WASHINGTON - Two months ago, several media polls examined the latest trend in pro-life legislation: outlawing abortions 20 weeks after fertilization, on the grounds that fetuses at this stage of development can sense pain. The polls suggested that pro-lifers held the upper hand in this fight. A new survey commissioned by Planned Parenthood challenges this conclusion. According to the organization, it shows that "when voters understand the real-world circumstances" of these abortions, "they overwhelmingly oppose" banning them.
Is that true? Let's look at the data. Planned Parenthood, to its credit, hasn't just issued a selective press release and a polished memo from its pollster, Hart Research. It has also provided the entire questionnaire and results to Slate. Here's an independent assessment of what the findings show.
1. The game plan. "Unlike other surveys," the pollsters write in their memo, "we measure voters' considered opinions about these 20-week bans by asking about the specific circumstances in which abortions should and should not be allowed." When the questions focus on these circumstances - giving voters "a chance to learn about the real-world consequences" of 20-week bans - "the margins in opposition to these bans are so significant that we think it is very likely that they would be voted down in a popular referendum in virtually any state."
The point of the poll and the memo, in other words, is to show pro-choice politicians and strategists how to frame the issue so they can win. The poll isn't a neutral test of public opinion. It's a model of how to steer the conversation. Pro-lifers want to steer it toward pain; pro-choicers want to steer it toward extreme circumstances.
2. The circumstances. With this strategy in mind, the poll demonstrates which circumstances you should talk about if you want to turn people against a 20-week ban. "Serious, long-lasting health problems if [the woman] carried the pregnancy to term" are persuasive: In that case, 66 percent of voters say abortion should be legal after 20 weeks. Rape or incest is another moving scenario: In that situation - on which previous findings were ambiguous - 61 percent say abortion should still be legal after 20 weeks. "Severe abnormalities" in the fetus are also effective: They persuade 58 percent to say abortion should still be legal.
3. No exceptions vs. no restrictions. This is the crux of the fight. A July poll by ABC News and the Washington Post put the question this way: "The U.S. Supreme Court has said abortion is legal without restriction in about the first 24 weeks of pregnancy. Some states have passed laws reducing this to 20 weeks. If it has to be one or the other, would you rather have abortions legal without restriction up to 20 weeks, or up to 24 weeks?" That question, which used the phrase "without restriction" twice, found a strong preference - 56 to 27 percent - for a 20-week limit. A subsequent Quinnipiac poll asked the same question and got almost the same result: 55 to 30 percent. People don't like tolerance without restriction.
But what if you ask them about intolerance without exception? That's a core strategy of the Planned Parenthood poll. Two of its questions ask about "blanket" 20-week bans. Two others ask about bans that "include" extreme cases such as rape or grave health risks to the woman. These questions reverse the findings of the Post and Quinnipiac polls, yielding anti-ban majorities of 53 to 59 percent.
4. Does the traditional pro-choice message work? Sort of. A July poll by NBC News and the Wall Street Journal told respondents that some people believe "20 weeks after fertilization is the point at which a fetus is capable of experiencing pain" while others believe "medical decisions should be between a woman and her doctor, and government should not be involved." When presented with those two arguments, a 44-to-37 percent plurality favored a 20-week ban.
The Planned Parenthood survey - taken by one of the two firms involved in the NBC/WSJ poll - challenges that finding. Its first question about a 20-week ban asks whether abortion should be legal or illegal at that stage when "a woman's doctor determines that the fetus is not yet viable and the woman and her family determine that her health and personal circumstances are such that she should not continue her pregnancy." In that scenario, 61 percent say abortion should be legal. The question is loaded with pro-choice phrases - "woman's doctor," "woman and her family," "personal circumstances" - but they're enveloped in language about health and medical circumstances. It's not clear to what extent the idea of personal decision-making, as opposed to unusual health circumstances, is moving respondents.
Later in the poll, a much clearer statement - "Blanket bans on abortions after 20 weeks go too far in interfering in the relationship between a woman and her doctor" - yields 59 percent agreement. But that message isn't pitted against a pro-life message, and by this stage of the survey, the respondent has heard seven questions or scenarios about specific tragic circumstances that might justify late abortions. At the end, respondents hear another statement: "It is not the place of politicians or the government to impose their views on the difficult personal decisions families have to make when a pregnancy involves a severe fetal abnormality or a woman's health could be permanently harmed if she carries the pregnancy to term." Fifty-two percent call this argument against a "blanket ban" very convincing, and another 20 percent call it somewhat convincing. But again, the pro-choice language is coated with references to health and fetal abnormality, and it isn't tested against a pro-life message.
5. Loaded scenarios. The poll never refers simply to fetal abnormalities. Each time, in a total of five questions, it describes the abnormalities as "severe." In two questions, it specifies abnormalities that "would cause fetal death or extreme disability." Likewise, it asks not simply about health but about cases in which the woman, if forced to carry the pregnancy to term, would face "serious health risks," "serious, long-lasting health problems," or "serious, permanent health problems." Why does the questionnaire phrase these scenarios so gravely? Because otherwise, you don't get a reliable majority against the ban.
The pollsters note: "By a 20-point margin (54 percent to 34 percent), voters oppose a ban that would make an exception only for the life of the woman after 20 weeks (without a health exception). But 'amending' this restriction to allow for other circumstances does not alter voters' objections to such bans. For example, by a similarly strong 17-point margin (53 percent to 36 percent), voters also oppose a ban that allows exceptions for both the life of the woman and limited cases involving her health." The pollsters portray this as evidence of solid opposition to a 20-week ban. But if you think women's health is an important reason to permit abortions, the failure of this exception to influence respondents is alarming. They also omitted from their memo the results of a question about another scenario, in which "a woman's doctor determines that the woman would be permanently unable to become pregnant again if she carried the pregnancy to term." In that case, a plurality but not a majority (45 to 38 percent) said abortion after 20 weeks should be legal. These are warning signs that in the political fight against 20-week bans, speaking broadly about women's health won't be enough.
6. Viability vs. pain. No published poll, to my knowledge, has asked respondents to choose between the rationale for drawing a line at 20 weeks post-fertilization (pain) and the rationale for drawing the line two weeks later (viability). (Gestational age is estimated from about two weeks prior to fertilization, on average, so viability occurs in the vicinity of 22 weeks post-fertilization or 24 weeks gestation.) The Planned Parenthood poll doesn't ask this question, either. It includes a scenario in which "a woman's doctor determines that the fetus is not yet viable," but that scenario also entails unspecified "health and personal circumstances." What would voters say if you asked them to choose between imposing a deadline at the onset of pain capability (whenever that is - most medical authorities dispute that it's at 20 weeks) and imposing a deadline at viability? That's the question any neutral pollster should want to ask.
7. Who has the upper hand? Previous surveys indicated that pro-lifers held the advantage here. Planned Parenthood's survey doesn't refute that. It shows that if you're pro-choice, you can win the debate. But it will be hard. If voters simply choose between 20 and 24 weeks, you'll lose. You have to focus the debate on extreme scenarios. You have to hope pro-lifers don't amend their bans to exempt those scenarios. And you have to deal with the fact that in the window between 20 and 22 weeks, abortion really is unrestricted. You have to demonstrate that people don't abuse that freedom of choice - that they have particularly good reasons for terminating pregnancies they've carried that far. Otherwise, on the whole, the public seems willing to step in.
Saletan (@saletan) covers science, technology and politics for Slate.