In this segment, taken from his 2008 talk at the Museum of Modern Art, New York, for his exhibition The Printed Picture, Richard Benson walks us through the use of planographic printing in photography. He gives an overview of collotype, pochoir, the nature of offset printing, the offset duotone, and the offset tritone. Pictures associated with each of the main themes presented in the segment can be found by clicking on any of the section links below.
Photo offset lithography. Carlo Taylor. Barry Torgerson’s ’38 Chevy from The 12 Port News magazine, September/October 1989. 1989. 8 x 6 1/4” (20.3 x 15.9 cm). The Museum of Modern Art, New York. Gift of Richard Benson © Inliners International, Inc.
A single-color 100-line-screen black halftone with blue spot-color on coated paper. To reproduce photographs, offset uses the same halftone screen as letterpress.
The offset press has ended up with three cylinders, stacked one above the other. On top is the plate cylinder, around which is wrapped a thin aluminum plate holding the image in a polymer coating. Ink and water rollers ride on the plate, providing the ink/water balance necessary for the image to form on it. Beneath the plate cylinder is the blanket cylinder, which is wrapped in a cloth blanket with a thick, smooth rubber face. The ink from the plate transfers to this blanket with each revolution. Beneath the blanket runs the “back” cylinder, a polished steel cylinder with grippers to hold the paper, which bends around the cylinder as it is printed. All three cylinders are connected by a gear train so that they run in perfect synchrony.
A sixteen-times enlargement of a section of a magenta contact screen, similar to the one used to make the magazine cover illustrated on the previous page. These screens, made photographically from the older glass halftone screens, were colored magenta to allow contrast control in the reproduction by filtering the light used in the copy camera.
The introduction of the rubber blanket between plate and paper has six crucial effects, which we must list: (1) The blanket can “kiss” the paper relatively lightly, transferring ink without distorting the sheet. (2) The soft rubber can conform to, and print beautifully on, a rough-surfaced paper. (3) The blanket acts as a moisture barrier, keeping the water spread on the plate away from the paper, so that it does not swell and change size. (4) The paper—abrasive over time—never touches the delicate surface of the plate. (5) The image is reversed a second time, so that a right-reading, emulsion-down halftone negative, produced directly in a copy camera, prints correctly. 6. Small alterations to the size of the printed image can be made by changes in the underpacking of plate and blanket.
A section of the halftone enlarged the same amount to show the remarkable sharpness of the offset process. In use the screen is angled at forty-five degrees, which gives the least-apparent dot pattern to normal vision.
These mechanical advantages are coupled with a tremendous change in the halftone that was possible as soon as it was printed lithographically. The old relief plates that were used to print halftone-based reproductions could never hold a screen ruling finer than about 150 dots per linear inch, because of technical problems in etching, inking, and transfer. The new lithographic plates, holding images on a plane metal surface with no relief, can easily maintain a cleanly printing screen ruling of 300 dots per linear inch. Such a fine dot is nearly invisible to even the best naked eye, and completely invisible to anyone over forty years old. In offset printing the halftone finally drops out of sight, as ink is laid down in patterns so fine that we fully accept them as tonal.
Collotype. Photographer unknown. Shingle-style house. c. 1910. 12 11/16 x 11 3/4” (32.2 x 29.9 cm). The Museum of Modern Art, New York. Gift of Richard Benson.
Flat-plate photogravure came about as a modification of the older technique of etching. The plates, paper, ink, and presses were in place; all that was required for photogravure was the development of a method to etch photographic information into the printing plate. The next ink-and-photographic process to come along, the collotype, was completely new. Being planographic—printing from a flat surface—it was related to stone lithography. It shared with that process the characteristic of using the incompatibility of oil and water to describe the printing areas, but beyond those two similarities collotype was a brand-new creature, unlike anything before it. Collotypes were made as early as the 1860s, barely twenty years after photography was invented, and the process was well established in France by the mid-1880s. Commercial collotype died in the 1960s, and when I first started work as a printer, in 1966, at The Meriden Gravure Company in Meriden, Connecticut, it still had three collotype presses running, doing book work.
Detail of Collotype. Photographer unknown. Shingle-style house. c. 1910. 12 11/16 x 11 3/4” (32.2 x 29.9 cm). The Museum of Modern Art, New York. Gift of Richard Benson.
