Italic

An italic is a second alphabet, not a tilted copy of the first. The style comes from a sixteenth century chancery hand, and the letters still carry the pen: the two storey a collapses to one, the e tips its bowl, the f grows a descender it does not have upright, and entry and exit strokes appear where a roman has flat terminals. The slope is a side effect of writing quickly, which is why it is the least reliable way to spot one.

The genuine sibling is the oblique, a roman mechanically slanted with no redrawing. Sans families often ship one and call it italic (Helvetica’s is an oblique), while others ship both and mean different things by each. Worse is what a browser does when a family has neither: it fakes a slant by skewing the roman outline, which thins the strokes on one diagonal and leaves the letters looking sheared rather than written. A variable font can expose the difference as data, with slnt for a pure angle and ital for the switch to the drawn italic.

The jobs are conventional and worth keeping straight. Italic marks emphasis inside running text, where a bold word would out-shout the paragraph around it, and it also names things: book and film titles, ship names, genus and species, words quoted as words, and a term the first time it is defined. In CSS the distinction between the two is the one HTML already draws between em (stress, which changes the meaning of the sentence) and i (a different mode of text, which does not).

Interfaces should use it sparingly. A run of italic longer than a sentence or two is slower to read on screen, small sizes lose the fine entry strokes to antialiasing, and an italic set in a UI face is usually the synthesized shear rather than a drawn face, since the family was built for labels rather than for prose. Where the goal is hierarchy rather than emphasis, a step of font weight does the same work without changing the texture of the line.