The Argument from Embarrassment, Inverted

There is a well-worn move in New Testament apologetics called the criterion of embarrassment. It runs: if a text reports something awkward for its own cause, the authors would not have invented it, so it is probably historical. Nobody makes up material that damages their case.

It is deployed constantly. Jesus is baptised by John — awkward, since it implies subordination. The disciples are dim and cowardly. Jesus is executed by the shameful method reserved for slaves and rebels. Women are the first witnesses to the empty tomb, in a culture that discounted female testimony. Jesus cannot perform miracles in his hometown. On the cross he asks why God has forsaken him.

Each of these, the argument goes, is a detail no propagandist would invent, so each is a mark of authenticity.

As a historical tool the criterion has a legitimate narrow use. As it is actually deployed in apologetics, it does not work — and it fails for a reason that generalises far beyond this case.

Every Movement Generates Awkward Material

The criterion assumes that awkward details are rare and therefore diagnostic. They are neither.

Consider what any real movement accumulates. The Millerites predicted the world would end in 1844. It did not, and the failure is preserved in their own literature as the Great Disappointment — deeply embarrassing, faithfully recorded. The movement survived and became the Seventh-day Adventists. Joseph Smith's early followers documented his failed treasure-digging and legal troubles. The Jehovah's Witnesses' own publications record a series of failed end-date predictions. Marxist parties preserved detailed accounts of splits, betrayals and inconvenient theoretical failures.

These are not accidental leaks. Movements record embarrassing things routinely, for perfectly mundane reasons: the events were widely known and could not be quietly dropped; the community already had the story; a rival faction preserved it; or the material was reframed as a test of faith and thereby made useful.

If embarrassing details are what every movement generates, then their presence tells you that you are looking at a real movement with a real history — which nobody disputes. It does not tell you that any particular claim in the vicinity is true.

Awkwardness Is Not Stable

The criterion also assumes we can tell what embarrassed the original authors. This is much harder than it looks, and the judgement is usually made from the wrong century.

Take the crucifixion. To a modern apologist it looks maximally awkward: a shameful execution, the opposite of what you would invent for a hero. But the texts do not treat it as an awkward fact grudgingly admitted. They treat it as the theological centre of the whole system — a deliberate sacrifice, foretold, freely accepted, the mechanism of salvation. A detail that the authors themselves present as the triumphant core of their message is not obviously an embarrassment they were stuck with.

The women at the tomb work the same way. The claim is that female testimony was discounted, so nobody would invent it. But the narrative function of that detail is not weak — it is a reversal, and reversals of expected status are among the most characteristic literary devices in these texts, which are full of first-shall-be-last inversions. A detail that fits the author's dominant literary pattern is poor evidence of unwilling reportage.

And the baptism: Mark reports it plainly, Matthew adds John's protest that he should be baptised by Jesus instead, Luke mentions it in passing, John omits the event. That sequence is exactly what we would expect if a known story was progressively managed — but the criterion needs the earliest version to be unmanaged reporting, and it cannot establish that from the discomfort alone.

What the Criterion Can Actually Show

There is a real principle in here, and it is worth separating from the overreach.

The legitimate version is comparative and modest: within a body of material where an author had clear motives to shape the story, a detail that cuts against those motives is somewhat less likely to have been shaped. Historians use this to reason about the relative reliability of parts of a single source — and used that way, in the narrow form, it is a reasonable tool.

What it cannot do is bear the weight the apologetic use puts on it. Note the pattern of application: the criterion is invoked for the baptism, the crucifixion and the empty tomb, and then quietly not invoked for the material that is not awkward — the miracles, the resurrection appearances, the fulfilled prophecies. But a criterion that promotes only the awkward material leaves the central claims exactly where they were. Even granting every embarrassing detail as historical, you have established that a man was baptised, that his followers were unimpressive, and that he was executed. Those are not the disputed claims.

There is a second, deeper problem. The criterion asks what an author would or would not invent — but the alternative to deliberate invention is not accurate reporting. It is transmission: decades of oral circulation in a community, in which stories are reshaped by memory, retelling, translation and the needs of the audience, with nobody at any point deciding to lie. Most of what is wrong in ancient sources got there this way. The criterion is built to detect fabrication and is blind to the process that actually does the damage.

The Same Move Everywhere Else

Once you see the structure, you find it well outside religious apologetics.

Conspiracy theories thrive on it: why would they admit that if it weren't true? Institutions release limited unflattering information precisely because it purchases credibility for everything else in the document. Advertisers name a small product flaw — "our seats are uncomfortable, but our fares are low" — because a concession makes the rest more believable. Politicians confess a minor, well-chosen failing.

In each case a piece of self-damaging content is doing persuasive work rather than evidential work. And the fact that it works so reliably as a rhetorical device should make us suspicious of treating it as an epistemic one. Awkwardness is not a costly signal, because generating it is cheap and it happens by default.

What Would Actually Help

The instinct behind the criterion is sound: we should look for evidence that is hard to fake. The mistake is thinking awkwardness qualifies.

What would qualify is what historians look for everywhere else — multiple independent sources not derived from one another; contemporary rather than later attestation; physical evidence; testimony from hostile or disinterested parties with no stake in the narrative; documents written for a purpose other than persuasion, like tax records and letters about other things.

That is the real reason the criterion of embarrassment gets such heavy use in this particular field. Where the independent, contemporary, disinterested evidence is thin, arguments must be built from the internal features of the documents themselves — and the criterion is the most sophisticated-looking tool available for that job. Its prominence is a symptom of what is missing, not a substitute for it.

A story containing awkward parts is just a story that has been told by real people for a long time. That is what stories look like. It is not a mark of truth, and treating it as one gets the direction of inference exactly backwards.