As If By Chance: Part I

Sketches of Disruptive Continuity in the Age of Print from Johannes Gutenberg to Steve Jobs

Drawing Hands by M.C. Escher, 1948

Everything existing in the Universe is the fruit of chance and necessity.

Democritus, circa 400 BC

Necessity is blind only so long as it is not understood.

G. W. F. Hegel, 1817

It is notable that some inventions in the history of print technology are recorded as having been achieved by chance. In accounts written at the time of the inventions as well as in historical studies, majorbreakthroughs in printing have been attributed to accidental events. Much in the same way schoolchildren are taught that the natural scientist Isaac Newton discovered the law of gravitation afteran apple fell from a tree upon his head, significant inventions in the history of printing are said to be the result of lucky mistakes.

Perhaps the two most well-known examples of this phenomenon are found in accounts of the late-eighteenth-century invention of lithography by Alois Senefelder and the early-twentieth-century inventionof offset printing by Ira Washington Rubel. In both cases, the technical advances made by the inventors areoften explained as having been accidental.

Here are two citations:

Lithography was invented around 1796 in Germany by an otherwise unknown Bavarian playwright, Alois Senefelder, who accidentallydiscovered that he could duplicate his scripts by writing them in greasycrayon on slabs of limestone and then printing them with rolled-on ink

Department of Drawings and Prints, The Metropolitan Museum of Art, October 2004

Offset printing, also called offset lithography, or litho-offset, in commercial printing, widely used printing technique in which the inked image on a printing plate is printed on a rubber cylinder and then transferred(i.e., offset) to paper or other material. The rubber cylinder gives great flexibility, permitting printing on wood, cloth, metal, leather, and rough paper. An American printer, Ira W. Rubel, of Nutley, New Jersey,accidentally discovered the process in 1904 and soon built a press to exploit it.

The editors of Encyclopædia Britannica, July 1998

Readers of these passages would not be blamed for thinking that Senefelder of Bavaria, Germany, in 1796 and Rubel of Nutley, New Jersey, in 1904 were the beneficiaries of pure luck or that they fortuitously stumbled their way into print technology history. However, this would be an incorrect or, at best, an incomplete way of understanding the contributions of these two innovators.

Why does the word “accidentally” appear in the above accounts of historic inventions that took place more than one hundred years apart and, together, established what is known as offset lithography, a technology that revolutionized the printing industry and remains today the dominant method of transferring ink to paper? Why is it that stories of accidental invention—even from authoritative sources like the Metropolitan Museum of Art and Encyclopædia Britannica—persist for both men, despite ample evidence that Senefelder and Rubel were in pursuit of innovation and striving to improve the printing process through the methods of ingenuity, experimentation, and science that prevailed during their respective lifetimes?

Finding answers to these questions requires investigative journey. While it may be a fact of popular interest that Senefelder and Rubel are known as much—or even more—for the accidental way they arrived at their achievements than they are for the significance of the achievements themselves, it is also a fact that invention by happenstance has occurred in history more often than is generally known. Since the “accidental” attribution tends to overshadow and mystify the progress attained—in printing as well as other industries—it is instructive to examine these two inventions in their socio-economic context and to locate the place of Senefelder and Rubel within the whole history of printing. Such an examination shows that their accomplishments were absolutely necessary advancements.

To untangle the riddle of accidental invention in the specific cases of Senefelder and Rubel, it is necessary to: (1) investigate the historical record and review the facts of what is known about the men and how they invented lithography and offset printing; (2) look outside print technology and into the prevalence of “serendipity” more broadly in the history of scientific and technological discovery; (3) explore the source of the need for the legends of accidental discovery in human progress; (4) make a theoretical analysis of the two-sided and contradictory content of “accidents” in general; and, (5) return to Senefelder and Rubel and show how their inventions were manifestations of disruptive continuity in the history of printing.

