By Sebastien Hayez. Published August 28, 2023


​VoxAtyp-I :
why it’s not enough, why it’s difficult to do better ?

In January 2023, we had analyzed a bit of the history of typographic classifications.

The Vox-ATyp-I classification remains the only typographic classification shared by the whole world: it was adopted, with some modifications, as a German standard (DIN 16518) in 1964, as a British standard (British Standard 2961:1967) in 1967, and as a French standard (NF Q60-007) in 1977.

Developed in 1954 by the Frenchman Maximilien Vox (1884-1974), it was adopted by the ATypI in 1962. But the association announces on its website that this classification will be de-adopted on April 21, 2021.

The reasons are:

“At a regular meeting held online on March 18, 2021, the ATypI board of directors passed two resolutions—one formalizing the de-adoption of the Vox-ATypI system and another directing the establishment of a working group to study the feasibility of developing and collaborating on contemporary typeface classification systems that embrace the wealth of writing systems from around the world.”

The merits of this initiative are not in question. While the designers of modernism worked almost exclusively on typefaces for the Latin writing system, countries using other writing systems were relegated to the sidelines. Worse, the meager presence of non-Latin typefaces reinforced the presence of transcriptions systems based on the Latin alphabet.

Two years after this announcement, there is little evidence of work on a new classification system to replace the Vox-AtypI. Thus, the imperfect or failing classification still remains the reference from the point of view of the standard norms than the profession.

Several points are important to emphasize in order to understand why the Vox-AtypI classification remains an effective, though imperfect, tool.

Before Vox's classification, another french graphic designer, Francis Thibaudeau (1860-1925) published a french manual of typography in 1924 with its own classification.
Thibeaudeau classification remains a taxonomy based in the form of the serif, quite close to Vox's classification but with fewer categories.


The domination of a Western economy over the whole world is a fact that one can deplore, but that one cannot dispute. Even if Asia and China in particular are competitors, the Chinese ideogrammic writing system is difficult to learn for a Westerner as well as for an Asian. The speed of learning the alphabet, Latin as well as Greek or Cyrillic and the ease of use of these writing systems thus allow the Chinese to facilitate the pronunciation of the language via the transcription in Latin characters, the pinyin.

One could say that the invisibilization of non-Latin writing systems in favor of the Latin script is similar to the loss of influence of national and dialect languages in favor of English. Globalization and capitalism imply a standardization that is as beneficial as it is detrimental to the cultural richness of each language and each writing system.

However, commercially, the typographic market is naturally drawn to marketing typefaces that can cover the largest market. Similarly, training in type design is more prevalent among countries where Latin script is covered. Culturally, learning typographic design for non-Latin writing systems for a Latin designer is a significant challenge. The eye needs to be more than accustomed: it needs to be immersed in the script, its history and its practice.

By focusing more on the Latin script, the Vox-AtypI classification covers a small field of writing systems, but a large field of typefaces on the market.


To understand the modelling logic of a classification is to understand that it is a question of bringing out the logic that links each item. Thus, the classification of living beings brings out the common points and morphological differences corresponding to the Darwinian evolution of the tree of life: between common ancestors, various branches separate and show the natural evolution during life.

Classification should not be a goal, but a method of visualizing structural principles. Typography is an applied art practice. In this, it is part of a creative process where freedom is one of the possible parameters. To classify in a logical way what turns out to be illogical is thus an all too human utopia. If I decide to create a typography where each glyph is a character without any link with the next one, I would have to create a "Miscelaneous" category thus breaking the rule of each classification:

"The system should be minimal: maximum parsimony is the best proof of the efficiency of the system. If you need 5000 categories to classify 5000 objects, then it's not a classification.".

Classifying a typography can be artificial since it would be useless to classify a piece of music by trying to include the musical genre (classical, baroque, jazz, pop, etc.), the instruments used (piano, violin, classical orchestra, voice, etc.), the form of composition (opera, concerto, etc.) and other indications of rhythm for example.

To classify a typography according to the will of inclusiveness dear to ATypI would mean crossing parameters as different as the writing system, the structure of the serifs, the angle of the axis of the barrels, the writing tools, etc. So many parameters that are difficult to combine. So many parameters that are difficult to combine.

Jean Alessandrini (1942-) developped Vox's classification but with a conbinatorial mood. Codex 1980 is able to describe a font with a great accurenty and fantasy too.


The Vox-AtypI classification is more of a standard than a tool. In fact, it has not been used for many years and I have not come across a foundry website or a catalog hierarchically classified by the Vox-AtypI for a long time.

Foundries are in the business of providing an editorial line. No book publisher structures its collections based on library classification standards.

The Vox-Atyp-I classification is therefore a theoretical tool used only by type designers and graphic artists to define or describe basically a font. Paleographers have other classification criteria and do not seek a categorization encompassing all parameters.

It would be tempting to use CEDARS+ as a descriptive tool, but again, describing is not cataloging, creating a branch for a typeface, is not classification.

That the AtypI manages to create a classification is highly probable, that it is coherent seems unlikely, that it is used seems utopian.

Typographies are identities inscribed in the letter. As such, they carry a constellation of criteria whose impressionistic composition gives the illusion of a solid unity.