Collotype prints were backward, reversed left to right, unless the copy negatives from which they were made were reversed through a prism. Occasionally photographers would shoot with glass plates put into their cameras backward (that is, with the emulsion facing away from the lens) to produce an original camera negative from which to make the collotype plate. The sharp clarity of the print on the previous page (detail shown to the left) makes me quite certain that that was done for this picture.
On my first visit to the company I was given a tour of the shop, and my guide, the ancient owner of the plant, let me know that it had long since been accepted that collotype could only be properly practiced by workers of Teutonic extraction. (I’m not kidding about this belief—the three pressmen were named Allendorf, Zande, and Brecklin.) The process was complicated, terribly unpredictable, and erratic at best; but when everything went well, a fine collotype could hold its own against the best of photogravure. Collotype presses were huge, holding a flat glass printing plate on a massive bed that ran on steel rollers, like a railway truck.
The bed moved beneath a set of ink rollers and then under a cylinder big enough to hold the large sheet used for book work. The paper wrapped around this cylinder came into direct contact with the printing plate, and the ink was transferred in a single impression. This arrangement—of the paper directly meeting the plate—was the weak link in collotype, because it was a lithographic process, using a moist plate, and the dampness striking the paper inevitably changed the paper’s dimensions. This doomed collotype to be a single-impression process, since the sheets could never dependably be registered for multiple impressions. We find some color collotype, but it is rare, and almost never in perfect register. The old-timers who printed collotype used to say that the way to do a fine limited edition in the process was to print twice as many sheets as needed, spread them out on the floor, and pick out the best ones. There was some truth in this: because collotype plates were dampened manually during the printing, the tone of the impressions varied from sheet to sheet.
Collotype. Rembrandt. The Officer’s Wife. 1636. 10 7/8 x 8 1/4” (27.6 x 21 cm). The Museum of Modern Art, New York. Gift of Richard Benson.
Collotype was used for process-color reproductions but the color impressions seldom registered properly. When correctly done, as in this reproduction of Rembrandt’s painting The Officer’s Wife, process collotype could be great, but pochoir is the more common form in which collotype was used to make color pictures.
Collotype was often used to print the black and white skeleton for hand-colored prints. Called “pochoir” prints, these were usually made on good acid-free paper, then colored with stencils. They were most often made for the decorative market but also have a long history of turning up in books. The technical problem in hand-coloring is that the monochromatic framework that holds the applied color must be pale in those areas to be colored. This means, for example, that a pale blue or red tone, which would normally become a light gray in a black and white reproduction, must print as white—the gray would make the applied color “dirty.” Collotype was suited to pochoir for two reasons: its random grain was invisible to the eye, so there was no halftone dot to make the owner think it was a plain old reproduction, and the printing plate was generated from a continuous-tone negative, which could be made with a filter that eliminated most of the dominant color in the original.
Pochoir. William Blake. When the Morning Stars Sang Together from the Book of Job. 1825. 8 x 6 3/8” (20.3 x 16.2 cm). Printed by Trianon Press. 1976. The Museum of Modern Art, New York. Gift of Richard Benson. This print is a hand-colored collotype proof of a Blake engraving that was colored by someone other than Blake himself.
If the print being reproduced was heavy in the reds, a red filter could be used to make the negative and those areas would be light, allowing the hand coloring to show. The difficulty here was that, if the print had any blues, greens, or cyans in it, they would be rendered dark, making the hand-coloring for those areas worse. This could be somewhat countered by making exposures through two filters—we called it “split filtering”—but even so a good deal of handwork was necessary to get all the colors to disappear. When process color printing was developed, using the three subtractive primaries, it turned out to need a fourth printer using black ink. The task of eliminating color densities in this black printer was a huge challenge.
Detail of Pochoir. William Blake. When the Morning Stars Sang Together from the Book of Job. 1825. 8 x 6 3/8” (20.3 x 16.2 cm). Printed by Trianon Press. 1976. The Museum of Modern Art, New York. Gift of Richard Benson. The pale and open black and white impression of the collotype is visible in this enlarged detail.
In the older pochoir a solution had been cobbled up through imaginative filtering, careful developing of the negative, and then skilled handwork to complete the job, but when color was printed from the primaries, and a skeletal black was added, making this black printer right was almost impossible. One of the most amazing benefits of digital technology in the printing trades is that the computer can look at any point in a picture, evaluate the color data present there, and easily eliminate any values that are not neutral gray. By doing this across the entire data field of the picture, the printer can generate a file in which only the neutral values are present—the perfect attenuated-black basis for a color print. This new technology would have been perfect for pochoir but it became available at the very moment that hand-coloring disappeared altogether from printmaking.