The concept of disruptive continuity applies to the development of printing—as well as all human technical progress—because it acknowledges that each innovation owes its emergence to the accomplishments of others that came beforehand; that significant innovation could not take place without innumerable connections to the past. At the same time, disruptive continuity also recognizes that each new breakthrough represents a sharp departure from the past. It is a transition point forward that expresses the future in ways that were previously impossible and could not have been accomplished but for the spark of genius embedded in the new innovation.

As this introduction will go on to explain, it is at this nexus point of discontinuity from the prior gradual progression and the moment of a leap into the future that the phenomenon of accidental invention occurs. To understand how unanticipated events, which are rooted in antecedent accomplishments, can and do become transformed into significant innovations is to understand the mechanism by which the old era of technology is superseded by that of an entirely new era of progress.

Finally, by developing a socio-historical-technical analysis of nearly six centuries of print communications—based on the theory of disruptive innovation—significant conclusions can be drawn about the future of ink-on-paper media within the new environment dominated by online, mobile, social and streaming content delivery systems.

* * * * *

The investigative journey begins with an examination of the work of the two printing innovators who are frequently remembered as accidental inventors. It is fortunate that, in the case of Senefelder, an account written by the inventor himself is available and, in the case of Rubel, there exists two technical explanations, an anecdotal account and a posthumous tribute to the inventor written by a close business partner at the time of his death.

The invention of lithography

In 1817, at the urging of his colleagues, Alois Senefelder wrote down the story of his life along with a detailed description of how he invented lithography by experimental methods. He also provided a step-by-step technical guide for those wishing to learn and practice the art also known as “printing from a stone” or “stone printing.” Senefelder’s account was published one year later in the German volume entitled Vollständiges Lehrbuch der Steindruckerey (A Complete Course of Lithography). The work was translated into English by J.W. Muller and published by The Fuchs & Lang Manufacturing Company in New York in 1911 as The Invention of Lithography.

The relevant passages from the 1911 English text are found in the first chapter, “Section I: History of Stone Printing, Part I: From 1796 to 1800.”

As mentioned in the above quote from the Metropolitan Museum of Art, the young Alois Senefelder was an aspiring playwright and was motivated to start a printing firm so that he could publish his own works. Senefelder wrote that he was familiar with the procedures of the letterpress printing process of his day, “I had spent many a day in the establishments,” and that “it would not be hard for me to learn.” Senefelder also had a “desire to own a small printing establishment myself” because—having studied both public finance and law for three years at the University of Ingolstadt—he wanted to “earn a decent living” and “become an independent man” by going into business.

However, it was economic reality—a lack of the financial resources required to become a printer—that drove Senefelder down the path of innovation. As he wrote, “If I had possessed the necessary money, I would have bought types, a press and paper, and printing on stone probably would not have been invented so soon. The lack of funds, however, forced me to other expedients.”

Senefelder gave details of three different approaches he took in an effort to replicate the letterpress method without the ability to purchase the technologies that were readily available to others with the requisite capital resources. These were:

  1. To etch letters in steel and then “impressing them on pear wood, in which the letters would show in relief, somewhat like the cast type of the book printers, and they could have been printed like a wood-cut.” He abandoned the approach, “I had to give up the whole thing through lack of implements and sufficient skill in engraving.”
  2. To purchase “enough types to set one column or folio” and transfer the letters “to a board covered with soft sealing-wax, and reproduce the relief plate thus obtained in stereotype form.” Although this method was a technical success—especially after he began “mixing finely powdered gypsum with the sealing-wax” and “made the latter harder than the ordinary type composition”—Senefelder was unable to move forward because, “even this exceeded my financial power.” He gave up on this plan, “especially as I had conceived a new one during my experiments.”
  3. To learn “to write out ordinary type letters exactly, but reversed” with “an elastic steel pen on a copper plate covered in ordinary manner with etching surface” and these plates would be given to copper-plate printers for the press work. Here, Senefelder had difficulties because, though he learned quickly the skill of writing in reverse, “I could not correct the errors made during writing” because the “accessories of copper-plate engravers, especially the so-called cover varnish, were quite unknown to me.”

