Is the far left to blame for more political killings in America?

SINCE THE assassination on September 10th of Charlie Kirk, a right-wing activist, President Donald Trump has said that “most of the violence is on the left”. Pam Bondi, the attorney-general, posted on X that “we’ve watched the radical left normalise threats, call for assassinations, and cheer on political violence”. And J.D. Vance, the vice-president, claimed, while hosting Kirk’s podcast in his stead, that supporting violence “is not a ‘both-sides’ problem”, but rather one where the left has a “much bigger and malignant problem”.

Untangling how much one side is to blame for political violence is tricky. The suspect in Mr Kirk’s murder does indeed seem to have been motivated by politics. Court documents quote texts from Tyler Robinson, who has been charged for the shooting, saying he “had enough of [Kirk’s] hatred.” Mr Robinson’s parents told authorities that he had become “more pro-gay and trans-rights oriented”.

Describing the bigger picture of American political violence is a challenge, however. Trying to catalogue political violence is inherently subjective: analysts must determine which forms of violence count as political and assign ideological labels to attackers or victims. Nevertheless, the data sets available suggest that extreme liberals are not committing the majority of political violence in America.

The Prosecution Project, led by Michael Loadenthal of the University of Cincinnati, analyses felony criminal cases involving political violence to see which ideologies are most common. The project attempts to create a taxonomy for crimes that seek “a sociopolitical change or to communicate” to outside audiences, says Mr Loadenthal. Its data show that extremists on both left and right commit violence, although more incidents come from right-leaning attackers. The Prosecution Project uses indictments and court records to ascribe ideology. But those who commit violence often leave a messy trail of resentments that defy easy classification, and some are clearly mentally ill.

While hosting Kirk’s podcast, Mr Vance said that “the data [are] clear, people on the left are much likelier to defend and celebrate political violence”, citing evidence from YouGov, a pollster. The survey showed that 25% of Americans who identified as “very liberal” said political violence could sometimes be justified, compared with just 3% of those who identified as very conservative and 11% of Americans overall. Although Mr Vance is correct that these data are clear, such polling can be affected by the timing of the question.

Previous polling by YouGov reveals that partisans are far more likely to describe political violence as a big problem in the aftermath of an attack directed at one of their own (see chart). The survey cited by Mr Vance was conducted immediately after Kirk’s murder. Many commentators assumed that he was the victim of a political assassination by an extremist liberal. This is likely to have framed the question in the minds of respondents, inflating the gap between liberals and conservatives in the short-term.

A risk in the aftermath of Kirk’s assassination is that violence begets more violence. Research suggests that partisans overestimate support for violence in the other party by as much as a factor of four. When corrected on their misconceptions, support for violence falls by a third. The current levels of political violence are nowhere near what they were in the 1960s, notes Lilliana Mason of Johns Hopkins University. She also sees a different trend: attacks against political figures to get attention, not to advance a cause. ■