By Sebastien Hayez. Published January 02, 2023

History of Type Classification

Type classification: ordering type creation

As a graphic designer, one of my main passions is reading graphic design books, from monographs to theoretical essays. Then, I really love to classify my bookshelves in a semantic way.

Well, I guess graphic designers are good at creating order, hierarchizing and revealing structure. It’s finding the inner structure in texts and pictures in order to make sense for the reader. Classification is the science or art of creating order between objects or ideas: from the periodic table of elements, to the Diagnostic and Statistical Manual of Mental Disorders (DSM), classification is the map displaying the connections between elements in a rational way. And it’s always easy to find what you want when looking at a map, isn’t it?

In typography, type classification is useful for:

1. revealing how types are designed;

2. revealing the different styles or design approach;

3. a better communication when describing and choosing a type

4. composing your font pairing.


The biggest part of this chapter is taken from Classification Typographique by René Ponotin Communication et Langages, issue 81, 1989. 

When needed, the first word is the original term in the original language. An english traduction is proposed after the /.

1921, Francis Thibaudeau (France)

1929, Marius Audin (France)

- Gothiques / Black Letters
- de forme / titling
- de somme / body text
- bâtardes / fraktur
- Romaines / Roman
- Elzévirs / Elzevirs
- Type (as model type)
- de transition / Transition
- Didot / Modern Face
- Type (as model type)
- Egyptienne / Egyptian
- Latine / Latin
- Antique
- Cursives / Cursive

1952, Maximilien Vox (France)

- Humanes / Venetian
- Garaldes / Old-Face
- RĂ©ales / Transitional
- Didones / Modern Face
- MĂ©canes / Egyptian Slab
- Linéales / Sans serifs
- Incises / Incised Latin
- Manuaires / Hand drawn display
- Scriptes / Scripts1

1953, Berry-Johnson (UK)

18 categories

1953, Balding & Mansell (UK)

- VĂ©nitiennes / Venetian
- Caractères anciens / Old Faces
- Style Ancien / Antique Style
- Modernes / Modern Face
- Sans SĂ©rif / Sans Serif
- Egyptiennes / Egyptian Slab
- Gothique / Black letters
- Cursive / Cursive
- Fantaisie / Fancy

1953, Bastien (UK)

- Romain vieux style / roman old style
- Romain moderne / roman modern
- Caractères gras / bold
- Romain type décoratif / roman type ornated
- Caractères sans empattement / sans serif
- Caractères sans empattement de type décoratif / sans serif ornated
- Égyptiennes / egyptian
- Caractères d'écriture à effet de pinceau lourd et léger / bold heavy brush
- Écritures conventionnelles / conventionnal writting
- Calligraphiques / calligraphic
- Blackletters
- Cursives

1955, John C. Tarr (UK)

38 groups in 11 categories

1957, Aldo Novarese (Italy)

- Lapidaires / Incised Latin
- Médiévaux / Blackletters
- VĂ©nitiens / Venitians
- Transitionnels / Transitional
- Bodoniens / Modern Face
- Ornées / Ornate
- Égyptiennes / Egyptian Slab
- Linéaires / Sans Serifs
- Fantaisies / Fancy
- Écritures / Script

1958, R. H. Munsch (France)

10 categories.

1962, Vox-AtypI, Maximilien Vox (International)

- Humanes / Venetian
- Garaldes / Old-Face
- RĂ©ales / Transitional
- Didones / Modern Face
- MĂ©canes / Egyptian Slab
- Linéales / Sans serifs
- Incises / Incised Latin
- Manuaires / Hand drawn display
- Scriptes / Scripts
+ Gothiques /Black Letters
+ Non-Latin

1963, Pellitteri (Italy)

- Linéaires / Sans Serifs
- Rectiformes / Slab
- Anguliformes / Angular-shapes
- Curviformes / Curvy-shapes
- Dégradés / Gradient
- Contrastés / Contrasted
- Scripts / Scripts
- Brisés / Brocken
- Ornementés / Ornate
- Hybrides et aberrants / Hybrids & aberrants

1978, Jacno (France)

- Linéale / Sans Serif
- Romain Ancient / Ancient Roman
- Romain Moderne / Modern Roman
- Egyptienne / Egyptian Slab

1979, Codex 1980, Jean Alessandrini (France)

The difficulty of explaining Alessandrini’s classification remains that the word used to describe the font are neologisms in order to stay away from Vox’s classification.