Senefelder then “labored desperately to overcome the difficulty” and tried three sub-methods within this “elastic steel pen” approach:

(a) Having “attained much chemical knowledge” during his days as a student, Senefelder began working with “spirits of wine and various resinous forms” and “oil of turpentine and wax” as methods for making corrections on the copper plate. However, he abandoned these materials because the chemical solution frequently became heavily diluted and “caused it to flow too much and dissolve the etching surface, at which time several well-done parts of the engraving were ruined.”

(b) Still determined to work with copper plate, Senefelder experimented with a wax and soap mixture as a material that could be used for correcting mistakes. He used, “a mixture of three parts of wax with one part of common tallow soap, melted over the fire, mixed with some fine lampblack, and then dissolved in rainwater, gave me a sort of black ink with which I could correct faulty spots most easily.” But this path “presented a new difficulty” in that he had only a “single little copper plate,” and, after he “pulled proofs at the house of a friend who possessed a copper-plate press,” he had to spend “hours again laboriously grinding and polishing the plate, a process which also wore away the copper fast.”

(c) To get around the limited copper plate resources, Senefelder transitioned to experimentation with “an old zinc plate of my mother’s,” that was “easier to scrape and polish.” However, “the results were very unsatisfactory,” because the “zinc probably was mixed with lead,” and he did not have a “more powerful acid” that could penetrate it.

Finally, Senefelder moved on to transferring a printed image to paper based on “a handsome piece of Kellheimer stone.” He explained, “The experiments succeeded, and though I had not thought originally that the stone itself might be used for printing (the samples I had seen hitherto of this Kellheim limestone were too thin to withstand the pressure exerted in printing), I soon began to believe that it was possible. It was much easier to do good work on the stone than on the copper.”

He began working “in order to use the stone just like copper” and trying “all possible kinds of polishing and grinding without attaining my purpose completely.” Senefelder wrote that moving from copper or zinc plate to printing from a limestone did not immediately result in the invention of lithography, “I had invented little that was new, but simply had applied the copper-plate etching method to stone.” And furthermore, “I was not the first discoverer of stone-etching, nor of stone-printing; and only after I made this new discovery which I will describe now, which led me from the engraved to the relief process, with my new ink, might I call myself the inventor of an art.”

In the midst of his detailed survey, Senefelder made it clear that he decided to write his account in 1817 in order to set the record straight, “I have told all of these things fully in order to prove to the reader that I did not invent stone-printing through lucky accident, but that I arrived at it by a way pointed out by industrious thought.”

However, he went on to say that his experiments with etched, i.e., mechanical and relief and not yet chemical, processes on stone “were entirely checked by a new, accidental discovery. Until now I had invented little that was new, but simply had applied the copper-plate etching method to stone. But this new discovery founded an entirely new form of printing, which basically became the foundation of all succeeding methods.” [Emphasis added]

Senefelder then recounted his well-known story of accidental invention:

I had just ground a stone plate smooth in order to treat it with etching fluid and to pursue on it my practice in reverse writing, when my mother asked me to write a laundry list for her. The laundress was waiting, but we could find no paper. My own supply had been used up by pulling proofs. Even the writing-ink was dried up. Without bothering to look for writing materials, I wrote the list hastily on the clean stone, with my prepared stone ink of wax, soap, and lampblack, intending to copy it as soon as paper was supplied.

As I was preparing afterward to wash the writing from the stone, I became curious to see what would happen with writing made thus of prepared ink …

My further experiments with this relief plate succeeded far better than my previous ones with etched letters. The inking in was much easier, and hardly one quarter of the force was necessary for making impressions. Thus the stones were not so liable to crack, and, what was the most important for me, this method of printing was entirely new, and I might hope to obtain a franchise and even financial aid.

It would take further experimentation with the stone by Senefelder to finally arrive at the invention of lithography, “Even this method, new in 1796, still was purely mechanical in its purpose, whereas the present printing method, which I began in 1799, may be called purely chemical.”

