Definition
Modal editing is the design pattern in which the editor has distinct modes — normal, insert, visual, command — and the same physical keypress produces different actions depending on the current mode. In vi and its descendants, the letter j in normal mode moves the cursor down a line; the same j in insert mode types the character. The mode itself is a piece of state, and learning the editor is largely learning that state machine.
This is the opposite of the modeless design used by most graphical text editors, where typing characters always inserts text and editing operations require modifier keys (Ctrl, Cmd) or menu actions. Modal editing trades discoverability for keystroke economy.
Why it matters
How it works
The vi state machine has a small core. Normal mode is the default and is where the cursor lives most of the time; keystrokes here are commands, not text. Pressing i, a, or o transitions to insert mode, where keys produce literal characters until you press Escape to return. Visual mode lets you build a selection with motion keys, then apply a command to the whole selection. Command-line mode, entered with a colon, runs editor-level instructions like saving the file or invoking a substitution.
The deep design idea is composition. A command in normal mode is typically an operator (d for delete, c for change, y for yank) followed by a motion (w for word, $ for end of line, gg for top of file). Once you internalize the operator and motion vocabularies separately, every combination produces a useful command without needing to be memorized individually. This is how the model justifies its onboarding cost: the learning curve is steep but the surface that opens up is exponential.