Riff: The Data We Create by Looking

A Riff on Attention, Evidence, and the People Who Disappear

A note before we begin: this isn’t Essay 10. That arrives next Friday, on schedule. But sometimes something surfaces between movements that needs its own riff.

This piece is about a grammar we rarely name — how what we look at becomes “data,” and how entire forms of suffering disappear when no one is watching.

If that sounds abstract, the story that follows is not.

This week the CDC overhauled the childhood vaccine schedule, cutting routine recommendations for six diseases based on what experts say is a misreading of Danish data. The same week, a senior White House official explained the administration’s posture toward Greenland and Venezuela: “We live in a world that is governed by strength, that is governed by force, that is governed by power. These are the iron laws of the world since the beginning of time.”

The iron laws of the world since the beginning of time. That’s one grammar.

My wife is a gynecologic oncologist. Each day she treats women—often poor, often uninsured, often showing up late because they couldn’t access care—dying of cancers a vaccine could have prevented. Some argued with her about vaccination even as COVID took them during treatment—or took the family members who were their lifeline through it. Not because they were stupid, but because wealthy ideologues had spent years teaching them to distrust the very system that might have saved them. The propaganda of the comfortable becomes the death sentence of the vulnerable. And when the vulnerable die outside the system, they don’t appear in the data that would indict the propaganda.

If you’ve been following the series, you’ll recognize the themes—Sage and Saint—ESSAY 9, the shame-pride architecture—ESSAY 6, forbearance—ESSAY 2 , the Pharaoh economics—ESSAY 8 underneath. If you’re new here, welcome to the middle of a conversation. The links should help you find your way back to earlier movements if you want them.

---

A few years ago, I learned something about crime statistics that changed how I think about evidence.

Criminologists have long understood that “reported crime” and “actual crime” are different things. When you increase police presence in a neighborhood, you detect more offenses. The data then shows “more crime” in that area, which seems to justify the increased policing that generated the data in the first place. The measurement changes what gets measured.

This isn’t a conspiracy. It’s a structural feature of how detection works. A broken taillight in one neighborhood leads to a stop, a search, a drug charge, a data point. The same broken taillight in another neighborhood leads to nothing—no stop, no search, no charge, no data. Multiply this by millions of interactions, and you get statistical patterns that appear to reveal something about the neighborhoods themselves rather than about the distribution of attention.

I’ve been thinking about this lately because I keep encountering arguments that follow a similar structure in completely different domains.

When Measurement Creates the Problem It Claims to Find

Consider the question of whether some intervention—a medical treatment, a policy, a practice—causes harm. One way to study this is to compare people who received the intervention to people who didn’t and see which group has more problems.

The immediate difficulty: people who receive interventions and people who don’t are usually different in systematic ways before the intervention. They differ in how often they interact with institutions that detect and record problems. They differ in resources, geography, education, and dozens of other factors that influence outcomes independently.

What I’ve been noticing is the direction this bias runs. The poor and marginalized are simultaneously more vulnerable to disease and harm and less visible in the data that records it. They’re excluded from the healthcare ecosystem by cost, by geography, by immigration status, by the justified distrust that centuries of medical exploitation have earned. They get sick and suffer and die—but not in ways that generate data points, because they never entered the system that would record their suffering.

So when you compare an “intervention group” to a “non-intervention group,” you’re often comparing a population whose problems are fully captured by the healthcare system against a population whose suffering is structurally invisible. The intervention group’s harms appear in the dataset. The non-intervention group’s harms never existed as far as the data is concerned—not because the harms didn’t happen, but because no one was watching when they did.

The populations rendered invisible aren’t random. They’re structured by intersecting systems—race, class, geography, documentation status, disability, historical trauma. The invisible denominator has a demographic profile. And sophisticated statistical analysis, applied to this structurally biased data, produces what looks like rigorous finding but is really rigorous laundering of the original bias.

The machinery is elegant, I’ll grant. We’ve constructed measurement systems exquisitely calibrated to detect the problems of those we were already watching while remaining perfectly blind to the suffering of those we’d prefer not to see—and then we congratulate ourselves on our empiricism.

Long ago, in what now feels like another life, I had a career in economics and business that gave me the chops to deploy those statistical tools. I can still appreciate elegant methodology when I see it. But the most important thing I learned in all those years of quantitative analysis wasn’t a technique. It was a phrase: garbage in, garbage out. Rigor in analysis cannot compensate for bias in data collection. You can calculate precisely with numbers that don’t mean what you think they mean.

The technical vocabulary can actually obscure this. Someone might list impressive-sounding methods—Cox proportional hazards, chi-square tests, multivariate regression—and conclude that because the analysis was rigorous, the findings must be valid. But rigor deployed on biased data produces rigorous bias, not truth. And when the bias systematically undercounts the suffering of the marginalized while fully capturing the problems of the visible, the “finding” will systematically favor conclusions that let us off the hook for solidarity.

When Inquiry Becomes Performance

What interests me isn’t the technical debate in any particular domain. It’s the question of how we know when we’re actually inquiring versus when we’re performing inquiry while seeking confirmation.

