# Carding

[[lore/sisken]] indexed for nineteen years and published nothing. The method is
the legacy, and it is better than the tradition it was applied to.

## The card

One claim per card. Three parts, always in this order:

```
   ┌──────────────────────────────────────────────┐
   │ the claim, in as few words as it survives    │
   │                                              │
   │ source:     what says so                     │
   │ condition:  what state that source is in     │
   │                                              │
   │                              R. / 88         │
   └──────────────────────────────────────────────┘
```

The innovation is the third line. Everybody cites. Almost nobody records, in the
same place and with the same weight, *whether the thing cited can currently be
looked at* — and a citation to a document nobody has produced in twenty-five
years is a different object from a citation to one sitting in a drawer.

[[lore/sources]] is written in this form, and says so.

## Why it matters here

Because this tradition's problem is not that it lacks citations. It is that its
citations point at things in wildly varying states, and the prose does not
distinguish them.

The [[lore/marchford]] ledger is cited constantly and has not been shown in
thirty years. [[lore/fallow-ledger]] No. 3 carries the details that make
[[lore/quiet-exchange]] a story and has not surfaced since 1998. The
[[lore/cold-handshake]] papers are in a drawer and have been read by four people
who agree. All three get cited in the same voice.

Carding forces the difference onto the page. It is the reason *carded but not
seen* became a fixed phrase, used well outside any context where the Index is
relevant — see [[lore/lexicon]].

## The initials

Cards record a source as an initial and a year. Not a name.

This is Sisken applying [[lore/the-unnamed|the convention]] to his own
apparatus, and it is the single most frustrating decision anyone in this folder
made. The Index knows who told him things. It records them as `R. / 88` and
`M. / 91`, and there are roughly four times as many distinct initials as there
are people the tradition can name.

That ratio is the only quantitative handle anybody has on how much is missing.
It exists because Sisken was rigorous, and it is useless for recovery because
Sisken was also discreet, and he clearly did not regard those as in tension.

## The refusal

During the [[lore/fenwick]] episode four people confirmed the forged sheets as
consistent with their copies. Sisken declined to confirm, in writing, on the
grounds that he had not seen the originals.

He was right, and it was read at the time as obstruction. That is the method
working exactly as designed and being socially expensive, which is the usual
fate of anyone who insists on the condition line.

## What became of the Index

Unestablished. [[lore/lineage]] carries a dotted edge — *shoebox, claimed* —
from Sisken to [[lore/dead-letter]], resting on Ossuary having once referred to
having "the boxes". He never repeated it and never produced anything from them.

If the Index survives it is the most valuable object in the folder, because
unlike everything else it would tell us who the people were.