The following can drawn from the above summary of Senefelder’s own account of his invention: (1) Senefelder began in 1796 by experimenting and practicing with multiple materials and chemicals as he sought to develop an affordable mechanical printing process that was less capital intensive than the letterpress method; (2) he insisted that he did not invent lithography “through a lucky accident” but by way of “industrious thought”; (3) he stated that his efforts to come up with an alternative mechanical method to letterpress “were entirely checked by a new, accidental discovery”; (4) he told the story of how, while working with a limestone as a mechanical image transfer base, he wrote a laundry list upon the stone and from here new possibilities then occurred to him; (5) it would take three more years of further experimentation with the limestone before the “purely chemical” printing method was discovered in 1799 that become known as lithography.

It is highly significant that in his own account Senefelder presented two different and internally contradictory explanations for how he made his breakthrough. In one sentence, he wrote that he did not invent lithography by “lucky accident” but by “industrious thought” and, in another sentence, he said his experiments with mechanical methods on limestone “were entirely checked by a new, accidental discovery” that subsequently led to his invention of the “art” of the purely chemical method of printing.

This shows that Senefelder was perplexed in his attempt to explain the two opposing determinations that both appeared to him as true. Since he could not have expressed the genuine relationship between accident and necessity in the invention of lithography in a clear and scientific manner, Senefelder instead gave two separate and mutually conflicting explanations for how it happened.

It becomes plain from this that it is Senefelder himself who is responsible for two different stories: one stating that he invented lithography by an “accidental discovery” and another that he arrived at stone-printing not “through lucky accident” but by deliberately experimental methods. While this explanation appears to confound rather than clarify matters, Senefelder’s contradictory elaboration provides an important clue to solving the riddle of why stories of chance discovery have come to predominate.

Adrian Frutiger (1928–2015): Univers and OCR-B

Adrian Frutiger: May 24, 1928 – September 10, 2015
Adrian Frutiger: May 24, 1928 – September 10, 2015

Adrian Frutiger died on September 10, 2015 at the age of 87. He was one of the most important type designers of his generation, having created some 40 fonts, many of them still widely used today. He was also a teacher, author and specialist in the language of graphic expression and—since his career spanned metal, photomechanical and electronic type technologies—Frutiger became an important figure in the transition from the analog to the digital eras of print communications.

Frutiger was born on May 24, 1928 in the town of Interseen, near Interlaken and about 60 kilometers southeast of the city of Bern, Switzerland. His father was a weaver. As a youth, Adrian showed an interest in handwriting and lettering. He was encouraged by his family and secondary school teachers to pursue an apprenticeship rather than a fine arts career.

Adrian Frutiger around the time of his apprenticeship
Adrian Frutiger around the time of his apprenticeship

At age 16, Adrian obtained a four-year apprenticeship as a metal type compositor with the printer Otto Schlaeffli in Interlaken. He also took classes in drawing and woodcuts at a business school in the vicinity of Bern. In 1949, Frutiger transferred to the School of Applied Arts in Zürich, where he concentrated on calligraphy. In 1951, he created a brochure for his dissertation entitled, “The Development of the Latin Alphabet” that was illustrated with his own woodcuts.

It was during his years in Zürich that Adrian worked on sketches for what would later become the typeface Univers, one of the most important contributions to post-war type design. In 1952, following his graduation, Frutiger moved to Paris and joined the foundry Deberny & Peignot as a type designer.

During his early work with the French type house, Frutiger was engaged in the conversion of existing metal type designs for the newly emerging phototypesetting technologies. He also designed several new typefaces— Président, Méridien, and Ondine—in the early 1950s.

San serif and Swiss typography

San serif type is a product of the twentieth century. Also known as grotesque (or grotesk), san serif fonts emerged with commercial advertising, especially signage. The original san serif designs (beginning in 1898) possessed qualities—lack of lower case letters, lack of italics, the inclusion of condensed or extended widths and equivalent cap and ascender heights—that seemingly violated the rules of typographic tradition. As such, these early san serif designs were often considered too clumsy and inelegant for the professional type houses and their clients.