Collotype. Photographer unknown. Plate from the souvenir book U.S. Steel Cruiser Boston (New York: E. H. Hart, 1888). c. 7 x 8 3/4” (17.8 x 22.2 cm) each.
Variations in print color is one quirk of collotypes. For some reason the old collotype printers were remarkably casual about the color of ink that they used. In any given publication we can find prints ranging from neutral to brown, green, or blue. The three reproduced here, to the left and on the following pages, have been taken from a small book titled U.S. Steel Cruiser Boston, published by E. H. Hart in 1888. These prints are green and brown but the book has other plates that run the full gamut of collotype colors. When I was a young printer, and working in a shop that still did collotype, the pressmen made quite a fuss about the ink, and even in the 1960s imported it from Europe, where we all assumed it was made by grinding pigment, bats’ wings, and frogs’ legs into secret formulas guarded through the generations. Perhaps no one paid any attention to color, since getting the prints to turn out decently was such a far greater problem.
Collotype. Photographer unknown. Plate from the souvenir book U.S. Steel Cruiser Boston (New York: E. H. Hart, 1888). c. 7 x 8 3/4” (17.8 x 22.2 cm) each.
Collotype imitating albumen: Chemical photography was always expensive because the prints had to be handled individually and their materials were costly. Editions could be printed from negatives, and there was no theoretical limit to their numbers, but the reality of labor and material costs meant that any edition larger than a hundred or so required some sort of ink printing. The woodburytype filled this need, but collotype could similarly be made to imitate the chemical photograph and was less expensive. The example on the next page is from a trade magazine that describes the print accurately as a “collotype imitation silver print.” The ink has been mixed to have a purplish-red cast, to imitate the old gold-toned albumen print, and the paper is a highly polished clay-coated sheet. I had an old-time photographer tell me that if you went into a large photo studio in Paris early in this century and ordered an edition of fifty or more prints, you were just as likely to receive collotypes as actual chemical prints.
Collotype. Arthur Yallop. Gorleston Harbour. c. 1900. 6 3/4 x 9 3/8” (17.2 x 23.8 cm). The Museum of Modern Art, New York. Gift of Richard Benson. Made as a sample of the process, this collotype was printed in The Penrose Annual as an advertisement for the publisher Morgan & Kidd.
One useful collotype technique was double-rolling, used to produce richer collotype prints without the need for multiple impressions. The plate was inked with a stiff, dense black ink and then immediately reinked with a soft, colorful one. Once double-inked, the plate was printed in a single impression.
Collotype. Photographer unknown. Oxford, St. Mary the Virgin’s Church, Porch. 1900. 5 7/16 x 7 7/16” (13.8 x 18.9 cm). The Museum of Modern Art, New York. Gift of Richard Benson.
The result looked much like the later duotones printed by photo offset, which tend to have a color cast in the light values and a strong, neutral black in the deepest tone. This method was often used for fancy postcards.
Photo offset lithography. John Szarkowski. Wainwright Building. c. 1954. 12 1/2 x 9 1/4” (31.7 x 23.5 cm). Printed by Cantz, Germany, from separations by Robert Hennessey. Cover design by Jerry Kelly. Jacket for John Szarkowski, The Idea of Louis Sullivan (Boston: Bulfinch Press, 1956, reprint ed. 2000). Courtesy Little, Brown and Company.
Photo offset lithography duotone. Tod Papageorge. New Year’s Eve at Studio 54. 1978. 13 1/16 x 8 3/4” (33.1 x 22.2 cm). The Museum of Modern Art, New York. Gift of Richard Benson © Tod Papageorge.
Photo offset lithography. Artist unknown. Label for fruit box, Independent Brand Northwest Pears. c. 1950. 6 15/16 x 9 13/16” (17.7 x 25 cm). The Museum of Modern Art, New York. Gift of Richard Benson ©Washington Fruit and Produce.
“Photo offset” lithography first came into widespread use shortly after World War II as a means of printing colorful labels for fruit and vegetable boxes. They could be produced very inexpensively in short runs, without the need for the costly copper or zinc plates that would have been required for letterpress printing.
