Leaving is usually described as an intellectual event and experienced as a logistical one. The arguments are what people talk about afterwards; what actually changes on the Monday is who you eat with, what the week is shaped around, and in a significant minority of cases where the money comes from.

Those are three separate problems and they have different clocks. Treating them as one problem — which is what 'deconstruction' as a word encourages — is why people spend two years arguing themselves in circles while the practical situation quietly deteriorates.

First: the thing nobody plans for is the network

Robert Putnam and David Campbell's American Grace (2010) found that religious participation is among the strongest predictors of civic engagement and social support in American life — and crucially that the effect runs through congregational friendships rather than through belief. People with more friends inside a congregation showed the effect; people who believed intensely but knew nobody did not.

Read that in reverse and it is a warning. What you lose on leaving is not primarily a doctrine. It is a standing weekly appointment with several dozen people who knew your name, noticed your absence, and would have brought food if something happened. Most adults have no other institution that does this, and the replacement is not automatic.

Peter Berger (1967) supplies the mechanism and it cuts both ways: a plausibility structure is the social base that keeps a worldview credible, and beliefs are sustained by being spoken aloud among people who share them. Removing yourself from the structure is why the belief stops feeling obvious. It is also why the next thing you believe will be whatever the next group of people you see every week believes — so choosing that group deliberately is a more consequential decision than choosing a position.

The practical version: before you resolve anything doctrinal, put two recurring commitments in the calendar that involve the same people repeatedly. Not events. Recurring, with named individuals who will notice. This is the single highest-return thing to do in the first six months and almost nobody does it, because it feels unrelated to the question they think they are asking.

Second: the money, if you were paid for it

If the institution was also the employer, you have a career transition and an identity transition arriving simultaneously, and they will contaminate each other. Keep them apart on paper even though they are tangled in practice.

Start with an inventory, because the standard move — assuming the experience does not count — is factually wrong. A working week in ministry typically contains: public speaking to a non-captive audience on a fixed schedule; one-to-one conversations where the job is to find the real problem behind the stated one; coordinating volunteers you cannot fire or pay; running a budget; and fundraising from people under no obligation to give. Those are, respectively, the core skill of training and enablement, of discovery in sales and research, of programme and project management, of operations, and of development and revenue.

That is career capital, and it is unusually transferable. What does not transfer is the vocabulary and the credentialling, which is a presentation problem rather than a capability problem. The sentence to practise is the one that states the function without the setting: not 'I pastored a church of 300' but 'I ran weekly programming for 300 people with a volunteer staff of 40 and a budget I raised myself.'

One caution against the obvious move. The fastest-looking route is adjacent employment — a nonprofit, a denominational body, a Christian school — because the vocabulary carries over and the hiring manager recognises you. It is often the right bridge and it is frequently a way of keeping the old plausibility structure while believing you have left it, which postpones the thing you left to do.

Third: the belief question, which can wait longer than it feels like it can

Heinz Streib and colleagues (2009), in the only large cross-cultural study of deconversion, found it is typically gradual, relational and multi-dimensional rather than a single argument decisively lost. Their cases move through loss of specific beliefs, loss of trust in the community, emotional disengagement and moral criticism — usually not in step, and over years.

The useful implication is that the state you are in is not a problem to be closed this month. It is the normal shape of the process, and the pressure to arrive somewhere definitively is mostly social — it comes from people on both sides who want to know which team you are on.

The trap Streib's data makes visible is oppositional identity: a period in which the new position is defined entirely by what it is against. It is extremely common, it feels like clarity, and it is still the old framework running with the signs flipped. The test is whether you can state what you now think without mentioning what you left.

What this does not mean

It does not mean the arguments do not matter. It means the arguments are not urgent, and treating them as urgent while the network and the income go unattended is the common failure, not a hypothetical one.

It does not mean leaving is the right answer. Nothing here argues for a conclusion; Pew's Religious Landscape Study (2023–24) found the long rise in the religiously unaffiliated has levelled off at around 29 per cent of US adults after two decades of growth, which is a fact about a population and says nothing about what any individual should do. The failure of a plausibility structure is not evidence either way — it is the reason the question became live, which is a different thing.

And it does not mean the sunk cost counts. Years of training, a degree, a career built inside one institution — none of that is recoverable under any option now available, so it is not a reason to stay and not a reason to leave. It feels like the heaviest input in the decision and it is the one input that is formally irrelevant.

The first ninety days

Two recurring commitments with the same named people, in the calendar, before anything else. An inventory of the week you used to run, written as functions rather than as titles. One conversation with somebody who made the same move three or more years ago, which is long enough that they are past the oppositional phase and can tell you something useful. And a decision to stop relitigating the belief question on a weekly basis — not because it is settled, but because it does not pay rent.