Monday, June 10, 2013

Kitten torture part 3: a scientific theory of morality

(Third in a series)

In this third and final post I want to tackle head on this problem that Bertrand Russell described in 1960:
I cannot see how to refute the arguments for the subjectivity of ethical values, but I find myself incapable of believing that all that is wrong with wanton cruelty is that I don’t like it.
Russell can surely be forgiven for not finding the answer to this problem because it relies on scientific advances that were not made until after his death in 1970.  Specifically, I believe that Richard Dawkins's theory of memes (1976) and Robert Axelrod's theory of the evolution of cooperation (1981) are both necessary and sufficient to produce a theory of morality that is both scientifically valid (which is to say, based on evidence) and consistent with most people's moral intuition.  In particular, it provides a principled argument against wanton cruelty, while still allowing the infliction of pain for a higher purpose.  It is similar to the theory Sam Harris describes in The Moral Landscape, but without some of the problems that his theory has.

Note that what I'm proposing here is different from what I wrote about back in 2008.  Back then I was just describing Axelrod's work on how moral intuition can evolve.  Here I am going further, proposing an affirmative theory of morality that is consistent with Axelrod (and Dawkins) but goes beyond merely observing that our moral intuition is generally reliable (because sometimes it isn't).

Let me start by being explicit about my premises.  If you don't accept these premises, you probably won't accept my hypothesis.

Premise 1: We (by which I mean humans) are alive, that is, we are the products of Darwinian evolution.  Our bodies are the direct phenotype of a collection of genes, but it is those genes, not our bodies, that are the reproductive unit on which natural selection acts.

Premise 2: In addition to our bodies, our genes also produce an extended phenotype, or indirect phenotype, comprising all of the effects that our bodies have on our environment: our clothing, our houses, our artwork.  All of these are products of our genes no less than our bodies.  The only difference is that there is additional steps in the process of producing the indirect phenotype than the direct one.

Premise 3: We humans are also hosts for a new (in evolutionary time scales) kind of replicator, called a meme.  Like genes, memes are self-replicating units of information that undergo a process of Darwinian evolution.  Unlike genes, memes are encoded in the synapses of the neurons of our brains, and more recently, in our extended phenotype: on stone tables, papyrus scrolls, paper, hard drives, and in the design of our artifacts.

Premise 4: Darwinian evolution is amoral (neither moral nor immoral).  It doesn't "care" about anything in the sense that we humans do.  It simply obeys the laws of physics to optimize a particular quality metric: the ability to reproduce.  But (and this is key) what Darwinian evolution optimizes is not the ability to reproduce bodies, but rather the ability to reproduce information.

Premise 5: It is fundamental to the nature of life that it draws boundaries around itself to separate itself from its environment.  One of the very first tricks that the earliest self-replicating molecules learned was how to make a cell wall.  Distinguishing between "us" and "them" is intrinsic to being alive.

Premise 6: Despite the fact that boundaries are fundamental to life, the nature of those boundaries is extremely flexible.  The direct phenotype of a single genome can include multiple discontinuous entities (as in an ant colony).  A single physical boundary can include multiple distinct genomes (as in the case of symbionts).

Premise 7: The information that reproduces and undergoes biological Darwinian evolution on earth is encoded in DNA, but that is merely an "implementation detail".  It doesn't matter to the evolutionary process how the information is encoded, only that it replicates, mutates, and undergoes a process of selection.

Those are all the premises.  Now a few key observations:

Observation 1: Humans are not autonomous from the perspective of the process of life.  A single human in isolation cannot reproduce, and even a mating pair would probably have trouble surviving in the absence of some kind of support structure.  At a minimum, to have a robust shot at long term survival requires a tribe or a village at a minimum.  In this regard, we have more in common with ants than is generally appreciated.

Observation 2: Even a single human body is not an autonomous whole.  We contain symbiotic microorganisms (mainly in our digestive tracts) without which we could not survive.

Observation 3: Unlike most other creatures, we humans can make conscious choices about where we draw the line between "us" and "them."  We can even draw multiple lines for different purposes, around our immediate family, our tribe, our race, our country, our species, even around all living things.

Observation 4: Our memes are also symbiotic entities.  They can't live without our brains (because our brains form the ecosystem in which they exist) and in turn they provide our genes with significant reproductive advantages relative to species without brains, or whose brains are not hosts for memes.

Observation 5: Usually the reproductive interests of genes and memes coincide, but not always.  For example, the meme for birth control is potentially disastrous for our genes.  So it should not be too surprising that there exist human brains with an instinctive revulsion towards birth control.  Such an instinctive revulsion would be a defense mechanism evolved by genes to combat the deleterious effects (from the point of view of our genes) of the birth control meme.

This last observation is the key.  I believe that this occasional conflict between the interests of genes and memes is responsible for most if not all of the genuine disagreements over morality.  They all come down in the end to a conflict between the reproductive interests of memes and those of our genes.