Relief printing (as halftone-based letterpress), gravure, and collotype gave the printing trades three terrific systems for reproducing photographs. These processes had their heydays from about 1880 until 1960—fully eighty years—and they were directly responsible for photography’s move out of the darkroom, with its chemical print, and into its modern role as the great visual data-transport system of today. Through the early years of the twentieth century, another process slowly took form that would prove to be the giant of all picture-printing processes. This was photo offset lithography, which, in its final form, brought together lithographic chemistry, lightweight and inexpensive printing plates, and simple, completely photographic data input to the printing press. I will try to be restrained in describing offset (as we casually called it), but it is both my trade and my chief love of all the ink-printing methods, so even while being cautious I will probably say too much about it.
Detail of Photo offset lithography. Artist unknown. Label for fruit box, Independent Brand Northwest Pears. c. 1950. 6 15/16 x 9 13/16” (17.7 x 25 cm). The Museum of Modern Art, New York. Gift of Richard Benson ©Washington Fruit and Produce.
The fruit label was drawn by hand, then printed with solid colors and coarse halftone screens. The printer used rough approximations of the subtractive primaries, but there are two blues instead of the usual cyan; one blue has a lot of red in it and the other, lighter one is quite green. Many package labels carried specialized colors, which almost always struck the eye with an intensity that halftone-generated mixes of the primaries could not. This technique, called “spot color,” is used to this day in fancy color printing.
More pictures have been printed in offset than by all other methods combined, so a careful examination isn’t such a bad idea. Today we are also witnessing the peak of the process’s development; it is fully integrated with the computer, and through the next years, as digital printing takes on smaller jobs, the offset behemoth will most likely live on, producing superb printed editions of books, catalogs, and magazines. I think it will be around for quite a while yet. The process had its origins in chromolithography.
Photo offset lithography. Shinobu Ishihara. Plate from Ishihara’s Tests for Colour-Blindness (Tokyo: Isshinkai, 1966). 1966. Printed by Isshinkai. 3 7/8 x 3 7/8” (9.9 x 9.9 cm). The Museum of Modern Art, New York. Gift of Richard Benson © 1959 Kanehara Shuppan Co., Ltd. Tokyo. Developed in the 1930s by Dr. Ishihara in Japan, these beautiful arrays of colored dots, printed by offset, could accurately hold the subtle colors needed to determine slight individual differences in color vision.
Lithography, printing right off the stone surface, could lay down a beautiful solid layer of ink. As soon as light-sensitive metal plates replaced the stones, the halftone, fully refined in letterpress printing, could move into lithography—the old, laborious translation of pictures by hand could be mechanized through the halftone screen. The metal plate and the halftone were vital to the rise of offset, but the central innovation, the one that made it king, was the introduction of the offset blanket. Printers had long known that when a fully inked roller was used to transfer ink to a printing plate, the roller, after the transfer, retained a negative image of the plate, since the ink in the image areas had been removed from it. This observation made clear that a rubber roller could hold an ink image.
Photo offset lithography. Shinobu Ishihara. Plate from Ishihara’s Tests for Colour-Blindness (Tokyo: Isshinkai, 1966). 1966. Printed by Isshinkai. 3 7/8 x 3 7/8” (9.9 x 9.9 cm). The Museum of Modern Art, New York. Gift of Richard Benson © 1959 Kanehara Shuppan Co., Ltd. Tokyo.
If a clean roller was passed over an inked litho stone, then rolled again onto a sheet of paper, a positive image could be picked up and transferred to the paper. This practice of transferring, or “offsetting,” an image became the core of photo offset lithography: a metal plate was inked, it printed onto a rubber blanket, and that blanket then transferred the image to the paper. The ramifications of this practice were revolutionary.
Photo offset lithography duotone. Tod Papageorge. New Year’s Eve at Studio 54. 1978. 13 1/16 x 8 3/4” (33.1 x 22.2 cm). The Museum of Modern Art, New York. Gift of Richard Benson © Tod Papageorge.
