Models of' and 'Models for': On the Relation Between Mechanistic Models and Experimental Strategies in Molecular Biology

Molecular biologists exploit information conveyed by mechanistic models for experimental purposes. In this contribution, I make sense of this aspect of biological practice by developing Keller’s idea of the distinction between ‘models of’ and ‘models for’. ‘Models of (phenomena)’ should be understood as models representing phenomena and they are valuable if they explain phenomena. ‘Models for (manipulating phenomena)’ suggest new types of material manipulations and they are important not because of their explanatory force, but because of the interventionist strategies they afford. This is a distinction between aspects of the same model; in molecular biology, models may be treated either as ‘models of’ or as ‘models for’. By analyzing the discovery and characterization of restriction-modification systems and their exploitation for DNA cloning and mapping, I identify the differences between treating a model as a ‘model of’ or as a ‘model for’. These lie in a cognitive disposition of the modeler towards the model. A modeler will look at a model as a ‘model of’ if he/she is interested in its explanatory force, or as a ‘model for’ if the interest is in the material manipulations it can possibly afford