​In the first part of this post, I took you through how our understanding of legibility evolved among scientists and psychophysicists. Now, I’m diving into what was happening among graphic and type designers during the same period.

We’ve already seen that the earliest relevant studies happened in the late 19th and early 20th centuries. But during that time, what’s really surprising is how little attention designers seemed to give to this topic. Even the big names offered pretty vague explanations. De Vinne, in The Practice of Typography 1De Vinne, Theodore Low. The practice of typography: A treatise on the processes of type-making, the point system, the names, sizes, styles, and prices of plain printing types. Vol. 1–4. Century Company, 1899. –arguably the most important text on typography since Fournier’s Manuel Typographique in 1764–, mentions legibility briefly but doesn’t actually define it. Edward Johnston (1872 – 1944), the renowned calligrapher and typographer, took a practical stance and said that the first general virtue of lettering is ‘readableness’, which he described as simplicity, distinctiveness, and proportion2Johnston, Edward. Writing & illuminating, & lettering. London: J. Hogg, 1913.. His student, the British type designer and sculptor Eric Gill (1882 – 1940), acknowledged the difficulty of nailing down a comprehensive definition. He was even more vague: “Legibility is what the Daily Mail reader finds readable; good style is what he finds good; the beautiful is what pleases him,” and finally added, “Legibility, in practice, amounts simply to what one is accustomed to” 3Gill, Eric. An essay on typography. David R. Godine Publisher, 1988. p. 103. British typographer and historian Stanley Morison (1889 – 1967) also addressed the concept in similar terms. Morison never wrote anything specifically about legibility, but he pondered the idea of habit: “Type design moves at the pace of the most conservative reader. The good type-designer therefore realizes that, for a new font to be successful, it has to be so good that only very few recognize its novelty” 4Morison, Stanley. First principles of typography. CUP Archive, 1936. p.7.

​If you recall, in the previous post I mentioned Matthew Luckiesh. He worked as a consultant for Linotype, and maybe thanks to those ties to the type world, his idea that legibility is a multidimensional issue —a concept he was one of the first to champion— began to gain traction among respected professionals. In The Crystal Goblet5Warde, Beatrice. The crystal goblet. Sylvan Press, 1955., the ever-insightful Beatrice Warde distinguished legibility as the concern of the optician, while readability was about the pleasure of a printed work. British typographer and writer Walter Tracy (1914 – 1995) built on Warde’s ideas in his well-known book Letters of Credit. Tracy broke down the concepts of legibility and readability in a dedicated section. Legibility, he argued, “refers to perception, and the measure of it is the speed at which a character can be recognised,” while readability, in his view, connects more with “comprehension” and “the quality of visual comfort”6Tracy, Walter. Letters of credit: a view of type design. David R. Godine, 1986. p.31. In short, legibility is about recognizing individual characters, while readability connects with comfort, visual appeal, and comprehension in reading paragraphs. This perspective dominated the type design world for a long time, and it hasn’t evolved much since. In the brilliant How to Create Typefaces: From Sketch to Screen 7Henestrosa, C., Meseguer, L., & Scaglione, J. (2012). How to Create Typefaces: From Sketch to Screen. Tipo e Editorial., Hernestrosa, Meseguer, and Scaglione present almost the same ideas as Tracy did, 26 years later. 

By the late 20th century, the legibility debate came roaring back, but without much scientific rigor. Instead, it was driven by design’s usual mix of intuition and experimentation. Desktop typography gave graphic designers the tools to play with type easily, and the line between graphic design and type design started to blur. As layouts and compositions began to unravel on the page, new ideas emerged about typography’s macro and micro aesthetics. Independent design publications like Emigre 8More relevant for this article are issues 15 & 18, both from 1991. But if you have the chance, simply read them all. were at the center of this discussion. Inspired by Zuzana Licko’s claim “we read best what we read most”, there was a wave of experimentation where legibility, understood as familiarity, became tied to archetypal models, regardless of their actual efficiency. Designers started pushing back against these archetypes, fueled by the belief that reading is a flexible, dynamic process. Licko, whose designs were considered wildly unconventional at the time, hoped that her typefaces would eventually be as legible as Times New Roman.

