Generated by Codex with GPT 5.6 Sol XHigh

Political opponents can disagree bitterly without inhabiting different moral universes. In Scientific American’s March 2025 article “How Immoral Are Our Political Opponents?”, psychologists Curtis Puryear, Emily Kubin and Kurt Gray argue that Americans routinely mistake partisan conflict for evidence that the other side approves of basic wrongdoing. That error can turn disagreement into disgust and make conversation feel pointless before it begins.

A large moral mirage

The authors begin with familiar evidence of distorted partisan perception. Democrats and Republicans exaggerate one another’s demographic traits and animosity, while the two-party system and outrage-driven social media make extreme examples unusually visible. The resulting picture is not merely that opponents are mistaken, but that they are malicious.

To test that belief, the researchers surveyed more than 600 self-identified Democrats and Republicans about six acts: wrongful imprisonment, tax fraud, embezzlement, animal abuse, viewing child pornography and cheating on a spouse. Almost every participant rejected every act. Approval was below 5 percent for the few behaviors that drew any assent, with no notable difference between parties.

Participants then estimated how many members of the opposing party would approve. Their average guess was about 23 percent - at least four times the observed rate. The overestimate persisted in another version of the survey designed to discourage deliberate exaggeration, and it remained even when participants were paid to answer accurately. The mistake therefore appears to be more than casual partisan posturing: many people seem to hold a sincerely warped model of their opponents’ basic moral standards.

From misperception to avoidance

That model matters because perceived immorality changes how engagement feels. Across further studies described in the article, people who regarded the other party as less moral were more willing to endorse dehumanizing descriptions and less interested in talking with or understanding an opponent. If someone is assumed to tolerate cruelty or exploitation, withdrawal can seem principled rather than polarizing.

The researchers also examined about 5.8 million posts made from 2013 through 2021 by 5,806 users on X, then called Twitter. Liberals and conservatives alike used labels such as “thief,” “sociopath” and “murderer” more often in political discussions than in nonpolitical ones. In the early 2010s, such language appeared at similar rates in talk about celebrities and political opponents. Its use against political opponents rose in 2016 and stayed elevated through the end of the dataset.

The social-media analysis documents a pattern, not its cause. It cannot show that online language changed users’ beliefs, that those users represent the broader public or that every use of a hostile word was a literal moral judgment. Likewise, the short article does not provide sampling details, effect sizes or full methods for most of the studies it summarizes. Its findings are therefore strongest as evidence of a recurring perceptual bias, not as a complete explanation of U.S. polarization.

Common ground, narrowly understood

One experiment points toward a modest intervention. Participants learned that a potential conversation partner belonged to the opposing party but shared their condemnation of wrongs such as tax fraud and animal abuse. Those given this concrete moral information were more willing to interact than people who did not receive it. Establishing common values before discussing a contentious issue may make an opponent seem like a person with conflicting judgments rather than a person without a conscience.

The common ground should not be overstated. Agreement on plainly condemned acts does not settle disputes over which groups are vulnerable, what counts as harm or how competing harms should be weighed. The experiment also does not show that a brief reminder produces durable cooperation, changes policy views or works in a heated real-world encounter. It identifies an opening for conversation, not a cure for political conflict.

The article’s most useful lesson is consequently precise: people should distinguish evidence of disagreement from evidence of depravity. Before treating an opponent’s position as proof of moral emptiness, they can test that assumption against specific, shared principles. Political divisions will remain, but removing an imaginary division over basic right and wrong may make the real ones easier to face.