Why Omni-Traits Make Your God Impossible
Here’s a question that seems simple until you press on it: if a being knows everything, does it actually know anything?
Think about what knowledge requires. To know that the sky is blue is to distinguish it from every way the sky is not. Knowledge is a figure against a ground. It only carries meaning because it excludes something—error, ignorance, the vast space of what’s-not-the-case. Strip away that contrast and you don’t get perfect knowledge; you get noise. A signal with no background is indistinguishable from silence.
This is an indication that the “omni” traits assigned to the classical God don’t maximize the properties they’re supposed to describe; they destroy them. And once you see that, the entire architecture of the classical deity starts to collapse, not because God is too mysterious to understand, but because the concept is structurally self-defeating.
What Personhood Requires
Before examining what omni-traits do, we need to be clear about what they’re being applied to. The classical God isn’t just a force or a substrate. Theism insists on a person—a being with will, knowledge, intention, and the capacity to act and relate. Prayer makes no sense addressed to a mathematical principle. Worship implies a “who.”
So what does personhood actually require?
First, it requires a boundary. To be a self is to be this and not that. Identity is contrastive at its core. “I” only means something because there’s a “not-I” it excludes. A being without boundaries isn’t a very expansive self; it has no self at all, because selfhood is precisely the operation of distinguishing inside from outside.
Second, it requires perspective. A person is located, not necessarily physically, but logically. To have a point of view is to be here and not everywhere simultaneously. Perspective means some things are closer, some further; some known directly, some inferentially; some relevant, some not. Remove location, and you remove the conditions that make a viewpoint possible.
Third, it requires lack. This is the most counterintuitive requirement, but it’s the most important. Agency—the capacity to want, choose, and act—is only possible where there’s a gap between the current state and a desired state. You can only will something if there is something you don’t yet have. Desire is structurally dependent on absence. A being that already possesses all outcomes, all power, and all knowledge cannot want anything, because want requires an unfulfilled condition. Without that gap, there’s no will, just a static totality.
Fourth, it requires an Other. Identity is forged through relation. A self is defined partly by what it encounters, what resists it, what it can’t simply absorb. If there’s nothing outside the being—no genuine Other—then there’s no friction, no encounter, no relationship; there’s only infinite solipsism.
These aren’t arbitrary philosophical preferences. They’re the minimum structural conditions for what we mean when we say “person.”
Now run the omni-traits through that filter.
What the Omni-Traits Actually Do
Omniscience doesn’t maximize knowledge. It collapses the conditions that make knowledge possible.
Knowledge is contrastive. It requires a distinction between what’s known and what’s not. For finite knowers, this distinction is real and consequential; we carry maps of reality that are always incomplete, always corrigible, always in tension with the territory. That tension is what makes “knowing” an activity rather than a state.
For an omniscient being, the distinction vanishes. There’s no unknown to contrast with the known. There’s no error to define correctness against. The information is total, which means it’s undifferentiated, and thus, it functions less like knowledge and more like the logical equivalent of white noise—everything present, nothing distinguished.
You can try to save omniscience by arguing that God understands logical negation: God knows all truths and knows all falsehoods as falsehoods, providing internal contrast. But this only relocates the problem. That internal contrast still isn’t the contrast between knowing and not knowing; it’s just a complete inventory. An omniscient being doesn’t come to know anything. It doesn’t reduce uncertainty. It doesn’t update. It simply is the total set. That’s not a knower. That’s a database without a user.
Omnipotence doesn’t maximize agency. It eliminates the conditions that make action real.
Action requires resistance. When we say someone did something, we mean they exerted force against friction—physical, social, logical, or otherwise. The doing is defined by the overcoming. An omnipotent being encounters no resistance by definition. There’s nothing to overcome, no effort required, no gap between intention and outcome.
Without that gap, “action” becomes a meaningless term. If willing and achieving are identical—if the distance between desire and result is exactly zero—then there’s no act, only an eternal fact. The being doesn’t do things. Things simply are, in whatever configuration the being’s will coincides with. That’s geometry, not agency.
Consider the stone paradox—can God create a stone so heavy it cannot lift it? This is usually treated as a clever puzzle, but it’s pointing at something deeper. The standard deflection is that God can only do what is logically possible, not what’s self-contradictory. But that concession defeats itself: all it establishes is that God can only do what God is able to do, bounded by his own nature. That’s not omnipotence. That’s a very powerful being with structural limits, which is exactly what the omni-prefix was supposed to eliminate.
