The Grid
One hundred and twenty traditions, answering the same twenty-three questions. The point is what happens when you sort it: a column ordered by any question puts traditions beside each other that may have never been compared. Every cell opens into an essay explaining what that tradition holds and why.
Suggested ways to read the grid
- Strange Bedfellows — seventeen structures that unconnected traditions arrived at separately, with no borrowing and no shared ancestor. They are also this atlas's control group, and the harder point: resemblance alone proves nothing.
- Read These Together — pairs that disagree in a way that explains both. Read either row alone and it is a position; read the pair and it becomes a choice, with a cost.
- Arterial Thinkers — minds that shaped many rows while founding none, and after them the Instrument Makers, who shaped the questions this atlas asks rather than any tradition in it.
How to use Theology Grid
- Tap any column header to sort every tradition by that dimension.
- Pin (⊕) any rows to hold them together at the top of the table, above every sort.
- Hide (−) any row to take it out of the table. It stays hidden until you click its family button at the top, which brings that whole family back — so you can clear a crowded column down to the handful you are comparing, then restore them a family at a time.
- Tap a tradition’s name to open its full profile; the buttons below filter by family.
- Cells show only each answer’s lead term. Tap any cell for the question its column asks, the complete entry, commentary on that tradition, and a short guide to reading the axis.
- Matching cells are not necessarily identical under the surface — tap both and see whether they are really the same.
Key terms, before you start
Every column carries a label like theodicy & ethics or soteriology. Those are the scholar's names for the questions. Here they are in plain English — you will not need any of them to read a cell, and the essays are written so that you never have to come back here.
- ontology — what exists at the bottom, underneath everything else.
- metaphysics — what reality is made of.
- theology — the account of God or of whatever is highest.
- epistemology — how anybody knows any of it.
- soteriology — the account of how a person is rescued, freed or completed.
- theodicy — how a good and powerful God can permit a world full of horrors.
- eschatology — the account of how things end.
- teleology — what the whole thing is for.
- christology — the account of who or what Christ is.
- axiology — what a tradition holds to be worth something.
- historiography — how the record itself came to be written, and by whom.
Three of the axis names are this atlas's own, and are used in a specific sense throughout:
- Ground — the self-existent bottom layer: whatever a tradition says was simply there, uncaused, with nothing beneath it.
- Ultimate — the highest reality a tradition actually points at, which is not always the same as its God.
- Metadivine — whatever stands above or behind the gods: a fate, a law, a substance or an order that even they do not control.
About the twenty-three axes
GROUND — Is there a self-existent bottom layer? One / Dual / Many / Prior-Stuff (world precedes gods) / None / Hidden-usurped / Unasked.
ULTIMATE — Its nature: Person / Consciousness-beyond-personhood / Both-stacked (personal operations, trans-personal essence) / Process / Law / Substance-plural / Personified Fate.
INTERFACE — How the ultimate is reached: direct address, via interfaces (avatars, sefirot, Manifestations), an impersonal apex over a personal middle tier ("the Stack"), or no operative personal layer.
WORLD — The world's status relative to the ground: appearance (to see through, or commissioned), God's body, real creation, self-existent, strategic or hostile construction.
VALUATION — Modifier: how the world is valued, independent of its status. Maximum → Negative.
METADIVINE — What, if anything, stands above the operative gods: fate, law, filed ordinances, the Way, the Pleroma, nothing.
TIME — Plotted-linear / Progressive-unbounded / Doomed-with-coda / Cyclical / Maintained / Dynastic / Everywhen / Imminent-suspended / Unplotted.
EVIL — The diagnosis: privation, adversary (created or uncreated), ignorance-craving, karmic physics, imbalance, pollution, misplaced divinity, indicted management, inside job, conceded defeat.
SALVATION — The target: reconciliation, liberation, awakening, theosis, purification, refinement, name-and-lineage, ease, none sought.
GRACE ↔ EFFORT — Who does the saving: pure effort (Jain pole) → cooperative (monkey) → grace-dominant (cat) → pure grace (Shinran / Luther pole).
CHANNEL — Is revelation open, closed-plus-interpretation, closed with scheduled reopening, suspended, reopened, esoteric, empirical, or participatory?
MEMBERSHIP — Lineage / civic / universal-missionary / universal-non-missionary / initiatory / elect-by-nature.
SUCCESSION — The engineering: closed-by-design, hereditary, apostolic office, collegial-elected, text-plus-community, reincarnation-lineage, living-prophet office, method-not-office, unresolved.
PURITY — The system type: sin, karma-deposit, contamination-without-intent, militarized, sanctuary-centered, collapsed-moralized, minimal.
ANTHROPOLOGY — What a person is: divine spark, created soul, karmic soul, no-self, natively pure, sprouts of goodness, crooked timber, fallen, covenant partner, workforce.
BODY & GENDER — Embodiment stance (celibacy-honored / householder-mandated) crossed with gender architecture (symmetric / asymmetric / hierarchical).