So my theory of morality is simply this: where there is conflict, the interests of memes should trump the interests of genes.

This is a humanist (with a small "h") theory, at least for the moment, because at the moment only the (extended) human phenotype can be the host for memes (as far as we know).  But it has two significant advantages over the Humanist (big "H") theory of morality, at least as it is stated in the Humanist Manifesto.  First, it is humanist, but it is not axiomatically humanist.  It is not species-ist.  It is axiomatically meme-ist.  It just so happens that we humans are the only species that can be hosts for memes, but that is only true as far as we know at the moment.  It could change.  When and if it does, we won't have to struggle with whether or not some alien species or a conscious computer "is human" because "human" is not part of the definition, just an observation about the current state of our knowledge about the world, a circumstance which we should expect to change.

The second advantage that my theory has is that it offers an answer to Russell's problem: a principled account of why wanton cruelty is immoral: it's bad for memes.  The kind of mindset that allows one to take pleasure in torturing kittens is generally not the kind of mindset that is conducive to thinking deep thoughts.  Propagating memes is not easy.  It requires that one's basic physical needs be met.  It requires a certain level of physical security.  It is advanced by certain kinds of technology (like computers and the internet, not so much by nuclear weapons) and, in many cases, by the presence of companions who might be members of other species.

It's a positive, uplifting message.  It is, essentially, a pro-life position, but one which broadens the definition of life to include memes, and resolves gene-meme conflicts in their favor.  I think it's the Right Answer.  There's just one minor problem that still needs to be resolved, and that is that calling oneself a "meme-ist" sounds really weird.


Sven on Senta II said...

Definitely a worthwhile topic even if I'm not sure exactly what your goal is; is it only to find a label for yourself ?

I have to think a bit before writing my counter. I know what the counter is but have to think about how to phrase it so it doesn't get lost in poor phrasing.

Ron said...

> I'm not sure exactly what your goal is

I don't know how I could have made it any clearer:

"I want to tackle head on this problem that Bertrand Russell described in 1960: I cannot see how to refute the arguments for the subjectivity of ethical values, but I find myself incapable of believing that all that is wrong with wanton cruelty is that I don’t like it."

In other words, I want to provide a better response to the claim that atheism necessarily entails moral relativism.

Don Geddis said...

"where there is conflict, the interests of memes should trump the interests of genes."

That's ... interesting. Certainly unusual. And probably novel.

It's hard to leap on board, without seeing a lot more examples worked out.

My initial concern is: just like genes use your physical body as a mere tool to reproduce themselves; so do memes do the same thing. I suppose this isn't quite what you said, but an ethical rules like "do whatever is best for memes" doesn't seem very attractive either.

You've already raised the example of birth control. That's (possibly) good for me as an individual, but bad for my genes. So the genes lose.

Similarly, religion is a very successful meme. Islamic suicide bombers happily sacrifice their individual lives, for the glory of the meme. Is this not an example of meme interests trumping gene interests? And yet, I'm still not a fan.

I guess I'm having a hard time seeing what your short phrase actually means, in practice. I'm going to need some help, in order to operationalize it for specific real-world situations.

Ron said...

Doing what is good for memes in general is not the same as doing what is good for any particular meme. What I'm proposing is doing what is good for memes *in general*, not doing what is good for any particular meme. So we can generally reject suicide bombing as immoral because it destroys "habitat" for memes (human brains) with no offsetting benefit to memes in general. The benefit (such as it is) is only for one particular meme (Islam).

But your point is well taken. I'll write another followup with more case studies.

Don Geddis said...

Interested in the followup. Particularly interested in the resolution of any potential conflict between genes/memes, and individual success. In much the same way that my genes want me to reproduce and (eventually) die, but I personally strive to control my reproduction (via technology), and avoid dying as much as I can (indefinitely, if possible).

Your writing here about genes vs. memes, seems (at first blush) to leave out individual selfish goals.

Ron said...

This is actually an easy one to answer: your genes want you(r body) to die. Your memes don't. They spent a lot of time and effort getting into your brain, and when you die, a lot of them die with you. The longer you live (and the more you blog :-), the more opportunities the memes in your brain have to spread.

Don Geddis said...

Sorry, I wasn't clear enough in my last comment. I didn't mean to ask you to resolve the specific question of individual death. (That may well be a gene/meme tradeoff.) Instead, I was trying to use it as an analogy, that my genes want my body to die, but I choose not to adopt those same goals, on my individual level.

My intuition suggests that there should be similar conflicts with memes. Things that memes "want", which I may choose (on an individual level), not to adopt as my own personal goals.

I'm not yet comfortable, adopting what I think I hear you saying, which is something like: what is Good and Moral, is to do whatever is good for memes.

My question is: surely there are easy examples, where something that is good for memes, is bad for me individually. (I was trying to use Islamic suicide bombing as an example, but you didn't address it in the way I hoped. Perhaps you can think of a better example.)

Just hoping you'll address a (theoretical?) conflict between "good for memes" vs. "good for me".