Genuine inquiry has a particular structure. It requires asking: What would I expect to see if I were wrong? It means actively searching for evidence that might disconfirm what you’re inclined to believe, not just collecting evidence that supports it.

The philosopher of science Karl Popper called this falsifiability—the idea that a claim is only scientific if you can specify what would prove it false. But falsifiability isn’t just a criterion for theories; it’s a disposition of the inquirer. Am I actually open to being wrong, or am I constructing an elaborate architecture of apparent rigor around a conclusion I reached before I started?

There are tells. The person who announces “I take no position” and then spends the next hour explaining why one side is obviously winning has taken a position—they’ve just learned that certainty sounds more respectable when dressed in the vocabulary of open-mindedness. The framing that treats one view as needing to “help us out” while the other is simply “what the evidence shows” has already done the work that argument was supposed to do. Methodological objections flourish when findings are unwelcome and evaporate when findings confirm. And counter-evidence gets diagnosed as conspiracy with a speed that suggests the diagnosis was prepared before the evidence arrived.

These are the marks of what we might call inquiry theater—the performance of open-minded investigation in the service of foregone conclusions. The theater can be quite elaborate. The production values are often excellent. But theater it remains, and what it reveals is not the truth about the question but the truth about the performer: that something more than evidence is at stake, and that something has already settled the matter.

This is a lot of ground, and it’s okay if the technical details are starting to blur. What matters isn’t the vocabulary, but the question underneath it: why some forms of suffering never register as claims on us at all. If we miss that, everything that follows will sound abstract. If we see it, the rest begins to make sense.

Why This Matters

I notice this pattern across domains that have become identity-marked. Positions on contested questions become badges of tribal belonging, and what looks like debate about evidence is actually negotiation of social identity. The vocabulary of science or scholarship or reason gets deployed, but it’s operating according to a different logic—the logic of group loyalty and self-definition.

Theological language has surprised me with its usefulness here.

The Christian tradition I inhabit has a word for the state of being curved in on oneself, unable to receive correction from outside: incurvatus in se—ESSAY 7. Luther used it to describe sin—the fundamental curvature that bends everything, including our best efforts, back toward self-reference. I’ve been discovering how the phrase applies not just morally but epistemologically too. There’s a kind of intellectual self-enclosure where we become incapable of genuine inquiry because our conclusions are load-bearing for our identity. To be wrong would be to lose ourselves.

I’ve been writing elsewhere about shame and pride—ESSAY 6, and I’ve come to see that what happens in contested public debates has deep roots in how our nervous systems work. Shame, the neuroscience tells us, operates like firmware—it fires in milliseconds through subcortical circuits, below the level of conscious thought. Pride, by contrast, is more like software—something that gets constructed over years through relational experience.

This asymmetry matters. When a position becomes identity-bound, when being wrong about it triggers not just intellectual discomfort but that deeper firmware response of shame, we’re no longer dealing with a disagreement that better evidence can resolve. The nervous system narrows what I’ve been calling the river—ESSAY 6—the range within which we can actually think, stay present, and remain open to correction. Outside that river, threat responses dominate. We defend. We attack. We shut down. But we don’t learn.

Augustine—ESSAY 7 saw this clearly centuries ago. He distinguished between amor sui—the curved-inward love that makes the self its own ultimate reference point—and caritas sui—proper self-care that flows from first being loved by something beyond ourselves. The first generates what we might call hubristic pride: defensive, comparative, needing to outperform others to feel secure. The second generates authentic pride: grounded in reception, capable of genuine contribution precisely because it isn’t scrambling for worth.

Hubristic pride cannot receive. That’s its structure. To receive would be to admit dependence, which threatens the whole defensive architecture. So we take, we acquire, we even accept—but we cannot receive in the way that acknowledges we might be wrong, might need correction, might learn something that destabilizes what we thought we knew.

Two Grammars of Selfhood

But there’s something even deeper than the shame-pride dynamics—a grammar of selfhood that determines whether solidarity even registers as intelligible, or whether it arrives already coded as offense.

Two grammars—ESSAY 9 use the same vocabulary but mean opposite things:

One grammar—call it the Sage—understands the self as self-constructed through discipline. Freedom means freedom from externals: from interference, from claims others might make, from vulnerability itself. The goal is an inner citadel, a fortress of autonomy where no outside force can compel me. Worth comes through achievement, through what I’ve made of myself by my own effort.

The other grammar—call it the Saint—understands the self as received through participation. Freedom means freedom for love and service. Vulnerability isn’t weakness but the very site where grace enters. Worth comes not through performance but through belonging—through being held in relationships I didn’t earn and can’t control.

The irony of the Sage’s inner citadel is that it’s a prison with the lock on the inside. We’ve decorated it so tastefully that we’ve forgotten we can’t leave. Freedom from turns out to mean freedom from connection, freedom from vulnerability, freedom from the relationships that might actually make life worth living. The Stoic achievement of needing nothing from anyone is indistinguishable from the tragedy of having no one who needs you. Call it autonomy if you like. Loneliness is the more honest word.

