What is the relationship between letters-as-objects and the psychological states of the writer and subsequent reader? The psychologist James Gibson [5] takes a relational approach to perception, proposing a ‹direct perception› that includes the idea that objects (in this case letters) have a set of ‹potentialities› linked to a set of possible actions. He called these «affordances». For Gibson these affordances are not always simply embedded in an object, but can arise out of a mutual relationship between the object and the agent (the agent being the writer in one instance and the reader in another). Considering Emma Hauck’s letters in terms of Gibsonian ‹affordances› allows us to consider the possibility that the letter might not only be an object through which to communicate, that it might not need a recipient or reader (her husband). Instead the letter could be a trace of the act of writing, with the act of writing ‹affording› relief, or excitement. This way of perceiving the letters turns my discomfort about them remaining, unsent, in the Prinzhorn Collection on its head: it suggest that they may not have been inscribed with the intention of ever being sent to a reader. Of course, I will never know what those letters afforded Hauck, or how she intended them to ‹be› (sent to her husband, destroyed, kept, to be artworks). My perception of Hauck’s letters are as much influenced by my ‹situated cognition› as their instantiation was influenced by her ‹situated cognition› at the time of writing. By ‘situated cognition’ I mean that any cognition of the letters is embodied, situated and distributed. I can make some guesses about Hauck’s situated cognition at the time of writing, but any ‹model› I have for her state will necessarily be partial: I assume that as she wrote while in an asylum she had very limited control over her movements and environment, but I know nothing of the detail of her embodied state (medication, clothes, comfort levels for example) nor of her environment (temperature, noise, the kind of people and surfaces around her) nor of her psychological state. My situated cognition as a reader of her letters is inflected by my experience of being the unwilling recipient of letters written by someone with psychosis. My feelings of fear, anger, desperation on reading those particular letters, made me want to ‹arm› myself with knowledge about what such letters ‹mean›, and Hauck’s letters were sufficiently removed from my own experience that I could consider them in a less emotionally charged state. So, the unwelcome handwritten letters I received might, like Hauck’s, be presented as ‹symptoms› or expressions of psychosis (indeed this was the position that the legal team, defending the man who stalks me, took as a way of dismissing the threatening content of those letters). By defining them in this way they become models, when we take the word to mean a «phenomenon that accounts for its known or inferred properties and may be used for further study of its characteristics: a model of generative grammar; a model of an atom; an economic model.» [6]