Method
When a reader asks what I actually do with a document, the honest answer is the same regardless of which one it is. I work it through six pillars, in order — identification, archive strategy, authentication, historical context, valuation, verification — with one pre-step in front of them. The pre-step is confirming a buyer or an institutional interest before the deepest work begins. I learned that discipline the hard way on an earlier project, and it now sits ahead of everything else.
1. Identification
The catalog usually names one signer. The arbitrage almost always lives in the others.
Read every signature, every date, every clerical mark, every physical attribute — not just the marquee name. A document filed in the mind as “a Hancock autograph” doesn’t get asked new questions for twenty-eight years. The discipline is asking, every time, what is actually on this page?
2. Archive Strategy
Most of the research happens in correspondence with institutional archivists, and most of that correspondence fails because it’s misdirected before it’s sent. Five working rules: write to the institutional inbox, not an individual; pull the finding aid yourself before writing; ask no more than two specific questions per email; ask for everything the warm conversation allows; and never mention auction or sale — the framing must be research, always.
3. Authentication
The gate. Without a viable authentication path, even a great find is uninvestible. Three moving parts to negotiate before submitting: LOA scope (every signature, not just the famous one), database coverage (which authenticator holds which signer), and exemplar sourcing (identify the institutional source of handwriting exemplars in advance and provide them in the briefing). A written briefing to the authenticator with historical context and specific questions produces a stronger letter at the same cost.
4. Historical Context
The work that turns a document into a primary source. Research what happened the day before, the week before, and the month before the document date — macro events often explain why a routine instrument was issued at an unusual time. Map the relationships between the people on the page, not their individual biographies. Check whether the named parties appear in a diary, letterbook, or account book that might corroborate the transaction. And remember that lesser-known figures often have richer archival records than their fame suggests — the most original research usually lives there.
5. Valuation
Most consignors think of one buyer. The method insists on at least three. Multiple signatures create multiple collector audiences. Corporate archives and institutions are distinct buyer categories that have to be named explicitly — “banking-history collector” is not the same as “Barclays Group Archives.” Academic corroboration lifts realized value at auction and opens direct-sale paths that wouldn’t otherwise exist. Institutions often pay more, net of fees, than auction does — and most consignors don’t approach them. Anniversary and timing windows are real market intelligence.
6. Verification
The discipline that distinguishes research from research-flavored speculation. Every quote in the working notes carries a source-level tag — manuscript-verified or secondary. Every figure is independently verified, with the access date recorded. A running open-items list stays separate from the main writeup so nothing unsupported slips into the final version. And verified negatives are positively-valued information — when an archive replies “we don’t hold that period,” that’s a finding, not a failure.
One thing verification produces that took me longer than it should have to internalize: research is a risk-asymmetry lever, not a bid-ceiling lever. Deep research doesn’t change what other bidders will pay. It changes the resale ceiling on the back end, the probability that an academic-corroboration path activates, and the institutional direct-sale floor. Same maximum bid. Different expected value across the distribution of outcomes.
Why the method, not the document, is what this is about
The pillars are the same on a $400 check that turned out to be worth $600 at auction, on a $245 sight draft that turned out to be a primary source on the morning after a colonial riot, and on a Hamilton letter that turned out to be a document Founders Online lists as “Letter not found.” What changes is the document.
That’s why this is research, not dealing. A dealer holds a document and works to sell it. An academic holds a question and works to publish on it. The six-pillar framework holds the document as a small bounded problem and works to understand it — completely, transparently, with every claim sourced — and only then asks where it should go.
If you’d like to read the essays this method produces, subscribe to The Counting House — one essay by email each Wednesday.