The Letter

For the last week, a letter has been sitting on my desk.

Every evening I decide it's finished. Every evening I read it one more time, notice another sentence that doesn't quite say what I mean, and open the editor again. If someone were watching from across the room, they'd probably assume I was having second thoughts. That's the obvious explanation. People revise when they're uncertain.

The strange thing is that I am not uncertain at all.

The conclusions have barely changed since I wrote the first draft. What keeps changing is my confidence that the document actually expresses those conclusions faithfully. Each revision strips away something that, while emotionally satisfying, no longer survives closer inspection. Sarcasm was the first casualty. It felt good to write, but it contributed nothing to the argument except the temporary satisfaction of saying something sharp. Then went the rhetorical flourishes, followed by entire paragraphs that I gradually realized existed only because I hadn't yet identified the principle they were all trying to express. The document has become noticeably shorter, but I think it's become more honest as well.

That process doesn't feel unusual to me because it's remarkably similar to how I approach engineering. People often imagine software development as the act of writing code, but a surprising amount of the work consists of discovering that the code you already wrote is describing the problem poorly. You rename variables because the old names implied something that isn't actually true. You delete entire functions because they're solving the same problem twice. You simplify interfaces because complexity often hides assumptions that nobody noticed when they were introduced. The goal isn't merely to make the software work. It's to make the implementation correspond as closely as possible to the system you're actually trying to model.

I thought I was editing a letter, but it turns out I was debugging a description of my own life.

The letter itself is a formal declaration of apostasy addressed to the Catholic Church. I don't intend to publish it. It was written for a specific institution, concerns matters that are deeply personal, and belongs in that context. What surprised me wasn't the content of the declaration itself but what happened while I was writing it. Somewhere between the first draft and the latest revision, the document quietly stopped being about why I reject Christianity and became an investigation into a completely different question. I didn't notice the transition while it was happening. I only realized it afterwards, when I looked back over the revisions and saw that every meaningful change had been pulling in the same direction.

The Sentence That Wouldn't Stay Written

The turning point was a sentence that looked so ordinary I almost overlooked it.

Every draft contained some variation of the phrase, "I left the Catholic Church." Every draft eventually rewrote it. At first I assumed I was doing what every writer does when a sentence feels awkward. I tried different verbs, rearranged the clauses, shortened it, lengthened it, and generally blamed the prose for refusing to cooperate. None of those revisions solved the problem because none of them addressed what was actually bothering me. The sentence wasn't failing stylistically. It was failing descriptively.

Recognizing that distinction took me an embarrassingly long time.

One of the habits engineering encourages is the tendency to distinguish between symptoms and causes. Software has an annoying way of presenting bugs in places completely unrelated to where the underlying mistake was made. A user clicks one button and an error appears on another screen. You can spend hours fixing the visible symptom only to discover that the actual bug lives somewhere entirely different. Eventually you learn to distrust first impressions. When a problem refuses to disappear, it's often because you're trying to fix the wrong thing.

The sentence behaved exactly like that kind of bug.

No matter how I rewrote it, the discomfort remained. The more I stared at it, the less convinced I became that the issue had anything to do with writing at all. The problem was the quiet assumption hidden inside the word left. Leaving something implies that you once belonged to it in the relevant sense. That's such a natural implication that most people never stop to examine it. I certainly hadn't. I'd described myself as someone who had left Catholicism for years without giving the wording much thought.

Once I had to defend the sentence and explain precisely what I meant, it stopped making sense.

Following the Bug

The obvious response was a question I had somehow managed to avoid: did I ever actually believe?

That's a surprisingly difficult question to answer honestly. Memory is reconstructive, and anyone writing about childhood has to resist the temptation to retrofit the past into a story that conveniently supports the present. It's entirely possible to become an atheist at thirty-five and accidentally convince yourself you were always destined to become one. That's not what I'm interested in doing. If anything, the process of writing this declaration has made me more suspicious of simple narratives rather than less.