More than a solid classification Codex 1980 is a modular classification: a font can be described as various categories.

- Simplices / Sans serif
- Emparectes / Slab
- Emparectes à congés / Rounded slab
- Deltapodes / Triangular serif
- Deltapodes à congés / Rounded triangular serif
- Filextres / Modern face
- Filextres à congés / Rounded modern face
- Claviennes / Nail shaped serif
- Romaines / Incised Latin
- Onciales / Uncials
- Gestuelles calligraphiques / Gestual calligraphic
- Gestuelles brossées / Gestual brushed
- Germanes / Black Letters
- Exotypes / Latin alphabet inspired by foreign writing systems
- Ludiques / Playful
- Machinales computrices / Computing machines
- Machinales cathodiques / Cathode-ray machines
- Machinales modulaires / Modular machines
- Hybrides
- Transfuges / Transfers (i.e.: various weight in the same style)
- Diagones / Italics or Obliques
- Stenciliennes / Stencil
- Aliennes / Non-latin writing systems

2000, Robert Bringhurst (USA)


While reading all these classifications, the surprising point is that different rational aspects are gathered in the same system. Historical styles, tools and writing systems are gathered inside the Vox-Atyp-I classification. Even if its creator, Maximilien Vox argues that a typeface can be classified using more than two categories, today’s font are totally different from what was possible in 1952.

In 2022, a single font file can embed a latin alphabet, other writing systems, a set of dingbats and the variable possibilities offer also interpolable variations like weight and axis variations, if not other possibilities.

In a more contemporary way, the classification proposed by Jean Alessandrini in 1979, the Codex 1980, is in a way more flexible because it uses various parameters to classify typefaces.

The debate is still vivid in the typographic circle where youngsters and elders are still trying to resolve this enigma with a vibrant romantic and amateurish passion.

A classification is a system based on rational criteria. It can’t be possible to establish a classification based on the aspect (sans or serif), the tool used (latin incised), the historical influence (Venetian, Old faces, Transitional), etc. A unique method is needed, a scientific approach. 

During the medieval times, some animals were considered close to some others regarding their appearance. But scientifically, a rhinoceros is close to the horse, as both belong to the Perissodactyla with various other animals.

More than the physical aspects, we need to know deeply the links between the objects of study.


In 2021, the AtypI association declared that it left the Vox-AtypI classification beside because of a need for a more inclusive system incorporating the non-latin writing system. The moral purpose has to be recognized as a first step from an occidental point of view to an intercultural vision. While we are waiting for the release of a new type classification system, let’s look further at how a scientific approach can bring some solution.

Scientific classification

A classification system is an organized and hierarchical system of categorizing objects. It involves a set of interacting elements that are hierarchical (the value is not moral but mostly of the order of filiation, of antecedent between a first object and the one that follows).

Categorisation is a mental activity that consists of placing a set of objects in different categories (classes, types, taxa) according to their similarities or common criteria.

Defining these similarities or common criteria amounts to defining identical identities, forms or typologies. The question is on what basis these similarities are based and identified.

Classification can be observed with two points of view:

1. Conceptual classification: "Classification is intrinsic to the use of language, hence to most if not all communication. Whenever we use nominative phrases we are classifying the designated subject as being importantly similar to other entities bearing the same designation; that is, we classify them together. Similarly the use of predicative phrases classifies actions or properties as being of a particular kind. We call this conceptual classification, since it refers to the classification involved in conceptualizing our experiences and surroundings"

2. Systematic classification: "A second, narrower sense of classification is the systematic classification involved in the design and utilization of taxonomic schemes such as the biological classification of animals and plants by genus and species.”.

