CeleryKills

On the Continuity Between Ritual Interpretation and Data-Driven Prediction

Algorithms show up in my life like quiet priests of a new order. Not robes, not incense, just a feed that knows what I might click before I do. I don’t always trust them, but I keep coming back, which is its own kind of confession.

I remember a morning in Seattle where the rain was doing its usual soft insistence and my phone buzzed with a notification I did not ask for. “You might like this.” It was a prediction, politely disguised as suggestion. I tapped it anyway, because of course I did. The screen opened onto a sequence of content that felt less like browsing and more like being gently steered. Somewhere between curiosity and compliance, I noticed the logic had already anticipated my hesitation, like a parishioner who knows exactly when I will reach for the pew.

Later that day, I watched a friend check a weather app with the seriousness of a ritual. Not casual scrolling. Consultation. The forecast was not just information, it was guidance for whether the barbecue would happen, whether the day would unfold one way or another. “Forty percent chance of rain,” they said, and paused as if listening for a second voice beneath the number. The system had already done its work: it translated atmospheric noise into structured prediction, then handed it over as something legible enough to organize behavior. I found myself thinking, this is prophecy, but with a UI and a terms of service page nobody reads. Anthropology calls this pattern-making under uncertainty. Information science calls it entropy reduction. Spirit, if you let it speak, calls it a sign. And I am not entirely convinced those are different languages anymore.

Algorithms as modern divination: data-driven systems replicate the predictive and interpretive role once held by magical and religious practices. The parallel is not decorative. Ancient diviners read entrails, stars, and patterns in scattered signals, then translated uncertainty into narrative. Today’s systems do something eerily similar, except the entrails are logs and the stars are vectors. The interpretive step persists. A model produces a probability, and someone, somewhere, converts it into action. Forecasts about markets, weather, elections, and even personal behavior function as structured guesses dressed as inevitability. “The algorithm says” has the cadence of “the oracle speaks,” and I notice how easily people lean in. I do too, sometimes, which is the uncomfortable part.

There’s a historical undercurrent that refuses to stay buried, and it becomes sharper when viewed through the Western entanglement of witchcraft, magic, and Christianity in the essay Witchcraft, Magic, and Christianity: A Western Entanglement (CeleryKills, 2026). The argument there traces how ritual practice was never fully displaced by doctrinal authority, only reclassified. What gets labeled as superstition often persists as informal epistemology, especially when institutional systems need ambiguity to remain legible without admitting it. That dynamic sits right underneath modern algorithmic culture. The algorithm does not arrive as a clean break from divination. It arrives as a continuation of the same tension: a patterned way of reading uncertainty that is simultaneously disavowed and relied upon. Readers can recognize the structure if they look closely. A system is declared rational, then treated as if it carries a kind of sanctioned foresight anyway. The old distinction between forbidden magic and authorized knowledge starts to look less like a boundary and more like a shift in vocabulary.

What follows from that is not nostalgia, but a recognition of how prediction functions as a social technology. In the historical context laid out in that essay, practices of divination survived when they were embedded within acceptable frameworks, even when their mechanics remained unchanged. The same pattern shows up in forecasting systems today. They are not only generating predictions, they are organizing attention around those predictions, turning probabilistic outputs into structured expectations that people act on. The reader should notice how easily this slips into everyday life. A forecast, a recommendation, a risk score, all of them quietly instruct without overt command. The question that lingers, and it is not a rhetorical flourish but a structural one, is whether the authority attached to these systems comes from their accuracy or from the cultural need to believe that uncertainty can be disciplined into something readable. Either way, the effect is the same: uncertainty does not disappear. It becomes legible enough to obey.

Anthropologically, this is familiar territory. Divination has never been about pure prediction. It is about making the chaos of the world legible through a system that feels non-arbitrary (Park, Divination and Its Social Contexts, 2003). The Yoruba Ifá system, for instance, encodes vast combinatorial complexity into interpretive patterns, producing meaning from randomness through disciplined procedure (Bascom, Ifa Divination, 1969). That structure is not far from a modern classifier that segments noise into categories. The difference is not that one is rational and the other mystical. The difference is mostly in the packaging, and in who gets to decide what counts as evidence. I find myself asking, are we any less ritualistic when we tune a model than when we cast bones?