The old relief trick of printing a black and white picture in two impressions was quickly adapted to photo offset lithography. The delicate touch of the blanket, and its effect of keeping water away from the paper, allowed extremely accurate registration: presses were made that could print a black layer on a sheet one day and then perfectly superimpose a gray pass on the next one. This kind of printing—called “dry trap”—also allowed strong matte blacks to be printed on soft paper. Whenever the old letterpress books had included halftones, they had had to be printed on a hard, clay-coated sheet, necessary for a smooth transfer of ink between the metal plate and the paper. Because of the softness of the blanket, offset could print on any surface, giving rise to a new, uncoated sheet—“offset stock”—that had the same appearance as the paper used in beautiful old letterpress books holding type only.
Detail of Photo offset lithography duotone. Tod Papageorge. New Year’s Eve at Studio 54. 1978. 13 1/16 x 8 3/4” (33.1 x 22.2 cm). The Museum of Modern Art, New York. © Tod Papageorge. The black impression, shown here by itself, is always printed first, because ink transfers best to an unprinted sheet and a superb transfer is required in order to hold a decent black value.
Offset presses can handle paper in big twenty-eight-by-forty-inch sheets, often with eight book pages per side, or sixteen pages in all. When folded down, such a sheet produces a “signature” in a book. Printers were initially sloppy about the tonal structure of their duotones. The early two-impression illustrations, done in the 1960s and ’70s, were tonally rich, because of the two layers of ink, but they didn’t always replicate the tones of the original very well. Gradually printers unlocked the puzzle of how to expose and develop the halftone negatives so that each color emphasized different parts of the picture scale and, once printed together, looked close to the original. In the case of Tod Papageorge’s image, the black printer was made contrasty and showed exaggerated detail in the shadow portions of the picture. The gray printer showed little distinction between tones in the dark areas but had terrific accuracy in the light values.
Detail of Photo offset lithography duotone. Tod Papageorge. New Year’s Eve at Studio 54. 1978. 13 1/16 x 8 3/4” (33.1 x 22.2 cm). The Museum of Modern Art, New York. © Tod Papageorge. The gray impression is printed on top of the black. Because the gray ink is transparent and lighter in value, it does not obscure the shadow detail rendered by the black printer.
In the print from the two combined, the shadow detail holds, because the gray of the second impression doesn’t bury it, and the light values are good, because there is little black ink in those areas, so the smooth gray steps are clear and accurate. This solution was necessary because there was no method for exposing and developing a single halftone negative that could accurately portray all the tonal steps from black to white. The carefully crafted duotone was a solution to this problem. When properly made, the duotone had a marvelous benefit unavailable in the darkroom: the black printer controlled one end of the tonal scale and the gray printer the other. The ink for each impression could be adjusted on press, so each end of the scale could be independently tuned. This was a tremendous control, allowing reproductions to be accurately pinned down as they were being made on press.
Photo offset lithography tritone. Mathew B. Brady Studio. Senator and Mrs. James H. Lane. 1861–66 (Printed by Richard Benson and Thomas Palmer, 1985). A multiple-impression proof sheet for Photographs from the Collection of the Gilman Paper Company (White Oak Press, 1985). From an albumen silver print from a glass negative, 8 13/16 x 7 3/4” (22.4 x 19.7 cm). The Metropolitan Museum of Art. Gilman Collection, Purchase, Alfred Stieglitz Society Gifts.
The duotone was followed by more complicated printing, in three, four, and even five colors, all to make beautiful reproductions of black and white photographs. These additional colors were particularly needed for the reproduction of photographs from the nineteenth century if the rich variety of their original chemical color was to hold. In the case we see here, a black impression was used to generate the darkest tones in the print, but then, instead of a simple gray, the other inks were warm in color to match the old purple/red of the albumen original. Early black and white photography can also be reproduced with process color (as in my book The Printed Picture), but the finest reproductions cannot be done that way.
Photo offset lithography. Richard Benson. Bend Boat Basin. 1985. 9 1/2 x 7 1/2” (24.1 x 19 cm). A tritone print in black, dark gray, and light gray ink. © Richard Benson.
The tritone, printing from three halftones in various shades of gray ink, does much the best job. I make this whole thing sound like a cakewalk but it is actually very difficult to do well. Reproductions never look just like the originals, and all serious ink reproduction of photographs is based on the understanding that the reproduction has to look “right” in its own context—of ink on white paper in a book. This rightness often requires that values in the original be altered in the reproduction. Many early photographs, for example, have turned quite yellow, and if this color is slavishly followed, the reproduction simply looks wrong. Printing requires an understanding of a concept of equivalence: the most accurate replica one can make of anything is seldom a literal copy, but rather a new thing that gives the viewer the impression that it looks like the original.
