Researchers at the MIT Libraries have managed to read a 300 year-old letter without even opening it.

On July 31st, 1697, Jacques Sennacques sent a letter to his cousin Pierre Le Pers, a French-born merchant living in The Hague. He needed a certified copy of the death notice for Daniel Le Pers. In order to ensure the message reaches its destination unaltered, he did what everyone at the time did, and ‘letterlocked’ the document. This practice involved the use of complex folding to prevent a third-party from tampering the letter.

He did a very good job of that: the letter has remained locked ever since. But, through the wonders of modern science, we’ve been able to take a cheeky peep at Jacques’ private correspondence without leaving a trace.

Reading through the cover

“Virtual unfolding is a computational process that analyzes CT scans of folded letterpackets and creates a flattened image of their contents,” the team explains.

“Our virtual unfolding pipeline generates a 3D reconstruction of the folded letter, a corresponding 2D reconstruction representing its flat state and flat images of both the surface … and each letterpacket’s crease pattern.” 

Letterlocking is a somewhat lost-to-time tradition. In an age before the Internet, reliable post offices, or envelopes, people used letterlocking to ensure their correspondence wouldn’t be tampered with. It’s basically a fancy folding technique used throughout the globe at one point that involves folding a letter in such a way that it couldn’t be opened without getting torn. If a note reached its destination torn, the recipient would know that someone opened it and potentially read or tampered with its contents.

But now, that’s exactly what we’ve done, without ripping the paper in any way.

The researchers from MIT and King’s College London used cutting-edge dentistry X-ray machines to produce 3D scans of the letter, to see exactly how the paper was folded. An algorithm developed by one former and one current MIT student then used this data to produce images of the letter’s crease patterns, and even readable images of its contents.

“Letterlocking was an everyday activity for centuries, across cultures, borders, and social classes,” said Jana Dambrogio, the Thomas F. Peterson Conservator at MIT Libraries and one of the authors of a paper.

This approach has been used on scrolls, books, and documents with one or two folds in the past, but never on something as complex as a locked letter.

While the team could have simply ripped the letter open to read it, they wanted to preserve it as it is. Most of all, they wanted to keep its folds and creases in the exact pattern it was given, as a way to preserve an excellent example of letterlocking. As far as we know today, the first examples of this practice come from the Vatican Secret Archives, in documents dating back to 1494.

“One important example is the hundreds of unopened items among the 160,000 undelivered letters in the Prize Papers, an archive of documents confiscated by the British from enemy ships between the 17th and 19th centuries,” the study reads. “If these can be read without physically opening them, much rare letterlocking data can be preserved.” 

Before the researchers’ computational analysis, they only knew the name of the intended recipient written on the outside of the locked letter. 

The paper “Unlocking history through automated virtual unfolding of sealed documents imaged by X-ray microtomography” has been published in the journal Nature Communications.