EDIT — The kavannah signature: what this tradition characteristically does to what it inherits.
CAPITALIZED VALUES — Throughout the grid, values set in italics mark the cells where a tradition is most distinctively itself — its signature answers, worth noticing first.
A NOTE ON METHOD — The twenty-three axes deliberately cut across registers a philosopher would keep apart: some questions are ontological (Ground), some theological (Nature of the Ultimate), some epistemological (Channel), some practical (Purity). Asking every tradition every question is a comparative device, not a metaphysical claim — some cells report a tradition's central concern, others its considered shrug, and the difference matters. The cells compress; the profiles carry the hedges and name the live scholarly disputes. Comparability has a price, and this atlas pays it knowingly. Each axis is therefore tagged with the philosophical register it belongs to — ontology, epistemology, soteriology, sociology — visible whenever a cell is opened, so that the crossing of registers is labeled rather than smuggled. One admission rule governs the grid itself: a seat requires a lived system — a picture of reality, a diagnosis, a path, and a carrier community. Thinkers who reshaped religion without founding one — Kant, Hegel, Schopenhauer — appear instead in Part III, as arteries.
Profiles
Every tradition in the atlas, grouped by family — and the arterial thinkers at the end, who shaped many of these traditions while founding none.
Reading a profile
Each entry opens with a short essay on what makes the tradition distinctively itself, followed by its axis readout. Tap any axis line to unfold that column’s plain question and the full commentary.
Dotted terms open the glossary. Blue links lead to related traditions.
The families run along the path of human migration — out of Africa, then West Asia, South Asia, East Asia, Oceania, Europe, the Americas — so the sequence carries no rank.
Arterial Thinkers are minds that shaped many rows while owning none — a row here needs a community, a practice and a way authority passes at a founder’s death, and these fail that test on purpose. Thinkers who do have a tradition of their own stay in their own row instead.
Instrument Makers follow them, set apart. These four shaped no tradition here at all — what they shaped is the description: the vocabulary and the questions comparative religion now uses, this atlas included. An artery flows into a tradition; these flowed into the thing that measures one.
Use the Also buttons to read either group on its own, or to set them aside.
The axis controls below apply here and to the Grid alike.
The Influence Audit
Documented arteries of influence, each graded for how well the descent is actually established. Both columns sort by tap. How a claim gets graded, and the convergence controls below that keep resemblance from counting as descent.
Reading the grades
SOLID — mainstream consensus. LIKELY — majority view, some dissent. POSSIBLE — a real parallel, causation unproven. DOUBTFUL — probably convergence rather than contact. REVERSE — an identity forged by opposition rather than inheritance.
One pattern is worth carrying through the whole table: receiving traditions tend to borrow the form and transform the content. Influence almost never means copying.
On the naming: Israel (biblical) marks borrowings received while the biblical text was still taking shape; Judaism (Second Temple / rabbinic / medieval) marks the later receiving eras. The distinction is chronological rather than polemical.
Convergence Controls, in a second table below this one, are the control group — structures that recur between traditions with no contact at all. Their presence is the reason resemblance alone is never treated as descent. A wider set of the same cases, drawn from across the atlas, is in Strange Bedfellows.
Convergence Controls
The control group: structures that recur across unconnected traditions — carved by the logic of the terrain, not transmission. Their presence calibrates the audit: resemblance alone proves nothing.
Findings
These findings are personal. They are not authoritative, they are nowhere near exhaustive, and every one of them can be sharpened or broken. The links lead to the clearest examples.
How a pattern earns a place here
They are patterns that seem to emerge — to one reader, at least — when the traditions are set side by side. There are surely more waiting to be found, and Theology Grid exists to make that kind of work faster.
Each must cite at least one exhibit in the corpus, and four things disqualify a candidate. Every one of these has cost this atlas a finding:
Circular — guaranteed by the atlas’s own construction, so it would vanish on a different choice of rows or axes. Single-instance — one episode generalized into a law. Untestable — asserting a correlation the grid does not record. Sample-dependent — true of the row-count rather than of the world.
A finding that fails audit is amended in place, with the amendment stated, rather than quietly dropped. One has been retired outright.
- The absolute does not stay abstract.
Traditions that refine God beyond every attribute find, again and again, that the ordinary worshipper wants a face, and reintroduces one where the theology left a gap.
The refining is genuine. A God described as beyond being, beyond category, beyond every name is not a placeholder for a vaguer one — it is the endpoint of centuries of careful argument, and the theologians who arrived there meant it.
And then the formless re-condenses into something that can be loved and addressed. A Buddha of Light with a name to call (Amitabha). The radiant emanations of the unknowable Infinite, through which Kabbalah gives the undescribable a structure one can actually pray toward. Beneath an impassible God who cannot be moved, the saints, whose friends can — and, on the other side of a boundary neither would want to blur, the shrines and the asking of intercession that grew up under a God declared unlike anything whatever.
