Backtracking and have to: Maintaining a Unified Analysis of Conditionals
This dissertation investigates the properties, meaning, and licensors of backtracking counterfactuals. In particular, backtracking has long been seen as problem for a unified account of counterfactuals as backtracking counterfactuals are counterfactuals whose antecedent time temporally follows the consequent time. Such a conditional is a problem for any account of counterfactuals that relies on similarity in the past and not the future to calculate which possible worlds are most similar. As such, backtracking counterfactuals have largely been avoided in analyses of conditionality.
Recent efforts have been successfully made to include some backtracking data into a wider theory of counterfactuality (Arregui 2005, Schulz 2007) but even these accounts do not encompass the full range of backtracking data. In Chapter Two of this dissertation, I will present the background assumptions for why backtracking is problematic as well as present backtracking data that Arregui accounts for, as this data must be accounted for in any theory of backtracking.
The focus of Chapter Three is new data. I present new instances of backtracking counterfactuals that challenge the current theories of backtracking. Furthermore, I present data from non-backtracking counterfactual conditionals that show that, contrary to the implicit assumptions of earlier theories, backtracking is not magically saved by an extra layer of modality. There are several strategies that make backtracking possible, and in each, no part of the denotation of the counterfactual is specific to a backtracking construction. I demonstrate this with data from forwardtracking (standard) counterfactuals, as well as non-counterfactual conditionals.
Chapter Four presents the formalism of existing accounts of backtracking as well as accounts of counterfactuals and conditionals more broadly. I examine the shortcomings of Arregui's account of backtracking for the data presented in Chapter Three and explore the tools from other areas of linguistics and counterfactual research that will guide my analysis, most importantly ordering sources, focus alternatives, and aktionsart classes. Chapter Fives applies these tools to present an updated analysis of backtracking counterfactuals. I will also account for an adversity reading that arises in some non-backtracking counterfactual conditionals.