In the autumn of 1985, off the outer arm of Cape Cod, a diver brought up a lump of concretion roughly the size of a bucket. It looked like a rock. Everything down there looks like a rock. Under the crust was bronze, and cast into the bronze were words: THE WHYDAH GALLY 1716.
Before that lump surfaced, the site was a scatter of interesting material off Wellfleet. After it, the site was the Whydah, the slaver taken by Samuel Bellamy and lost in a northeaster in April of 1717 with almost every man aboard, and it has been described ever since as the first pirate shipwreck anyone authenticated by name.
Weigh the bell as metal. A few hundred pounds of corroded bronze, and corroded bronze is not a commodity anyone gets excited about. Melt it down and the proceeds would not cover a single decent day at sea. Against the coins from the same scatter it is a rounding error.
And the bell is the reason anybody can say what those coins are.
Diagnostic and valuable are different objects
That is the whole of this letter, and it holds nearly everywhere. The object that identifies a site and the object that pays for the site are almost never the same object.
A gold ingot tells you about gold. It is honest about its weight and mute about everything else. Meanwhile the material that actually speaks tends to be worthless on any market: a bell or a gun with a founder's mark, pewter with a touchmark, a broken plate whose fabric and glaze put it in one factory in one decade, ballast stone that tells you which coast the ship loaded on, olive pits and fish bone and weevil-drilled grain that tell you the route and the season and how badly the crew was eating by the end. A coin does its most interesting work not as bullion but as a clock, because the newest date in an assemblage sets the earliest possible year the ship could have gone down.
The same asymmetry produced the strangest object ever raised from the sea. The Antikythera wreck gave up bronzes and marbles that any collector of the period would have recognized instantly. It also gave up a corroded lump that sat in a museum until it split and turned out to hold a geared calculating machine, and nothing in the ancient written record had prepared anyone for the idea that such a thing existed. Nobody lifted that lump because it was beautiful or salable. It came up because it was there and someone recorded it.
The uncomfortable part
Here is the strongest counterargument, and it is aimed directly at people like me.
Identification is not a purely scientific act when a commercial operation performs it. A named wreck is worth more than an unnamed one. Attribution multiplies the price of every object on the site, sometimes by an order of magnitude, and it does that for reasons the market is quite open about. So when a company like this one says it recovers the bell first because the record matters, notice that the record also makes the payday. The incentive to document and the incentive to profit are pointing the same direction, and I cannot separate them by telling you my intentions are good.
The academy has made this criticism for decades, including about the Whydah project specifically, and the criticism is not stupid. Entangled incentives are real incentives. Sooner or later they diverge, and the interesting question is what happens on that day.
The test, and it is observable from outside
So do not take anyone's word for it, mine included. There is a test, and it does not require trusting a single person's motives.
Every operation on earth lifts the bell. The bell is glamorous, the bell is the press release, the bell pays for itself many times over. The test is what happens to the four hundred objects that will never appear in any sale, catalog, or photograph.
Those five lines are unglamorous and they are the entire difference between an excavation and a very well-documented removal. They are also, every one of them, a cost with no revenue attached. The bell funds itself. The seeds never will.
The bill this comes with
So the standard this letter obligates the company to is narrow and specific, and it is meant to be checkable.
The diagnostic material gets budgeted before the season at the same rate as everything else, not funded out of whatever is left when the season goes well. It gets conserved on its own timeline rather than the sale's. It gets published whether or not the results flatter anybody, including us. And the assemblage stays coherent, because an object that has been separated from what it was lying next to has been quietly downgraded from evidence to decoration, and no one can undo that later.
I keep coming back to the shape of the thing. A bucket of rock with a name inside it, worth almost nothing by weight, holding up the meaning of every gold object on the same seabed. That is not a charming exception. That is how the record works, most of the time, on most sites.
The cheapest object on the site is usually the one that makes the rest of it true. Lift it first, and pay to keep it.