A Dmitri Rebuttal by Messaging Expert Anat Shenker-Osorio

Political messaging expert Anat Shenker-Osorio breaks down the art of reframing the debate for progressives to win.

People in line to vote in Bartow County, Nov. 26, 2022 in Cartersville, GA.
People in line to vote in Bartow County, Nov. 26, 2022 in Cartersville, GA. Photo: Elijah Nouvelage for The Washington Post via Getty Images

Why is “fund the policE” a losing message for Democrats? That’s the question and type of messaging Deconstructed is exploring this week with Anat Shenker-Osorio, the founder of ASO Communications and host of the podcast “Words to Win By.” Grim and Shenker-Osorio discuss why certain messages do well and others fail, why reframing the debate is essential, and how to craft messages that resonate and win. They also revisit Grim’s recent conversation with political strategist Dmitri Mehlhorn and unpack why messaging that gravitates toward the right and center is not effective.

[Deconstructed intro theme music.]

Ryan Grim: So, a couple of weeks ago we had onto the program Dmitri Mehlhorn, who is a kind of tech executive and a Democratic operative who has become one of the more influential, but little known, figures in Democratic politics. And he has strong ideas about how Democrats ought to campaign and how they ought to message, and we got a lot of feedback from our interview with him which, if you haven’t listened to yet, you should go back and listen to that one first before you listen to this one.

And one of the people that he namechecked the most throughout that interview was Anat Shenker-Osorio, who’s going to join us today?

Anat, thank you for agreeing to come on. Appreciate it.

Anat Shenker-Osorio: Thanks for having me.

RG: Were you surprised at all that he brought you up so many times throughout that conversation?

AS: Yeah. I think that I was surprised at the number of times. I also think I was surprised that he did so in such a complimentary way, despite the fact that we definitely have places where we don’t agree. So, yeah, I was appreciative of that. 

RG: I actually interviewed Dimitri about a year ago for an extended period of time, but the sound quality wasn’t very good, and so we put off doing it again. And just, one thing became another, and it took upwards of, I think, a little over a year for us to reschedule and get back on the horn.

But I went back and looked at the earlier interview – and we can play some of that throughout here – but I wanted to get your reaction, to start, to the way that he frames how he views politics. And then I want to go through the ways that he brought you up, and see if you can work through what some of those disagreements were.

Here he is talking about how he got into politics after Donald Trump won; he was kind of shocked into action. And he’s starting to think through what it was that broke in the Democratic Party. And so, he said:

Dmitri Mehlhorn: These massive disagreements come from framework differences, conceptual differences. And we’re like, okay, something broke here to allow Trump to become president, and what broke is not, we just didn’t get enough money to Guy Cecil, right? Like, something deeper broke. And what are the things that broke? And how can we – and we have a deadline here. If this guy gets a reelect, it’s a very, very big difference between the first term and the second term.

And, by the way, if he comes back, it’s even worse, but put that aside for a second. But like, you look at the history around the world of somebody like that taking power in a quasi-democratic way, and like, if they lose the first time, there’s a little bit of a breathing room where civil society can organize some resistance. And, you know, to some extent that happened already, but that was our focus.

RG: And so, then he talks about going into Virginia. And I – and correct me if I’m wrong – maybe you helped him with some of this? He talks about going into Virginia in 2017.

DM: One of the things we did is, we invested in all 100 Virginia House of Delegates in 2017. It’s a very easy baseline comparison, and Mark Herring ran in both races; he’s the attorney general. So, Mark Herring, 2013, Mark Herring, 2017, absolute numbers, margins. That gives you a baseline for every one of the hundred districts. And then you can test a bunch of interventions in a place where campaign finance laws are limitless, as long as you disclose.

RG: It’s so loose.

DM: Throwing money. And so, just see like, what works, what doesn’t. And the idea’s that seemed most promising, we tried to scale in the hundred closest 2018 battlegrounds. And by “we,” I mean his money, my money, but also a lot of friends, and kind of the network of venture capital in Silicon Valley, New York, etc.

