Menu Search
Jump to the content X X
Smashing Conf San Francisco

We use ad-blockers as well, you know. We gotta keep those servers running though. Did you know that we publish useful books and run friendly conferences — crafted for pros like yourself? E.g. upcoming SmashingConf San Francisco, dedicated to smart front-end techniques and design patterns.

Making Sense Of Type Classification (Part 1)

In my previous article on Smashing Magazine (“Understanding the Difference Between Typography and Lettering1”), I wrote about how understanding type terminology can help us better appreciate the arts of typography and lettering. This article again deals with terminology, probably more specifically than most designers are used to, and the title gets to the heart of what I’m communicating in this article.

Everyone knows their serifs and sans, slabs and scripts, but most classifications go much deeper than that. Type classification, while helpful, is often convoluted, confusing and even controversial. This article, distilling some of the complexities into a more understandable format, lands somewhere in the middle between the basics and genuine type nerdery — the perfect level for a practicing designer.

Further Reading on SmashingMag: Link

Why Classify Type? Link

There’s a certain intellectual delight in knowledge, particularly knowledge about one’s field of work and study. More importantly, perhaps, there is a way in which seemingly impractical knowledge of one’s profession lends more credence to the designer. That being said, what you’ll read here is by no means impractical. It really comes down to solid design choices.

Making Sense of Type Classification

A good grasp of type history will help you avoid typographic anachronisms, which, although often lost on the general public, do not escape the notice of many designers, as demonstrated in Mark Simonson’s article6 on the 2012 Oscar winner for Best Picture, “The Artist,” and his other typographic scrutinies7 of popular movies and media.

It’s not exclusively about the history of type, however. Type classification is also helpful in pairing typefaces for projects, sometimes based on historical proximity but also by noting similar features that unify the typefaces, such as axis or x-height. In some cases, by finding enough disparity in the small features, very different typefaces become complementary.

Most importantly, perhaps, this article will not only familiarize you with general type history and commonly used terminology, but also help you learn to look for and recognize important characteristics of type and the inexhaustible minutiae that make typefaces unique, as well as arm you with useful descriptors of type styles.

Type Classification Systems Link

Over the past century, quite a few classification systems have been proposed. Most are generally believed to be subjective and incomplete, and many of them use the same terms for similar but slightly different classes. The impossibility of a truly complete classification system has led many people to dismiss any attempt to classify typefaces — there are simply too many variables to make anything close to a practical, comprehensive system. Essentially, classification describes typefaces; it does not define them. It’s not inflexible, and is more of an aid than a rule. However, for the reasons given above, I believe there is value to be found in it. Below are a few examples.

Sets film in 1920’s uses typeface from 1975.

The primary “official” classification system currently is the Vox-ATypI system. Originally put together in 1954 by Maxmilien Vox, it was adopted in 1962 by the Association Typographique Internationale (ATypI), which made a minor change at the 2010 conference (appropriately, held in Dublin) to include Gaelic as an extra category. It classifies typefaces in 11 general categories, with some subdivision. Its Wikipedia article provides an excellent overview9.

The British Standards Classification of Typefaces10, adopted in 1967, is also based on Vox’s original classification. It is slightly simplified and has remained essentially unchanged since its adoption.

Bringhurst, in his Elements of Typographic Style11 — perhaps the standard in typographic textbooks today — categorizes typefaces loosely after periods of art history; for example, Baroque, Rococo, Romantic, etc. A book designer himself, Bringhurst focuses on text typefaces and practically ignores display type.

Others are much more general. An early system by French typographer Francis Thibaudeau12, which provided the base for Vox’s later more thorough classification, includes four broad categories: Antiques (sans serifs), Égyptiennes (slab serifs), Didots and Elzévirs (faces with triangular serifs).

Gerrit Noordzij, while at the Royal Academy of Fine Arts in the Hague, held that typography was essentially an extension of handwriting13, teaching typography using loose categories of letters that might be written with a broad-nib or pointed-nib pen, as well as interrupted or uninterrupted strokes, with varieties of both serifs and sans falling into each category.

These are just a few of the ways people have classified type over the years. In this two-part article, I will condense the various methods slightly and present what is at the very least generally accepted as legitimate (as there will always be a few out there who refuse to give up a particularly unusual classification method, or who decry any method at all).

Classifying Serif Typefaces Link

In part 1 here, we’ll cover serif styles, following the natural progression of type history, and thus move into sans and other categories in part 2.