This is not a failure of rigor, and the same thinkers frequently hold both — the theologian who proves God has no attributes is often the one who writes the hymn. It is what devotional practice does with an abstraction, and the traditions that go furthest into the abstraction are the ones with the most practice at it.
- Closed channels reopen.
When a tradition declares revelation finished, its people find ways to be spoken to anyway.
The declaration is usually deliberate and load-bearing. Closing the canon settles who may speak for the tradition, and prevents anyone from trumping a settled question with a fresh vision.
It does not stop the traffic. Pentecostal prophecy flowers inside a sealed canon. Sikhs open the Guru Granth at random and take the hymn at the top of the page as the day's answer. Chabad Hasidim consult their late teacher's collected letters, opening a volume and reading what is there. Tibetan Buddhism formalized the whole problem with terma — teachings hidden by a past master and recovered centuries later, so that genuinely new material arrives while nothing is technically being added. And village gods quietly resettle the margins of Jain temples, in a tradition whose liberated teachers cannot hear a petition.
The need to hear from heaven now, and not only long ago, appears to be permanent — and the ingenuity spent on satisfying it without reopening the canon is one of the more inventive bodies of work in this collection.
- The two-truths pattern keeps being rediscovered.
Wherever a tradition pushes divine presence or emptiness to its limit, the same architecture appears.
The problem it solves is a real one. Push the ultimate claim hard enough — everything is empty, or everything is God, or the self does not exist — and you undercut your own ability to say anything, including the teaching, the path, and the difference between wisdom and confusion.
The solution is two levels. One truth for daily life, in which the world and its distinctions are entirely valid: chariots exist, people have names, actions have consequences. And one ultimate truth in which those distinctions do not hold up under analysis. Both are true, for their own purposes, and neither cancels the other.
Buddhist philosophy formalized it, Advaita reached it from the other direction, Chabad built it into a doctrine of creation, and Islamic theology developed it over the status of God's speech. Each built the seam independently — which suggests the tension it manages belongs to the territory rather than to any one map.
- Remove everything above God, and God's own structure must carry the weight.
Older cosmologies gave the gods a ceiling: fate, cosmic law, an order even heaven obeyed.
The monotheisms removed that ceiling on principle. Nothing may stand above God — not destiny, not a law he did not author, not a rule he must operate inside.
But the questions the ceiling once absorbed did not disappear. They moved inside.
How can God be one, yet genuinely relational? The Trinity. How can God's speech be eternal, yet a book that exists in time? The uncreated Quran. How can an infinite God leave any room for a world to exist in? Tzimtzum — the divine self-contraction that opens a space.
In each case a paradox now does load-bearing work at exactly the point where the older cosmos kept its ceiling. These are not confusions. They are the visible marks of where the weight went.
And Spinozism marks the finding's limit. Surrender the personality instead of the ceiling — a God who is necessity itself, with no will and no preferences — and no paradox is required at all. Which shows what the knots are for: they are the price of keeping supremacy and personhood together.
- Grace and effort form a fork that every tradition eventually meets.
Does the infant monkey cling to its mother, or is the kitten simply carried?
South Indian devotion split over exactly that image, and the two schools are still named for it. The monkey holds on; the kitten is picked up by the scruff and does nothing at all.
The fork reappears everywhere. Shinran in Japan and Luther in Germany, with no contact and three centuries apart, both chose the kitten — and both arrived there by the same route, having concluded that human beings are too compromised for their own effort to count. Jainism stands alone at the far pole, where no grace exists because no dispenser exists, and the soul must do every inch of its own climbing.
Unconnected traditions keep arriving at the same crossroads, which suggests the crossroads is real. What splits them is not evidence but a prior judgment about how much a person can be relied on — and that judgment is made before the theology starts.
- Succession decides survival.
How a tradition hands on authority shapes its fate more than almost any doctrine.
The Sikh decision of 1708 gave the office of Guru permanently to a book and a community, abolishing the succession crisis by design. Chabad's chair has stood empty since 1994. The tulku lineage transmits an office through recognized rebirths and is now contested by a state claiming authority over reincarnation itself. And the exam-canon staffed an empire for thirteen centuries by making authority a qualification anyone could earn and nobody could inherit.
Institutions are theology with a lifespan.
The completed grid adds the opposite pole, and it is the finding's most useful half. The deepest durability in the atlas — San, Aboriginal, Diné — comes from no office at all: knowledge distributed across many trained people until there is nothing central to seize. Colonial authorities across two continents discovered repeatedly that destroying institutions did not stop practice.
So durability has two solutions. Armor the chair, or abolish it and make everyone a partial heir.
- Missionary faiths spread widely and are fragile at the root; bounded peoples stay small and endure.
Buddhism converted half of Asia and vanished from India, its homeland; Jainism never left India and never died.
The trade-off is consistent. A tradition that recruits must be portable, which means it must be separable from any particular people — and what is separable can also be displaced.