Omnipresence doesn’t maximize presence. It eliminates the perspective that makes presence meaningful.
To be present somewhere is to be absent somewhere else. Presence is defined by location, and location is defined by contrast. If I’m here, I’m not there. This isn’t a limitation; it’s the logical structure of what “being somewhere” means.
An omnipresent being is equally everywhere, which means it has no here, which means it has no perspective—no location from which things appear closer or further, more or less relevant, this way or that way. The self that was supposed to be present everywhere has instead dissolved into the spatial fabric itself. It isn’t in all places. It is all places. And a place is not a person.
The Abstraction Argument
The most precise way to state what’s going wrong: omni-traits are limit concepts, not instantiable properties.
Consider infinity. Infinity is an enormously useful mathematical concept. It tells you the direction in which a sequence is heading, describes the behavior of functions at their boundary conditions, and it’s indispensable in calculus, set theory, and logic. But you cannot arrive at infinity. It’s not a very large number you eventually reach if you count long enough. The moment you try to treat it as an actual quantity rather than a directional concept, the math breaks down.
Omni-traits work exactly the same way. “Omniscience” is useful as a limit concept—it describes the direction in which knowledge points as ignorance approaches zero. “Omnipotence” describes the asymptote toward which power tends as resistance decreases. These are valuable conceptual tools for thinking about knowledge and power as abstract quantities, but they’re not properties a being can possess, for the same reason infinity is not a number you can possess.
The moment you try to instantiate them in a person—to say “this entity has omniscience” the way a person has a name or a perspective—you’ve committed a category error. You’ve tried to place a limit concept inside a bounded entity. The limit concept wins. The bounded entity disappears.
Others Have Noticed This
Charles Hartshorne is the most significant and interesting voice here, because he was a theist. He believed in God and still concluded that the classical omni-God was incoherent as a person. His process theology argued that a being incapable of change, learning, or being genuinely affected by the world is not a living being—it’s a static abstraction. To preserve God’s personhood, Hartshorne argued, God must be finite in relevant ways: capable of relation, of growth, of genuine encounter with an Other.
A theist concluded that omnitude destroys personhood. That’s worth sitting with.
Patrick Grim made the point more formally on omniscience specifically, using Cantor’s theorem to argue there can be no coherent “set of all truths” for any knower to hold. The structure of truth itself, he argued, outruns any possible container, including an omniscient mind.
Both are pointing at pieces of the same problem. The full structural argument is this: omni-traits don’t describe a maximally perfect person; they describe the conditions under which personhood becomes impossible.
A Finite God Makes More Sense
The question my analysis raises isn’t whether God exists. It’s whether the dominant conception of God is even internally coherent; and the answer, on structural grounds, is no.
A being with genuine omni-traits is not a very powerful, very knowledgeable, very present person. It’s the substrate of reality—the total sum of information, the complete causal structure of the universe, infinite space itself. Call it what you want, but it has no face. It can’t want anything, know anything in the way knowing requires, or act in any sense that distinguishes act from fact. It’s not a person; it’s a description of everything, which is to say, a description of nothing in particular.
The “omni” prefix doesn’t elevate a person to perfection. It dissolves the person into an abstraction.
But here’s what’s interesting: a finite God actually makes more sense of the record. A God who doesn’t know what his creations will do—because he isn’t everywhere and can’t foreknow the future—is a God who runs experiments. Like a scientist who breeds the mice and watches what they do, God fathers his children and discovers who they become. That God has reasons to test, reasons to feel wrath, reasons to grieve. The problem of evil stops being a philosophical embarrassment and becomes an expected feature: of course there’s suffering a finite God cannot stop. He made the conditions, but he didn’t script every outcome.
This version of God is coherent in a way the classical version never was. Not all-knowing, but deeply knowledgeable. Not all-powerful, but formidably capable. Not everywhere, but perhaps very far-reaching. A being with a perspective, a will, and genuine stakes in how things turn out.
The classical God of omnitude isn’t theology. It’s mathematics that forgot it was mathematics.
Which leaves the question classical theism has never seriously answered: if you strip away the omni-traits that make God logically impossible, what remains? Perhaps something stranger and more interesting than the textbook version—a God small enough to be a person, invested enough to have a will, and limited enough that what happens here actually matters to him.