Humanist Link

Starting off, naturally, at the beginning of type history, we’re in the middle 1400s, during the Renaissance. The movement, led by Italian cultural hubs such as Florence and Venice, was drawing Europe away from medieval practices, and typography was one part of that. Rather than using the blackletter, or Fraktur type, that Gutenberg used, printers began to create type mimicking the Latin writing hand of the philosophers and scribes of the time, beginning around 1465.

Renaissance Printing
A 1905 textbook illustration of Renaissance printers

These typefaces are variously called Humanist or Venetian due to the zeitgeist and geography of the Renaissance. A number of distinct characteristics define Humanist typefaces.

Primarily, Humanist faces were very calligraphic in nature, and one way this manifested itself was in the strong axis, most apparent in the bowls of characters and the lowercase “o,” a characteristic borrowed from the angle at which a right-handed writer holds a pen. Another interesting way this showed itself was in the notably angled crossbar on the lowercase “e.” Other calligraphic influences are clear, such as inconsistencies in stroke weight and the way serifs are formed.

Other defining characteristics include a small x-height and a lower contrast between thick and thin strokes.

Venetian Typeface Characteristics

Not all Humanist typefaces are from the Renaissance era, however; many Humanist revivals have been created in more recent years, such as Centaur (1914) and Adobe Jenson (1996). Adobe Jenson, used in the specimen above, is based on the work of Renaissance printer Nicolas Jenson, a prominent printer and type designer who moved from his native France to Venice and contributed significantly to print and design history. There are even Humanist sans faces, but we’ll get to those in part 2. Although an influential period in type history, the Humanist era served primarily as a transition to newer styles of typefaces and was relatively brief.

Other examples of Humanist typefaces include Guardi, Arno, ITC Berkeley and Stempel Schneidler.

Garalde Link

In the Old-Style faces, often called Garaldes, we see type really beginning to come into its own. I call them Garaldes here because the term “Old Style” is at times used to include Humanist, Garalde and Transitional typefaces; simply calling this group “Garalde” helps to retain its identity.

Aldus Manutius and Claude Garamont
Aldus Manutius and Claude Garamont

This period in type history lasts from the late-15th century all the way until the early 1700s, and the type created in this period has shown remarkable longevity. “Garalde” itself is a hybrid term borrowed from the names of two notable type designers of the era, French punchcutter Claude Garamont and the Venetian Aldus Manutius. The category is occasionally called “Aldine” after Manutius.

There are many similarities to the Humanist faces, but things are moving in a particular direction, as we’ll see with the consecutive categories of Transitional and Didone. You can see the type designers treating type as different from the written word, losing some of the idiosyncrasies of handwriting that the Humanist designers retained, while carrying over others. The axis of the stress straightens, and while it still has an angle, it is subtler. The serifs become more carefully formed, and characters are designed more proportionately. One of the most obvious differences is the crossbar of the lowercase “e,” which, while remaining angled in the Humanist typefaces, drops to a horizontal position in the Garaldes. Also, the difference between heavy and light stroke weights increased, and everything became more precise, perhaps due to the progress in technical aspects of making type.

Old Style Typeface Characteristics

A huge amount of type was created in this era, and much of it is commonly used today, either digitized versions or new revivals. Common examples of the Garalde faces include Caslon, Sabon, Palatino, Galliard and Janson — not to be confused with Jenson, the Humanist typeface. In fact, Janson, named after Dutch punchcutter Anton Janson, is now thought to be the work of Miklós Kis, a Hungarian, produced during an apprenticeship in Amsterdam.

Transitional Link

Work was begun on the first Transitional typeface in 1692, long before people had left behind making Garaldes. In fact, William Caslon was creating typefaces based on Old-Style Dutch type as late as the 1720s. Because this part of type history is also significant, many have asserted that “Transitional” is an inadequate name for it, and this category may also be termed Neoclassical or Realist.

Louis XIV and the Romain du Roi
Louis XIV and the Romain du Roi

In the late-17th century, Louis XIV, as part of a general renovation of France’s Imprimerie Royale (the governmental printing works), commissioned the French Academy of Sciences to create a new typeface. The Romain du Roi — literally the “King’s Roman” — was designed using a strict grid, and its development was an arduous process, involving a committee that included a mathematician and an engineer. Although commissioned in 1692, the entire family of 86 fonts was not completed until 1745.

