Abstract
We inquire into the notions of ‘boundary’ and ‘cluster’ in the fields of medical genetics, pharmacogenetics, and population genetics. First we show that the two notions are not well discussed in literature. Then we propose a promising explication of them, in which we argue that clustering is always ‘property laden’, that is, fundamentally dependent on decisions about the properties to be taken into account. In particular we suggest three different kinds of properties that have a role in these decisions. That is, we conclude that boundaries and clusters among humans depend on our way of considering nature. Concepts of ‘race’ and ‘ethnic group’ are discussed too, since they are the most used clusters among humans.