Browsing by Author "Pöyhönen, Samuli"

This master's thesis deals with human scientific classifications and the philosophical frameworks that have been used to describe them. The main objective of the study is to argue contrary to some recent suggestions that these human kinds cannot be fruitfully described with conceptual tools taken from the tradition of natural kinds. Medical classifications and classifications made in the modern behavioral and social sciences constitute an interesting (but so far widely neglected) field of study for philosophy of science. The origin of these classifications is typically in everyday life, but as scientific objects they also play part in the epistemic practices of science. This double-bind makes human scientific classifications and the corresponding phenomena in many ways different from the objects of the natural sciences: Unlike the classifications in the natural sciences human kinds have moral connotations and are often tied to a specific socio-cultural setting. However, there has not been much philosophical research focusing on the nature of classifications in the human sciences. When human kinds have been studied the common starting point for analysis has been to approach them with the conceptual tools developed for classifications in the natural sciences. The research question of this thesis is therefore: Can the theory of natural kinds be successfully extended to apply also to the classifications made in the human sciences? According to my view the answer is, briefly, no. In chapters 2-4 I describe several influential theories of natural kinds and try to reveal the connection between questions of classification and the problem of induction. I aim to point out the function of the concept natural kind in the contemporary philosophical discussion. By comparing different theories of natural kinds I show why Richard Boyd's account of natural kinds as causal homeostatic property clusters is the most promising starting point for extending the theory of natural kinds to the human sciences. In chapters 5 and 6 I introduce the central opposition of the thesis: According to some philosophers of science, such as Paul Griffiths, Boyd's theory of natural kinds should be extended to cover all scientific kinds including human scientific classifications. I call this view the ecumenistic theory of classification. On the other hand the theory of looping effects of human kinds introduced in the work of Ian Hacking gives good reason to suspect the plausibility of this ecumenistic extension. By further analyzing these opposing views in chapter 7 I show how the complex feedback loops emerging between human scientific classifications and the classified people constitute a counter-example to the ecumenistic theory of classification. Along with this critical objective of the study I also briefly sketch a possible synthesis of the ecumenistic view and Hacking's theory. I suggest that despite their problems these theories could together form a starting point for a typology of different kinds of classifications. I claim that such a typology could be a useful analytical tool for characterizing for example the controversial human scientific classifications that have until now often been dubbed socially constructed classifications. In the concluding chapter of the thesis I also assess the current state of the concept natural kind and claim that in many cases using this overdetermined concept creates more confusion than clarity.