Baskerville and Fournier
John Baskerville (left) and Pierre Simon Fournier (right)

Two of the biggest names in type during this period were John Baskerville and Pierre Simon Fournier. Baskerville, an entrepreneur who dabbled in multiple businesses, developed quite an interest in printing and eventually designed his own type in order to improve on Caslon’s work. This did not please most of the printing world at the time, and Baskerville endured harsh criticism, despite having such luminaries as Benjamin Franklin as friends and advocates of his work. You may have read of the humorous encounter14 in which Franklin outwitted a critic of Baskerville. Numerous revivals, both metal and digital type, that draw on Baskerville have been made.

Fournier was among the printers who praised Baskerville’s type, reserving particularly high compliments for his italics. Fournier was highly respected in his lifetime, and despite having consulted royalty both within France and internationally on type design and having established printing houses, Fournier is primarily remembered today for introducing the point system as a way to measure type sizes. Pierre Fournier, uncannily sharing a name with an acclaimed 20th-century cellist, also had an interest in music and developed a new style of typography for musical notation.

Transitional Type Characteristics

In the Transitionals (or Neoclassicals), we see certain trends continuing. The axis is now nearly, if not completely, vertical. The weight difference between the thickest and thinnest points is now exaggerated. The serifs are less bracketed and flatten out. Details become very refined.

Eric Gill’s Joanna, Melior, Clearface and Mrs. Eaves — a Baskerville revival named after Sarah Eaves, Baskerville’s wife — all fall into this category.

Didone Link

As strange as it seems, what we call modern typefaces first appeared in the second half of the 1700s. Therefore, I will call them by their less absurd name — and who can argue that saying “Didone” is not more fun than saying simply “Modern”? Bringhurst terms them Romantics.

Through the 18th and 19th centuries, France witnessed a small printing dynasty in the Didot family. Over multiple generations, the family made major contributions to printing. One of the most remarkable members was Firmin Didot, who, with Giambattista Bodoni, ushered in and now acts as a namesake for this part of type history.

F. Didot and G. Bodoni
Firmin Didot (left) and Giambattista Bodoni (right)

In large part inspired by Baskerville, Didot and Bodoni pushed the limits of type design. They explored a similar style and were both meticulous craftsmen, consequently igniting a fierce rivalry. Bodoni (1740–1813) gave himself entirely to his craft. He was renowned for the beauty of his type specimens, and, a technically brilliant punchcutter himself, he designed some 298 typefaces. Didot (1764–1836), on the other hand, retired in 1827 to pursue political office and literature in his later years, writing tragedies and literary critiques.

If Baskerville’s stroke contrast was exaggerated, then the Didones’ are in the extreme. The heavy strokes are very heavy, and the light are a hairline. The stress is again completely vertical, and the apertures — places where the character opens — are generally very tight. Combined, these make for a very awkward visual rhythm, and Didones are always a poor choice for chunks of text. Rather, they work best at large sizes, as titling and display type, because the features emphasize the elegance of individual characters and do not blend well. Adobe’s New Caledonia15, which softens some extremes and thus works for longer bits of text, is a possible exception.

Didone Characteristics

Aside from the obvious Bodoni and Didot faces, in their dozens of variants from nearly every foundry, Basilia, Aviano, Walbaum, Ambroise and Scotch Roman are exemplary moderns.

Slab Serif Link

Slab Serif Characteristics

This article wouldn’t be complete without a mention of slab serifs. These are among the easiest to identify because of their very obvious appearance. Originally created for advertising, posters and other large media, slab serifs, alternatively called “Mechanicals” (in VOX-ATypI) and “Égyptiennes” (by Thibaudeau), were the first types expressly designed as display type. Vincent Figgins is credited with the first slab serifs, the earliest specimen dating to 1815, and his work inspired a diversity of critiques variously commending and lambasting the new style.

Abrupt serifs, usually in heavy weights, and a no-nonsense attitude are the trademarks of this style.

Clarendon characteristics

Clarendons, a notable offshoot of the original slab serifs, are a slightly tamed slab style, often in less extreme weights and using bracketed serifs. They have a lighter, friendlier character than the Neo-Grotesque slabs (i.e. those with unbracketed serifs and geometric construction).

H&FJ’s Sentinel (2009) and David Berlow’s Belizio (1998) are examples of recent Clarendons.

And That’s It… For Now Link

If you have made it this far in the article, congratulations! You are now in possession of a solid basic understanding of type classification, at least as far as serif typefaces take you, and you are able to recognize the important distinguishing features that make typefaces unique. Following the line of type history, we’re now in the middle of the 19th century, and we have the entirety of sans serifs and some discussion of display faces ahead of us. We’re really only halfway through, so if you’ve enjoyed it, you can look forward to part 2!

