# Theory

The shared foundation. Both books and all six essays rest on these two files.

| File | What it does |
|---|---|
| **`EVIDENCE.md`** | Every load-bearing claim sorted into four tiers, with sources. **Governs what may be asserted anywhere in the project.** |
| `THEORY.md` | The Relevance Theory itself — the argument, narrowed to what the evidence supports, with the cut material and reasons at the end. |

## The tiers

| Tier | Meaning | How to write it |
|---|---|---|
| **A** | Well-supported, multiple sources | State it, cite it, move on |
| **B** | Contested — real dispute in the literature | Present both sides *in the text* |
| **C** | Plausible, untested — the author's hypothesis | Mark as hypothesis, in the text, every time |
| **D** | Speculative or personal | Memoir or metaphor; cannot carry an argument |

**The failure mode this exists to prevent is a Tier C claim written in a Tier A voice.** The original contribution — the three-distance relevance model — is Tier C in its entirety. Saying so plainly, in the text, is what makes the rest of the book believable.

## Two things in `EVIDENCE.md` worth knowing without opening it

**There is an internal contradiction that has to be resolved before drafting.** The theory treats default-mode activity as the thing modern life starves us of, and therefore good. But mind-wandering predicts *lower* momentary happiness (Killingsworth & Gilbert 2010), rumination is default-mode-linked, and the best experiment in the literature (Bratman 2015) found nature helping by *quieting* self-referential thought. "We need more reflection" and "nature helps by reducing it" cannot both stand. The resolution — that the target is flexibility, not more of either mode — is chapter 7's job in Book One.

**Several claims were cut and should not creep back.** The foreign-policy argument, the weight-loss protocol, the diagnostic test's numeric scoring, the "I"/"me" labels for the networks, and the seasonal network mapping. Each has its reason recorded at the end of `THEORY.md`. If one reappears in a draft, that draft has regressed.