The first phototypesetting machine

In the 1950s machines began to appear in the United States and Europe that set text in phototype, challenging the hot-metal machines that had been setting the type for books, magazines, and advertising since the late 19th century. Their production target was offset printing, which used a film image of the type to make a flat plate. Lithography. The three-dimensional metal type required for letterpress printing was no longer needed.

The phototype concept had been around for a while, but the first machine is usually identified as the Lumitype (see Wikipedia), invented in Lyon, France in the late 1940s.

But this history ignores events two decades earlier. In 1924 Nobuo Morisawa (1901–2000) and Mokichi Ishii (1887–1963), filed a patent for a phototypesetting machine that could set Japanese.

That first machine was operated by hand and set display type. It was really an astounding achievement since the language has a minimum of 2,000 glyphs—and the current standard calls for more than 20,000. But each fits in a square, unlike Latin glyphs. The two partners set up a factory and started selling the machines in 1929. Morisawa, an engineer from a working-class family, was the one who realized a Japanese phototypesetter would have the advantage of set character widths, but there were the minimum number was 20x the Latin character set at the time. He adapted the mechanics of the Nippon typewriter which had appeared in 1917, replacing the tray of metal characters with a film grid containing 2,400 glyphs. (Additional glyphs were available on a second grid.)

Tech and design

Ishii focused on the type design, and his fonts are still used and admired today. With wealthy parents, he provided the capital and the connections to get the firm started. And the designs were essential. The first company was called Ishii Phototypesetting Institute.

There was a market: Lithographers who printed art and advertising posters; agencies who used drawings and lettering for ads that were then made into photoengraved “zincs” for letterpress printing. But it would be 30 years before offset lithography challenged letterpress for big publishers.

In Japan there was even greater inertia in the technology of metal type and letterpress, due to investments in equipment, infrastructure, and labor. Somehow the new company survived through the stress of the 1930s (economic depression and war in China and Russia), but the partners split up.

WWII obliterated type inertia in Japan, and after the war Morisawa rejoined Ishii and began to work on faster typesetters, with keyboards. But the two fell apart again; in 1948 Morisawa started his own company. Ishii changed the name of the original firm to Sha-ken.

Japan’s phototype era

Both were positioned to take advantage of the postwar boom, as the by Japanese printers and publishers embraced offset to rebuild. Phototype added momentum, since there was no reason to re-invest in the tons of 3D metal type and giant composing rooms when phototype was a well-developed alternative. But Sha-ken took the lead, since it had Morisawa’s machines, and held onto the fonts that Ishii had drawn or commissioned.

Morisawa built a new machine at the end of the 50s that could set the proportional fonts for kana, the Japanese writing system based on sounds—and of course Latin fonts.

Japan held onto phototype for the next 30 years. Sha-Ken was the leader, and built a big studio, continuing to dominate Japanese type design after Ishii’s death in 1963.

The next big change was the advent of desktop publishing in the mid-80s. Adobe came to Sha-Ken to license their fonts for Postscript, but Ishii’s descendants turned them down.

Nobuo Morisawa understood the inevitable changes in technology, realizing the next step was fonts for screens. The company quickly agreed to work closely with Adobe, a relationship that continues.

The next three decades saw Morisawa emerge as the leader in Japan. The company replaced its hardware business model with type design and font licensing. And it moved beyond Japanese to both Asian and Latin scripts. The prestigious biannual Morisawa Design Competition recognizes the best typefaces in all languages. The company made strategic alliances with type foundries in other countries in the last decade, and acquiring the leading foundry in Taiwan, Arphic, and a prominent foundry in the U.S., Occupant (a founding partner of Type Network).

To close the circle, Morisawa has agreed with Sha-Ken to release, in the OTF format, the fonts drawn by Ishii for Nobuo’s first machine.

Meanwhile, in the West

Europe and the United States lagged Japan for two decades. To fact-check my assertion that Morisawa produced the first successful phototypesetting machine, I did some research. I found earlier references to the idea of phototype. Akihoko Morsiawa, now the company’s chair, has pointed to an ad in the Penrose Annual for something called Photoline. (I could find nothing about it.)