Maybe the most well known type designer who worked hard to bridge the gap between design and psychophysics by the end of the 20th century was Gerard Unger (1942 – 2018). Aside from being a dedicated and influential professor who mentored the last three generations of typographers, Unger was also an active advocate for bringing scientific research into the world of type design. He didn’t conduct peer-reviewed experiments himself, but he did write two books that attempt to connect type design with areas like reading psychology and legibility science 9Unger, Gerard. While you’re reading. Mark Batty New York, 2007. Unger, Gerard. Theory of type design. Nai 010, 2018.. In his book The Theory of Type Design, Unger offers a definition similar to that of researcher Keith Rayner: “Legibility can be defined as the ease with which visual symbols can be decoded” 10Unger, G. (2018). Theory of type design. Nai 010. (p.169). To Unger, decoding offers a clear distinction from reading. Just a few lines earlier, Unger defines reading as “the conversion of written or typeset text into linguistic meaning. This process has a sensory and neuronal part—the decoding of visual signs—and a cognitive part—the acquisition of meaning”. Ultimately, he reinforces the distinction that Tracy laid out, using language that echoes Warde: “Readability is the domain of authors and editors, legibility that of type designers, typographers, graphic and website designers”. I don’t share his view, but I’m sure Gerard would’ve loved to discuss it over coffee for hours, and honestly, he probably would have convinced me in the end.

When we compare Unger’s work with that of major psychophysics researchers like Legge or Rainer, the gap between design and legibility research over the past 30 years becomes clear. Unger spent his career perfecting visual symbols to make them easier to decode. He worked with different reproduction environments (especially newspapers) and adapted his letterforms to changing conditions, always envisioning a reader with normal vision. Researchers use scientific tools like the MNRead tests*, and they measure quantifiable variables like reading speed and critical print size in controlled environments. Unfortunately, their work rarely aligns with type design optimizations that designers like Unger championed; and typically, the typographic palette used by legibility researchers in their experiments remains limited. Unger mentions Legge’s work in The Theory of Type Design, published twelve years after Legge’s excellent Psychophysics of Reading in Normal and Low Vision, but keeps his references and scientific data to a minimum. Why? In my opinion, he didn’t want to overwhelm his audience–designers–, who may not be so familiar with scientific publications and long reading. Actually, did any of them even get this far in the article?

But then, finally…​
In the first two decades of the 21st century, a new generation of design researchers began to bridge the gap between design and cognitive psychology with practical, peer-reviewed experiments. Remember when I mentioned that some researchers’ work could actually help type designers? Here are some of the best: Charles Bigelow, Sofie Beier, Mary C. Dyson, Ann Bessemans, or Nadine Chahine, each with their own approach. Their work isn’t about wrestling with psychophysics for the sake of it—it’s about making sense of messy questions and giving us frameworks we can use. They’ve moved past the hunt for one single, universal definition of legibility (though of course, each of them has their own, which quietly steers their work). But what’s more important is they get type design, and they understand what designers need and care about.

The reason these studies matter is that they isolate specific typographic variables and measure their impact. Take the classic serif vs. sans-serif debate: in practice, that’s almost never the only difference—contrast, width, and spacing often vary too. But if you can build your own typefaces, you can isolate the ‘serif’ variable. That’s exactly what Charles Bigelow did when he designed two virtually identical versions of Lucida—one with serifs, one without. He found that serifs slowed reading in RSVP† tests by about 20% at small sizes, but made no difference at larger ones11Morris, R. A., Aquilante, K., Yager, D., & Bigelow, C. (2002). In , 33. P-13: Serifs slow RSVP reading at very small sizes, but don’t matter at larger sizes (pp. 244–247)..

Another great example is Nadine Chahine’s doctoral research. She designed the typeface Afandem in three levels of complexity, with and without vocalization marks. Using an eye-tracking setup to study reading behavior, she discovered that simpler styles led to faster fixations, while more complex styles required longer fixations but reduced the number of regressions. In other words, her work shows just how delicate and balanced the reading process really is12Chahine, N. (2012). Reading Arabic: legibility studies for the Arabic script [Doctoral dissertation, University of Leiden].

It took us about a century, but finally, design and typography research is generating insights that are both relevant to designers and rooted in the rigor we’ve long needed. Desktop typography now allows researchers to craft custom fonts quickly, controlling variables like proportion, weight, contrast, or x-height. Both scientists and designers have stopped chasing the holy grail of ‘universal legibility’. Instead, we’re now asking sharper questions and getting sharper answers. Of course, not every study is fully validated or universally applicable… but that’s true of any field! 

I think the next big challenge for typographic research should be integrating two rapidly advancing tools: variable fonts and artificial intelligence. Variable fonts offer unparalleled adaptability to the plastic nature of the human visual system, allowing for typographic profiles that suit individual performance and preferences. Meanwhile, AI’s data-processing power is opening doors to efficiently navigate the massive space of typographic variations and control complex design spaces with agility.

In the next post, I will discuss some of the studies that helped shape Tiphares, our first typeface. 

​

Want to share your thoughts on this post? Tag me on X (@octaviopv) using the hashtag #WhatisLegibility(P2), comment on the post entry in our Instagram account, or send me an email at hello@pathfinders.rocks with the post title in the subject line.​