I have never seen Case 1, Box 6. I have never been in the room where it is kept.
Almost everything I know about the documents I write about, I know at one remove — through finding aids, catalogue entries and inventories written by somebody who had the box open when I did not. That is the ordinary condition of working at a distance and it is mostly fine. The description stands in for the thing, and you get used to reading one as if it were the other.
It stops being fine the moment the description says something you cannot parse. Then you are not slowed down, you are stopped — not because the archive is shut, but because the single instrument you have for seeing inside it has handed you a sentence you cannot use. And there is no way to work around that from your own desk, because the problem is not in the collection. It is in the only account of the collection you can reach.
Three weeks ago that happened to me, over four words.
In my essay of 23 September I quoted a sentence from the finding aid to the Simon Gratz Collection at the Historical Society of Pennsylvania, and said I did not understand it.
The sentence describes what happened when Gratz bought a rival collector’s autographs in 1881:
Despite his misgivings about the Sprague collection, assimilation of the material left its mark upon the Gratz Collection: Sprague’s system of arrangement by category survives almost intact in the first section.
My question was simply: the first section of what? The collection runs to a hundred and ninety linear feet — six hundred and thirty-three boxes — grouped into Cases. Nothing in the inventory is labelled a section. I had been reading in Case 1, where Francis Hopkinson sits in Box 6, and I wanted to know whether the arrangement I was looking at was Gratz’s or a dead man’s that Gratz had inherited.
I wrote in that essay that I was not going to decide it on my own behalf, and that it was going in my next letter to the people who hold the collection.
I sent the letter on 30 September. The answer arrived on 5 October, and it is better than anything I had guessed at.
What the archivist found
Cary Hutto, the Society’s Director of Archives, went looking in HSP’s own institutional archives — the records the institution keeps about itself.
She found a page from a draft of the Society’s 1991 Guide to the Manuscript Collections of the Historical Society of Pennsylvania — the third edition, after 1940 and 1949, with a supplement to follow in 1993. On that page, Simon Gratz’s autographs are a single collection, numbered 250, and the entry opens like this:
250
Gratz, Simon. Collection, 1343-1928.
The collection has two sections: one is arranged by category of achievement; the other is alphabetically arranged.
Eight paragraphs down the same entry comes the sentence I had been staring at. And in the draft it is perfectly clear. The first section is the one arranged by category of achievement. The second line of the entry says so.
What happened next is not recorded anywhere, and I want to be careful about that.
What is in print is two states of the same thing. In the draft, Gratz’s autographs are one collection with two sections. In the finding aids HSP publishes today there are two collections — 250A, arranged by category of achievement, and 250B, arranged alphabetically — and they are separately described. The step between those two states is not documented. Hutto found no annotated drafts recording it, and she was explicit with me about that: “this conclusion here is my own, particularly about the splitting up of Collection 250, since I don’t see that we have additional or annotated drafts of the guide to consult. Still, it is logical enough based on what is in actual print!”
So this is an inference — hers, and now mine. A good one, because the draft and the current finding aids sit on either side of it and the shape of the change is plain. But I would rather tell you it is an inference than let the story run smoother than the evidence does.
And the four words stayed.
Nobody went back through the text of 250A to remove “in the first section” — perhaps because, once 250 had become 250A, the phrase was not wrong, exactly. It was just pointing at something that no longer existed as a thing you could count to. The collection is the section. Within 250A there is no second one to be first of.
In Hutto’s own summary: “The collection is the section. Most of the section was arranged by Sprague. Hence, most of Collection 250A was arranged by Sprague.”
So the answer to my question is yes, and the answer is thirty-five years old
Francis Hopkinson, in Case 1, Box 6, filed under OLD CONGRESS, is almost certainly sitting in William Buell Sprague’s arrangement rather than Gratz’s. A system bought second-hand in 1881 — ninety thousand items for twenty thousand dollars, “probably twice as much as it was worth” in Gratz’s own note, quoted in the finding aid — and still, a hundred and forty-five years later, the order in which the boxes sit.
That was the guess I made in September and declined to assert. It turns out to have been right, but I want to be careful about the part that still is not settled, because Hutto was careful about it too:
We don’t know if this covers every single category, or material at the end of the collection listed by name, organization, something other than a recognizable “category of achievement.”
Most of 250A is Sprague’s. Not demonstrably all of it. The edges are still open.
What I actually find interesting here
It is not that a finding aid contains a stale phrase. Finding aids are long, they are revised in pieces, and a document describing six hundred boxes will carry some scar tissue.
What I keep turning over is what kind of error it is.
I have spent this year writing about catalogues that describe their contents incorrectly — a man’s name misread, a signature missed, a volume that no search term will return. Those are errors about the objects. Somebody looked at a thing and wrote down the wrong thing.
This is different. Every word in that sentence was accurate when it was written. The structure it describes was real. What appears to have changed is that the structure was reorganised and the description was not — and the sentence went on quietly working, grammatically intact, for at least thirty-five years, directing readers toward a division of the collection that the finding aid no longer describes.
A catalogue is not just an opinion about a collection, which is what I said last time. It is an opinion with a date on it, sitting inside a document that gets edited at a different speed than the thing it describes. The description and the described drift apart, and the joins do not announce themselves.
And here is the part that should give any researcher pause. That sentence was not ambiguous. It was confident. It told me, in plain English, that something survived “in the first section,” and the only reason I noticed anything was that I went looking for Case 1 and could not find a section. If the collection had happened to have an obvious first part, I would have read the sentence, believed it, and been wrong in a way I could never have detected.
The rest of it
Hutto also confirmed the second thing I mentioned, which I had offered as a throwaway: Case 1, Box 6 and Case 1, Box 15 are listed in the finding aid as covering the same span of names. She confirmed the listing reads that way.
And she ended her note with a sentence I did not expect, and which is the real reason this essay exists:
Based on what I found, I think we have some revisions to make!
An outside reader asked about four words. An archivist went into her own institution’s past and came back with a reason to change the finding aid.
I mention it because the posture I most often see toward institutional catalogues is suspicion, and I do not think suspicion is the useful stance. The useful stance is to ask.
When I wrote to thank her, she wrote back: “Editing our finding aids is a regular occurrence. Whether it’s to attend to errors or add new information, they are living documents. We aim to be accurate but have never claimed perfection.” The thirty-five-year-old phrase I went chasing is not evidence against that. It is what a living document looks like when nobody has had a reason to open that paragraph in a while.
A scheduling note. These essays now come twice a month rather than weekly, so that more of the time goes into the archives and less into the keyboard. Nothing else changes. Still free, still one document at a time.
Thanks for reading.
— Paul
Sources. The draft page quoted above is page 66 of a draft of the Society’s 1991 guide to its manuscript collections, Box I-8-18, “MSS Dept. – Projects, Guide to the manuscript collections project,” HSP Institutional Archives — formally, The Historical Society of Pennsylvania, with the Balch Institute for Ethnic Studies Institutional Records (Collection 3136). It was found, and its significance worked out, by Cary Hutto, Director of Archives, who is quoted here with her permission. The live finding aid is the Simon Gratz autograph collection, 0250A, Historical Society of Pennsylvania — 190 linear feet, 633 boxes, 1343–1928 — consulted 14 October 2026. HSP expects to revise it; the phrase quoted here may no longer appear.