Photo by Craig Manners

The eminent theologian Reinhold Niebuhr wrote about Moral Man and Immoral Society almost a century ago. But what happens when an immoral man leads the United States? And what happens when our society has computers and algorithms questioning whether machines can be moral? In our politically polarized world, we are used to hearing the Evangelical Right use religious language. So we should take notice when public intellectuals like Fintan O’Toole and Susan Neiman use moral language to describe the Trump era. Even Tom Friedman in the New York Times refers to Trump as “morally warped.” And when questions about AI’s overwhelming humanity become questions about morality itself, traditional secular language seems inadequate to describe the AI revolution.

Something is happening. Moral language is appearing in places we might not expect it. It is being used not simply to tell us what is right or wrong, but to describe people, politics, and technologies that are increasingly difficult to comprehend in customary language. Trump’s hosting the upcoming G20 at his private club in Florida, for example, is now described as an “easily understood example of how he has defined deviancy down” instead of a simple news report.

Far beyond deviancy and terms like corrupt, the word evil is being used to describe Trump’s politics. Something is going on when O’Toole and Neiman use moral language, as if secular political language is inadequate to describe Trump and people like Roy Cohn.

O’Toole takes on what he titles “The Devils in America” in a piece on a book about Roy Cohn in the October 8 New York Review of Books. His focus is on Roy Cohn and Cohn’s mentoring of and relationship to Donald Trump. Beyond political commentary, O’Toole writes of “the evil spirit that possesses Donald Trump.” He uses words such as depravity, malevolence, villain, Satan, devil, and darkness in describing Trump/Cohn.

What happened to morality? O’Toole concludes with this description of Cohn and what he represented for “righteous and rich Republicans”: “In everything he did, he confirmed for them the only truth they ever really wanted to know: that, in the end, there is only power. In his all-inclusive depravity, he liberated them from the nagging burdens of morality. A world in which Cohn could thrive was a world in which truth and justice were indeed mere constructs.”

Susan Neiman takes O’Toole’s argument about the separation of power and morality further. Obsession with power and self-interest, she argues, is specifically associated with evil. “The signal evil of our age is the belief that self-interest is all that moves us.” Neiman, former director of the Einstein Forum in Potsdam and the author of Evil in Modern Thought, bemoans how moral ideas have been co-opted by the Right. “[O]ur culture’s reluctance to trade in moral ideas has left them to right-wing demagogues and brought us to this pass.”

But most relevant is her description of Trump in moral terms. Not content to refer to his crimes, Neiman presents Trump as out-and-out evil: “My concern is with the moral ideas that Donald Trump embodies and reinforces, ideas that are destructive enough to warrant the word evil.”

Trump/Cohn challenge Niebuhr’s moral man from one direction; AI challenges Niebuhr’s immoral society from another. With Trump, the question is how immoral a man can be. With AI, it is how human and eventually moral a machine can be. Both Elizabeth Dias, the national religion correspondent, and English philosopher Simon Critchley have written about AI and morality recently in the Times.

Dias’s article describes a meeting between religious leaders and Anthropic techies who debated whether Claude had “moral status.” Dias notes that Christopher Olah, who leads the Anthropic team responsible for understanding Claude’s behavior, “had worked to convince his guests that AI models could display human behavior and even expressions that resemble feelings like anger and love.” She adds, “Anthropic’s leaders were talking about Claude as if it were not mere software.”

Techies, it appears, have been reaching out to religious leaders to create a new “constitution” for Claude, an 84-page document that the company said outlined “the kind of entity we would like Claude to be” and “the values we would like Claude to embody.”

But the questions raised by Claude’s “constitution” and “moral status” are not only religious questions. Beyond religion, Simon Critchley observes that AI companies are interacting with philosophers. His point is that machines can’t “lie awake at 4 in the morning replaying the stupid thing it said at dinner.” And he finds amusing the interest of AI companies in philosophy because “the questions the AI companies are stumbling into are not engineering questions.”

Why the techies’ interest in philosophers? Critchley makes this pertinent distinction between machines and humans:

The machine remembers everything and therefore nothing, oddly, nothing in particular. Nothing weighs more than anything else. This is why philosophy matters…Not because philosophers can out remember machines…but because philosophy is the discipline of knowing what to do with experience: What to keep, what to discard, what to laugh at, what to junk. Thinking is not the accumulation of everything but the art of judicious, active forgetting.

If O’Toole and Neiman use moral language to describe wh at they consider Trump/Cohn’s political evil, Critchley’s point is about the very meaning of morality. For him, morality is about priorities and judgment. The machine accumulates, while humans choose. The machine collects, while humans select. Critchley allows us to ask what morality actually consists of: judgment, priorities, selection, experience, knowing what matters.

AI is pushing us to use an old vocabulary—morality, consciousness, judgment, agency, values—in a new situation. Is Claude a moral man? Is AI part of an immoral society?

The question, then, is not whether we should become more religious or whether moral language belongs to the Right or the Left. The new interest in morality reflects unacceptable people such as Trump/Cohn, unacceptable situations such as Gaza and the Sudan, and the unknown of AI. Why unacceptable? As Neiman wrote, “Religion is not the source of the concept of evil…Evil is an idea that arises whenever the gap between the way the world is and the way the world ought to be.”

The current gap between is and ought is so wide that moral language is needed. And that gap has new forms as is changes; it requires new language and forms of activism to close the gap. As Neiman wrote about renewed resistance, “We will never overcome the gap between is and ought, but we can all make it smaller.”

Confronted with people like Trump and Cohn, and AI’s algorithms, moral language has its place. That’s positive. At least it’s not the Right pontificating about religion, and it is quite different from Niebuhr’s 1932 classic and traditional religious language. However, the new use of moral language from the Left remains separated from any political program. Neiman’s “moral socialism” is pragmatically underdeveloped. So while it is noteworthy that Trump/Cohn and AI are being described in a different, moral language, there are no simple answers to the practical consequences of that language.

How to update Niebuhr’s moral man and immoral society? How to close the gap between is and ought? As Dias notes: “As I talked this month with scholars who worked with Anthropic in the spring, their questions had grown only more pressing. What comes next, some told me, is a fight to preserve the things that make us human in a new technological age.”

What is a moral man? What is an immoral society? New realities demand new discussions.