From Stoichedon to programming: a concise history of monospaced typefaces

In Ancient Greece, the Stoichedon style of engraving already made use of a single width for each letterform, which could consequently be aligned vertically as well as horizontally within a grid. As all letters were fitted within the proportions of a square, they were placed at equal intervals along their respective alignment, regardless of word breaks, thus creating an orderly layout. Considered to be the dominant style of composition in Athens throughout the fifth and fourth centuries B.C., many have speculated over the reasons for this strictly regimented layout, yet the most convincing argument for its adoption so far appears to be aesthetic. Scholars have indeed described Stoichedon engraving as ‘a practical expression of Greek feeling for order and beauty’ [Austin, 1938: 119]; it is also believed that the inherent structure of the script itself, whose letterforms frequently called for the use of vertical strokes, was favourable to the similarly vertical alignment of characters and to their wide spacing. Nevertheless, this aesthetically pleasing grid layout was not without its drawbacks, as words often had to be divided across lines and ease of reading was compromised.

As social and economical evolutions within the Greek society called for the script to evolve and for the alleviation of the rigid grid, the Stoichedon style of engraving slowly declined at the beginning of the third century B.C.. The idea of fitting letterforms within a square, and of arranging these squares on a grid, however, has survived within Asian ideographic writing systems such as Chinese, Japanese and Korean , whose graphic arrangement bears some strong similitude with the Stoichedon layout.

Within the Western world, the need for monospaced letterforms does not seem to have reappeared until the introduction of mechanical typesetting, and of the typewriter in particular. In hand-set foundry type, punches could be engraved and type could be cast with any required width; end-of-line decisions were the responsibility of the compositor, who would decide when to break a line, whether hyphenation was necessary, and how much white space was required between words in order to justify a line. But as the task moved from the compositor to the machine with the mechanization of typesetting, the need for automated ways of calculating character width became evermore pressing. In the early 1880s the American Linn Boyd Benton foresaw the benefits of a width-quantifying system and devised a scheme he called ‘self-spacing type’ . While working on a machine to justify type automatically, Benton conceived a font in which the em square for each typeface was divided into six equal parts, which formed the basic unit width, and all letters and spaces in one typeface were cast on a multiple of this unit. In one version of the self-spacing type, the lowercase ‘i’ was cast on two units, the ‘e’ on three units and the capital ‘W’ was seven units wide, while the majority of characters were cast on four units. Benton patented his self-spacing type in 1883 and his system effectively reduced the time necessary for hand-justification.

In parallel to the development of self-spacing type, the end of the nineteenth century witnessed the growing success of the typewriter and of strike-on composition, which catered for a new sphere of commercial and administrative printing beyond the usual needs of the trade. Whereas the craft of typefounding enabled characters to be cast on any required width, strike-on composition generally relied on an escapement process controlled by a rack whose teeth were evenly spaced. Consequently, every time a key was struck, the machine systematically advanced the same width, thus requiring that all characters share identical horizontal proportions.

The need for all characters to fit within the same width necessarily affected the overall appearance of typefaces, whose traditionally wide and narrow letters were now expected to be identical. These constraints commonly resulted in dense characters such as lowercase letters ‘m’ and ‘w’ to be drawn extremely narrow and with tight sidebearings, while lowercase letters ‘i’, ‘j’, ‘l’ and ’t’ were inevitably wider and provided with looser spacing. One solution often used in typewriter typefaces was to design narrow characters such as ‘i’, ‘j’ and ‘l’ with large serifs so that they could occupy their full width more comfortably. Capital letters and numerals were also drawn with tighter proportions than would have been traditionally deemed acceptable in foundry type.

Despite these constraints, which contributed to typefaces for strike-on composition being immediately distinguishable from foundry type (although typefounders soon added to their catalogues monospaced designs imitating typewriter faces), the use of fixed-width characters offered a number of advantages in the composition process: it facilitated, for instance, the use of multiple columns and the integration of letters and numerals within a tabular composition. Moreover, typewriter typefaces were commonly available in two possible sizes, known as ‘Pica’ (which fit 10 characters within an inch), and ‘Elite’ (allowing for 12 characters per inch). Applied to a standard page composed on a typewriter, these sizes equated to 60–72 characters per line, which contributed to a comfortable reading experience.

