The primary concern of this dissertation is determining the distinction between semantics and pragmatics and how context sensitivity should be accommodated within a semantic theory. I approach the question over how to distinguish semantics from pragmatics from a new angle by investigating what the objects of a semantic theory are, namely expressions or utterances. In Chapter 1, I investigate the preceding question. I am approaching the question of how to draw the line between semantics and pragmatics by first considering what the objects of a semantic theory are. This is an under-appreciated point in this discussion. By beginning with this question I am approaching this from a new and important angle. For if we are interested in what literal meanings are assigned to, it is a necessary step to consider what the objects of a semantic theory are. This dissertation proceeds in answering this question over what the objects of a semantic theory are by arguing that they are expression types relativized to contexts, or expressions-in-context, by relying on arguments presented by David Kaplan. In addition it is argued that the way we approach language is expression based in that we often assign semantic content to sentences that are not a part of any obvious utterances. Moreover, it is then argued that semantics based on expressions-in-context provide a clear account of examples of the historical present without positing additional ambiguity. In Chapter 2, I approach a powerful objection to an expression based approach to semantics, that of true demonstratives, i.e. expressions the use of which require an actual demonstration. If semantics proceeds by assigning content to expressions-in-context, how can it assign different contents to "that" occurring more than once in the same sentence while maintaining that it has the same linguistic meaning? Stefano Predelli defends the view that by representing the linguistic meaning, i.e. character, of 'that' as a function from the context and an integer, representing the location of "that" in the sentence, to the demonstratum in that context, we can maintain a single meaning for all demonstrative occurrences of "that". I develop this idea further and compare it with Nathan Salmon's approach to demonstratives whereby the relevant demonstration is in the formal context rather than the demonstratum. I conclude by arguing that by taking expressions-in-context as the objects of a semantic theory, rather than utterances, we can still formally account for true demonstratives, and as a result show that the approach developed in the first chapter can handle this objection. In Chapter 3, I approach the question of how we might determine if apparent contextual influence on communication is semantically required or pragmatically influencing the communicated content. If we approach semantics as expressions rather than utterances, we must find a method for determining whether contextual features are triggered by the linguistic meaning of expressions. In the hopes of getting clearer on when context is semantically relevant I consider numerous tests that have been presented in the literature, but find them all wanting. These have been put forward by Cappelen and Lepore, Cappelen and Hawthorne, and Jason Stanley. I evaluate each in turn and conclude that although someone may be able to develop tests to provide some evidence for when an expression is context sensitive, no test is immediately forthcoming. As a result of this conclusion, the best that we can do is evaluate complete theories based on coherency and explanatory power. In Chapter 4, my final chapter, I apply the results set out in earlier chapters to a particularly problematic case study: negative existentials. Negative existentials present a pressing case because any purely semantic approach will find a problem with them. So, any solution to this problem will rest on the distinction between semantics and pragmatics. I develop a theory of negative existentials based on Gricean generalized conversational implicatures within the framework provided by the distinction made in Chapter 1, where semantics assigns contents to expressions while pragmatics to utterances as well as the results of Chapter 3, namely that if we cannot test an expression for context sensitivity we must evaluate a more overarching theory for coherency. I argue that building on the discussion of an expression-based approach to semantics we have a better formula for explaining intuitions regarding negative existentials. To do this, I argue that names originating in fiction refer to genuinely existing abstract artifacts allowing negative existentials to have truth-values. This results in an elegant semantics and the intuition that the negative existential is true is handled by the Gricean theory of conversational implicatures. (Abstract shortened by UMI.). [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]