Briem on spacing, italics, and the bold weight
Three chapters from Gunnlaugur SE Briem’s notes on type design cover the work that comes after the roman is drawn: making a bold weight, designing the italic, and spacing the result. All three are harder than they look and more forgiving than they seem.
What exactly is a bold?¶
“If you haven’t got an opinion about bold letters,” Briem writes, “now is the time to make up your mind.” Where do you add the weight? How thick do your hairlines stay? How do you handle the smallest counters — the inside of the degree sign, the interior of the lowercase ‘e’ — when they’re filling with stroke weight from all sides?
His answer: “It’s probably whatever you say it is, as long as it isn’t actually lighter than the regular weight. It is usually between two-fifths and two-thirds heavier.”
He shows three examples — 33%, 50%, and 67% heavier — and recommends going very bold for the exercise. The reasoning is practical: design a very bold master and a regular, and you can interpolate any weight between them. Extrapolating extra-black from regular and bold, on the other hand, “is a mess.” Start heavy, blend down.
At high weights, curves become nearly circular, counters shrink dramatically, and the character distinctions that give a typeface its personality can disappear. Keeping the letters legible and distinct at extreme weights is the central challenge.
Italic: not just a slanted roman¶
“Designing an italic is much the same as designing a roman. A few extra points are worth bearing in mind.”
Briem’s most important point: the italic must fit the roman. The two sometimes appear in the same word — common in technical and academic writing. If the italic’s letter placement is off, gaps open at the roman-italic boundary.
His example shows the letter ‘u’ in two positions: too far right in its character box (causing a gap after a roman ‘H’), and positioned correctly (the gap comfortable). The fix is a sidebearing adjustment, but it requires checking roman-italic combinations specifically, not just the italic in isolation.
For precision work on slanted letters, he recommends straightening them first. “Measurements are easier that way. Adjust. Recheck coordinates. Then slant the letters again.” Counterintuitive but sound: slanted geometry is harder to reason about than the same letter upright.
Spacing: the old argument and the new one¶
Jan Tschichold published spacing examples in 1952 that he called “legible and attractive” versus “a thicket of letters, frequent mistake.” He intended these as a lasting standard.
Time passed. In 1950, Stempel released Hermann Zapf’s Palatino with spacing Tschichold would have condemned — tight, striking, demanding. Forty years later, Palatino had changed: more conventional in proportions, but its spacing had grown even tighter. “What Tschichold condemned, Hermann Zapf practised. It carried the day.”
Briem’s point is not that Tschichold was wrong or Zapf was right. Spacing conventions are not fixed. They evolve with technology (what works in letterpress may not work on screen), with taste (ideal text density has shifted multiple times in 500 years), and with context (a display face at 72pt lives differently than a text face at 10pt).
What he offers instead of fixed rules is method: set your control characters first, link the dependent glyphs, test in running text, adjust by eye. Accept that the result belongs to its moment, not to eternity.