Very few terms have been used so habitually and carelessly as the word ‘experiment’. In the field of graphic design and typography, experiment as a noun has been used to signify anything new, unconventional, defying easy categorization, or confounding expectations. As a verb, ‘to experiment’ is often synonymous with the design process itself, which may not exactly be helpful, considering that all design is a result of the design process. The term experiment can also have the connotation of an implicit disclaimer; it suggests not taking responsibility for the result. When students are asked what they intend by creating certain forms, they often say, ‘It’s just an experiment…’, when they don’t have a better response.

In a scientific context, an experiment is a test of an idea; a set of actions performed to prove or disprove a hypothesis. Experimentation in this sense is an empirical approach to knowledge that lays a foundation upon which others can build. It requires all measurements to be made objectively under controlled conditions, which allows the procedure to be repeated by others, thus proving that a phenomenon occurs after a certain action, and that the phenomenon does not occur in the absence of the action.

An example of a famous scientific experiment would be Galileo Galilei’s dropping of two objects of different weights from the Pisa tower to demonstrate that both would land at the same time, proving his hypothesis about gravity. In this sense, a typographic experiment might be a procedure to determine whether humidity affects the transfer of ink onto a sheet of paper, and if it does, how.

A scientific approach to experimentation, however, seems to be valid only in a situation where empirical knowledge is applicable, or in a situation where the outcome of the experiment can be reliably measured. What happens however when the outcome is ambiguous, non-objective, not based on pure reason? In the recent book The Typographic Experiment: Radical Innovation in Contemporary Type Design, the author Teal Triggs asked thirty-seven internationally-recognized designers to define their understandings of the term experiment.

As expected, the published definitions couldn’t have been more disparate. They are marked by personal belief systems and biased by the experiences of the designers. While Hamish Muir of 8vo writes: ‘Every type job is experiment’, Melle Hammer insists that: ‘Experimental typography does not exist, nor ever has’. So how is it possible that there are such diverse understandings of a term that is so commonly used?

Among the designers’ various interpretations, two notions of experimentation were dominant. The first one was formulated by the American designer David Carson: ‘Experimental is something I haven’t tried before … something that hasn’t been seen and heard’. Carson and several other designers suggest that the nature of experiment lies in the formal novelty of the result. There are many precedents for this opinion, but in an era when information travels faster than ever before and when we have achieved unprecedented archival of information, it becomes significantly more difficult to claim a complete novelty of forms. While over ninety years ago Kurt Schwitters proclaimed that to ‘do it in a way that no one has done it before’ was sufficient for the definition of the new typography of his day — and his work was an appropriate example of such an approach — today things are different. Designers are more aware of the body of work and the discourse accompanying it. Proclaiming novelty today can seem like historical ignorance on a designer’s part.

Interestingly, Carson’s statement also suggests that the essence of experimentation is in going against the prevailing patterns, rather than being guided by conventions. This is directly opposed to the scientific usage of the word, where an experiment is designed to add to the accumulation of knowledge; in design, where results are measured subjectively, there is a tendency to go against the generally accepted base of knowledge. In science a single person can make valuable experiments, but a design experiment that is rooted in anti-conventionalism can only exist against the background of other — conventional — solutions. In this sense, it would be impossible to experiment if one were the only designer on earth, because there would be no standard for the experiment. Anti-conventionalism requires going against prevailing styles, which is perceived as conventional. If more designers joined forces and worked in a similar fashion, the scale would change, and the former convention would become anti-conventional. The fate of such experimentation is a permanent confrontation with the mainstream; a circular, cyclical race, where it is not certain who is chasing whom.

Does type design and typography allow an experimental approach at all? The alphabet is by its very nature dependent on and defined by conventions. Type design that is not bound by convention is like a private language: both lack the ability to communicate. Yet it is precisely the constraints of the alphabet which inspire many designers. A recent example is the work of Thomas Huot-Marchand, a French postgraduate student of type-design who investigates the limits of legibility while physically reducing the basic forms of the alphabet. Minuscule is his project of size-specific typography. While the letters for regular reading sizes are very close to conventional book typefaces, each step down in size results in simplification of the letter-shapes. In the extremely small sizes (2pt) Miniscule becomes an abstract reduction of the alphabet, free of all the details and optical corrections which are usual for fonts designed for text reading. Huot-Marchand’s project builds upon the work of French ophthalmologist Louis Emile Javal, who published similar research at the beginning of the 20th century. The practical contribution of both projects is limited, since the reading process is still guided by the physical limitations of the human eye, however, Huot-Marchand and Javal both investigate the constraints of legibility within which typography functions.

The second dominant notion of experiment in The Typographic Experiment was formulated by Michael Worthington, a British designer and educator based in the USA: ‘True experimentation means to take risks.’ If taken literally, such a statement is of little value: immediately we would ask what is at stake and what typographers are really risking. Worthington, however, is referring to the risk involved with not knowing the exact outcome of the experiment in which the designers are engaged.