Type classification seems to be inefficient regarding a conceptual classification. The systematic classification remains the only other method.

Perhaps the best example of such a classification system is the phylogenetics system for classifying the living organisms. This science is part of the systematics (or taxonomy)..

The inference of phylogenetic tree is based on:

1. Describe all the characteristics of the observed objects to classify.

2. Gather the criterion in order to establish various possible models. The best one, the optimal criterion, is the one that fits all the datas. When found, the system should fit these points:

2.a. 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.

2.b. maximum likelihood estimation: this statistical method estimates the probability for some parameters to fit a category.

3. But everything depends on the mathematical model chosen. For example, the variation of the fall of a water droplet is random. But seen on mathematical diagrams, it’s just fractal: not prediction but a model that appears because of the way we organize the datas.

Description & CEDARS+

It’s now time to try to describe a font. Let’s look at its aspect, its morphology.
The most complete font description system designed recently is called CEDARS+, acronym for its 6 basics parameters:

- 6 basics parameters: Contrast, Energy, Details, Axis, Rhythm, and Structure.
- 4 advanced descriptors: Hx, serifs, waist, advanced widths
- 7 descriptors for special glyph: a, c, e, g, K, M, R.

This description system remains completely incomplete. What about variable fonts, when all the 6 basic parameters can be linked together? What about multilingual Opentype font, when latin, greek and cyrillic alphabets are gathered with arabic and probably chinese ideograms? What about a fancy font inspired by Dada, Fluxus or Punk aesthetic, mixing hazardous styles in one font file ?


Maximilien Vox was classifying a font, movable type object based on the latin alphabet.

Just look at Univers by Adrian Frutiger. Designed as a set of 27 fonts, you can now purchase the whole collection in 2 files: roman and italic for extended latin glyphset.

The first step should be to look at writing system classification.

Click here to view the full size graphic

Remember that this map doesn't show the asian writing systems (chinese, japanese and korean) and probably ignore the rarest ones included in the extended latin glyphset.

This first step is quite complexe, so imagine going further…

A type classification should remain pragmatic, if you can’t use it, so why produce such a tool? If the living being are classified in a scientific way, nobody is looking for an animalia vertebrata tetrapoda mammalia placentalia carnivora feliforma felidae felinae felini, to be short, a cat. In the same way botanics are using vernacular names and botanical ones in latin.

In the same way, when you are looking for a book, you don’t follow any library classification system.

Because classifying the living is understanding its natural evolution, while type design is not natural, it’s a human activity linked with social contexts, art and techniques. Like any creative practice, working with science is a great deal for a new approach, for developing new tools, but Art can’t be observed as a rational practice: it’s deeply personal and subjective. Have we ever seen a scholar classifying Picasso's paintings with style criteria? The chronology remains enough.

Look at your favorite type foundries. Very few are indexing their fonts in the Vox-AtypI classification. When your font catalog is less than 50 font families, custom categories are efficient. When the catalog includes more than 50, categories and keywords work well.

Also because fonts are products, they are published, like books, regarding editorial criteria, and the criteria defined the whole difference between one type foundry and its concurrent. Your Helvetica look alike font can be categorized as Sans, Modernist, Grotesk, Neo Grotesk, Essential, Basic, etc.

Let’s work with linguists, mathematicians and type scholars for a new classification, but who would ever use it? Paleograph are using their own criteria. It would be useful for a very few people even if it would be a great task to solve such an equation with so many unknown. During this time, while the AtypI is working hard, everything continues in his own life. Like an old grammarian prescribing grammar rules, language and its users are smarter than that, type design is more powerful than classification systems.