This study asked what a scene written for comparison can show about idle talk at work, and how such a scene marks what it has made up. Its answer, from eleven GAEKJU scenes that are all synthetic, is that a written scene can show where a measure stops, and that GAEKJU's scenes mark their making well in their evidence, poorly in the judgments of their longer form, and in an unexpected way in their counter-conditions.
Discussion
The debate on imagined cases has two worries, and the scenes answer them differently. The thought-experiment debate asks, with Norton, where the information in an imagined case comes from. GAEKJU's scenes give a clear answer in their evidence. The information comes from one condition the writer has added, a timetable of danger, a listener who is an animal, a room without light, and from what follows from that condition. The scenes say which condition it is and what they did not test. Read this way, they are neither reports nor proofs. They are arguments about a measure, and their conclusion is limited and useful: counting talk by its words would miss the work talk does wherever that work depends on timing, on the body, on a missing listener or on a voice in the dark.
The vignette debate asks how far a response to a written scene can stand for conduct in a real one. GAEKJU's scenes have no responses to weigh. The writer of the scene is also the reader who judges it, and that is where the second finding lies. The eight longer scenes give their judgments in the past tense of events their own talk supplies: a reason revealed over biscuits, a driver who turned towards the station, a second-year who learned the clock of a song. Those judgments are true of the scene and of nothing else, and nothing in the scene could have turned out otherwise unless the writer had written it so. The three shorter scenes, the ones that label their evidence as synthetic, judge with a refusal: the scene alone cannot show the gain, the greeting alone cannot guarantee understanding. The difference between the two forms is not one of honesty in general, since the long scenes are scrupulous in their evidence. It is a difference at a single point, the move from what the scene has shown to what the scene is taken to have found.
The counter-conditions then complete the picture. Eight of the eleven name a blasting log, an hourly detection record, a colliery's safety log or a duty roster against which the reading could be checked. The settings are written, so those records do not exist, and the counters work as a signpost rather than a test. In Dancy's terms, an imaginary case can reveal the importance a feature can have in a real case, and no more; the counters, as this study reads them, are where the scenes say what the more would require. It is a careful gesture, and an unfinished one. A reader is told where to look, but the scenes do not say who would look, or when, and there is nowhere real to look.
Contributions
The study adds three things to the debate on imagined cases, and through them to the reading of GAEKJU's public records.
First, it describes a third kind of imagined case. GAEKJU's scenes are neither the thought experiment that tests a theory of nature or duty, nor the vignette that collects a respondent's judgment. They are written to test a measure, and what they can show is the edge of that measure. This study reads Gendler's definition, reasoning about an imaginary scenario to confirm or disconfirm a hypothesis, as fitting them once the hypothesis is taken to be a claim about counting.
Second, it separates two moments in a single synthetic record: the declaration of what was made up, and the judgment of what the made-up scene shows. In GAEKJU's eleven scenes the first is careful throughout, and the second is careful only in the short form. A synthetic record that judges its own invented ending as a result confirms itself; one that refuses to do so leaves the question where a reader can still take it up.
Third, it shows that the scenes place every test outside themselves, in records that a written scene cannot contain. This is the scenes' answer to the vignette debate's worry about the distance between a scenario and the situation of action: they do not close the distance, they point across it.
Limits
The limits are those of a small and strict reading. The eleven scenes are one kind of talk out of seven, and they were chosen by an exclusion that removed most of GAEKJU's synthetic scenes because they refer to the figures of a source record. The scenes that remain may differ in form from those left out, and nothing here describes the synthetic kind as a whole. This study quotes no sentence about a source record, and so it reads the scenes without that part of their evidence. Every quotation is this study's translation from the Korean, and another translation could shift a tense or a nuance. The test for formulaic repetition reads low on Korean text, so the scenes may share more form than the count shows. The works are read through their abstracts and stated arguments, and Dancy's paper through the abstract of its reprint; where this study goes past those notes it has said so. Above all, the scenes are synthetic. This study has said nothing about what happens in mines, barns, canteens, factories or control rooms, and nothing it says about the scenes should be read as a claim about those places or the people who work in them.
Further research
Three lines of work follow. The first is to read the other six kinds of talk in GAEKJU's synthetic scenes, and the scenes this study excluded, with the same attention to the two moments of declaration and judgment, to see whether the long form's habit of judging its own ending holds across the library. The second is to set the synthetic scenes beside GAEKJU's outside cases on the same kind of talk, since the outside cases rest on published studies and the scenes on writing: the comparison would show whether the studio marks what each kind rests on in the same way. The third is the use the vignette literature would suggest. The scenes could be shown, as written, to readers who work in comparable trades, and their readings collected; that would turn GAEKJU's scenes into vignettes in the sense of the survey literature that Finch describes and give the counter-conditions something real to meet, though it would also be a different study with its own ethical and practical questions.
The contribution, stated again at the end: a scene written for comparison can show where a way of counting stops, and GAEKJU's synthetic scenes show it well when they say what they changed. They are least reliable when the longer ones judge the ending their own talk supplies, and they leave every test in a record that no written scene can hold.