Process color in photo offset lithography. Photographer unknown. Betty White. c. 1990. 7 3/4 x 8 1/4” (19.7 x 21 cm). The Museum of Modern Art, New York. Gift of Richard Benson.
This is a printer’s test sheet, used to check for slur in the image caused by slippage in the press. The enlarged halftone gives a fine demonstration of how the four dot patterns in full-color process printing interact.
Color printing was the engine that drove the development of photo offset technology. Printers had long since solved some of the basic problems of color: one was how to angle the halftone screens to avoid moiré patterns; another was to develop filters and inks that fitted each other well enough so that separations made with one filter could be printed with an ink of the complementary color. This particular problem was never fully solved, since no pure inks have ever been developed, but many darkroom tricks, using intricate masking, did a pretty good job of fixing those errors. Color printing also came to depend on the large-format transparency. These color intermediates were photographed to make the reproduction, then were available on press to guide the printing. Not many museums were willing to send their Rembrandt (or Betty White) to a printshop for the purpose of comparison.
Process color in photo offset lithography. Photographer unknown. Betty White. c. 1990. 7 3/4 x 8 1/4” (19.7 x 21 cm). The Museum of Modern Art, New York. Gift of Richard Benson.
Since color was extremely complex, the preliminary or “prepress” work tended to be done in specialized houses—the transparencies were made in the field by professional photographers, and then the separations, masking, and conversion to halftones were done by the specialists. The printing took place at the companies that had the big presses. One drawback to this system was that the division of labor allowed each party involved to blame the others when things didn’t work out, as was often the case. In all my glowing descriptions I leave out the fact that the reproductions were almost always poor, whether made in color or black and white. The business of making a replica of anything is to some extent doomed from the start, and when the object being reproduced is a photograph, the job is nearly impossible.
In any multiple-screen print—one that uses more than one dot pattern—some dots are next to each other and some are on top of each other. If the inks are transparent, their respective locations make no difference to the eye of the viewer when seen at a normal viewing distance. This detail shows the basic pattern of four-color process printing using traditional halftone screens.
Despite its complexity, in the end color printing turned out to be easier to do well than black and white. It’s hard to pin down the reasons for this. One big one is the development of color proofing systems that could make pretty good prepress replicas of what the halftones would do. These methods used photo polymers in the subtractive primaries; the separation house could make a set for fifty dollars or so to give the customer an idea of what the reproduction would look like. There was never an adequate method for proofing a black and white duotone or tritone. Another reason for the success of color printing was that it could be done badly and still look good. Even if the print was too heavy, or too light, or somewhat out of balance, the colors’ interrelationships could still hold and the colors could be enticing even if inaccurate. The fact that the original was seldom seen near the reproduction helped too. But in black and white work, errors in weight and scale could remove whole areas of content, and tonal distortions could murder the picture.
Web photo offset lithography. Arthur Hochstein (designer) and Spencer Jones-Glasshouse (photographer). Cover of Time magazine, December 25, 2006/January 1, 2007. 10 7/16 x 7 7/8” (26.5 x 20 cm). The Museum of Modern Art, New York. Gift of Richard Benson. Courtesy the Editors of Time Magazine © Time, Inc. This cover, printed by web offset, has an aluminized plastic mirror stuck on in the monitor area, intended to reflect the reader’s face.
All the offset printing we have looked at so far was done on cut sheets of paper. The printing speeds gradually increased, from about 4,000 impressions an hour in the early presses up to 10,000 or more for the newest ones. If a book is to have 192 pages, and each press sheet holds sixteen pages (eight per side), then twelve sides must be printed. Even with long makeready times such a book can be done in less than a week as long as the edition is no more than a few thousand. When we get to magazines routinely printed in editions of hundreds of thousands, with a new issue weekly or monthly the math simply doesn’t work out. There has to be a faster way to print. The answer is web printing. For readers under thirty I must emphasize that this is not the World Wide Web we are talking about, but paper printed on rolls instead of sheets.
