In the new survey, which was conducted by Ipsos in partnership with ABC News using Ipsos’ Knowledge Panel, 52% approve of the move, compared to 47% who disapprove.
The question put before respondents about summoning the military only asked about violent demonstrations, and not peaceful protests.
Earlier this week, President Trump raised the possibility of dispatching active-duty troops to control the protests in cities across the country, triggered by the death of George Floyd, a black man, by a white police officer.
“If a city or state refuses to take the actions that are necessary to defend the life and property of their residents,” Trump said, in a warning to state and local leaders, from the White House Rose Garden, “then I will deploy the United States military, and quickly solve the problem for them.”
“That is why I am taking immediate presidential action to stop the violence and restore security and safety in America. I am mobilizing all available federal resources, civilian and military to stop the rioting and looting, to end the destruction and arson, and to protect the rights of law-abiding Americans, including your second amendment rights,” he continued, seeking to crack down on the civil unrest.
He stopped short of invoking the Insurrection Act, which empowers the president to send in U.S. military and National Guard troops within the country’s border in rare circumstances.
After Trump’s threat, the commander-in-chief’s own defense secretary, Mark Esper, broke with the president, at least for now, saying at a news conference on Wednesday, “The option to use active duty forces in a law enforcement role should only be used as a matter of last resort, and only in the most urgent and dire of situations.”
“We are not in one of [those] situations now. I do not support invoking the Insurrection Act,” he added.
The military, according to a Gallup survey, is the most trusted institution in American society, with 73% of Americans in 2019 saying they have either a great deal or quite a lot of confidence in the armed forces.
The second-most trusted institution, according to the survey, is small business, with 68% saying they have a great deal or quite a lot of confidence in mom-and-pop shops. But unlike those at the higher end of the spectrum, only 38% of Americans have a great deal or quite a lot of confidence in the presidency in 2019.
Still, the concept of using the military to control civilian unrest is particularly controversial.
Trump’s former defense secretary, James Mattis, in a remarkable and rare rebuke of the president he once served, wrote in an essay published in The Atlantic, “At home, we should use our military only when requested to do so, on very rare occasions, by state governors. Militarizing our response, as we witnessed in Washington, D.C., sets up a conflict—a false conflict—between the military and civilian society.”
“It erodes the moral ground that ensures a trusted bond between men and women in uniform and the society they are sworn to protect, and of which they themselves are a part,” he continued.
A fierce debate over Trump’s threat to send in the troops also played out over a New York Times op-ed, authored by Arkansas Sen. Tom Cotton, a conservative Republican who encouraged the president to invoke the Insurrection Act to control the “rioters,” which prompted widespread criticism.
But amid the wave of demonstrations commandeering city streets, some of which have turned violent, Republicans (83%), independents (52%), whites (56%) and Hispanics (60%) support sending in military troops to quell the violent protests.
Meanwhile, Democrats (72%) and black people (73%) disapprove of such a rare use of force to deal with the violence in near equal measure.
The newest numbers follow the release of an ABC News/Ipsos survey on Friday, which found that only 32% of Americans approve of Trump’s handling of the response to Floyd’s death, which sparked the more than 12 straight days of protests, while about two-thirds disapprove.
This ABC News/Ipsos poll was conducted by Ipsos Public Affairs‘ KnowledgePanel® June 3-4, 2020, in English and Spanish, among a random national sample of 706 adults, with oversamples of black and Hispanic respondents. Results have a margin of sampling error of 4.3 points, including the design effect. See the poll’s topline results and details on the methodology here.