Is "Harm" Framing Harmful?

Its implicit perfectionism can induce passivity rather than progress

I mentioned on Sunday that I was going to return to this messaging debate, and even though that debate feels like it is mostly over, I’m going to do that now, because I want to focus on an aspect that hasn’t been highlighted.

Most commenters on the below messaging sheet have highlighted one of two things: either the fact that it rejects “choice” language that has been central to the abortion rights cause for decades, or the fact that it moves the language notably to the left, abandoning the stated goal of making abortion “rare” most notably.

Sarah Ferris @sarahnferris

NEW: The Pro Choice Caucus has just sent out messaging materials to House Dems on Roe draft. One of the recommendations: Don't use "choice."

That’s not what I want to talk about, for two reasons. First of all, I’m not a marketing guy, and this language ultimately ought to be marketing language. If “safe, legal and rare” and “pro-choice” are helpful at moving voters in your direction, it strikes me as silly to abandon them—but that’s a marketing decision. From where I sit as a non-marketing guy, the proposed changes don’t seem crazy. The “and rare” part of “safe, legal and rare” always struck me as disingenuous; I prefer the honesty of “safe, legal and accessible.” “Choice,” meanwhile, always struck me as a trivializing way to discuss a matter as emotionally and morally fraught as abortion. “Decision” really is just a morally more serious word, regardless of its relative marketing value. But that’s just my reaction, and these certainly aren’t the kinds of decisions that should be made based on how they strike me.

As for the move to the left, I agree with Matt Yglesias that clever framing is no substitute for a policy agenda that a majority agrees with. But as I’ve articulated repeatedly (see here here and here), I’m not convinced there is a “vital center” to be captured on the abortion question. The intensity is all at the extremes, and I suspect the issue is genuinely low-salience for folks in the middle. If that’s the case, there’s just not much benefit to moderating (and also not as much benefit as some liberals hope to be had from a backlash to the reversal of Roe). We’ll see whether that dynamic changes in the wake of outright abortion bans in states like Texas where Democrats feel perpetually close to breaking through but keep falling short, and if so how.

No, what I want to focus on is something else: the use of the word “harmful” to describe the language being rejected. “Harm” is a word that has been spreading pretty widely of late. It has become the standard word for our culture’s proliferating formal apologies, as well as the standard way to describe why certain language or actions need to be avoided or changed. I think the increasing popularity of the world is unfortunate. Indeed, I think the word is doing a lot of, well, harm, not only to particular causes but to our ability to engage in discussion and debate at all.

To begin with, the word is incredibly loaded. In the messaging sheet above, it’s contrasted with “helpful” but unlike alternative contrast words like “unhelpful” or “counterproductive,” it doesn’t implicitly suggest a positive but misguided intent. On the contrary, it’s implicitly accusatory. As such, it’s almost designed to make anyone on the other side of a debate—someone who, for example, thinks “pro-choice” is a useful framing for getting legislation passed—defensive. Moreover, it collapses the probabilistic nature of cause and effect in the real world. It’s impossible to know precisely how any given language is going to be received by different interlocutors and audiences; we’re all making our best guesses, hopefully based on data, about what will be received well and what might be less effective, or could backfire, or confuse a message that we’re trying instead to clarify. Any of those would arguably be “harmful” to the cause of victory, but it’s the possible effects that are harmful, not the words themselves. Collapsing that chain of cause and effect and attributing “harm” to the words themselves obscures the actual process by which language influences the world, and not incidentally eliminates the space in which debate about that language ought to happen.

That points to the most damaging effect of this word choice, which is second-order: the way it induces passivity. The injunction “first do no harm” isn’t as easily followed as people sometimes imagine, but if you try to follow it what you quickly discover that the best way to avoid doing harm is to do less. Sometimes that’s a good idea: you don’t want to move an accident victim in circumstances where you might thereby exacerbate their injuries. But often it’s a terrible idea: you don’t want to let someone bleed out because the only thing you have for a tourniquet is a shirt that isn’t sterile. “Harm reduction” is an awkward phrase but a much better principle: you want to have reasonable confidence that your actions are helping rather than making things worse, as opposed to near-certainty that your actions will have no negative effects whatsoever.

Something “harmful” is something you simply shouldn’t do. So describing “pro-choice” and “safe, legal and rare” as harmful implies that the cause of abortion rights would be better served if instead of using these phrases, advocates said nothing. Could any advocate possibly believe that? Surely not. What they mean, surely, is that this phrasing is “suboptimal” or has “reached the point of diminishing returns” or something similar. Maybe it was once useful and is now actively counterproductive; maybe it was never the best framing and in retrospect abortion rights advocates should always have opted for a different rhetorical strategy. Framing the language as “harmful” erases all these distinctions and the acceptability of learning from experience in favor of a kind of perfectionism.

And as any perfectionist will readily tell you, it’s a psychological stance that is paralyzing, where you’re so afraid of doing something wrong that you can’t do anything at all. This is why perfectionists have so much trouble getting anything done; the prospect of doing it wrong the first time, having to amend, reverse course and redo is too threatening to their sense of self. I’m prone to that kind of perfectionism myself, and a big part of getting out from under is convincing yourself that it’s ok to get things wrong and try again—that is to say, that a mistake does no lasting harm. But that is precisely what “harm” framing denies. Meanwhile, the simplest way out of the psychological bind of perfectionism is to abdicate one’s agency to someone else. There’s psychological safety in numbers; if you make the same mistake as everyone else, and you didn’t really make a decision at all, then you can’t really be blamed for any “harm” that results. This is a different kind of passivity, and the kind of herd behavior it leads to can be very harmful indeed.

I don’t think the word “harm” has any more magic power than other words. We could get used to saying to ourselves, “well, I did some harm, but on balance I did more good, so it’s ok.” But we’d be swimming against a current that says it’s wrong to do harm. And that current is, fundamentally, a good thing; in most circumstances it’s not ok to knowingly do harm even if it’s for the greater good. I can’t, for example, murder someone so that their organs could save five other people’s lives.

We have a rich language capable of expressing all kinds of fine moral distinctions. When we flatten them in the name of “moral clarity” or “social justice” or some other purportedly higher purpose, we’re only harming ourselves and the causes we claim to believe in.