Web printing was first developed for letterpress newspaper work, where time was a central concern, and it gradually moved over to offset. When a press prints individual sheets, its speed is limited by the need to properly register each sheet moving into it. This cannot be done without stopping the sheet at some point, moving it against register guides, and then accelerating it up to the press speed. But it turns out that there is no speed limit for the actual printing itself; plate, blanket, and paper are quite happy moving ink around at far higher speeds than any sheet-fed press can manage. If the printing is done on a roll of paper, and with a workable, high-speed registration system, there is in principle no limit to how fast a press can print. That high-speed registration system turned out to be a stroboscopic light, which reads register marks printed on the paper and adjusts the tension of the web to keep things aligned. Web presses are so fast that heat drying is used so the paper can be cut and folded right in the press, which discharges full signatures at the other end. In order to get finished pages the press must print both sides of the roll; this is called “perfecting,” and the web press must have eight printing units instead of the four that a sheet-fed press needs for color work. A large web press may be 200 feet long, with the four or five printing units occupying perhaps a third of its length and the feeder, dryer, and folder taking up the rest. Web presses return profits when run continuously, and they even have devices built into them allowing the input rolls of paper to be changed without stopping them. A good web press can print 35,000 to 50,000 impressions per hour, on sixteen magazine pages at a time, with four-color pictures, using halftones of 200 dots per linear inch, printing on both sides of the web. The operators have to wear ear protection, and when things go wrong they do so really quickly, but it is web printing that makes all those magazines, newspapers, and catalogs possible.
Photo offset lithography. Joan Lyons. Untitled, from the portfolio Presences. 1980. 22 1/4 x 16 1/2” (56.5 x 41.9 cm). The Museum of Modern Art, New York. Gift of Richard Benson © Joan Lyons. Lyons is one of relatively few artists who have championed photo offset as a creative medium. This multipass print was derived from a twenty-by-twenty-four-inch Polaroid.
Not too many artists use photo offset lithography as their primary medium. I am one, having done my best work in ink on a single-color offset press (long since junked). Joan Lyons, who made the picture on the left, is another. Syl Labrot—now dead—did remarkable color work in offset, and Lyons, Labrot, Scott Hyde, and Carl Sesto pretty much make up the group I know about. I am sure there are others out there but probably not too many. This is not surprising, because production offset presses are large, expensive, dirty, and dangerous. They had one at Yale in the ’70s, in the graduate design department, but it nearly scalped a student once and they had to get rid of it. One has to be extremely careful around sticky ink moving on rotary cylinders at high speed. As if this wasn’t bad enough, the artist using this medium also needs to have a copy camera, vacuum frame, arc lamp, and tons of other bits and pieces to have a functioning offset shop.
Photo offset lithography. Scott Hyde. Folded cellophane – Polarized x 3. 1966. 8 1/2 x 4 3/4” (21.6 x 12 cm) © Scott Hyde.
It is unfortunate that the artist using offset is so rare, because it is a great medium. Ink can be piled up in multiple layers, editions in the few hundreds are easy, and if the inks are chosen carefully, work made this way can be extremely permanent. Small art shops exist for intaglio processes, stone lithography, and fancy letterpress, and some even have offset proof presses on hand, but the pure-offset art shop never really happened. It is even less likely today, because the new digital printing technologies lend themselves beautifully to atelier-scale work. A big inkjet printer, high-end Macintosh computer, and flatbed scanner together cost a fraction of the investment needed for a new offset press. A tremendous number of older presses are out there in daily use, but ultimately offset will most likely fade away except for large-edition work, done by web printing, and substantial book projects that can justify the initial investment in preparatory work. Maybe fine little two- or four-color presses will then be available cheap, and perhaps a few enterprising souls will grab them and set up private shops, to make art with this great medium even as the digital revolution takes over.
Photo offset press sheet. Carl Sesto. Press sheet. 1994. 15 3/4 x 11 1/4” (40 x 28.6 cm). From Ordinary Events (Boston: SMFA Press, 1994). The Museum of Modern Art, New York. Gift of Richard Benson © Carl Sesto. This uncut press sheet shows four pages of a book printed by the artist himself on a small offset press.
A good friend of mine has a printing company with, hidden away among the behemoths, a beautiful little nineteen-by-twenty-five-inch, five-color Heidelberg press. Every time I visit him I make a point of walking by that press to see if they are still using it. When new it must have cost upwards of $500,000, but once obsolete, it won’t be worth a nickel. The last time I visited the plant I surreptitiously paced off the length of the machine to see just how much room it took. I could just manage to fit it into my basement workroom.