A tradition transmitted by birth and practice is very hard to kill and very hard to grow. Judaism and the Parsis show the same shape: small, bounded, and still here after every empire that outnumbered them.
Haiti supplies the refining exception. The Vodou ounfò makes kin by rite rather than by blood — which matters enormously for a population whose descent lines had been deliberately severed by sale and separation. The Lakota hunka adoption ceremony records the same discovery independently.
So the rule bends on one hinge. Boundedness survives displacement exactly insofar as its kinship can be made, and not only inherited.
- Winning minds and surviving as an institution are different victories.
Half the human race prays inside an eschatology — resurrection, judgment, a final setting-right — that Zoroastrianism plausibly supplied.
Part III weighs the case rather than asserting it. What is not in doubt is the arithmetic: perhaps 125,000 Zoroastrians remain.
Neoplatonism is the pattern's Greek twin. Its concepts became the grammar of God for three monotheisms — the God beyond being, evil as the absence of good, the soul's ascent, the whole vocabulary of describing the divine by denial — and its own schools were closed by imperial decree in 529. Augustine records finding the doctrine of evil-as-privation in «the books of the Platonists», and being released by it from nine years of Manichaeism.
Ideas travel best in other people's vehicles. The traditions that supplied the concepts are gone or nearly so; the concepts are held today by billions, and the debt is invisible from inside — which is what it looks like when an idea has been fully absorbed rather than borrowed.
- Religions of place face a hard choice about scale.
Bound to a city or a landscape, a religion dies when its place loses meaning — as Greece's did when the free city vanished.
Greek religion had no creed, no scripture, and no membership separate from residence. It was an activity of the polis, and when Alexander and then Rome dissolved the sovereign city, a religion that was citizenship had nothing left to be.
Scaled up to a nation without being universalized, a religion loses the one thing a universal God provides: a standpoint above the nation from which the nation can be judged. That is the lesson of State Shinto, and the atlas's Civil Religion row exists to keep the question live in every language.
And yet the Yoruba-Atlantic story shows the rule can bend. A religion of place, carried across an ocean in chains, survived by hiding inside its captors' church — because its succession lived in trained people rather than in temples, and the priesthood walked onto the ships.
The anchor's type decides the outcome. Traditions bound to a political form die with the form. Bound to land or to kin, they can outlive the form's conquest — Zulu, Vodun, the Andes, and Aboriginal Australia through dispossession itself.
- What a tradition does to what it inherits is its signature.
Every religion in this atlas builds with borrowed materials, and none is diminished by the fact.
Israel took the region's flood story — in which the gods send the waters because humanity has grown noisy and is disturbing their sleep — and gave the disaster a moral cause. Christianity took Greek philosophy and made it creed. Islam took an ancient shrine and re-aimed it at Abraham. A Hasidic master took a French battle anthem and made it a prayer.
The borrowing is universal, and the atlas's Canaanite row exists to make it measurable: El's name became the Bible's ordinary word for God, and Baal's storm-titles transferred nearly verbatim.
The transformation is where each tradition's identity actually lives. What arrives is regional and shared; what a tradition does with it is not.
- Each tradition is shaped by the heresy it most fears becoming.
Watch what a tradition polices most anxiously, and you will find the position closest to its own.
Judaism guards against «two powers in heaven» — the reading that would make a second divine figure of the exalted angel, or the Word, or the Son. Islam guards against shirk, the associating of partners with God, which is the one offense the Quran calls unforgivable.
Kabbalah remembers the Sabbatians, whose messiah converted to Islam and whose followers concluded that redemption now required deliberate transgression — the mystical tradition's own logic, run one step past the cliff.
Christianity keeps re-condemning Marcion, who proposed severing Jesus from the God of Israel and cutting the Hebrew scriptures away. The condemnation has needed repeating, most recently against a twentieth-century movement that tried it again with state backing.
And Buddhism conducts a standing internal trial over whether Buddha-nature smuggles the self back into a tradition founded on there being none.
In each case the feared heresy is not a foreign error. It is the tradition's own strongest tendency, taken one step further than it can afford.
- Monotheisms that take both scripture and philosophy seriously settle into a two-story architecture.
Again and again the same resting point appears: a God fully personal in action — speaking, willing, loving — above whom, or within whom, stands an essence beyond every category.
The pressure producing it is easy to state. Scripture presents a God who acts, changes his mind, and can be addressed. Philosophy argues that whatever is ultimate must be beyond change, beyond composition, and beyond the categories that apply to created things. A tradition taking both seriously has to house them together.
The solution is two stories. Maimonides holds that nothing positive can be said of God's essence while the Torah's personal language remains binding. Eastern Christianity distinguishes God's unknowable essence from his energies, in which a creature can genuinely participate. Mature Islamic theology separates the divine essence from the attributes. The Kabbalists place Ein Sof, the Infinite of which nothing whatever can be predicated, behind the emanations through which it becomes addressable. And the Sikh creed opens by naming God formless and manifest in the same breath.