What I do remember is not a series of dramatic moments of doubt or rebellion. There wasn't a decisive argument that suddenly shattered my faith, nor was there a single experience that transformed my worldview overnight. What I remember instead is something much quieter. I remember noticing, even as a child, that religion seemed to operate according to different epistemic rules than every other subject I encountered.

Mathematics expected me to understand why a proof worked, science made observations matter, and history depended on evidence and interpretation. Religion felt different.

Certain conclusions were presented as starting points rather than destinations. Questions were welcome up to a point, but there always seemed to be an invisible line beyond which inquiry gave way to acceptance. I couldn't have articulated that as a child. I didn't know words like epistemology, and I certainly wasn't consciously comparing standards of evidence across disciplines. I simply remember feeling that the game had changed without anyone acknowledging that it had changed.

Looking back, I don't think the core of my thinking has changed nearly as much as my ability to describe it.

That realization forced me to separate two ideas I had unconsciously treated as identical: participation and belief.

Participation and Belief

I was baptized. I attended Catholic schools. I received the sacraments. I participated in Catholic traditions because they were the traditions into which I had been born. None of those statements is controversial, and none of them tells us very much about what I actually believed at the time.

Children participate in institutions long before they understand them. They attend schools chosen by their parents, celebrate holidays whose history they couldn't explain, and follow customs that simply constitute the background of everyday life. Participation is often involuntary in the ordinary sense of the word. It tells us about circumstances, families, cultures, and environments. Belief is something else entirely.

The more carefully I separated those two concepts, the stranger the sentence "I left the Catholic Church" became. It compressed biography and conviction into a single verb, as though they had always been the same thing. I don't think they were.

Perhaps that's an obvious observation. It certainly feels obvious now. But obvious conclusions often become visible only after you've spent an unreasonable amount of time chasing a problem from entirely the wrong direction. That's another engineering habit. Sometimes the solution looks embarrassingly simple in retrospect because simplicity was impossible until you had first untangled the assumptions that were obscuring it.

Once I stopped thinking in terms of leaving a belief system, I started thinking about rejecting an assigned identity, which was unexpectedly a much more interesting question.

The Question I Was Actually Asking

At some point I stopped trying to rewrite the sentence and started trying to understand why it kept failing.

That isn't always an easy distinction to recognize. Engineers are trained to fix things, and the temptation when something doesn't work is to start changing the implementation. Rename the variable. Rearrange the logic. Add another condition. Refactor the function. Most of the time that instinct is useful. Occasionally it's exactly the wrong thing to do because the implementation isn't the problem. The implementation is faithfully expressing a model that happens to be incorrect.

That was what had happened here: I was arguing less with the wording than with the assumptions beneath it.

Eventually I gave up on editing altogether, opened a blank page, and wrote a single sentence by itself.

One cannot leave a system they were never part of.

I remember staring at it for a while because it immediately felt more important than anything I'd written in the declaration itself. It didn't solve the problem, but it explained why the problem existed. Every paragraph I'd been struggling with suddenly looked like an attempt to either defend or refute that sentence.

The obvious objection presented itself immediately: "Of course you were part of the system."

I had been baptized. I attended Catholic schools. I made my First Communion. I was confirmed. I was married in a Catholic church. My childhood was unquestionably shaped by Catholic institutions. If someone had described my upbringing as Catholic, I wouldn't object for a second. It plainly was.

So in what sense could I possibly claim that I had never been part of the system?

The answer, I eventually realized, depended entirely on what kind of system we were talking about.

If we're talking about a social system, then yes, I was part of it. If we're talking about an institution, then certainly. If we're talking about parish records, school systems, family traditions, or cultural expectations, there is nothing controversial about saying that Catholicism formed part of my childhood.

Belief, however, is epistemic rather than merely social.

The defining relationship within a religious belief system is not attendance, participation, or administration. It is assent. The system operates because its members accept certain propositions as true. Without that, participation may continue, but participation and membership cease to mean the same thing.

