The Moral Life of Geeks

Develop the fundamental theory, algorithms, and software for the design and analysis of robust, high-performance, team-based, multi-agent cooperative control systems operating in dynamic, uncertain, adversarial environments.

That sounds like fun, where do I apply? As it turns out, the Office of Naval Research. It seems that most of the really interesting research -- especially in areas of intersection between computer software, biotech, and nanotech -- is funded today either by the Pentagon System or corporations. But what if you are, like most Monkeyfisters, a geek and a leftist? What if you are a person inclined to do technical work but also inclined to refuse to work for evil institutions, that is, institutions that cause undeserved harm?

The basic dilemma lies between, on the one hand, not developing one's innate and learned capacities, which can be a kind of harm to oneself and to others, and, on the other, developing one's capacities by working for evil institutions.

The dilemma has many forms. For example, most Monkeyfisters are or have been involved in developing free software, often because of moral considerations. And yet there is a moral tension: Richard Stallman wrote the GPL in order to give software people a way to share their efforts freely with neighbors. But one of the guiding principles of open source software is that licenses cannot discriminate against fields of endeavor. But what about fields of endeavor that are evil? What about writing software, or doing research, that will, directly or indirectly, be used to cause undeserved harm to others?

I owe myself and others a duty to develop my capacities; and one way I've chosen to do that is to be involved in the development of a free software infrastructure. But I also owe an obligation to refuse and resist cooperation with evil institutions. Under all standard free software licenses, anyone may use the fruits of my labor -- including evil institutions like the Pentagon, the US Armed Services, defense contractors like Boeing and United Technologies; online porn merchants; biotech corporations like Monsanto; and agents of globalization like the World Trade Organization. So in developing free software it appears that far from resisting cooperation with evil institutions, I'm may be directly or indirectly contributing to them.

I've only used free software development as a representative activity; what I've said so far about it applies to many kinds of technical R&D. Why shouldn't I discriminate against evil fields of endeavor? There are three standard responses:

1. Technology is amoral. The first response is that since technology is morally neutral, so as long as I don't put my work to evil ends, I'm not morally blameworthy.

2. Redefine the dilemma. The second response says that just because I can write software or do research that may be used by evil institutions, I don't have to. I could be a waiter or a farmer instead.

3. Applied technology v. basic science. The third response distinguishes between basic research and applied technology; in doing so, it claims that, since it increases human knowledge and is only indirectly, if ever, applied, basic research is morally praiseworthy, or at least not prima facie morally blameworthy, even if evil institutions ultimately use it to achieve evil ends.

The first response is flawed. It's simply not the case that all technology is necessarily amoral. Technology, like any other cultural artifact, doesn't just fall from the sky. It's always already embedded in, and inextricable from, social space, which is always already a political space, which, in turn, is always already an ethically-contested space.

I take this lesson from the work of David Noble and Steven Shapin. Technology, with very few exceptions, gets developed in our late Western capitalist era because its development gets funded by governments and corporations, often in partnership. Failing to take that social and political context into account when evaluating technologies, and the morality of one's participation in their development, is simply to fail to take account of all the relevant facts. While some technologies -- for example, computer-supported collaborative work (CSCW) -- can be used equally well for good or evil ends, technology itself is not necessarily amoral.

The second response is coherent, but problematic if you believe, as I do, that persons have a duty, to themselves and to some others, to develop their innate and learned capacities as a necessary condition of human flourishing. The second response is applicable in what we may call limit situations in which the only choice one has is either developing one's capacities in association with an evil institution or not developing them directly, if at all. What proponents of the second response fail to recognize is that limit situations are rare. In sum, then, the second response is a useful and valid one, but only in some rare situations.

The third response claims, essentially, that whatever the moral status of particular bits of applied technology, or engineering, basic research is at least only second-order problematic. While I agree that we shouldn't abandon all basic research, even when it's reasonable to assume that some of it will be used to achieve evil ends, it's unclear whether most technical people do basic research, or whether doing basic research funded by evil institutions should be done at all. The modern research university is obviously of crucial importance, but an ever-increasing majority of research done in universities is funded by the Pentagon and corporations. In short, that basic research is only second-order morally problematic can at best be ameliorative, not dispositive, of the basic dilemma. (And in the particular case of software geeks, most software development is more like applied tech than basic research, i.e., more like the development of, say, the Apache Web server than what Donald Knuth does, and so the third response isn't very helpful to the geeks.)

So how should technical people respond to this dilemma? I suggest three kinds of response, the first two of which are specific to the development of free software, while the third is generally applicable.

First, we need to reinvigorate moral debate about free software (and, by extension, about technology and intellectual property in general) by talking not only in terms of freedom, which Richard Stallman has done well, but also in terms of responsibility, that is, acknowledgment of one's duty to avoid cooperating with institutions that are evil. One way to do that is to talk about an Ethical Public License, at least as a thought-experiment. What might such a license look like? Is it legally possible to write binding software license that prohibits particular its use within fields of endeavor or particular types of institution? What kind of moral claims are involved in such a license? How far should one go to prohibit one's work from being used to cause undeserved harm? Could the resulting license still make a claim to be free software, that is, a tool of extending personal freedom?

Second, and this applies primarily to those of us who are both leftists and geeks, we need to challenge the wholly unreflective libertarianism of free software, and Internet, culture. Most geeks, I suspect, would not credit the dilemma I've described, if for no other reason than that most geeks are habituated libertarians, who don't think about their technical work in terms of social or institutional analysis.

Finally, we need, especially in America, to reassert democratic control over the kinds of institution that fund technology development and basic research, but particularly those that are ostensibly democratic: government and universities. In that way we may be able to reassert control over public corporations as well.

What good can come of reasserting democratic control? If governments, universities, and corporations were under democratic control, they could be harnessed to pursue ends that contribute to, rather than impede, human flourishing. Under strong, reinvigorated democratic control, the moral status of basic research becomes much clearer, since it becomes reasonable to assume that the applications of that research will be for good, not for harm. Democratic control of these institutions would make limit situations exceedingly rare, since it would tend to promote the pursuit of good ends over evil ones.

Technology has liberative potential, but only if it's controlled by democratic structures and institutions. And given the sorry state of American democracy at present, it's no wonder that geeks, engineers, and scientists of good will daily face difficult moral dilemmas. The solution to those dilemmas, and the key to harnessing technology for the good, is the reassertion of democracy in the face of its slow, ongoing demise.