ETAOIN SHRDLU (continued)

Continued from Part 1.

What the Shrdlu, Etaoin talking about?

Etaoin Shrdlu doesn’t come from the typewriter world, but typewriting history helps explain its emergence. Typewriters and hot-metal typesetting machines, used to push sequences of type molds together through keyboard action and then cast those moulds in hot metal, evolved during the same period.

One man, James O. Clephane, funded and helped beta test Sholes and Glidden typewriter designs in the mid-1800s and was a key mover in what became the Mergenthaler Linotype Company, the first maker of practical hot-metal typesetters. Clephane was a famous stenographer, friend of Abraham Lincoln and other presidents, in an era in which accurate notetaking and representations of someone’s speeches and of court proceedings could actually make you a celebrity.

His father was a printer, so Clephane had insight into that side of things. And, like telegraphers and other professions, he and other stenographers wanted a device that would let him turn his shorthand notes into something legible to others. But he also wanted the output to be persistent for record-keeping, and, most importantly, potentially reproducible without the additional costly, time-consuming, and often impractical step of typesetting.

This led him to fund several successive ventures, some of which were typewriters and some were keyboards that would result in a lithographic surface that could be used for printing or for making a typed-out papier mâché mold that could cast a full sheet (a plate) of type. None of that panned out until he met Ottmar Mergenthaler, a gifted machinist, who took the mold idea and radically transformed it. Instead of papier mâché, Mergenthaler’s Linotype would use brass matrices from which type was cast.

A typesetter sat at a Linotype, which had magazines (compartmentalized containers) fitted above it, fed by gravity, full of molds or matrices. As the compositor typed, molds fell through channels to line up in a composing area. When the line was finished, pressing a bar sent boiling lead rushing in to cast a line.

That’s all very interesting to those of us obsessed with type and typesetting, but what keyboard layout did Linotype use? QWERTY? No. It used one that predated Dvorak for relying on letter frequency. And two slightly diagonal rows close to the center of the keyboard were the 12 most frequently used letters in the English language in order of frequency: ETAOIN SHRDLU separately in upper and lower case.

Finishing a line with a flourish

Even as typewriters were consolidating around the QWERTY layout, plenty of other approaches remained in use, and Mergenthaler had his own ideas about everything. But it also made sense, because his Linotype magazines had to have a single matrix for each letter that would be set in a line. At the end of a line, an operator would trigger distribution, which used a remarkable system to return the matrices back into their proper slots in the matrices for re-use, but the compositor could be setting a new line while that happened.

The magazines were loaded with matrices corresponding to frequency and it would make the same sense that the keyboard would be organized around the same lines: the letters the keyboarder would type most often. Forget ergonomics: the machine was dominant and people would conform.

But we finally get to the crux: wherefore art though, Etaoin Shrdlu? When a typesetter would make a typo or other error, there was no backspace key. A matrix could be removed by hand, but that was inefficient, as was rearranging letters. And incomplete lines were harder to set because of how lines were justified, or had all their spaces made of equal width to fill a column flush on the left and right.

Instead, a typesetter would run their fingers down the rows e, t, a, o, i, n, s, h, r, d, l, u, adding spaces as necessary until the line was filled. I’m sure it was melodious, the matrices falling down in order, too, over the din of this mechanical wonder that occasionally strafed its operator with squirts of lead.

Because the Linotype set each line as a separate piece of metal, called a slug, he or she could have removed an offending line, or expected a proofreader to do so in organizations that went through proofing. But it was missed often enough that newspapers, magazines, and even Linotype-set books would have etaoin, shrdlu, cmfwyp, and other lines and combinations appear by accident. For fun, search on each of those terms in Google Books, and you’ll see what I mean, and see examples at the end of this post.

You don’t see these errors as frequently in books set in hot metal, because most books were set on a Monotype, a not-much-later competitor. Instead of casting a “line or type,” the Monotype was about “mono”: it cast each piece of type individually in a clever fashion. A typesetter sat at a keyboard that let them tap out pages at a time, which were punched as dot patterns in a roll of paper tape. This tape would be carried to a casting machine that read it (using pressurized air) to cast each letter from a matrix case. Errors could be fixed in a variety of ways, including swapping out individual pieces of type. And, in any case, the keyboard was QWERTY.

The Etaoin Shrdlu era ended as Linotypes and later competitors with the same approach hit the scrap heap. And digital phototypesetting systems allowed correction. I worked on one at my high-school newspaper that also set just one line of type at a time—but the keyboard had a delete key.

Examples of historical keyboard misprints

For your edification, more examples of Linotype insertions in old books and magazines.

And also read the bizarre poem, “Doctor, the Chloroform,” in The Typographical Journal (March 1912)