In virtue of what do scientific diagrams acquire their epistemic legitimacy? Which factors serve to validate schematic visual representations, rendering them useful and accepted components of scientific practice? This thesis addresses the epistemic legitimacy of scientific diagrams by investigating a variety of diagrams whose referents are “invisible”, that is, whose targets either cannot be seen, lack physical form, or have no material analogue. In focusing on such images, we shall gain insight into the factors that shape the forms that practicing scientists give to their diagrams and shed light on contemporary issues in the philosophy of scientific models and representations. In this work, common factors underscoring the epistemic legitimacy of scientific diagrams are identified through three in-depth historical case studies. First, we consider several diagrammatic approaches to visualizing chemical structure that emerged around the 1860s, especially the competing approaches of August Kekulé and Alexander Crum Brown, and investigate the factors that led to the enduring success of Crum Brown’s visual representations and the corresponding abandonment of Kekulé’s. Second, we examine a spectrum of stereochemical diagrams and material models produced from the 1870s to the early 20th century, particularly those produced by J. H. van ‘t Hoff, and consider the factors that determined the forms given to representations of three-dimensional structures of chemical compounds. Third, we explore the diagrammatic approaches taken by physicist Richard Feynman in his mid-20th century lectures on quantum electrodynamics, paying close attention to his diagrams’ stylistic commonalities and dissimilarities as well as their ability to mediate between various aspects of the practice of physics. Finally, this thesis concludes by considering several common factors regarding the epistemic legitimacy of scientific diagrams that can be identified in these case studies, including the importance of a bijective relationship between scientists’ understanding of their diagrams and of their diagrams’ referents, the utility of diagrams for productively reasoning about their referents, and ability of certain diagrams to reduce scientists’ cognitive burden, especially through visual similarities. These factors serve to unite divergent approaches to the philosophy of scientific models and representation and reorient contemporary debates concerning representation towards an integrated historical-philosophical methodology.

In virtue of what do scientific diagrams acquire their epistemic legitimacy? Which factors serve to validate schematic visual representations, rendering them useful and accepted components of scientific practice? This thesis addresses the epistemic legitimacy of scientific diagrams by investigating a variety of diagrams whose referents are “invisible”, that is, whose targets either cannot be seen, lack physical form, or have no material analogue. In focusing on such images, we shall gain insight into the factors that shape the forms that practicing scientists give to their diagrams and shed light on contemporary issues in the philosophy of scientific models and representations. In this work, common factors underscoring the epistemic legitimacy of scientific diagrams are identified through three in-depth historical case studies. First, we consider several diagrammatic approaches to visualizing chemical structure that emerged around the 1860s, especially the competing approaches of August Kekulé and Alexander Crum Brown, and investigate the factors that led to the enduring success of Crum Brown’s visual representations and the corresponding abandonment of Kekulé’s. Second, we examine a spectrum of stereochemical diagrams and material models produced from the 1870s to the early 20th century, particularly those produced by J. H. van ‘t Hoff, and consider the factors that determined the forms given to representations of three-dimensional structures of chemical compounds. Third, we explore the diagrammatic approaches taken by physicist Richard Feynman in his mid-20th century lectures on quantum electrodynamics, paying close attention to his diagrams’ stylistic commonalities and dissimilarities as well as their ability to mediate between various aspects of the practice of physics. Finally, this thesis concludes by considering several common factors regarding the epistemic legitimacy of scientific diagrams that can be identified in these case studies, including the importance of a bijective relationship between scientists’ understanding of their diagrams and of their diagrams’ referents, the utility of diagrams for productively reasoning about their referents, and ability of certain diagrams to reduce scientists’ cognitive burden, especially through visual similarities. These factors serve to unite divergent approaches to the philosophy of scientific models and representation and reorient contemporary debates concerning representation towards an integrated historical-philosophical methodology.