Manufacturers of typewriters were keen to provide their customers with a broad range of monospaced typefaces. Brands such as Olivetti, IBM, Underwood or Remington offered a diversity of designs, from serif to sans serif typefaces right through to blackletter and even script faces, with a number of manufacturers offering more than sixty different styles by the 1960s. [1]1. A number of these manufacturers even commissioned renowned designers to conceive typefaces specifically for typewriter composition: Cassandre designed Graphika for Olivetti, Goudy designed Remington Typewriter for Remington, and Adrian Frutiger produced a monospaced version of his Univers for IBM. However, Egyptian typefaces with rectangular serifs and little or no contrast between thick and thin strokes proved by far the most popular style and have come to be associated more prominently with typewriters than any other. Whereas triangular serifs and a pronounced contrast between thick and thin strokes posed the risk of tearing off the paper when striking a letter, the sturdiness of slab serifs proved well suited to the technology. The most successful of all monospaced faces from that era is undoubtedly Courier, which was initially developed by Howard Ketter for IBM in the mid 195os, and released on its Selectric Composer in 1961. Thanks to its wide x-height, its large counters and the subtlety of its curves and of its rounded slab-serifs, Courier proved the perfect candidate for typewriter composition. The fact that IBM did not patent the design also greatly contributed to its wide adoption by other manufacturers, and to its international success.

The possibility to use proportional spacing and variable unit-widths in strike-on composition was incorporated on the Varityper in 1947. In 1961, IBM launched the Selectric Typewriter, which was the first to feature a golf-ball typing head, and by 1967, its Magnetic Tape Selectric Composer enabled letterforms to fit one of nine possible unit-widths, thus reducing (yet not completely alleviating) the aesthetic restrictions of a single width on letterforms. Yet, as Alan Bartram noted in 1962, by that period, the look of monospaced fonts had come to be widely accepted as a style of its own:

A number of these manufacturers even commissioned renowned designers to conceive typefaces specifically for typewriter composition: Cassandre designed Graphika for Olivetti, Goudy designed Remington Typewriter for Remington, and Adrian Frutiger produced a monospaced version of his Univers for IBM.

To Bartram’s satisfaction, the introduction of proportional typefaces in strike-on composition did not sound the death knell for monospaced fonts, as the advent of digital technologies created a new demand for fixed-width types. Already in the 1960s, the need for machine-readable typefaces that worked in combination with Optical Character Recognition systems led to the creation of monospaced designs such as OCR-A and OCR-B, the latter being a commission from the European Computer Manufacturers Association to Adrian Frutiger in the 1960s. In the case of OCR-A, the requirements of the optical character recognition process called for all letterforms to fit within a matrix of 5 x 9 cells, resulting in an extremely stylised design composed essentially of straight strokes. The font contained only capital letters and numerals. OCR-B, on the other hand, included lowercase letters and was conceived by Frutiger to be much more pleasing to the eye, while remaining monospaced. The design of the typeface was also driven by the necessity for all letters to be clearly differentiated from one another, in order to avoid errors during the recognition process. Interestingly, OCR-B was used by a number of computer manufacturers who all required that characters share the same width, but who had their own specific requirements regarding vertical proportions. Consequently, OCR-B was developed in three different versions of varying height, while the width of every single character was identical across the range.

Monospaced fonts proved again a popular choice in the early days of digital typography, mostly because of the serious limitations in memory storage and graphical capabilities on early computers, terminals and printers. Fixed-width typefaces were particularly essential to text mode applications, which displayed the content of a program by using a string of characters rather than graphical cues, and addressed the screen as an orthogonal grid similar to the Stoichedon style of engraving. Because of this renewed demand for monospaced faces, Courier was one of the first typefaces – together with Times, Helvetica and Symbol – to be adapted to the PostScript format by Adobe in the early 1980s. [2]2. Various updated versions of Courier were released in subsequent years, including Courier New, which was bundled with Windows 3.1 in 1992. Other popular non-proportional faces offered as system fonts were to follow, such as Lucida Console, Monaco, or more recently, Consolas.

Thanks to advances in digital technologies and plentiful computer memory, monospaced typefaces have now ceased to be a technical requirement. Yet they have remained extremely popular in computer programming, as fixed-width characters enable greater control over the length of each line of code, and are also credited with providing greater differentiation between individual characters, leading to a more comfortable coding experience, and presumably making it easier to spot errors. As monospaced fonts have also come to be widely associated with programming, they are frequently used in publications and technical resources on the subject. They feature either as a visual cue to differentiate lines of code from more conventional passages of text, or as a stylistic strope to reference computerization and related ‘high tech’ topics.
Fixed-width typefaces have also found a broad range of uses outside the programming sphere, from the composition of scientific data and musical notation, [3]3. Monospaced faces are common in the tablature form of musical notation, which indicates instrument fingering instead of musical pitches. right through to the typesetting of screenplays, which are still today required to be set in 12 pt Courier as the easiest means to control the length of a script. Beyond practical considerations, they have come to be widely accepted as an autonomous typographic style with aesthetic qualities of its own.

Initially published in 2015 for Gemeli Mono type specimen, ISBN 979-10-93578-01-9.