Tittle

The tittle is the dot above a lowercase i or j, and that really is the whole definition. The word is a medieval one, from the Latin titulus, and it survives mainly in the phrase “to a T”, which was once “to a tittle”, meaning down to the smallest mark. Scribes added it in the eleventh century for a practical reason: in the dense blackletter hands of the period a bare i vanished into the neighbouring strokes of an m or a u, and a stroke over it told the reader where one letter ended and the next began.

Typographically it is a separate contour inside the same glyph, and its shape is a real decision rather than a default. It can be round, square, rectangular, diamond, or a slanted wedge, and its size and how far it floats above the stem are among the first things that distinguish one sans from another at small sizes. It usually sits near ascender height, and in a good many faces it is the tittle rather than the b and the d that establishes how tall the ascenders can be, because a mark colliding with the line above is worse than a short l.

The tittle is also the reason internationalized text handling has a famous edge case. Turkish and Azerbaijani treat dotted and dotless i as two different letters: ı is a letter in its own right and its capital is I, while the dotted i capitalizes to İ, an I that keeps its tittle. So uppercasing a lowercase i is locale-dependent, and code that assumes otherwise turns Turkish words into misspellings. This has broken real systems, most memorably where a case conversion applied to an identifier changed its meaning on Turkish devices. If you are uppercasing user text, pass the locale.

One last piece of trivia for the pub: the dot is called a tittle only on the i and the j. The dot over a Scandinavian å is a ring, the pair over ä is a diaeresis or umlaut depending on what it does, and the mark that turns n into ñ is a tilde. They are all diacritics; the tittle is the one that is simply part of the letter.