Rudolf Koch, Kabel, 1927
Rudolf Koch, Kabel, 1927

Paul Renner, Futura, 1927
Paul Renner, Futura, 1927

Eric Gill, Gill Sans, 1927
Eric Gill, Gill Sans, 1927

Along with the modern art and design movements of the early twentieth century, a reconsideration of the largely experimental work of the first generation of sans serif types began in the 1920s. Fonts such as Futura, Kabel and Gill Sans incorporated some of the theoretical concepts of the Bauhaus and DeStijl movements and pushed sans serif to new spheres of respectability.

However, these fonts—which are still used today—did not succeed in elevating san serif beyond headline usage and banner advertising and into broader application. Sans serif type remained something of an oddity and not yet accepted by the traditional foundry industry as viable in terms of either style or legibility.

In the 1930s, especially within the European countries that fell to dictatorship prior to and during World War II, there was a backlash against modernist conceptions. Sans serif type came under attack, was derided as “degenerate” and banned in some instances. Exceptions to this trend were in the US, where the use of grotesque types was increasing, and Switzerland, where the minimalist typographic ideas of the Bauhaus were brought by designers who had fled the countries ruled by the Nazis.

The Bauhaus School, founded in 1919 in Weimar, Germany, was dedicated to the expansion of the modernist esthetic
The Bauhaus School, founded in 1919 in Weimar, Germany, was dedicated to the expansion of the modernist esthetic

After the war, interest in sans serif type design was renewed as a symbol of modernism and a break from the first four decades of the century. By the late 1950s, the most successful period of san serif type opened up and the epicenter of this change emerged in Switzerland, signified by the creation of Helvetica (1957) by Eduard Hoffmann and Max Miedinger of the Haas Type Foundry in Münchenstein.

It was the nexus of the creative drive to design the definitively “modern” typeface and the possibilities opened up by the displacement of metal type with phototypesetting that brought san serif from a niche font into global preeminence.

Frutiger’s Univers

This was the cultural environment that influenced Adrian Frutiger as he set about his work on a new typeface as a Swiss trained type designer at a French foundry. As Frutiger explained in a 1999 interview with Eye Magazine, “When I came to Deberny & Peignot in Paris, Futura (though it was called Europe there) was the most important font in lead typesetting. Then one day the question was raised of a grotesque for the Lumitype-Photon [the first phototypesetting system]. …

“I asked him [Peignot] if I might offer an alternative. And within ten days I constructed an entire font system. When I was with Käch I had already designed a thin, normal, semi-bold and italic Grotesque with modulated stroke weights. This was the precursor of Univers. … When Peignot saw it he almost jumped in the air: ‘Good heavens, Adrian, that’s the future!’ ”

An early diagram of Frutiger’s Univers in 1955 shows the original name “Monde”
An early diagram of Frutiger’s Univers in 1955 shows the original name “Monde”

Final diagram of Frutiger’s 21 styles of Univers in 1955
Final diagram of Frutiger’s 21 styles of Univers in 1955

Originally calling his type design “Monde” (French for “world”), Frutiger’s innovation was that he designed 21 variations of Univers from the beginning; for the first time in the history of typography a complete set of typefaces were planned precisely as a coherent system. He also gave the styles and weights a numbering scheme beginning with Univers 55. The different weights (extended, condensed, ultra condensed, etc.) were numbered in increments of ten, i.e. 45, 65, 75, 85 and styles with the same line thickness were numbered in single digit increments (italics were the even numbers), i.e. 53, 56, 57, 58, 59, etc.

Univers was released by Deberny & Peignot in 1957 and it was quickly embraced internationally for both text and display type purposes. Throughout the 1960s and 70s, like Helvetica, it was widely used for corporate identity (GE, Lufthansa, Deutsche Bank). It was the official promotional font of the 1972 Munich Olympic Games.

