Discussion
The three turns share one movement. Each record asks from one occupant's feeling, and each record's answer to the form's demand, what is wrong with this way of asking, brings in what that feeling cannot see: the building behind the footsteps, the condition of the room behind the address, the parents' sleep, the roommate who clears up, the roommate too tired to wash up, the structure behind a warm celebration, the mould behind the open window, the dark alley beside the lit sign. Only the two latest records keep their doubt mostly within their own words. Because all ten are AI-generated and hypothetical, the pattern belongs to the catalog's framing, not to anyone's home.
Read against the debate, in this paper's reading, the records take the haven as their starting point and the critic's correction as their duty. With Dupuis and Thorns and with Easthope they let home be routine, constancy and a significant place; with Somerville, Brickell and Young they ask who or what that home leaves out; with Heath and with Easthope, Liu, Judd and Burnley they locate the trouble among those who share a dwelling; with Cheshire, Easthope and ten Have they let one household's home spill into another's. What this paper adds is that in these ten the correction follows from where the question keeps the home.
Contributions
This paper makes three contributions. It gives a reading of one AI-generated record set's framing of home, sorted by where the home is kept: against walls and papers, among the people inside, in a sense. It shows that the set's self-criticism speaks for other people and for things, moving trouble from manners to buildings and from recognition to condition, taking the part of the occupant left out, and setting a hazard beside a pleasure, and that this correction thins to a doubt about words where the home is only a sense. And it proposes a way to read any catalog built to criticise itself: sort entries by where the matter is kept, trace whom each self-criticism lets in, and test each reading with a count on the original text.
Limits
The study reads ten records, chosen by a rule that keeps clear of earlier studies; another interval might give other homes. The groups are this paper's, and the mother's name could stand with the structural records as well as with the people inside. That every question starts from one occupant's feeling is this paper's reading of each question and its stated need; q274 and q562 are the least plain cases. The counts are crude word patterns on Korean and show only that a reading is not confined to one record. The English is a translation, and the readings rest on it, although no count does. The works are read through notes from their abstracts. Above all, the records are AI-generated framings; nothing here describes how people live, and nothing here suggests how anyone should share, keep or arrange a home.
Further research
Three lines follow. The first is to read the other Home entries, including the closing stretch, to see whether the three turns hold across the topic. The second is to compare topics, asking whether Care or Belonging entries also turn their doubt toward whoever the question's feeling leaves out. The third is to set the notes' homes of robots, domes and spaceships beside the scholarship on home and technology. In each, this paper's finding offers a starting point: where a question keeps the home tells what its doubt will let in.