Different languages, one floor plan — and the staircase between the floors is in constant use.
- One system's ceiling is another's floor.
Nothing in the cosmos carries its rank on its face.
The clearest case is a single Greek word. Heimarmene — the fate written in the stars, the chain of causes running through the whole world — was enthroned by the Stoics as providence itself: the rational order to be recognized, trusted, and lived in accordance with.
The Gnostics kept the same cosmology and reversed the verdict. The identical astral machinery became the mechanism of a prison, operated by planetary powers, from which the soul must escape.
Same structure. Opposite valuation.
Elevation is assigned by the system doing the looking — which is a caution worth keeping whenever two traditions appear to be describing the same thing. They frequently are, and the description is rarely the disagreement.
- When institutions fail catastrophically, indicting the management becomes thinkable.
Gnosticism arose, most scholars believe, among devout people whose temple had burned and whose God had not answered.
The problem they faced is the sharpest form of the problem of evil, and it arrived as an event rather than as a question. If the power that made and governs this world is supreme and good, then what happened is unaccountable.
Their resolution was to deny the premise. The power running the world is not the highest one — it is a lesser maker, ignorant of what stands above it, and the god of this world is not God. Which explains the state of things without either abandoning the divine or excusing the disaster.
The move recurs, and the second clear instance is fifteen centuries later and unconnected. When Nikon's council corrected the Russian liturgy in the 1660s, part of the Old Believer response was to conclude that the apostasy had arrived through the church's own authority — evil as official correction, the institution that was supposed to carry salvation now the thing to be refused at any cost. The priestless wing drew the further conclusion that valid orders had ceased and the age of Antichrist had begun.
Both cases share a structure that is not merely the oppressed indicting their oppressors, which is common enough to be uninformative. It is narrower and stranger: the authority that ought by its own account to be legitimate is identified as the counterfeit — the maker of the world in one case, the church council in the other.
Modern conspiracy culture keeps the structure with the theology removed: a hidden truth, a sleeping populace, a saving knowledge available to those who see. That much is analogy rather than descent, and is offered as one.
One can decline the answer and still honor the wound that asks the question.
- The shape of time decides the force of the design argument.
To a mind that assumes the world began, a creator feels nearly self-evident: something must have started this.
To a mind that assumes the world is beginningless, the question is malformed. An eternal series does not need an ignition; it needs a structure that keeps producing the next state from the last — which is what dependent origination supplies, and what Jain cosmology supplies, and what neither requires a maker for.
The Mahapurana presses the point directly, a millennium before Hume: where was God before creation, and what produced the desire to act in a being said to lack nothing?
Neither side is being careless. Each has parked one brute fact and reasoned outward from it — a necessary Person, or a necessary World — and the intuitions divide at bedrock rather than at the argument.
Which is why the design argument feels overwhelming inside one time-shape and unintelligible inside the other, to people of equal seriousness.
A note on what kind of finding this is. The grid records each tradition's shape of time; it does not record how compelling that tradition's members find the argument from design, so there is nothing here to check this against. It is reasoning rather than a convergence observed across unconnected rows — and the difference is worth marking, since most of its neighbors are the second kind.
- What a tradition believes a person is determines what it believes salvation is.
Tell me your anthropology and I will tell you your rescue.
A divine spark imprisoned in matter calls for release — and Orphism, Gnosticism and Neoplatonism all supply an ascent.
A fallen soul, damaged in the will and unable to repair itself, calls for grace — and Augustine, Luther and Shinran all arrive at a rescuer.
An ignorant mind calls for awakening, which is why Buddhism's remedy is insight rather than pardon.
Innate sprouts of goodness call for cultivation, which is why Confucianism's remedy is ritual, education and fifty-five years of practice.
And a being polluted from outside, who has done nothing wrong, calls only for water — which is why Shinto rinses and the Diné re-place, and neither has any machinery of confession or absolution at all.
The quietest axis in this atlas writes the loudest one. What a tradition holds a person to be is rarely argued about and almost never contested from outside — and it determines, in advance, what salvation could possibly consist of.
- Expecting the end soon is a phase, not a home.
Communities that live at the edge of the imminent burn bright and build little.
The returning Christ, the emerging Hidden Imam, the messiah due this afternoon: while the end is weeks away, institutions are a distraction and property is a liability.
Then the delay forces a reckoning. Time acquires a longer plot, and institutions rise where urgency stood.
The moves are well documented. Adventism kept the 1844 date and reinterpreted what happened on it: not an arrival on earth but the opening of a judgment in a heavenly sanctuary, still under way. The Witnesses kept 1914 and placed the enthronement in heaven rather than on earth — then turned the interval itself into one of the most thoroughly measured missionary efforts anywhere. Shia Islam kept the Imam alive and hidden, and redistributed his authority to scholars for eleven centuries. And Nichiren's mappo is the elegant solution: an age that is always the last one, so the urgency never expires and never needs rescheduling.