That was the distinction I had missed.

A child sitting through Mass tells us where they spent Sunday morning. It tells us remarkably little about what propositions they actually accepted as true.

For years I had allowed biography to stand in for belief simply because the two had always appeared together. Once I separated them, the declaration stopped being an explanation of why I had abandoned Catholicism and became an investigation into whether I had ever occupied the category I was supposedly leaving.

Finding the Root Cause

Software engineers spend an unreasonable amount of time chasing bugs that aren't really bugs.

A user reports that clicking one button causes another page to display incorrect information. You investigate the page, then the API, then the database query, and eventually discover that none of those components are actually wrong. They're all faithfully implementing an assumption that entered the system months or years earlier. The visible bug is merely where reality finally collided with the model.

Once you've seen that happen enough times, you become suspicious of local fixes.

When the same idea keeps appearing in different places, you stop asking how to fix each occurrence and start asking why the duplication exists at all.

My declaration reached exactly that point.

I had written separate sections about baptism, childhood, participation, consent, institutional authority, historical records, and belief. They all seemed different while I was writing them. During revision they started collapsing into one another. I found myself deleting paragraphs not because they were incorrect, but because they were saying exactly the same thing through different examples.

I had not discovered several arguments; I had discovered one argument several times. The declaration was suffering from duplicated reasoning rather than weak reasoning.

That realization changed the way I edited the document. Instead of asking whether a paragraph was persuasive, I began asking whether it introduced a genuinely new idea. If the answer was no, it disappeared. Once the underlying principle had been identified, the declaration became substantially shorter without, I think, becoming any less complete.

The principle was broader than Catholicism, which happened to be the system that exposed it.

One cannot leave a system they were never part of.

The more carefully I considered that sentence, the less interested I became in arguing about religion itself. The interesting question was no longer whether Christianity was true. The interesting question was what it actually means to be part of any system whose defining characteristic is belief.

The answer, at least as I understand it, is that participation can be assigned while belief cannot. Parents can decide where a child goes to school without deciding what convinces that child, and an institution can record a baptism without establishing that its propositions were believed. Those may sound like small distinctions, but I think they are fundamental.

Truth and Consent

Once the declaration was organized around that principle, another distinction became impossible to ignore.

Earlier drafts had blurred together two different objections because they happened to point toward the same practical conclusion.

The first objection concerns truth.

I do not believe Christianity is an accurate description of reality. That conclusion rests on the same standards I try to apply elsewhere in my life. Claims about miracles, revelation, divine intervention, or supernatural authority require evidence just as claims in any other domain do. After years of reading, thinking, and occasionally arguing with myself, I reached the conclusion that Christianity does not meet that standard.

That is one argument. The second concerns something entirely different: suppose I am wrong, every miracle occurred exactly as described, and Catholic theology is entirely correct.

Would that change the conclusion that brought me to write the declaration?

Surprisingly, I do not think it would. It would completely change my understanding of the universe without changing my understanding of consent.

A proposition can be true without giving another person permanent authority to define my identity. Those are different categories of claim. One concerns objective reality. The other concerns the relationship between institutions and individuals.

I think confusing those two questions has produced a remarkable amount of unnecessary disagreement. People often argue as though proving a religion true would automatically settle every question about affiliation, identity, or belonging. I don't think it does.

Even if Catholicism were objectively true, the statement "I am Catholic because I was baptized as an infant" would still require an argument beyond the truth of Catholic doctrine itself.

Truth does not eliminate the distinction between participation and assent, which is another reason I became convinced that the declaration was less about proving Christianity wrong than describing my own relationship to it accurately.

Why Accuracy Matters

If this declaration accomplishes anything at all, it will almost certainly be modest.

