I recently attended two conversations under the aegis of the Templeton Foundation’s Big Questions initiative. Both dealt with whether God exists”one directly, the other obliquely, by asking whether science makes belief in God obsolete. Both were so deeply unsatisfying that I have to wonder whether this enterprise should more aptly be named “wrong questions.”

To be clear, I think that the Templeton initiative is praiseworthy. The problem begins with the way all such conversations are framed. God, contra the late Richard Rorty, is not a conversation stopper, but the question “Does God exist?” surely is. It tilts toward the atheist’s strength, for it assumes that the religious believer is committed to the existence of something akin to unicorns or gremlins, for which there is not the slightest bit of evidence. In asking “Does God exist?” the atheist challenges the believer to produce sufficient evidence to persuade him. The believer cannot. That which counts as evidence for the believer can be explained or explained away by the atheist.

Evidence can always be interpreted in a variety of ways, some of them unfavorable to the believer’s case. (If you claim that God spoke to you in a dream, how is that different, as David Hume asked, from dreaming that God spoke to you?) The atheist will charge that the believer’s interpretation is gratuitous or merely circular: The believer is simply asserting the truth (of Scripture, for example) of precisely that which he must demonstrate.

When modern science is brought into the discussion, the believer is hard pressed to find a bit of room left for God to occupy. Our contemporary ways of thinking about nature leave little space for God-as-hypothesis. Science has no place for untestable hypotheses. Once again, the way such science-and-God questions are framed”Does science render belief in God obsolete?”diverts believers from testifying to the God they claim to love. It squeezes them into making existence claims on behalf of a distant God in a universe that doesn’t have a place for him.

From the atheist’s point of view, that is fine. It matters not at all whether the believer is pressed to defend the existence of Zeus or the God of Israel; for the atheist, it’s all mythological and superstitious in the end. The believer is reduced to having to defend the God of Israel as if he were Zeus”just another fictive entity in a universe now known not to contain such things. When the believer protests that the sort of God the atheist asks to be convinced of is not his God, the atheist protests that the believer is diluting the tradition. If the believer invokes a notion such as metaphor to explain religious language, the atheist already feels that he has won. And so he has. For the question “Does God exist?” is immediately prejudicial to those who take seriously the God of Israel.

In fact, the God of Israel is not one about whom existence claims can be made in any straightforward fashion. This is both a theological assertion and, leaning on the work of Ludwig Wittgenstein, a logical one. Martin Buber, on the eve of the First World War, was asked by a visitor, an English clergyman who worried about Buber’s soul, whether he believed in God. As Buber waited with him for the train and the indefinite parting to which the war would condemn them, Buber answered, reluctantly, yes. The pastor was satisfied.

Later Buber ruminated on what he had said. He felt that he had erred. The God he encountered in prayer, awe, wonder, and the small graces of the everyday was not a being about whom one could speak in the third person. God was not the sort of thing about which one could say, yes, it exists or no, it does not exist. To speak in this way was already to be estranged.

God, Buber felt, could not be discussed but only addressed”and that in the second person as “you.” To speak of God as if one were speaking of a thing, however recondite and mysterious, or of a distant person, was to speak of nothing more than a fictive character. For Buber, it seems, the word God named nothing real. Rather, the use of the word God , in the context of address, absorbs one in a way of life that touches on the real. All that we can really say of God is what we can say to God.

Faith, in this view, is never a set of belief claims. It is a way of life marked by trust, by affirmation of the goodness of being, by the repudiation of despair, and by an infinite openness to others and their needs. Buber contrasted religion, invidiously, to faith. Faith needs no tall tales. Religion cannot exist without them.

There is much that is wrong with this thoroughly existential view. Historical religions, as complex cultural and moral systems, do not fare well under it. (Buber was not an observant Jew in any traditional sense.) Nor am I sure that it sustains rational coherence. But there is also something honorable and right about it. The refusal of the faithful to be boxed into the existence question, as if it were the one thing needful, rings true. The faithful know that their way of life springs from mystery and goes to mystery. Their God, in Maimonides’ astringent view, bears no predicates. We are as little able logically to call God just as to call him unjust ; it is piety, for Maimonides, not logic that guides us to use the former rather than the latter term.

God is one, in classical Jewish theology. This does not mean that God is numerically one, as opposed to two or three, but that God is unique: There are none like God. Nothing about which we know anything is in any relevant way like God. Therefore, our language categorically fails to touch the divine. To say of God that he exists or to say of God that he does not exist are both wide of the logical mark because they force God to be like all of those things and persons of which we can say “X exists.” That is, they force God not to be God. (Given this severely negative theology, Maimonides also had his work cut out for him to accommodate traditional faith.) Maimonides and Buber, in their very different ways, both articulate a truth: We cannot talk about God in the way that we talk about the creation. The question “Does X exist?” is a question about things and persons, not about the God of Israel.