The pattern has repeated from the first Christian generation, waiting for the parousia, to the present day. The clock, it turns out, designs the architecture.
- Courage under a known doom is a shared human summit.
Where the outcome is conceded, worth migrates to bearing — the one territory that cannot be taken.
The Norse reached it through defiance. Odin knows how Ragnarök ends, knows the wolf will kill him, and spends an eye and nine nights on the world-tree buying information that changes nothing. Valhalla is not a paradise but a muster: an army assembled for a battle already known to be lost, recruited on exactly those terms.
The Stoics reached it through trust. The order is rational and good, what happens is what had to happen, and freedom is the acceptance of it.
The Mexica reached it through duty. The sun runs down unless repaid, and the composed bearing of the offered was itself part of the offering.
Defiance of an order that will lose; trust in an order that is good; feeding an order so that it will not fail. Opposite theologies, one unmistakable dignity — and perhaps the clearest place in this atlas where very different maps mark the same mountain.
- The stack is humanity's default; the departures are the achievements.
A remote source acknowledged everywhere and petitioned almost nowhere, above busy intermediaries who receive the actual worship.
With the African family filled out, the pattern is continental. Nyame over the abosom. Olodumare over the orishas. uNkulunkulu over the amadlozi — and the same pairing from the Caribbean to Sumer to the Andes, where Viracocha shaped the peoples and then walked away across the sea.
The reasoning is consistent wherever it appears. A supreme creator is conceived as too vast and too impartial to be interested in one person's illness or one village's rain, while the intermediaries are specific, local, and reachable. So the high God is named constantly in speech and approached almost never in practice.
Five continents, one architecture.
And the row-count cannot show it, which is worth stating plainly. Seven rows here carry the whole African continent and three carry Oceania, while fifty come from the three literate families that generate many documented sub-schools. This atlas over-samples exactly what this finding calls the departures, because departures are what produce arguing successors. The claim is about human religion; the exhibits, not the tally, are the evidence.
Read against this baseline, the usual story reverses. Strict monotheism is not religion's natural endpoint but a demanding departure: the deliberate demolition of the middle tier, maintained against a pull that reasserts itself constantly, as Finding I records. And nondualism is a second departure, dissolving the top into the all.
The stack is what humanity builds when nobody is arguing.
- In the beginning, a tradition is everything; its modern history is a map of ground held, ceded, and contested.
The old traditions were their peoples' entire intellectual estate.
Medicine — the chantway. Science — Sumer's omen-reading ripening into astronomy. Philosophy: why is there anything at all? And history, law, and art besides. One institution held all the ground because no rival institution existed.
Modernity is the jurisdiction settlement, and it has run four ways at once.
Some ground is ceded gracefully — rabbinic law itself directs the sick to the physician, and did so long before there was a contest.
Some is contested openly: Darwin against the literalists, the healing revival alongside the hospital.
Some was never split at all. Where the chantway is still sung, medicine and liturgy remain one craft, and the sandpainting is a treatment.
And some flows back — the clinic borrowing meditation, therapy inheriting confession, the twelve-step meeting rebuilding the mutual-accountability structures that religious communities used to supply.
Which suggests the working question for any row in this grid is not «is it true?» but «what ground does it still claim — and who now holds the rest?»
- The known, unpracticed, is the same as the unknown.
Across unrelated families the diagnosis converges: the enemy is not ignorance but forgetting.
The obvious truth is the best-hidden one, and what is known but unpracticed does not stay known. The Talmud puts it sharply — whoever says «I have nothing but Torah» does not have even Torah (Yevamot 109b) — and the Mishnah adds that wisdom exceeding deeds does not endure (Avot 3:9).
So the machinery is mnemonic, and it says so in its own names. Dhikr is remembrance. Nām simran is remembrance of the Name. Nembutsu is Buddha-remembrance. Sati, now translated «mindfulness», means memory first — and what is being remembered is the teaching, held in view during exactly the moments when craving would carry it out of mind. Hesychasm keeps the remembrance of God. The Stoics drilled prosochē, attention, as the day's first exercise. And the Shema is recited twice daily on the explicit reasoning that once is not enough, even for the obvious.
Even the goal is framed as return rather than acquisition. Kashmir Shaivism calls liberation pratyabhijñā — re-cognition, knowing again what was never actually absent.
An Italian kabbalist opened his ethics manual by promising to teach nothing new, since precisely the well-known is what is hidden, and forgetfulness of it is universal (Luzzatto, Mesillat Yesharim, 1740). The same diagnosis keeps being rediscovered outside the traditions, by people with no stake in any of them — which may be the strongest evidence that it is true.
Religion, on this axis, is a technology of attention — and the striking thing is not its age but its independence. A dozen traditions with nothing in common arrived at the same diagnosis of the same problem, and built the same kind of machine against it.
