The scene at Pilate’s tribunal depends on a custom the Gospels present as familiar to everyone in the crowd. At the feast, Pilate is accustomed to release for the people one prisoner whom they request. The wording varies slightly across the four accounts, but the premise stays constant: a Roman governor, on this particular holiday, releases a prisoner of the crowd’s choosing, as a matter of established practice.

No such practice appears anywhere outside the Gospels.

Josephus, who wrote at length about Roman administration in Judea and about the volatile politics surrounding Passover in particular, describing riots, crowd violence, and the heavy military presence Rome kept in Jerusalem during the festival, never mentions a governor’s amnesty. Philo, who catalogued Pilate’s provocations against Jewish sensibilities in detail hostile to the point of documenting them for effect, says nothing about it either. No Roman legal source describes a standing practice of releasing a prisoner at a subject people’s request during a festival, and the practice would be, if real, a strange one for Rome to have adopted. Releasing a convicted criminal at the demand of a crowd is not how Roman provincial administration is otherwise known to have worked. It runs against the whole rationale of an occupying power maintaining order through the demonstrated consequences of resistance, particularly at a festival Rome already treated as the year’s highest security risk.

Scholars have proposed a handful of explanations for a custom this poorly attested. Some have looked for a parallel in later rabbinic material about releasing prisoners at festivals, without finding anything that lines up closely with the same institution. Others have proposed that the practice existed briefly, informally, under Pilate alone, and simply escaped notice in every surviving source that would otherwise have recorded it. I find that explanation strained past the point of usefulness. An unattested custom invoked to explain an otherwise unattested custom does not amount to an explanation. It amounts to two gaps standing in for one.

The simpler reading, and the one increasingly favored in critical scholarship, is that the Passover amnesty is a literary invention. The Gospel authors needed the crowd to choose between two men, and a customary release gave them the mechanism. Once the custom’s absence from every independent source is taken seriously, this stops being simply the boring solution left over after harmonization fails. It becomes a fact demanding its own question: why invent this particular mechanism, structured this particular way, to force this particular choice?

This is where I want to argue for something beyond what the absence of the custom, on its own, establishes. The absence tells us the scene did not happen as described. It does not by itself tell us why the author built it the way he did. For that, the content of the choice matters as much as its lack of attestation, and the content is this: a crowd is asked to choose between Jesus who is called Christ and Jesus Barabbas, and Barabbas is not a personal name. Bar-Abba is a patronymic, Son of the Father, the same relationship to the divine that Messianic kingship claimed for itself, since a king in the line of David was understood, going back to the royal theology of Psalm 2, as God’s son by adoption and by office. Christos, Messiah, and Bar-Abba, Son of the Father, are not two unrelated designations that happen to land on the same stage. They are two names for the same claim, one drawn from Greek theological vocabulary and one drawn from Aramaic patronymic form, both converging on the identical assertion: this man is the anointed king, the son of God, in the political sense of Davidic kingship and the theological sense of divine sonship at once.

I would read the amnesty scene, then, not as an invented mechanism for choosing between two rivals but as an invented mechanism for staging a convergence the author wanted an attentive reader to notice. The crowd is made to choose, on the story’s surface, between the guilty man and the innocent one. But the two men they are choosing between carry names that mean the same thing, and a Gospel author working in a milieu saturated with wordplay on names, on titles, on the theological weight carried by a patronymic, would have known exactly what he was doing by setting those two names against each other in the same breath, twice, in Matthew’s version of the scene. The choice the crowd makes looks like a moral failure on the surface. Underneath it, the author has built a scene in which the crowd is, without knowing it, being asked to choose between two versions of the identical title: Son of the Father, once in Greek theological dress and once in Aramaic patronymic form.

This is the argument my book is building toward directly. In The Gospel as Mirror: Inversion, Substitution, and the Suppressed History of the Passion, I argue that Jesus Barabbas and Jesus called Christ were a single physical man standing twice before Pilate under two names, and that the amnesty scene is the literary device the Gospel author built to display the convergence of two titles onto a single office, Messiah and King, Son of God and Son of the Father, using two names that mean the same thing in two languages to make a theological point the surface narrative was not built to state directly. The historical absence of the custom is the evidence. That the two names converge on a single office rather than resembling each other by coincidence is an inference the amnesty scene supports directly: an author capable of naming a character Bar-Abba in a scene about the identity of the true king was not doing so by accident. That the convergence extends to one man beneath two names, rather than two men sharing one title, is the further claim this book makes, and it depends on the wider argument developed across its chapters, not on the amnesty scene in isolation.

On this reading, the Gospel tradition preserved, at the exact moment it needed the crowd to condemn one man and free another, a name for the freed man that means precisely what the condemned man’s title meant, because the freed man and the condemned man were the same man carrying two names. That is not the kind of detail a historical custom would have produced by chance. It is the kind of detail an author builds when he is staging the identity of one man beneath the appearance of two.