I can well imagine an atheist at this point taking issue with this whole line of reasoning. It is one thing, he might say, to be unable to present credible evidence for the existence of your God. It is another thing to glory in your inability to do so. If God refers to nothing, why use the word at all? If God is unlike anything of which we can have knowledge, then what could justify your use of the term? We use words because they have meaning. It’s not meaningful to use meaningless words. This just proves my point, the atheist could conclude: The term God no longer has any conceptual work to do. We should pension the concept off to permanent metaphysical retirement.

The faithful need to argue at this point that the word still has work to do and that this work cannot be done with equal adequacy by any other term. Note how the question has shifted: We are no longer discussing whether God exists and going through the terminally unsatisfactory process of marshalling evidence or assembling proofs. (Even the medievals, who put great stock in proofs, knew that the intellectual project of proving the existence of God rested on faith. If faith was lacking, the proofs could not succeed in their work: fides quaerens intellectum .) We are now wondering whether a whole way of speaking is warranted. The task has shifted from a metaphysical one to a logical one: Are we warranted when we speak to”or, if we must, about ”God?

One way of answering this question, presumably the one favored by the atheist, is that we are warranted in speaking to or about God only if God exists. We are warranted in saying “It’s snowing outside” only when it is snowing outside. Absent a discernible set of facts, one would not be warranted in making assertions about them. The atheist always wants to bring us back to the apparent paucity of facts. But are facts the only court of appeal for whether statements are warranted? Clearly not. There are many different warranted uses of language in which the stating of facts is beside the point.

So-called performative utterances constitute such a case. “I christen you the good ship Lolly Pop” does not state a fact. It brings a new fact into the world, as does “I now pronounce you man and wife.” Whenever we use language to change the world rather than to describe it, we reach beyond the practice of stating facts. Affirmations such as “Love is stronger than death” look factual in their surface grammar but are not. They are not testable in the way that statements of fact normally are, but they are no less meaningful and significant for that. Their significance lies in how life is lived in light of the conviction that they express.

Similarly, poems and stories model reality in complex ways and create new realities at the same time. They do not merely state facts. Could not religious language be more like these usages than like the standard, pedestrian use of language as a means of stating facts (or alleged facts) about the world? What seems to rule this out, the critic would charge, is that most statements about God in Scripture”most statements made day in, day out by ordinary believers”purport to convey factual content. God acts like the name of a being, whose existence is in dispute.

Against this, I would argue that the faithful ought to claim that, despite how the word God has been used for much of its linguistic career, the continuing warrant for the use of the term has little to do with the ­straightforward statement of facts. God plays a role in a way of speaking that is constitutive of a way of life, without which the world would be poorer and darker. The work it does is not to name a mysterious being who may or may not exist. The word God does not make a claim about the furniture of the universe. Rather, to speak of God is to underwrite a form of life that allows us to respond with love and courage and hope to the mystery out of which we come and toward which we progress. That some of our ancestors took the language of God in a mythological way, as a set of existence claims, is undeniable, although even here great ancestors, such as Maimonides, saw the problems that inhere in such naivete.

Wittgenstein taught us that language belongs to groups, not to isolated minds. Language reflects communal practices. Much of the reality that terms mark out is specific to the communities that use the terms. As any learner of a foreign language knows, reading a newspaper in that language requires learning about social and political realities specific to another culture. The abstract question “Does God exist?” is the question of an isolated mind. It tears God out of the context of communities who pray, celebrate, and serve, and it reduces the term to a cipher.

As the atheists were speaking at the Templeton events I attended, I asked myself what holidays I would celebrate if I were an atheist. What kind of community could atheism sustain? What degree of continuity, if any, could a perfectly atheist Western civilization sustain with its own past? This is not to suggest that religion is warranted only on account of the social, functional tasks that it performs. All sorts of false and pernicious things can enhance social solidarity and mobilization. Rather, it is to point toward a truth: As communal beings, we have constitutive ways of speaking that locate us in a meaningful universe and give moral contours to our shared form of life. An adequate conversation between a person of faith and an atheist cannot afford to neglect the questions of what we can celebrate, what we can hope for, what we must remember, what stories we can tell our children, and why we should bring children into the world.

I suppose that the question “Does God exist?” will not go away, for many people find it crucially important. Nonetheless, I would like to see it put aside. I would rather hear questions such as these put to atheists and persons of faith: Could you please make sense of love, courage, hope, and virtue? If you think that belief must be tied rigorously to evidence, on the basis of what evidence ought one to live a life of love, courage, and hope? What stories bear truth for you, and on what basis do you believe they do so? Is truth separable from stories with respect to matters of human significance? Can persons in the end live without a sense of the sacred? Do we inevitably discover sacredness in (or ascribe sacredness to) something central to our lives as persons and societies? If so, are we not better off anchoring the sacred in historic patterns of thought and conduct than in fresh enthusiasms? Does science open up a vista for us of wonder and delight, an iron cage of technical problems and complex moral consequences, or both? Does faith? And in the end, within reason, how shall we decide?

Alan Mittleman is director of the Louis Finkelstein Institute for Religious and Social Studies and professor of Jewish philosophy at the Jewish Theological Seminary.