A similar definition is offered by the E.A.T. (Experiment And Typography) exhibition presenting 35 type designers and typographers from the Czech Republic and Slovakia, which coincidentally will arrive in the Netherlands shortly. Alan Záruba and Johanna Balušíková, the curators of E.A.T. put their focus on development and process when describing the concept of the exhibition: ‘The show focuses on projects which document the development of designers’ ideas. Attention is paid to the process of creating innovative solutions in the field of type design and typography, often engaging experimental processes as a means to approach unknown territory.’

An experiment in this sense has no preconceived idea of the outcome; it only sets out to determine a cause-and-effect relationship. As such, experimentation is a method of working which is contrary to production-oriented design, where the aim of the process is not to create something new, but to achieve an already known, pre-formulated result.

Belgian designer Brecht Cuppens has created Sprawl, an experimental typeface based on cartography, which takes into account the density of population in Belgium. In Sprawl, the silhouette of each letter is identical, so that when typed they lock into each other. The filling of the letters however varies according to the frequency of use of the letter in the Dutch language. The most frequently used letter (e) represents the highest density of population. The most infrequently used letter (q) corresponds to the lowest density. Setting a sample text creates a Cuppens representation of the Belgian landscape.

Another example of experiment as a process of creation without anticipation of the fixed result is an online project . Ortho-type Trio of authors, Enrico Bravi, Mikkel Crone Koser, and Paolo Palma, describe ortho-type as ‘an exercise in perception, a stimulus for the mind and the eye to pick out and process three-dimensional planes on a flat surface…’. Ortho-type is an online application of a typeface designed to be recognizable in three dimensions. In each view, the viewer can set any of the available variables: length, breadth, depth, thickness, colour and rotation, and generate multiple variations of the model. The user can also generate those variations as a traditional 2D PostScript font.

Although this kind of experimental process has no commercial application, its results may feed other experiments and be adapted to commercial activities. Once assimilated, the product is no longer experimental. David Carson may have started his formal experiments out of curiosity, but now similar formal solutions have been adapted by commercial giants such as Nike, Pepsi, or Sony.

Following this line, we can go further to suggest that no completed project can be seriously considered experimental. It is experimental only in the process of its creation. When completed it only becomes part of the body of work which it was meant to challenge. As soon as the experiment achieves its final form it can be named, categorized and analyzed according to any conventional system of classification and referencing.

An experimental technique which is frequently used is to bring together various working methods which are recognized separately but rarely combined. For example, language is studied systematically by linguists, who are chiefly interested in spoken languages and in the problems of analyzing them as they operate at a given point in time. Linguists rarely, however, venture into the visible representation of language, because they consider it artificial and thus secondary to spoken language. Typographers on the other hand are concerned with the appearance of type in print and other reproduction technologies; they often have substantial knowledge of composition, color theories, proportions, paper, etc., yet often lack knowledge of the language which they represent.

These contrasting interests are brought together in the work of Pierre di Sciullo, a French designer who pursues his typographic research in a wide variety of media.

His typeface Sintétik reduces the letters of the French alphabet to the core phonemes (sounds which distinguish one word from another) and compresses it to 16 characters. Di Sciullo stresses the economic aspect of such a system, with an average book being reduced by about 30% percent when multiple spellings of the same sound are made redundant. For example, the French words for skin (peaux) and pot (pot) are both reduced to the simplest representation of their pronunciation — po. Words set in Sintétik can be understood only when read aloud returning the reader to the medieval experience of oral reading.

Quantange is another font specific to the French language. It is basically a phonetic alphabet which visually suggests the pronunciation, rhythm and pace of reading. Every letter in Quantange has as many different shapes as there are ways of pronouncing it: the letter c for example has two forms because it can be pronounced as s or k. Di Sciullo suggests that Quantange would be particularly useful to foreign students of French or to actors and presenters who need to articulate the inflectional aspect of language not indicated by traditional scripts. This project builds on experiments of early avant-garde designers, the work of the Bauhaus, Kurt Schwitters, and Jan Tschichold.

Di Sciullo took inspiration from the reading process, when he designed a typeface for setting the horizontal palindromes of Georges Perec (Perec has written the longest palindrome on record, a poem of 1388 words which can be read both ways, see The typeface is a combination of lower and upper case and is designed to be read from both sides, left and right. (This is great news to every Bob, Hannah or Eve.) Di Sciullo’s typefaces are very playful and their practical aspects are limited, yet like the other presented examples of experiments in typography, his works points to previously unexplored areas of interest which enlarge our understanding of the field.

Although most of the examples shown here are marked by the recent shift of interest of European graphic design from forms to ideas, and the best examples combine both, there is no definitive explanation of what constitutes an experiment in typography. As the profession develops and more people practice this subtle art, we continually redefine the purpose of experimentation and become aware of its moving boundaries.