- Creation from nothing is a late idea, and a local one.
Ask what is at the bottom of things, and every ancient cosmogony we can read answers with material nobody made.
Sumer begins with Nammu, the primordial sea, from which the first gods arise. Egypt begins with Nun, the waters that were never abolished and still surround the ordered world on every side. Canaan's Sea and Death are as old as the order that fights them. Greece begins with Chaos, the yawning gap, and Norse cosmogony builds the world from the body of a murdered prior being.
None of these is an evasion. They are answers — and they agree.
Nor is the Bible an exception. Genesis opens with the deep already present, darkness over the face of the waters, and a wind moving over them: a picture of ordering rather than of origination, matching its neighbors exactly.
Creation ex nihilo — no preexisting matter, no raw material, nothing God worked with — was hammered out in the second century CE, largely against Gnostic and Platonist positions holding matter eternal. Its purpose is precise: to deny that anything whatever constrains God from outside, including the stuff.
The traditions that decline it are instructive. Jainism refutes a creator by argument and keeps an uncreated universe. The Buddhist rows have no floor at all. And Latter-day Saint teaching denies the doctrine by name, holding matter co-eternal with God and creation to be organization — a modern reversion to the ancient default, inside the family that invented the alternative.
So «why is there something rather than nothing?» is not religion's founding question. It is a specific philosophical achievement, with a date, an address, and an opponent — and it has become so standard that it now reads as the plain sense of texts that were read otherwise for a very long time.
One limit belongs on this finding, and it is the same limit Finding XIX states about itself. Every source above is written, and writing is about five thousand years old. What this establishes is that the earliest cosmogonies anyone can read agree with each other — which is a claim about the first records rather than about the first thoughts. Modern humans had been burying their dead with grave goods for something like a hundred thousand years before Sumer wrote anything down, and what those people believed about the bottom of things is not recoverable. The finding's core claim survives that limit intact: creation from nothing is datable, and the date is late. What it cannot claim is that nobody ever thought of it before.
- What licenses resistance is not the doctrine of providence but whether the standard can be captured.
The obvious hypothesis is that a strong doctrine of providence produces quietism: if the standing order is God's will, submitting to it is piety and opposing it is contending with heaven.
The corpus does not support it.
Providence is total in Confucianism, and the Mandate of Heaven licenses rebellion outright — a successful revolt demonstrates its own legitimacy. Providence is total in Marxism, whose dialectic nobody authored and nobody can suspend, and the conclusion drawn is that one must act. Predestination is at its strongest in the Reformed churches, which produced Europe's most developed theory of resistance to tyrants. And the Stoics identified God with fate and with nature outright, then built a freedom sized exactly to what tyrants cannot reach.
Four traditions with providence at maximum, landing in four different places.
What actually tracks the outcome is a different question: is there a standard of judgment the ruler cannot capture?
Where the standard is God's will read through a scholarly consensus that rulers patronize, appoint and fund, the default is obedience — not because rebellion is unthinkable but because it is unadjudicable. The Ashʿarite formula grants that God wills every event and approves only the good, so a rising is as much God's doing as the tyranny; and with both sides equally willed, who is right must be settled by some other authority, which the ruler generally holds.
Where the standard sits outside that reach, resistance becomes statable. Reason, in the Muʿtazilite and Zaydi reading, is available to anyone who can think. Results, in the Chinese reading, are available to anyone who can see a famine. Scripture, once every believer reads, is available to anyone who can read. History's direction, in the Marxist reading, is claimed by the party rather than the state.
And where the standard is one's own assent, as with the Stoics, the tyrant becomes irrelevant rather than illegitimate — which is sovereignty without politics, and the reason that school produced both counselors of emperors and men who died refusing them.
So the question to put to any tradition here is not how much it credits to God. It is who is competent to say that something has gone wrong — and whether that competence can be bought.
- Decidability and capturability are the same property.
A tradition that can settle an argument has an office empowered to settle it — and an office is a thing that can be seized, bought, or coerced. A tradition safe from seizure is safe because no such office exists, which is also why it can never settle anything. These are not two facts about religious institutions. They are one fact, seen from either side.
The exhibits run in both directions. Catholicism can define a doctrine and close a question, and its chair has been captured repeatedly — held at Avignon under French pressure for most of the fourteenth century, then claimed by three men at once, then run for generations as a dynastic asset. Sunni Islam vests authority in scholarly consensus, cannot be decapitated, and settles its disputes by attrition, several of which have been running for a thousand years. The Karaites refused every central office and could not thereafter fix a calendar: festivals fall on different days in different communities, because sighting the moon is a local act and no one can impose a common reckoning. The Reformers understood the trade, said so in writing, and judged the alternative worse.
The price is scale-dependent, which is the part worth keeping. A band of thirty has no doctrinal quarrel to adjudicate, no calendar to standardize and no state to face, so the San pay nothing for a design that costs the Karaites a great deal. What the absence of an office costs depends entirely on what the tradition needs to be able to decide.
