Monday, December 1, 2008

The Atheist's Definition of Morality

I know I've been writing on religion quite a bit lately, but I really want to examine one more thing before attempting to get back on subject (whatever that means).

Of everything Richard Spencer said, the one thing that still bugs the shit out of me, a week later, was his point on morality. I just couldn't believe he said that Atheists have no moral values, simply because we don't have the Bible to spell it out for us. That is the dumbest oversimplification I've heard for quite a while, with second and third easily coming from that same lecture.

Atheists are moral. Hell, even chimps are moral. What's less bluntly obvious to religious folk is our definition of morality, so I give upon you my definition of morality:

Σ[(Mcommunity-Msubject)/σ] < Threshold Function

The magnitude of the quantity [perceived average of one's community's moral values minus morality of subject divided by the standard deviation] is less that a threshold function.
  • Sigma: It goes almost without saying that morality is a multidimensional issue. To say that it can be reduced to a single number would be a gross over-simplification. One's view on murder could have very little to do with their view on abortion, or animal rights, or their views on energy consumption. This allows someone who is generally inline with the community seem to be a generally a good person, except when carbon footprints come up in conversation and he looks like a jerk. The mathematics for this is left as an exercise for the reader/philosopher. Many things will be on a [0, 1] analogue scale (Vegetarianism), while other metrics might very well be on a single (How many people can you murder before it's uncalled for?) or multidimensional plane (resource usage - gas, electric, space, etc). Hell, you could start coming up with metrics that require complex geometry, depending on how hard you work at it.
  • Mcommunity: The perceived average moral values of the community. Contrary to what has been said about Atheists, I can say that what Hitler did was wrong. I will go even farther, and point out that there was a large number of people who didn't think what he did was wrong. My definition still holds. His community's collective moral value happened to be skewed towards it's-ok-to-kill-Jews-because-it's-all-their-fault (ΔM is small, so they think its ok). My community's Mcommunity happens to lie closer to not thinking it's ok to kill people (ΔM is large, so we think Hitler is evil), but who can say what's right?
  • Msubject: The perceived moral value of the person being examined. Most axes of this point should understandably undefined. When you first meet someone, the first thing you start to do is (hopefully) rarely start grilling them on abstract moral dilemma. Only for close family members, significant others, and maybe your best friend would you have a good general idea where they really lie in this multidimensional morality space. For most everyone else, almost arbitrary values are used, based on Mcommunity, Mpersonal, and even apparent correlations between separate axes (ie gay marriage vs interracial marriage, animal rights vs vegetarianism, etc)
  • The standard deviation: The deviation of a community is going to give a good indication as to how open minded the group is. A community of only evangelicals is going to be less accepting of a homosexual than a vast array of people with one or two evangelicals mixed in, who might still condemn them, or might not, since they don't feel that have as much to stand on for support in the community on the issue.
  • The threshold function: I would imagine the threshold would be something quite like the sigmoid function. The exact coefficients would of course be personally defined, and could easily be influenced by the community as much as the perceived σ. This allows for smaller deviations in less important matters to not classify someone as a character-lacking jerk than deviations among seemingly more important matters, which can easily classify someone as such.  And notice that the sigmoid isn't a step function, so the result of the entire equation is going to be more statistical than binary, which means that you can think Hitler is truly evil down to his core, but that guy who talks shit about Sally behind her back is just a jerk.
So there is my engineer-based view on a philosophical matter which has become much too much based on religion for my taste. It still needs a bit of work to deal with the edge cases, but they're less obvious than miscongruencies like homosexuality or shellfish, or entire communities being completely inexplicable.


  1. Kenneth,

    Great to have someone from Dick Spencer's school (engineering, I mean) come up and hand him his comeuppance. The guy was such a complete tool, it wasn't even funny. Just sad, pathetic, distasteful, ugly, etc.

    I like that morality, in your definition, applies to both those for and those against Hitler's ideas. Since morality isn't unique to any one brand of community, we have to accept that sometimes people will act in conflict based on reasons they feel are very true indeed.

    That, of course, was the point of Dick's argument, that morality had to have a single source so that we would all be able to find a way to finally put an end to things like war and killing in the name of, etc, because we'd all agree about right and wrong. Once that happens, life will be good on earth and we can get on with living the way we should so we can all get to heaven.

    What a bunch of crap. Dick's small minded thinking is the very reason we have war and killing in the name of, but fat chance of ever getting him to realize that.

  2. Who defines the value of the M sub variables? You defined Hilter as having a small delta M. You said that yours is large. You assume that your moral number is higher than Hitler's. I agree that you have morals but how do we know this is true?

    Don't assume I am a theist.

    Do not answer with simple answers like, "because it is rational", or "because society says it is". That is a big gap in logic that I currently can't resolve. Right now, I haven't heard any acceptable logical and rational answer to the simple question. In the context of your blog, who decides the moral numbers?

  3. @Anon: Everyone personally defines the M variables as best they can, with the hope that everyone in a like-minded community will come up with similar results for M_community, which is really an average of all of the M_perceived of everyone you care about. At the same time, since even M_community is a personal number, this means it's not uncalled for for you and your parents to disagree on the morality of something (i.e. daughters bring home boyfriends the parents don't like). It can differ caused by different perceptions of those around you, or even by who's around you. When a kid starts spending time with "the wrong crowd," he starts to act amorally, because his M_community has become skewed from ours by the punks. Debates then become a matter of bringing other's M_community closer to yours by strongly asserting your M_personal, or your own M_community, or even someone else's perceived M_personal (play the devils advocate), to give the community a smaller variance. Feeling that most people around you don't differ on what you feel is the important issues (which carry larger magnitudes than smaller issues) gives you the sense of community, which you appreciate.

    I didn't say anything about me having a larger morality number than Hitler. Talking about these M variables as absolute numbers doesn't really mean anything. I said that a committed Nazi would classify his deviation from their M_community as much smaller than I would classify his deviation from MY community. That's why the Nazis supported him, but I decided that his delta M surpassed my threshold function, and thus classify him as not-so-great of a guy.

    And for the record, the mathematics are more of an analogy than something I seriously think people should use and think about on a daily basis.