Abstract/Summary

To what extent should we think of Mary Hesse as having been a scientific realist? Her early works were clearly intended to make space for a form of realism based on a fuller recognition of the role of models and analogies in science than existing empiricist views allowed. Her commitment to realism is explicit in her publications from the mid-1960s until at least the mid-1970s. But there were also some kinds of realism that she always distanced herself from. I investigate what all these versions of realism were, and why she endorsed some but opposed others. In the 1970s, Hesse’s growing conviction that there is no cumulation or convergence of theory across the history of science, together with the idea of the underdetermination of theory by data, led her to endorse what she thought of as a form of ‘pragmatism’ or ‘instrumentalism’. Had she turned against her earlier form of scientific realism, or perhaps all forms of scientific realism, and if so, why? Or did she think her pragmatism was compatible with her initial, model-based realism? I draw attention to a residually realist aspect of her views, one that should make us think of her as having suggested what others would later come to think of as structural realism. But I also argue that her approach to language might well jeopardise the idea that she was ever a realist.