Under Sage grammar, any external claim on me is categorically offensive. Not because I can’t afford it, but because it constitutes interference with my self-constructed autonomy. The request itself violates who I am.

This is why masking for others feels like tyranny rather than love. Why vaccinating for community health registers as imposition rather than participation. Why funding schools for all children—not just the children of my tribe—feels like theft rather than investment in shared flourishing.

Notice the achievement here. It takes considerable formation to experience a request to protect the vulnerable as an assault on your freedom. The Sage grammar has accomplished something remarkable: it has made care feel like violence and selfishness feel like liberty. We have worked hard to reach the point where “love your neighbor” sounds like oppression.

Under Saint grammar, these same requests aren’t offenses at all. They’re invitations to participate in what I was made for. If my worth comes through belonging rather than achievement, if freedom means freedom for rather than freedom from, then bearing cost for others’ sake isn’t diminishment—it’s fulfillment.

Why Solidarity Stops Making Sense

The theologian John Bowlin—ESSAY 2 makes a distinction I’ve found clarifying. Forbearance, properly understood, isn’t just patience or tolerance. It’s the charitable endurance of offense for the sake of love and reconciliation. It’s bearing injury without excusing wrong, staying in relationship, trusting that repair is possible.

Forbearance is what makes common life sustainable. It’s how we remain in community despite real wrongs. But forbearance is structurally inaccessible under Sage grammar.

If I’ve defined my selfhood as the absence of external claims, then to practice forbearance—to bear offense for love’s sake—would be to surrender the autonomous self I’ve labored to construct. The Sage cannot forbear because forbearing would mean letting an external reality make claims that the fortress was built to refuse.

And this is where the data problem and the forbearance problem converge.

The Sage grammar needs the invisible denominator. It needs not to see the marginalized dying outside the healthcare system, needs not to see the children failed by defunded schools, needs not to see the vulnerable harmed by the solidarity we refused. Because if we saw them—really saw them—their suffering would make a claim. And claims are what the constructed self was built to repel.

This is the dark comedy of the sovereign self: it must construct elaborate systems of not-seeing in order to maintain the fiction that it owes nothing to anyone. The autonomous individual turns out to require enormous social infrastructure to sustain the illusion of independence. We are radically dependent on the invisibility of those whose visibility would reveal our dependence.

This is why inquiry theater isn’t innocent. The performance of rigorous investigation that systematically undercounts certain suffering isn’t just bad methodology. It’s moral architecture. It’s building the epistemic conditions under which we never have to practice forbearance because we’ve arranged not to see the offenses we’re committing.

The structure is ingenious, really. First, define freedom as the absence of claims. Then, construct measurement systems that render invisible the people whose suffering would constitute claims. Then, produce “findings” from these systems that appear to justify the original refusal of solidarity. Finally, accuse anyone who questions the findings of being anti-science. The citadel builds itself, and calls the construction evidence-based.

Why Detachment Isn’t the Cure

The alternative isn’t neutrality or detachment. Everyone has commitments, priors, communities that shape what we notice and how we interpret it. The question is whether those commitments make us defensive or curious, whether they close us off from correction or open us to it.

I think about the difference between holding a position and being held by one. When I hold a position, I can examine it, stress-test it, let go of it if the evidence requires. When a position holds me—when my identity is invested in it—I’ll construct elaborate defenses against any evidence that threatens it.

The data that confirm me I’ll scrutinize charitably. The data that challenge me I’ll scrutinize skeptically. I’ll deploy methodological objections selectively—rigorous when encountering uncongenial findings, credulous when encountering congenial ones. And I’ll do all of this while believing myself to be simply following the evidence.

The theological tradition offers a diagnostic here: check for gratitude. Can I receive? Can I be genuinely thankful when someone shows me I was wrong, or do I experience correction as threat? Aquinas—ESSAY 7 called gratitude “a debt of love from which no man should wish to be free.” The phrase stuck with me—not least because we have made debt a curse word. The entire trajectory of modern freedom has been toward the cancellation of debts, the escape from obligation, the construction of a self that owes nothing to anyone. We have achieved the remarkable feat of making Aquinas’s vision of flourishing sound like servitude.

Aquinas saw what we’ve forgotten: the person who wishes to be free of the debt of gratitude has misunderstood not just ethics but reality. The self organized around refusing obligation cannot welcome debt, cannot bear the vulnerability of having been helped, must maintain the fiction of autonomous competence. And so it mistakes its bondage for freedom, its isolation for independence, its poverty for wealth.

Learning How to Be Wrong Together

We don’t escape these grammars by thinking better or choosing the right side.

We escape them by learning how to see—and how to stay—with what our seeing reveals.

The question underneath this riff isn’t whether we have biases. We all do.

The question is whether we’re willing to notice what our attention makes visible—and what it allows to disappear.

That willingness is not neutrality. It’s humility. And it’s the condition for any truth that might actually change us.

Next week, we return to the main movement—back to the cracks in the dikes, and to what presses through them when control finally fails.