A lettering font is an attempt to package a drawn alphabet so that it can be typed. The attempt is never complete, because a drawn alphabet is not a set of independent shapes — it is a set of shapes that a hand modifies according to what precedes and follows them. Most of the technology in a good comic font exists to reintroduce that dependency artificially.
From drawn alphabet to character set
The historical starting point is a set of pencilled guide lines and a nib. A letterer working by hand does not store an alphabet; the alphabet is stored as motor habit, and the shapes come out slightly different every time because the pen angle, the ink load and the available space all vary. Converting that to a font means making three decisions that hand work never has to make explicitly.
The first is which forms to freeze. A hand produces perhaps a dozen plausible versions of a capital S depending on context; a font must pick one as the default. The second is spacing. A hand fits letters by eye into whatever space remains; a font assigns each glyph a fixed advance width plus a kerning table of exceptions. The third is the character set itself — punctuation, accented forms, currency, dashes and the various quotation styles that hand lettering simply drew as needed.
The resulting file is judged on how convincingly it hides those three decisions. A font whose repeated letters line up in visible vertical columns has failed the first, a font whose capitals collide after a T or an L has failed the second, and a font that cannot set a language beyond English has failed the third.
The uppercase-dominant tradition
Comics lettering has been predominantly uppercase since the newspaper strip era, and the reasons are mechanical rather than aesthetic. Capitals share a common cap height and sit on a common baseline, which produces a rectangular block of text that packs efficiently into the irregular gaps an illustration leaves. There are no descenders to collide with the line below, so line spacing can be tightened. And reproduction at small sizes on absorbent newsprint favoured the simpler, more open shapes of capitals over the finer distinctions between lowercase b, d, p and q.
The trade-off is real. Continuous all-capital text is measurably slower to read at length, because word-shape recognition — the ragged outline created by ascenders and descenders — is unavailable. In comics this costs little, since dialogue arrives in short bursts inside balloons rather than in paragraphs; in a long caption it costs a great deal, which is why extended narration is increasingly set in mixed case. The full argument about reading speed is on the legibility page.
A convention specific to this tradition is the crossbar-I rule. In all-capital lettering, a plain vertical I is ambiguous against a lowercase l and a numeral 1. The convention restricts the crossbarred form — the I with serifs top and bottom — to the first-person pronoun and to initials such as a monogram, and uses the plain vertical everywhere else. Using the crossbar form inside a word is one of the most immediately recognisable errors in the field, and every serious comic font ships both glyphs on separate codepoints or behind a stylistic set so the distinction can be typed.
Alternates, ligatures and contextual substitution
The repetition problem is the defining weakness of a typed alphabet standing in for a drawn one. The standard remedy is to supply multiple drawings of each glyph and swap between them automatically. Three OpenType mechanisms do most of the work:
- Stylistic alternates supply parallel versions of a glyph, selected either manually or by feature. A comic font may carry three to six versions of the most frequent capitals.
- Contextual alternates select a version based on neighbours. The commonest rule replaces the second of two identical adjacent letters, so a doubled L or doubled O never appears as a mirrored pair.
- Ligatures replace a sequence with a single drawn shape. Standard ligatures resolve collisions such as an f followed by an i; discretionary ligatures in display faces do the decorative joins described on the logotype page.
A well-built comic font layers these so that ordinary typing produces varied output with no manual intervention. The measurable target is that no two identical glyphs appear within a short window of each other; the perceptual target is that a reader scanning the balloon cannot find a repeated shape without hunting for it.
Metrics, spacing and the invisible half of the file
Outlines get the attention, but metrics decide whether a font is usable. Each glyph carries an advance width and two sidebearings; the relationship between them is what produces even colour on the page. The traditional rule of thumb is that the space inside a letter and the space beside it should look equivalent, so a font with open round counters needs generous sidebearings and a condensed font needs tight ones. Because display faces are set large, kerning matters more than in text: pairs involving a capital A, T, V, W and Y against round or low-left letters can need corrections of several per cent of the em, where in a text face they would need almost none.
Vertical metrics decide line spacing. Comic fonts commonly set a tighter default line height than text faces because there are no descenders in an all-capital setting, and because balloon interiors are expensive space. Values in the region of 105 to 120 per cent of cap height are typical; below that, successive lines start to interfere optically even when they do not collide geometrically.
Encoding and coverage
A font is only as international as its character set. Codepoints, normalisation and the rules for how combining marks attach are defined by the Unicode Consortium, and any font intended for translated editions has to cover at minimum the accented Latin range, correct guillemets and the various dash and quotation conventions used across European languages. Comics are widely translated, and a font that forces a translator to substitute a different family mid-series destroys the visual continuity the lettering was chosen for.
For terminology, Google Fonts' glossary is a serviceable free reference to the standard vocabulary, and foundry publishers such as Monotype document the technical specifications behind variable and OpenType features. Fonts built specifically for comics work, including families with the alternate and contextual machinery described above, are published by specialist houses such as the foundry at Blambot.
Whichever route a project takes, the underlying skeleton still has to be right. The vocabulary for judging that skeleton is set out under letterform anatomy, and the historical categories that most display families descend from are described in the serif field guide.