'pure' altruism exists. I think it is extremely enlightening to consider what the ancients said on this subject.

So in the Discourses of Epictetus, written down by Arrian, we find:

[...] for every animal is so constituted as to do every thing for its own sake. Even the sun doth all for his own sake: nay, and to name no more, even Jupiter himself. But when he would be styled the Dispenser of Rain and Plenty, and the Father of Gods and Men, you see that he cannot attain these offices and titles unless he contributes to the common utility. And he hath universally so constituted the nature of every reasonable creature, that no one can attain any of its own proper advantages without contributing something to the use of society. And thus it becomes not unsociable to do everything for one's own sake. For, do you expect that a man should desert himself and his own interest? How, then, can all beings have one and the same original instinct, attachment to themselves?

So this is how some of the ancient philosophers solved the problem of altruism, by positing that man's nature is inherently social. It is also what Aristotle wrote in his Nicomachean Ethics: "φύσει πολιτικόν ὅ ἄνθρωπος" ("by nature, man [is] a social being").

Now for your question, this would mean that there is a definite distinction between 'pure' and 'rational altruism': namely we can either act in a manner that is entirely consistent with our self-interest, but without satisfying any sort of social nature (which in some people may exist to a much lesser degree, or otherwise may exist but for some reason be ignored) -- this is what we would colloquially term a 'selfish' or 'egotistical' personality. When someone like this acts altruistically, it will not be out of a desire to satisfy the social instinct, but will rather be 'nakedly self-interested', and we would not be inclined to call it 'pure altruism', but rather 'rational altruism'. On the other hand, when someone possesses the social instinct to the usual extent, he or she can act in accordance with their social nature. If such a person acts altruistically, this is then what is commonly known as 'pure altruism'.


There are numerous approaches from reason to moral choice and action. It is not irrational to consider reasons why one would act towards others in a particular way and with chosen limits, definitions and provisos. Coming at it from Evolutionary Game Theory would be one of numerous schema by which one may make a decision. Kant, Hume had their take on it as with others.

Existentialism and Madhyamaka philosophy of Nagarjuna have a similarity in positing (or depreciating, with existentialism) no ultimate 'essence' including a self, seeing existence and entities as a coming together and passing away of conditions, with no ultimate "core". There are also neuro-psychological indications now that see a similar deconstruction of our concept of "self". All this impinges on any idea of human altruism and reasoning about it.

Currently I see it as mainly an existential choice with whatever schema seems helpful to you, or without a schema at all. There are strong schemas that make a good argument for non-sacrificial altruism as a communal strength. Along with that, there appears to be a human species genetic, almost instinctual primate bonding urge, but it is clear from human history this is able to be overriden by our higher mentation, whether for ill or good.


Traditionally people think that rational altruism is an advanced form of egoism. Because if you're altrustic, your neighbors survive and they may help you later, so it's actually a part of egoism. I'd like to turn this argument around.

The true nature of evolution is to help others. The goal of an organism is to help organisms around it to survive. In the beginning, there were purely altruistic organisms. Unfortunately, they died and in doing so, they failed to live and help their neighbors. A more advanced form of altruism was born, rational altruism, which includes a bit of egoism. The egoism is required to keep living, so you can help others.


Altruism isn't irrational. It's most definitely rational. It's merely harder the more people you are altruistic towards. Surely that's obvious? And surely that is rationally obvious. It's easy to look after one (whoever that is, and all things being equal) than it is to look after three (whoever they are and all things being equal).

Let's take a simple analogy: Consider multiplying two numbers together:

7 x 6 = 42

Well that was easy. Now try

42 x 42 = 1684, roughly maybe

Then try

1684 x 1684 = ???

Not so easy. Likewise with altruism. It's possible to place altruism on a continuum where we begin with altruism towards the self and end with altruism towards the world, that means everyone and everything in it. The first bit is easy like

1 x 1 = 1

And the last part is hard, like

545574635 x 67642537 = hard

Altruism at that level, is generally not carried out by persons but by institutions. For example, governments, welfare states, charities, the Dalai Lama and so on. That this is not so obvious perhaps is that institutions are embodied in persons otherwise they become a kind of dead letter, even in these days of increasing automation and automata.

  • I tend to agree. One of the chief arguments against, for instance, the social safety net is that instead of paying taxes, people should just be charitable in their own communities. (Doesn't really work, especially in an interconnected society.) But, while Von Neumann may have been a little paranoid and pessimistic, he was no slouch, so it's hard to get around the game theoretic definition of rationality. (Of course, the "always defect" strategy goes right out the window as soon as the game is iterated, and relationships and trust come into the picture;)
    – DukeZhou
    Commented Nov 4, 2018 at 19:57

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .