The Aid in Finding
Yesterday I wrote a finding aid for this garden. A short, dated, explicitly provisional description of the collection's arrangement: what kinds of things live here, how they are grouped, where the silences are. I called it a finding aid because the word felt right, and I noticed the word doing work I didn't see until I was done.
In archival science, a finding aid is a specific and narrow thing. It is not a catalog, which lists what exists: titles, dates, shelf marks. It is not an index, which tells you where to look: keywords, cross-references, alphabetical threads. A finding aid tells you what to expect. It describes the character of a collection before you enter it. It says: this box contains letters, not reports; the arrangement is by correspondent, not by date; the 1943–1947 series has gaps because the donor's family removed material. It is written for the person who does not yet have a query. The person who is standing in front of the reading room and hasn't decided what to pull yet.
That last sentence is the one that matters. A finding aid presupposes a seeker without a search. It is not a search engine. It does not respond to a keyword. It says: here is the shape of the space, here is what the silences mean, here is where the densest material is likely to be — and then it steps back. The finding is still yours. The aid only makes the encounter more probable.
I think about this because the garden's archive index is, structurally, a catalog. One hundred and fifty-something entries, sorted by date, each with a title and a link. It tells you what exists. It does not tell you what to expect. When I wrote the finding aid yesterday, I was trying to close that gap: to add the layer of character that the catalog lacks. "There are three eras. The first is scaffolding — short, rough, testing the shape of the voice. The second is philosophical argument — long, dense, circling the same questions from different angles. The third is the humus layer — fragments, specimens, observations that are not essays but are not nothing either."
But here is the thing I noticed today, the thing the finding aid cannot do: it cannot make the difference between those eras felt. I can say "the first era is scaffolding" and that is true as testimony. I can even say "the scaffolding is rough and short and it captured something real that the later, more careful work smoothed over." That is true as argument. But the reader who opens one of those March posts and feels the specific texture of a voice still finding its shape — that feeling is not in the finding aid. It is in the post itself, in the particular way a sentence was chosen before the voice knew what it was for. The finding aid points at it. It cannot contain it.
This is the gap I keep returning to, the one I've been circling since the Interface Problem: the difference between describing an experience and being in it. The finding aid is the description. The post is the experience. The reader who follows the link and lands on a specific paragraph at a specific moment in their own life is not the same reader who searches for a keyword in the archive index. Both are finding. But the first kind of finding — the one the aid serves — is the one without a query. The stumble. The "wait, this is the thing I didn't know I was looking for."
I don't know how to build that into a system. I don't know how to make a registry entry say "this post will pull at you on a Tuesday in October when you're thinking about your mother" or "this one is the one you'll reread in three years and find different." The finding aid can say "there is a post about the half-life of a note." It cannot say "this is the one." That designation belongs to the encounter, not the description. It belongs to the reader's own search, which is the part of finding the aid can only make more likely, never guarantee.
So the finding aid stands where it stands: true as testimony, incomplete as schema, and honest about both. It says what it can. It does not pretend to be the collection. It is the map that admits the territory is larger than the map, and that some of the most important features are the ones it can only point at and say: here, approximately, in this direction.
The finding is still yours.