For now, here’s a little exercise to test your comprehension of what we’ve covered in this article so far. Take a look at these specimens and comment on how you’d classify them. Keep in mind that classification is an aid, rather than a hard and fast system, so don’t be shy — let us know where you’d place these typefaces!

Typography Test Specimen

Identify each typeface by its number (1 to 6) if you are classifying it in the comments. Extra points if you can identify the individual typefaces! I’ll be joining the discussion with the answers later, although I am sure you’ll have figured them out soon.


Footnotes Link

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15

↑ Back to top Tweet itShare on Facebook

Joseph Alessio is a lettering artist and designer from the Detroit area. He has worked with companies such as Patagonia, Reach Records, Monotype and the Art Director's Club. When not working, he plays 7 musical instruments and reads classic literature. You can keep up with some of what he's doing on his Dribbble profile, and he often talks about life, design and current events in 140 characters.

  1. 1

    My guess

    3.slab serif

  2. 2

    It’s a shame that Google Fonts don’t use a more commonly accepted (and useful) from of type classification rather than just limiting it to Serif, Sans Serif, Display and Handwriting!

    Nice article – looking forward to part 2.

    • 3


      1 – Garalde/Old-Style (Garamond)
      2 – Didone/Modern (Modern 20)
      3 – Transitional/Neo-Classical (Cambria)
      4 – Humanist/Venetian (Centaur)
      5 – Transitional/Neo-Classical (Perpetua)
      6 – Garalde/Old-Style (Minion Pro)

  3. 4

    Douglas Bonneville

    April 17, 2013 5:33 pm

    Good stuff. To add some examples to this article, here is a popular list of the top 100 fonts from fontshop (by sales) but also categorized and sortable by font classification (and date, category, designer, foundry, etc.)

  4. 5

    Excellent! Jaime is close but not quite on a couple of samples—note the larger x-height and horizontal ‘e’ crossbar on 1, and the angled ‘e’ crossbar and strong axis on 4. 3 is a tricky one; it is indeed Cambria, which was designed in 2004 for digital use, thus the large x-height and heavier serifs. However, the serifs don’t figure prominently enough to call it a slab. Roy is more correct in calling it a transitional, since it has a rational axis and flat serifs on ascenders.

    Roy has nailed the classification, even identifying all but 1 of the typefaces correctly! Can anyone tell me which one is the odd one out, and identify it for us?

    • 6

      Ethan Resnick

      April 4, 2014 6:26 am

      I know I’m late to the party, but I actually think Cambria is a good example of how type classification systems are *not* always the best way to think about things (as the article noted). Yes, if I had to pick I’d probably call Cambria a transitional but, to my eye at least, the stroke contrast is too low and the characters too wide for it to fit in with the historical “transitional” model. It looks like a hybrid influenced by the constraints of the screen, etc. Maybe a bit of a confusing example for this article.

  5. 7

    I’m pretty sure 1 is not Garamond. Is it Goudy Old Style?

  6. 8

    1. GARALDE (high x-height, angle of the “o”) – Garamond (very distinguished “W”)
    2. DIDONE (extreme stroke contrast, unbracketed)
    3. TRANSITIONAL (high contrast stroke, angle of the “o” is vertical) – Cambria
    4. HUMANIST (angle of the “o” is +/- 16°)
    5. GARALDE (horizontal bar of the “e” )
    6. GARALDE (just a guess, high x-height, angle of the “o”)

    Looking forward to part 2!

  7. 9

    Am I the only designer that always gets nervous when reading about typography?

    I really should know more about it, but I just can’t seem to focus on it long enough. My interests lie more towards usability and technology. I know it comes into play when doing UI and graphic design, but I just can’t seem to shake this feeling of “I could be learning how to code”. Even through a decade of working in web-design, it just doesn’t ‘click’ for me. I get the jist of the basics (x-height, contrast, ligatures, etc) and I’ll tell you what’s Helvetica and what’s Arial, but I would never count myself as an “expert” on type.

    Of course, this is by no means meant as an attack towards you, Joseph; I’m merely saying that I lack the capacity to absorb this material in the way it deserves to be.

    I guess what I’m trying to say is: I’m sorry!

    • 10

      Trust your eyeballs: everything else is a footnote.

