This document presents an overview of the ethical issues associated with the case of Richard Machado and his hate mail. If we use the ImpactCS framework (see below) the Machado case certainly highlights some important in almost all of the columns of ethical issues defined by the framework. In addition, some of these issues need to be addressed as more than simple individual ethical decisions about whether to send hate mail or not; we will need to look to the group, national, and global issues involved.
The ImpactCS approach to ethical analysis was devised by a panel of ethicists, computer scientists, and social scientists. The point is that any particular computing system can be analyzed from both the perspective of social analysis and of particular ethical issues. The grid you see below was designed by the panel to serve as an analytic tool in thinking about any system. The idea is that each of the ethical issues can be analyzed at each of the levels of social analysis. If you click on the colored cells in the framework, you will be linked to a discussion of that ethical issue in the Machado case.
If the theme of the socio-technical analysis for this case was clashing cultures, a reasonable theme for the ethical analysis is about competing values. Does Machado's right to free speech override the right student's have to a safe environment to go to school? Do individuals' right to privacy override the need of those running a network to track down people using electronic methods (including surveillance)?
Sticking with values
Notice first how the language moves quickly back and forth between rights and values. We can usually describe some right an individual has as being based in some thing we value (e.g. the right to privacy is based in the value we place on autonomy). And so philosophers often derive rights in this way and suggest that discussion on the level of values is more clear. . But I prefer a teaching oriented reason to move rights discussion to the level of values discussion: it helps people keep a more open mind. Rights are often talked about as though they were absolute and inviolable. "I have an unconditional right to privacy and to freedom of speech," a student might say. But the right to freedom of speech can get in the way of the right to privacy, and if they both will not budge, then our discussion stops. If we move instead to saying we value each of these things, then we can ask how much we value it, and how we balance it against other, competing values.
Finding free speech in the ImpactCS framework
Finally, this case also provides an interesting critique of the adequacy of the framework. Defenders of Machado's right to send insulting email to people (let's leave death threats out of it for the moment) will most likely phrase their defense in terms of a right to free speech. A close look at the framework does not include free speech rights anywhere and some consideration of these is essential to understand this case.
So, should we include this particular item as a column in the framework? I think we should do so only if we want to privilege that right above any of the other multiplicity of rights we can think of. For instance, the United Nation's 1948 Universal Declaration of Human Rights has 30 articles in its text. Twenty-seven of these articles enumerate rights that the United Nations Assembly adopted. Some of these articles appear to list ten or more rights, leaving us with a full sufficiency of rights when fully enumerated. I have no objection to this full enumeration of rights, but listing all of them in a framework designed to help us understand the primary ethical issues in computing would be impractical.
In this document, I will discuss free speech rights under the heading of "Access," meaning access to a public hearing of one's position on an issue. Another solution might be to add a column titled "political issues" to the framework in which free speech, democracy, and cultural differences in political values might be addressed. But I will wait to recommend this until more than one case has shown difficulty in fitting the ImpactCS structure.