Richard Hollerith

The moral framework used by very many admirable people including Eliezer has every human being having intrinsic moral value — usually the same intrinsic value as every other human being. In contrast, I advocate a moral framework in which human beings (and other intelligent agents) have zero intrinsic moral value. Since humans are so far the only intelligent agents we know about, humans have great instrumental moral value.

Once greater-than-human engineered intelligence comes onto the scene, under my framework, the (instrumental) value of a human goes way down whereas in the more popular framework, the (instrument) value of a human remains high. This sharp difference is why I have chosen to concentrate my advocacy efforts in the singularitarian community: it is where advocating my moral framework will tend to make the most difference.

In the current era, the era before the explosion of engineered intelligence, my moral framework leads to many of the same general answers as the more popular framework, and Eliezer’s most recent series of posts, the series that culminates in a post titled Ethical Injunctions is a good example: I claim that my moral framework gives generally the same answers as that series of posts gives.

Even the most rational humans are only barely competent at rationality. For this and other reasons, the way for a human to maximize his contribution to the universe (which is the purpose of life according to my moral framework) is to devise and execute a large number of simple plans rather to devise a complicated long-term plan and then to spend the rest of his life implementing the long-term plan. The optimal strategy leans more in the direction of trying to make every action and every decision a contribution-maximizing action or decision and less in the direction of complex plans whose successful achievement will require many subplans and many actions and decisions.

Well, there we have the first objection to suggestions that we should relax ethical injunction X in order to achieve worthwhile prosocial goal Y: any plan that contains a subplan that requires violating an ethical injunction will tend to be a complicated plan, and again, since humans are not good enough at rationality to succeed at complicated plans, they should usually avoid them.

Contributing to the universe is hard. The vast majority of possible actions and possible plans have a negative or destructive effect on the universe. In contrast, interfering or ruining someone else’s plan for contributing to the universe is easy. Destruction is easy; creation is difficult — for information-theoretic reasons. Also, protecting or enhancing a person’s ability to create is hard whereas it is easy permanently to destroy that ability — by injuring or infecting or stealing from or defrauding him. In the future, we will collect all these ways of reducing a person’s ability to create under the word “harm”.

My moral framework is “agent-indifferent”: it does not matter which human or non-human agent makes a contribution to the universe: what matter is the fact of the contribution. Consequently, any plan that entails the intentional harming of a person or the intentional interfering with a plan of the person will tend to be a suboptimal plan because any help the intentional harm might give to one’s own plan has to exceed the harm done to the other person’s plans.

Well, the reader might ask, What about destructive people? There are people whose effect on the universe is indisputably negative.

My answer is that it is hard to arrive at an accurate judgement of a person’s destructiveness because people are complicated. That does not mean that one should not try to arrive at such judgements, but rather that one should realize that such judgements are unreliable, so that one should hedge one’s bets and refrain from actively harming the people one has (tentatively) judged as destructive. You should distance yourself from them, refuse to cooperate with them, refuse to teach them and ignore them, but not harm them. (Of course, there are a few exception, e.g., if you happen to work in the district attorney’s office.)

If you are living your life right, you will rarely know enough about or get close enough to destructive people to be able to harm them even if you wanted to because you will have surrounded yourself with the most non-destructive (creative) people you can find. (And by the way, that is easier if you work as an entrepreneur rather than an employee because entrepreneurs can more often choose whom they work with than employees can.)

And it is obvious that intentionally harming the most creative people you can surround yourself with is almost never the choice that maximizes your contribution to the universe. (Remember that creativity is defined as the ability multiplied by the willingness to make a positive contribution to the universe and that increasing the creativity of the people around you is an extremely potent way to maximize your own creativity and contribution.)

Lying to people to further a plan of your own is call fraud, and is fraud is definitely a kind of harm. More precisely, a human is not good enough at rationality to devise a non-destructive plan that includes fraud as a subplan.

This is straight out of the writing of John David Garcia, by the way (except for the bit about humans’ being barely rational and the appeal to information theory). Garcia who started publishing in 1970.

This entry was posted on Tuesday, October 21st, 2008 at 10:57 am and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.