Empirical and Pragmatic Grounds of Scientific RepresentationMatthiessen, Dana (2023) Empirical and Pragmatic Grounds of Scientific Representation. Doctoral Dissertation, University of Pittsburgh. (Unpublished) This is the latest version of this item.
AbstractThe central thesis of this dissertation is that the ability to reason and learn about the natural world using models can be explained in terms of the practices that warrant researchers to integrate models with accounts of their data-gathering procedures and act on their behalf. I argue that a model only functions as a representation with respect to a target phenomenon when this phenomenon is a plausible member of its domain of application and when the model can be used to characterize this target from data. I argue that this requires, first, that the model can be compared to data and second, that the model be integrated with an account of the process by which this data was produced from the target phenomenon. I provide an account of the representational accuracy of models based on their integration with a theory of technique and subsequent comparison with data patterns. On the same basis, I provide an account of the pragmatic representational content of models in terms of the set of practical inferences they license as a supplement to the empirical programs within the model’s domain of application. Historically, one often sees a back-and-forth negotiation where a model-based target characterization and a data-gathering practice are iteratively tuned to one another. Models are routinely informed by empirical results in the process of their construction and adjusted in response to them. Conversely, models add depth to target characterizations and fill out theories of technique in ways that alter data-gathering procedures. From this perspective, we can understand how a model’s representational content might gradually accrue to it and allow for finer distinctions in data outcomes. I present an extended case that tracks the development of X-ray crystallography and its use for the characterization of the molecular structure of proteins. Ultimately, what is presented here is intended as a robustly pragmatist account of scientific representation. That is, one that does not only tie model use to purposes, but also to the realm of human action. Share
Details
Available Versions of this Item
MetricsMonthly Views for the past 3 yearsPlum AnalyticsActions (login required)
|
![[feed]](/images/feed-icon-32x32.png){kind=icon}