At FiveThirtyEight yesterday, David Leonhardt entered the thicket of public opinion on abortion policy with the goal of making some defensible generalizations, with some success. But on the crucial question of what kinds of restrictions the plurality of Americans who don’t fall neatly into “always legal” and “always illegal” camps would support, he may not have gone deeply enough.

He does establish pretty clearly that public opinion on abortion is quite stable (despite the occasional hyper-ventilating of both activists and pollsters themselves over marginal trends that reflect either outlier data or subtle changes in survey wording); does not break along gender lines (which pro-choice folk sometimes have a hard time accepting); and generally supports, by a comfortable margin, the status quo, including the preservation of Roe v. Wade.

Leonhardt also usefully warns against polling on abortion restrictions that don’t match the restrictions actually being discussed, which has often occurred during the big wave of state restrictions that has occurred after the national 2010 GOP landslide:

On abortion, as with almost any issue, short poll questions and answers do not capture the full complexity of people’s views. The standard Gallup questions about trimesters do not mention exceptions, for instance. And although a recent National Journal poll may seem on the surface to suggest that a plurality of Americans support the restrictions that the Texas Legislature passed last weekend, the bill includes details ­— no exceptions for rape or incest, the effective closing of most of the state’s clinics ­— that most Americans would quite likely oppose. On the flip side, some people who oppose a broad abortion ban after 20 weeks of pregnancy also oppose abortion access in some instances, like, for example, for economic reasons or in cases when prenatal screening shows the fetus to have a nonfatal condition.

What I’d add to his discussion of restrictions and exceptions is this: a stable majority of Americans appear to support legalized abortion but a plurality have what might be called “when” and “why” concerns that often conflict. The “when” concern, reflected in diminishing support for abortion rights as pregnancies develop, may have been sharpened by rapidly growing familiarity with ultrasound images, which make fetal development less of an abstraction governed by “trimesters.” “Why” concerns are hazier because they often cannot be effectively legislated: e.g., “convenience” abortions or “abortions as birth control” or “sex-selection” abortions. Who, other than the woman involved, could possibly know? So “why” concerns more often cut in the opposite direction, in the form of exceptions to “when” restrictions.

Leonhardt notes that rape/incest exceptions are very popular, which of course is why there are constant battles within the GOP and even within the anti-choice movement over including them in restrictive legislative or constitutional proposals. But that’s also true of health exceptions (even in the case of late-term abortions), which anti-choicers typically oppose and even demonize as a “loophole” that allows medical professionals (especially psychiatrists) to create “abortion on demand” via highly permissive judgments. (This was the anti-choice truism John McCain expressed in his famous Saddleback Church debate with Barack Obama in 2008 when he put the term “women’s health” in air-quotes in discussing abortion policy).

Since current constitutional law on abortion (as reflected in the reigning precedent, Casey v. Planned Parenthood, though it was eroded somewhat by Justice Kennedy’s paternalistic view of “women’s health” in Cathcart v. Gonzales) heavily emphasizes the health exception, its popularity is a very big deal, and a very big problem for anti-choicers. So narrowing it with definitions that exclude non-lethal health threats or mental and emotional distress, or inverting it by making “health and safety” regulations of abortion providers the ostensible rationale for restrictions, has become a regular feature in restriction proposals. None of these issues, particularly when you consider the practical impact of “health and safety” regulations on the early-term abortions that large majorities of Americans want to keep legal, have really been weighed in public opinion surveys.

Leonhardt’s conclusion is that the complexity of public opinion on abortion shouldn’t obscure the fundamental fact that Americans (or at least a large plurality of them) are deeply “conflicted” on abortion policy. That’s true if the any position other than absolute support for or opposition to legal abortion in every circumstance is considered “conflicted.” But I’d say majority support for a regime of legal early-term abortion without restrictions and legal late-term abortions if they meet several “why” exceptions–including rape, incest, and most crucially life and health of the woman–is reasonably clear and consistent, even if antichoicers manage to come up with devious ways to undermine that regime. Since that remains the constitutional status quo and thus (at the federal level, at least) the law of the land, the “deeply divided” and “conflicted” CW about abortion opinion misses the mark.

Ed Kilgore

Ed Kilgore is a political columnist for New York and managing editor at the Democratic Strategist website. He was a contributing writer at the Washington Monthly from January 2012 until November 2015, and was the principal contributor to the Political Animal blog.