This is reasoning from exhibits rather than a correlation measured in the grid, and the distinction matters here. The Succession column records how authority passes; nothing in the atlas records whether an interpretive dispute can be adjudicated, so the pattern cannot be counted — only argued and illustrated. Finding XV carries the same label for the same reason.
- Reformers who keep the Ground and the Metadivine but change the Ultimate are called a sect. Those who change the Ground and the Metadivine are called a new religion.
Three of this atlas’s questions are in play. The Ground asks what the deepest layer of reality is. The Metadivine asks whether anything stands above the gods that even they answer to. The Ultimate asks what kind of thing sits at the top — a person, a process, a law, many beings. The first two are the structure a tradition assumes. The third is its account of God’s character.
Reformers rewrite the third and leave the first two alone. Reform Judaism revised the liturgy, the calendar, the dietary law, the messianic hope and the boundary of membership, and did not touch the floor or the ceiling. Protestantism removed the priesthood, the sacramental system and the interpretive office, and did not touch them either. The Zaydis and the Ismailis divide from the Shia mainstream and from each other on the imamate, on law and on cosmology, and agree with it about both. Across the twenty-seven parent-and-revision pairs in this collection, the two differ on most questions and hold these two in common four times in five.
Six revisions did move the frame, and what became of them is the finding. Two moved one half and stayed: the Karaites shifted the ceiling and remain Jewish, if marginally; Swaminarayan shifted the floor and is securely Hindu. Four moved both. Yiguandao is classified as a new religion rather than a reform of Chinese popular religion. Tenrikyo describes itself as a new religion rather than a Shinto sect. The Latter-day Saints are the most disputed case in their family, and the dispute is precisely about whether they are inside it.
So the word «sect» appears to be tracking something structural. A tradition may say almost anything new about what God is like and still be counted a wayward child. Say something new about what lies beneath the world and what stands above the gods, and the parent stops using its own name for you.
Two things this does not cover. Gaudiya Vaishnavism moved both and is securely Hindu, which the pattern does not explain and which is left standing here rather than argued away. And the whole shape belongs to families that police a doctrinal boundary — it holds almost without exception among the Jewish, Christian and Islamic pairs, and fails in both East Asian pairs tested. A tradition with no creed to be outside of has no frame to guard.
Glossary
Every tradition-specific and technical term in this atlas, defined with no prior background assumed. Dotted-underlined terms anywhere in the atlas link here. Alphabetical; search below.
Sources
Where the claims come from. Every grid value that rests on a particular text, edition or study is pinned to it here.
How the atlas cites
Citation is at the precision each thing allows. Primary texts get chapter and verse where the numbering is stable; events get standard scholarship or primary documents; interpretations get a named scholar; a tradition’s self-description gets the tradition’s own source.
The comparative work — the findings, the convergences, the artery grades — is the atlas’s own, marked as such and never dressed in borrowed authority.
Where a source’s own standing is doubtful, the doubt is cited as part of the source: a saying the hadith critics grade baseless, a prophecy nobody can locate in Garvey, a gospel known only through its prosecutors.
The riskiest pinpoints were checked against the primary sources: the Guru Granth’s page numbers, the Hidden Words’ numbering, the 1885 platform’s fifth plank, and the Brazilian census — which, re-run in 2022, tripled one row’s count and moved the ledger with it.
What these sources can and cannot underwrite
They support claims: that a tradition holds a doctrine, that a text says a thing, that an event falls where it is dated. Pinned phrases tie a specific wording in the atlas to where it came from.
They do not support the grid values. A value like «Both-stacked» or «Prior-Stuff» is this atlas’s classification of a sourced claim, and no citation settles whether the classification is right. Of 2,760 values, 58 carry a pin directly.
So a reader checking a claim should find it here. A reader doubting a column is doubting the atlas’s taxonomy, which is argued in the essays and answerable only by argument — see the method note linked from the footer.
Build Your Own Religion
Twenty-two questions, answered as if you were designing a religion from scratch. At the end, the traditions here that answered the same way — and the places where nobody did.
By default every answer counts the same, so a match means this tradition shares the largest number of your answers. Under each question is a slider for how much that question matters to you, and a dealbreaker box — the nuclear option, which immediately disqualifies every religion not matching your exact answer to that question. Use the sliders only if you want the results weighted toward certain questions.
What this is
Twenty-two questions. Answer them as if you were designing a religion from scratch, for
yourself, with an open mind and nothing to defend — not as a test of what you already
believe.
When you finish, this finds the traditions in the atlas that answered the same way.
It is not a personality test and it does not tell you what you are. It compares
structures: these are the people who made the same choices you did, and you can go
and read what came of them.
Some traditions do not have a recorded answer to every question. Those are scored only
on the questions they do answer, so nobody is penalized for a gap in the record.