Frutiger explained the significance of his creation in the interview with Eye Magazine, “It happened to be the time when the big advertising agencies were being set up, they set their heart on having this diverse system. This is how the big bang occurred and Univers conquered the world. But I don’t want to claim the glory. It was simply the time, the surroundings, the country, the invention, the postwar period and my studies during the war. Everything led towards it. It could not have happened any other way.”

Computers and digital typography

Had Adrian Frutiger retired at the age of 29 after designing Univers, he would have already made an indelible contribution to the evolution of typography. However, his work was by no means complete. By 1962, Frutiger had established his own graphic design studio with Bruno Pfaffli and Andre Gurtler in Arcueil near Paris. This firm designed posters, catalogs and identity systems for major museums and corporations in France.

Throughout the 1960s, Frutiger continued to design new typefaces for the phototypesetting industry such as Lumitype, Monotype, Linotype and Stempel AG. Among his most well-known later san serif designs were Frutiger, Serifa and Avenir. Frutiger’s font systems can be seen to this day on the signage at Orly and Charles de Gaulle airports and the Paris Metro.

The penetration of computers and information systems into the printing and publishing process were well underway by the 1960s. In 1961, thirteen computer and typewriter manufacturers founded the European Computer Manufacturers Association (ECMA) based in Geneva. A top priority of the EMCA was to create an international standard for optical character recognition (OCR)—a system for capturing the image of printed information and numbers and converting them into electronic data—especially for the banking industry.

By 1968, OCR-A was developed in the US by American Type Founders—a trust of 23 American type foundries—and it was later adopted by the American National Standards Institute. This was the first practically adopted standard mono-spaced font that could be read by both machines and the human optical system.

However, in Europe the ECMA wanted a font that could be used as an international standard such that it accommodated the requirements of all typographic considerations and computerized scanning technologies all over the world. Among the issues, for example, were the treatment of the British pound symbol (£) and the Dutch IJ and French oe (œ) ligatures. Other technical considerations included the ability to integrate OCR standards with typewriter and letterpress fonts in addition to the latest phototypesetting systems.

Comparison of OCR-A (1968) with Frutiger’s OCR-B (1973)
Comparison of OCR-A (1968) with Frutiger’s OCR-B (1973)

In 1963, Adrian Frutiger was approached by representatives of the ECMA and asked to design OCR-B as an international standard with a non-stylized alphabet that was also esthetically pleasing to the human eye. Over the next five years, Frutiger showed the exceptional ability to learn the complicated technical requirements of the engineers: the grid systems of the different readers, the strict spacing requirements between characters and the special shapes needed to make one letter or number optically distinguishable from another.

In 1973, after multiple revisions and extensive testing, Adrian Frutiger’s OCR-B was adopted as an international standard. Today, the font can be most commonly found on UPC barcodes, ISBN barcodes, government issued ID cards and passports. Frutiger’s OCR-B font will no doubt live on into the distant future—alongside various 2D barcode systems—as one of the primary means of translating analog information into digital data and back again.

Frutigers Sign and Symbols 1989
Frutiger’s 1989 English translation of “Signs and Symbols: Their Design and Meaning”

Adrian Frutiger’s type design career extended well into the era of desktop publishing, PostScript fonts and the Internet age. In 1989, Frutiger published the English translation of Signs and Symbols: Their Design and Meaning a theoretical and retrospective study of the two-dimensional expression of graphic drawing with typography among its most advanced forms. For someone who spent his life working on the nearly imperceptible detail of type and graphic design, Frutiger exhibited an exceptional grasp of the historical and social sources of man’s urge toward pictographic representation and communication.

As an example, Frutiger wrote in the introduction to his book, “For twentieth century humans, it is difficult to imagine a void, a chaos, because they have learned that a kind of order appears to prevail in both the infinitely small and the infinitely large.  The understanding that there is no element of chance around or in us, but that all things, both mind and matter, follow an ordered pattern, supports the argument that even the simplest blot or scribble cannot exist by pure chance or without significance, but rather that the viewer does not clearly recognize the causes, origins, and occasion of such a ‘drawing’.”