The aim of this text is not to present all the different systems of notation used in theories of so-called ‘formal semantics’. One could ask whether such a project would be feasible or even sensible. Not only do the notations used in ‘formal semantics’ vary from framework to framework, they also change with the development of the corresponding theories and from article to article — as can be expected of any living field of inquiry. The first objective of this text is to introduce some of the most currently used notational devices in Montague semantics, which can still be considered to be the reference theory in formal semantics. In this way, this text hopes to give some necessary background for the other entries on formal semantics. A second aim is to answer a question reminiscent of one that is asked in conversation analysis. There the question is: how can we find a transcription that reflects all the relevant factors of an utterance, meta- and paralinguistic clues included. Here a parallel question could be: given a natural language sentence, how do we transcribe or translate it into a formal language? Is there some kind of ‘automatic’ translation procedure that provides the interpretation of any natural language utterance? In order to answer this question, this paper will closely follow Cann (1993), an excellent description of what such a translation looks like. In the end, however, it will briefly analyze the role of context, to show that such a translation cannot be as ‘mechanical’ as is sometimes suggested and that it is impossible to neglect the role of context in deciding which proposition is expressed by a natural language sentence