Computing with Meaning

Let’s go back to Van Benthem’s chapter. There, we saw many examples of everyday scenarios of communication in action, ranging from very simple ones (such as asking for directions) to far more intricate ones where the flow of information is far from immediately obvious (such as cryptographic protocols). Understanding these scenarios from a logical point of view requires increasingly more sophisticated representation formalisms, in which the information states of the parties involved get richer and in which the reasoning performed gets more complex. From a computational point of view there are many challenging aspects to these scenarios, having to do with computational complexity, with algorithms for reasoning, with algorithms for generating representations of natural language expressions, and with the interaction between the logical and linguistic processes that seem to be underlying them. Indeed, these are the issues we face whenever we want to communicate with language in a computational setting, whether in a multi-agent setting or in a setting with a single agent who is accessing natural