La volonté machinale: understanding the electronic voting controversy

ing from technology and law According to Niklas Luhmann, observation is distinguishing indication. This means that risks are perceived based on available distinctions. They are revealed in the interaction of existing distinctions with the environment. If existing distinctions are inadequate, monsters may appear, and these may lead to changes in our distinctions. We would like to be able to prevent monsters by pro-actively reflecting on our categories in risk assessment. Technology and law are externalisations of existing distinctions. They are therefore relevant if we wish to implement reflection on categories. We should write down which implicit distinctions are present in technology and law, and thereby get an overview of which sides of distinctions are being marked at present. Technology and law serve as mirrors of our distinctions in this approach: they tell us something about ourselves. Thereby, they can themselves contribute to updating our ethical framework to the developments they are part of. They can tell us that we could have made different choices, by presenting us with categorical challenges. Can we mark the other side? In order to reflect on our distinctions in technology assessment, we should investigate existing technological artifacts, scientific theories and legal arrangements. Then, we write down associations that categorise the specific legal or technological feature within our existing cultural framework. We try to find counterconcepts to the concepts that we wrote down, and thereby challenge the existing categorisation. On a general level, this chapter has shown how challenges to our concepts can be identified by looking into the “other side” of authenticity, or the “other side” of availability. Identity can be more than something to verify, and availability will mean a different thing in the context of multi-channel voting. However, this approach can also work on a lower level. For example, if we look at the RIES system, we find that one can verify how one’s vote has been counted. Possible associations with this procedure include the concept of proof from logic. In RIES, the presence of your vote is proved by providing a witness (your vote), as in constructive logic. As a counterconcept, we have a proof without a witness, as in classical logic. Could this be used as an alternative? Such systems already exist in the scientific literature. Are these more desirable, more practical or cheaper? An example from law is the Dutch e-voting regulation. There is a requirement that the way of storing the vote should not allow reconstruction of individual choices (see page 33). Possible associations with “way of storing” include memory and software. If we mark the other side of the distinctions, we may find something like processing and hardware. Is it relevant to include clauses on these aspects as well? Or instead a more general formulation? Such reasoning could have led to the idea of tempest attacks, associated with processing and hardware, or otherwise to a more general formulation of the requirements that would have covered the issue. 168 A Categorical Challenge to Democracy As categorical challenges, both new technological options and new legal options can emerge from this process. These constitute alternatives for existing arrangements. Finally, we evaluate the benefits of alternatives that are revealed by means of this process. This includes evaluation of desirability, feasibility and cost. Reconstructive technology assessment can help us to reveal pro-actively what existing distinctions hide, and how this affects our understanding of the new technology under consideration. Precise methods for implementing this process require further research. Of course, both technology and law need not be implemented in the real world before this process can take place. Letting go of Nature In reconstructive technology assessment, we recognise that technology is not merely an external reality, but rather part of a process to which we contribute by means of our categorisation of phenomena. This replaces the distinction between nature and culture by an interactive process, in which many stakeholders can have a role. Even scientists are no longer occupied with Nature, but they join forces in the replacement of confidence with trust. This makes possible different separations of powers, for which Latour already gave some indications (section 4.4). If this helps to reveal different risks by means of different distinctions or categories, instead of amplifying the Nature of existing ones, the discussion on e-voting and other technologies may take a quite different form.