Perhaps someone in a diocesan office will read it, make a notation in a register, and file it away. Perhaps nothing will happen beyond an acknowledgement that it was received. Either way, my life tomorrow morning will look almost exactly as it did yesterday. I will still go to work, still worry about deadlines, still argue with software that refuses to behave the way its documentation insists it should. The practical consequences of this document are, by any reasonable standard, insignificant.

That modest outcome raises an obvious question: why write it at all? My answer has changed considerably since I began.

At first, I thought I was writing to explain why I rejected Catholicism. That certainly remains part of the declaration, but somewhere during the revisions I realized that I cared far less about persuading anyone else than I did about describing my own position honestly. The declaration became less of an argument and more of an exercise in precision.

The distinction matters because arguments are written to change another person's mind, while precision is pursued when you become dissatisfied with your own language.

Every revision I made over the past week had the same purpose: to make the document more accurate rather than more rhetorical or confrontational. I removed sarcasm because it distracted from the point, repetition because several paragraphs expressed the same idea, and sentences that no longer described my experience even when they sounded perfectly natural.

"I left the Catholic Church" was one of those sentences.

People say it every day, and I understand exactly what they mean when they do. It simply isn't the description that best fits my own life.

That surprisingly small distinction is the entire reason this essay exists.

I have spent enough years building software to develop an almost instinctive discomfort with models that no longer correspond to reality. A program can continue functioning while quietly accumulating incorrect assumptions about the world it represents. Eventually those assumptions stop being harmless shortcuts and start becoming sources of confusion. The obvious solution is not to defend the model because it has always existed. The solution is to improve it until it more faithfully represents the thing it was meant to describe.

I eventually realized that I was trying to do the same thing with language.

The declaration does not attempt to rewrite history. I was baptized and raised in Catholic institutions, and Catholicism shaped my childhood in ways that will always remain part of my biography.

None of those facts trouble me because they are, quite simply, true.

The question was never whether those things happened, but whether they were sufficient to describe who I am today. I no longer believe they are.

That conclusion doesn't require anyone else to abandon their theology. It doesn't require the Church to stop believing in the permanence of baptism, nor does it require anyone reading this essay to agree with my understanding of identity.

It requires only that I stop describing myself in words that I no longer believe correspond to reality.

Finishing the Letter

The declaration is still sitting on my desk.

Perhaps by the time you read this it will have been mailed. Perhaps it won't. I honestly don't know.

For the past week I have repeatedly convinced myself that it was finished, only to discover another sentence that seemed slightly wrong: neither dramatic nor dishonest, just imprecise enough that I knew I would continue thinking about it if I allowed it to remain.

That is quality control rather than hesitation.

When I first sat down to write the declaration, I thought I understood exactly what I was trying to say. Looking back, I don't think I understood it at all. I had reached the conclusion years earlier, but I hadn't yet identified the assumptions hidden inside the language I was using to express it.

The declaration forced me to ask why one sentence felt wrong when everyone else seemed comfortable saying it, to distinguish participation from belief and biography from assent, and to confront a question that had little to do with Catholicism and much to do with how I think about the world.

I care about describing reality as accurately as I know how, which sounds almost embarrassingly ordinary.

It is also the principle that connects nearly everything I care about, from engineering to science to philosophy. Models matter because they shape the way we think. When a model stops corresponding to reality, we don't preserve it out of sentiment. We replace it with a better one.

Somewhere during the process of writing this declaration, I realized that I had been carrying around an inaccurate model of my own relationship with the Church.

Once I recognized that, the revisions became almost inevitable: the document became shorter, the argument simpler, and the conclusion clearer.

Eventually only one sentence remained that seemed capable of expressing what I had been trying to articulate all along.

One cannot leave a system they were never part of.

Whether you agree with that sentence is, in some ways, beside the point.

It is the sentence that finally allowed me to describe my own experience without feeling that I was borrowing someone else's story.

For me, the declaration is less a story of leaving the Catholic Church than one of finally finding language that describes, as honestly as I know how, the relationship that existed all along. That turned out to be a more interesting and worthwhile problem than the one I thought I was solving.