Spiritual framing complicates this further. Traditional prophecy does not only predict. It binds. It tells you what the world is doing and what you should do about it. Modern forecasting systems inherit that dual function. A risk score is not just a number; it is a directive. A recommendation engine does not merely describe preference; it shapes it. There is a quiet theology in that feedback loop, one that assumes the world is readable and therefore actionable. I can’t help but hear a faint echo of the idea that “the map is not the territory,” except here the map is actively rewriting the territory (Korzybski, Science and Sanity, 1933). If the map starts deciding where you walk, what then is the territory?

Information science gives the most stripped-down version of this dynamic. Prediction is compression. To forecast is to reduce uncertainty by encoding patterns in past data into a model that generalizes forward (Shannon, A Mathematical Theory of Communication, 1948). But compression is not neutral. It selects, discards, emphasizes. A model is a kind of epistemic lens, and every lens bends. In practice, these systems do not eliminate uncertainty, they reassign it. They turn a fog into a set of probabilities and call it clarity. That move feels almost alchemical. I’ve watched teams treat a 0.73 confidence score like a verdict. Why does a number feel so much like knowing?

Prophecy as structured prediction: modern forecasting systems mirror traditional forms of divination in their attempt to render uncertainty legible. The structural similarity is simple enough that it’s almost funny. Both rely on constrained rulesets, both produce outputs that exceed their inputs, and both demand interpretation. The oracle and the algorithm occupy the same conceptual slot: they organize ambiguity into something that can be acted upon without collapsing it entirely. The legibility is the product, not the truth.

Early prophecy worked less like a forecast and more like a negotiated reading of signs. The Delphic oracle did not hand down a clean variable with confidence intervals. It delivered language that needed interpretation, often deliberately ambiguous, and that ambiguity was part of the system’s power (Park, Divination and Its Social Contexts, 2003). The Priestly divination traditions, whether reading omens or stars, operated within constrained grammars of meaning, where the “output” was inseparable from the interpreter’s training and the surrounding ritual context. Readers should notice the difference: the prophecy is not a single number, it is a structured uncertainty that requires a human to complete the circuit. I keep thinking about how that completion step is where authority quietly forms, not in the prediction itself but in who gets to interpret it.

Modern forecasting systems compress the same structure into something more mathematically disciplined. A weather model does not speak in riddles; it speaks in probabilities, often with deceptively precise formatting. Yet the interpretive burden still exists, just displaced. A 40 percent chance of rain looks like clarity until someone decides whether that means cancel the picnic or gamble on dry grass. In financial forecasting, the same pattern repeats: risk scores and predictive analytics translate noisy data into actionable signals, but the action only becomes real when someone reads the signal as guidance rather than noise (Shannon, A Mathematical Theory of Communication, 1948). The contrast with earlier prophecy is not that one is mystical and the other scientific. It is that the older systems made interpretation visible, while the newer ones hide it behind the smooth surface of quantification. The readers can feel the difference, even if they don’t always name it. Who, exactly, is doing the interpreting now, and why does it feel less like consulting an oracle and more like following instructions that arrived without a speaker?

Why use anthropological, spiritual, and information science explorations together? Because each one refuses to close the loop on its own. Anthropology shows the continuity of ritual and pattern-making. Spiritual frameworks expose the interpretive commitments hidden inside prediction. Information science keeps pulling us back to the mechanics, the bits, the entropy, the cold arithmetic. Take one away and the picture flattens. Keep all three, and the question starts to shift. Not “Is it true?” but “What kind of system is this that makes uncertainty feel like something you can hold?”

I keep thinking about that. And I keep thinking about how often I refresh a page, waiting for a system to tell me what I might already know, just in a more convincing form.

References

• Bascom, William. Ifa Divination: Communication Between Gods and Men in West Africa. 1969.
• CeleryKills. Witchcraft, Magic, and Christianity: A Western Entanglement. 2026.
• Korzybski, Alfred. Science and Sanity. 1933.
• Park, Robert. Divination and Its Social Contexts. 2003.
• Shannon, Claude. A Mathematical Theory of Communication. 1948.