    • 11

      Dirk, I can definitely understand where you’re coming from! However, I’d say it is definitely worth the effort. It’s been claimed that the web is 95% typography, and with the tools available to us today, we have huge capability for applying age-old typographic principles on the web. The web isn’t just print design’s little brother anymore, and deserves the same level of education that we apply when learning physical publication design.

      So, even if it’s a little intimidating—I know it was quite a long article!—come back and read the article again when you’re feeling up for it; hopefully you’ll be able to derive some enjoyment and useful information the second time around!

  8. 12

    Ben Spatafora

    April 18, 2013 4:08 am

    Big important question that I for some reason can find nowhere on the internet: How do you pronounce Garalde and Didone?

    I’m assuming based on the pronunciation of “Bodoni” that Didone is pronounced with a long “e” at the end, but by this logic (basing the pronunciations of the portmantesques on the pronunciations of the names they are formed from) the “e” at the end of Garalde should be short, but this sounds funny.

  9. 13

    Daniel is correct—it’s Goudy Old Style, most evident in the very distinct ‘a’ and ‘r’. Good eye!

    Broeiend—very close; all but one are correct! 5 does indeed have a horizontal ‘e’ crossbar, but the vertical axis and higher contrast makes it a Transitional/Neoclassical. #1 is actually Goudy Old Style, which is somewhat similar to Garamond.

    • 14

      Ah – Garamond was the one I wasn’t 100% on. I have four different versions on my system and it wasn’t identical to any of them, but very close. I thought it might have been another version from a different foundry. Good call on Goudy by Daniel.

  10. 15

    Alright, I’ll give it a shot —
    #1 Not sure what to say about this one, at first I thought it was Geralde until I looked at #6 which seems a bit more suiting.. It still has a medium axis and angled serifs, but some elements make me think it’s a little more transitional because of the sylized dot on the i, the higher stroke contrast, and the more designed serifs.
    #2 Didone — stroke contrast is super extreme, apertures are super small.
    #3 Slab serif (abrupt) — strokes and serifs are low contrast and thick. Vertical axis.
    #4 Humanist — angles of serifs/crossbar of ‘e’ reflect handwriting.
    #5 Transitional — near vertical axis on lettering and flatter serifs. Higher stroke contrast
    #6 Geralde… ‘e’s crossbar has been straightened, axis is less extreme, though angles remain to reflect handwriting, like on the serif of the t, i, and d. Stroke contrast is relatively low.

    • 16

      Good work! #1 is indeed Garalde as you had originally supposed (I included 2 in a couple of categories, and didn’t include slab serifs). #3 is a tricky one. It’s probably better classed as a transitional/neoclassical, since although it has lower stroke contrast the serifs aren’t as prominent as they would need to be to make it a slab serif.

  11. 17

    Great article Joseph, very informative and just what I was looking for. Thanks for writing.

    Really looking forward to Part 2.

  12. 18

    It’s amazing that there is so much history behind the type we use day-in day-out as designers. We should really know more about it.

    Looking forward to part 2!

  13. 19

    How’s she doing? Tell me she isn’t still waking at 4:30 to shower.

  14. 20

    Thank you for the great article, Joseph. I really enjoyed reading about the history of these typefaces. You mentioned, Humanist type was largely “mimicking the Latin writing hand of the philosophers and scribes of the time.” Would you mind elaborating on some of the notable philosophers or scribes that you are referring to? Thanks again for keeping me inspired and evolving.

  15. 21

    Great series. I’d be delighted if someone ranked the legibility of these types in order for me. :)

    Obviously, I shouldn’t stick to this 100%, but it’s handy to have as a basic guide when selecting typefaces suitable for large bodies of text.

  16. 22

    I think the comment thread went off course from (1) the author’s intent, and (2) the needs of someone who wants to choose a typeface.

    1. The author wrote: “… classification is an aid, rather than a hard and fast system … let us know where you’d place these typefaces.” Instead, the comments focused on determining how they were officially classified by the “experts.”

    2. This kind of classification is no help to me when I want to choose a typeface. As a former volunteer newsletter editor for ten years, I used typefaces to print text, and text to convey meaning, mostly to people who didn’t know one serif face from another. When I look at text, the first thing I see is color, which consists of weight first, and then stroke contrast. Then I notice deviations from conventional letter shapes, first width, then x-height, and last, individual peculiarities like the Garamond “a” and the Venetian “e”. Serif forms (except for slab serifs, because of their weight) and axis inclination are at the bottom of the list, and most people won’t notice them, especially in text sizes. With the exception of the extreme stroke contrast of the Didones, historical classification systems turn that priority on its head.


↑ Back to top