Here is a structural problem that every storyteller eventually runs into. A protagonist alone on a stage cannot explain anything. They can think, but thought is invisible; they can talk to themselves, but the audience will only tolerate so much of that. The plot, meanwhile, requires that certain things be known: what the plan is, why it is risky, what happened offstage, and what the hero is actually feeling underneath the competence. The cheapest available solution has been the same for about two and a half thousand years. Give them somebody to talk to.
The oldest job in the building
Greek tragedy had a whole apparatus for this. The chorus commented, the messenger reported violence that could not be staged, and the nurse or attendant stood beside the principal absorbing confidences that the audience needed to overhear. Aristotle, writing about how tragedy is put together in the Poetics, subordinates character to plot in a way that still reads as bracingly unsentimental: the arrangement of the incidents is the soul of the thing, and persons exist to make the incidents happen. Read that way, the attendant with three lines is not a minor character. They are load-bearing. The information has to arrive somehow, and a person is a more graceful delivery mechanism than a placard.
French neoclassical tragedy industrialised the arrangement. By the seventeenth century the confidant was practically a staffing requirement, with major figures each issued a listener whose function was to be told things at the start of an act. The convention drew mockery almost immediately, and the mockery was deserved when the listener had no interior life of their own, existing only to say the equivalent of go on. But the criticism identifies exactly the right fault line. The sidekick is born as a device. Everything interesting that has happened to the role since has been a matter of what you build on top of a device that the story cannot do without.
The squire and the fool
Two medieval and early modern figures did the building. The first is the squire, whose comic apotheosis is Sancho Panza, attached to a master whose view of reality he does not share. The pairing is a machine for generating meaning at no narrative cost. One character sees giants, the other sees windmills, and the reader is handed both readings simultaneously without anybody having to explain the joke. The squire is not merely an audience for the knight; he is a second measuring instrument pointed at the same events, and the discrepancy between the two readings is the book.
The second figure is the licensed fool, and the fool solves a problem the confidant cannot. Somebody has to be able to tell the powerful character that they are wrong. A courtier cannot, an enemy will not be believed, and a friend saying it plainly ends the friendship and often the scene. The fool can, because the fool's speech is discounted in advance as nonsense, which is precisely what makes it sayable. The Folger Shakespeare Library's critical essay on King Lear takes the play's Fool seriously as a moral instrument rather than an ornament, and the same library's survey of fools across the plays shows how consistently the position is used: the character with the least status has the most licence, and uses it to say the thing everyone else is managing not to say.
That is a substantial promotion from the neoclassical listener. The fool is no longer a receptacle for exposition. The fool is the play's conscience, hidden in the one role nobody has to take seriously.
Four jobs, one chair
Modern criticism has tried to sort out what the position actually does, and the tidiest breakdown identifies several distinct jobs that tend to get bundled together. A study of the role in twentieth and twenty-first century popular fiction, opening a book-length treatment of the subject, separates the sidekick as narrative gateway, as devil's advocate, as comic relief and as foil. They are worth pulling apart, because a companion doing all four at once is usually doing none of them well.
The gateway is the position most people think of. A companion who does not understand the world is a reader-shaped hole in the story, and every question they ask is a question the audience wanted asked. This is why detective fiction so often narrates from beside the detective rather than inside them: the method must remain opaque until the reveal, so the perspective has to belong to someone who cannot see it either. The device is old and remains unbeaten. Put the camera behind the wrong eyes and you get suspense for free.
The devil's advocate is different work entirely. Someone has to say that the plan is mad, so that the protagonist's decision to proceed becomes a choice rather than a reflex. Without that voice, recklessness looks like the character simply not having thought about it. The foil is different again, and quieter: a companion whose contrasting temperament makes the lead legible. You cannot see that someone is unusually driven until you stand them next to somebody normal. Comic relief, the fourth job, is the one that ate the others for most of the twentieth century, and it is the reason the word sidekick still carries a faint smell of condescension.
The promotion
Something shifted over the past few decades, and it was probably economic before it was artistic. Long-form serialised storytelling changed the arithmetic. A film has two hours and can afford exactly one interior life; a series running to fifty or eighty hours cannot sustain that, because the protagonist's inner weather has been fully surveyed by the end of the second season and the show still has to fill the fourth. The obvious source of new material is the people standing next to them.
So the ensemble absorbed the sidekick. The former comic companion acquired a family, a competing goal, a moral position at odds with the lead's, and eventually episodes of their own. The structural signature of the change is the rotating point of view: whole instalments handed to a character who, under the older arrangement, would have been in the background asking clarifying questions. Scholarship on character has been catching up with the same shift, and the Cambridge Companion to Narrative Theory's treatment of character reflects a general move away from ranking figures by prominence and towards mapping the relations between them, which is a better description of how contemporary casts actually work.
Why we like them better
The uncomfortable part, for anyone who has ever tried to write a hero, is that audiences frequently prefer the companion. There are structural reasons, and they are not really about charm.
A protagonist carries the story's moral load. They must want the right things, learn the right lesson, and arrive somewhere by the end, which places heavy constraints on how strange, petty, inconsistent or funny they are allowed to be. The companion has no such duty. They can be wrong without the story collapsing, they can have a hobby that never becomes relevant, and they can hold an opinion the narrative never endorses or punishes. That freedom reads on the page as specificity, and specificity is what makes a character feel like a person rather than an argument.
There is also a matter of position. The companion stands roughly where the audience stands: adjacent to the extraordinary thing rather than being it, impressed, sceptical, occasionally exasperated, and required to keep functioning around someone whose talents they cannot match. Very few people identify with being the chosen one. Almost everyone recognises the experience of standing next to someone difficult and remarkable and deciding, again, to stay. That is the sidekick's real subject, and it has always been the more interesting story of the two.