One early European attempt was the Uhertype in the late 20s, a manually operated photo composition device designed by Edmond Uher, a Hungarian living in Switzerland. It was a sturdy vertical machine that looked a lot like the first Morisawa typesetter. And it set text, like Morisawa’s second-gen machine. It is remembered today because Uher commissioned original typefaces in the early 1930s for itsfont disks. The Bauhaus designer, Joost Schmidt, did a contemporary sans serif (since digitized). Jan Tschichold, the great typographer, produced several others for the Uhertype. One has been revived and was named Tschichold. A few machines were built, and a specimen book survives, but the venture was a failure.

Then there was the Rutherford Photo-Lettering machine in the U.S., patented in 1936. It was complicated and manually operated. Sales to the limited market of engravers and lithographers were slow. The machine came into its own in the 1950s when one of the inventors, Ed Rondthaler, set up the typesetting company, Photo-Lettering in New York, with a rich catalog full of classics and contemporary designs.

The same year in Britain, George Westover, a Monotype manager, filed a patent for the Rotofoto textsetter, which used punched tape from Monotype keyboards for input. Its film fonts were on a grid. A prototype machine was installed at the London College of Printing, but Monotype abandoned the effort.

First generation

The manufacturers, Mergenthaler Linotype and Monotype, thinking that customers were not ready to abandon the familiar casting machines, tried to adapt them for phototype. The first called, appropriately, Fotosetter, was based on the Intertype (a clone of the Linotype machine) and appeared in 1946. One was installed at the Government Printing Office in Washington, and the same year it was used to set a pamphlet for the National Gallery. But that was the end of it.

In 1950, Mergenthaler brought out a similar device called the Linofilm, and Monotype followed with the Monophoto in 1955. These first-generation photosetters took the old metal matrices and fitted them with tiny photos of characters. When I saw a Linofilm, I thought, “How can these little pieces they didn’t, and this effort by the old companies was doomed.

Second generation

Lumitype, known in America as Photon, was the first successful second-generation photosetter, selling its first machine in the early 50s. The machine was computerized and had vastly fewer moving parts than a Linofilm. The initial system used a keyboard connected to the typesetter. Fonts were photo negatives on a disk that rotated as fast as a wheel on a car. The computer would pick out a character, adjust a lens to change the size, fire a strobe light, and direct the image to join a line of characters roll of photo paper. After setting to the right width, the roll would advance, and the machine would start on the next line. The second model Photon, with separate keyboards, could set type at the rate of 28,000 characters per hour.

The old guys saw the change was working and brought out second-gen Linofilm and Monophoto machines, with networked keyboards. Interest from publishers and printers was warming up in the 50s since their equipment was wearing out. Some of their big customers invested in the new systems.

Third generation

The trend was clear: Better computers, fewer moving parts. More powerful, simpler systems, such as the Mergenthaler Linotron in 1960, used cathode ray tubes (CRTs) to image type from film fonts, eliminating lenses and whirring disks. The type images were getting sharper.

In 1970 a new Linotype hot-metal machine cost $30,000. It was probably not a coincidence that their smallest, computerized phototypesetter, called the VIP, was also $30,000. This left room for companies like Compugraphic, Varityper, and A/M (Addressograph Multigraph), which made smaller and cheaper machines for small printers and design studios. Because the old guys would not license their fonts, these newcomers offered some (purposefully?) bad imitations.

Next, fonts went digital. Also in the 1970s, Dr. Rudolph Hell’s Digiset typesetter (Videocomp in the U.S.) fed font bitmaps directly to the CRT. Autologic used that concept to take over the newspaper market, by providing a study, fast, and economical machines for newspapers.

So, no more film fonts. By the 1980s the only photo in new typesetting equipment was the film or paper used to image columns and pages. An era was ending. We were looking at type on screens, without knowing that was the future. Morisawa was one of the first type companies to realize this, just as it was the first to bring phototypesetting to market.