RG: And so, how did you first get to know Dmitri Mehlhorn? And what was this iterative process that they went through to try to figure out what worked in these primaries? 

AS: Let me take the first part first. So, I became acquainted with Dmitri because I work frequently with a crew of folks at a really, really brilliant set of strategists and content creators that work under the Shingle frameshift. And, basically, what we do together is, we take an issue – could be electoral, could be a kind of standard progressive organizing issue – and we create a whole bunch of pithy 30-second digital ads, and then we test them through a randomized controlled trial. Dmitri has been really, really supportive of that project over the years, and that’s how I came to know him.

Our more specific interaction was around a project that I was helping lead in 2020 across battleground states in the Midwest, where we were taking these really high-performing ads that were using a framework that I helped co-create, alongside Ian Haney López, who famously wrote the book Dog Whistle Politics, and Heather McGhee, who at the time was running a thinktank called Demos. The name of that framework is the Race Class Narrative, it’s the basis for a lot of the messaging advice that I dispense out and the sort of campaign approach that I tend to take.

So, we had made a bunch of – we shorthand it RCN, Race Class Narrative – ads.

RG: Can you give people a shorthand of what the Race Class Narrative is? I think most listeners of this program probably are familiar with it by now but, for those who aren’t…

AS: So, the Race Class Narrative is essentially set up to – I would hope finally, although there is no finally, it seems, in politics, we seem to be having the exact same conversation over and over and over again. At least that’s what it feels, it’s very Sisyphean. But, to finally settle this perennial question that I think Dmitri asks as well, which can be summed up as, “Are we doing turnout? Are we doing persuasion?” Which is, in and of itself, ironically, racially coded speech for, “Are we going to put our eggs in the flipping-white-people basket, or are we going to put our eggs in the turning out and mobilizing our base?” Which is overrepresented, as you know, by people of color, by young people, by single women, by queer folks, etc.

And my take on this argument is, and has always been, that turnout is persuasion. If your words don’t spread, they don’t work. And if your base is not engaged and repeating your message, then it turns out the middle is never going to hear it, because it is a very, very noisy world.

So, what the Race Class Narrative does is, it refuses to fall either on the side of colorblind economic populism, which is some people’s solution. Basically, don’t talk about race, don’t talk about gender, don’t talk about so-called cultural issues. And it also doesn’t seed the class debate.

So, a Race Class Narrative message begins with a shared value that expressly names – or, if you’re looking at a visual ad, depicts – race. It then calls out the other side and their villain in a very specific way. It essentially exposes the dog whistle, talks about how they are deliberately dividing us, or deliberately shaming and blaming Black people, new immigrants, trans folks, take your pick; whoever they’ve got at the top of their punch list that day or that hour. And then it expressly calls for cross-racial solidarity.

I can give you an example of an RCN message.

RG: Yeah, that’d be great.

AS: So, do you want to pick a topic, or do you want me to?

RG: Let’s say, Republicans are pushing to add work requirements to food stamps in order to reach a debt ceiling deal. 

AS: What a far-fetched idea. Where could you possibly have taken that example from? Great.

RG: Ripped from the headlines.

AS: So, I will produce a message. It will need to be copyedited because, as you can verify, I’m making this up from scratch. So, it would sound like this: no matter what we look like or where we come from, most of us all know all too well what it’s like to see a loved one struggle to make ends meet.

Most of us believe that, in America, all of us should be able to put food on the table and be home in time to eat it. But, today, MAGA Republicans want to take food away from our families, forcing us to prove that we deserve the bread that we require. They hope that by getting us to point our finger in the wrong direction, we won’t notice that they’re handing kickbacks to their wealthy donors from the wealth that our work creates. By joining together and demanding that we pay our bills, we can ensure that all of us have what our loved ones need.

That, more or less, is the three-part structure. So, the opening value where we state and name, explicitly, race as the access of division in which they’re trafficking. That we all want this kind of shared thing. Second sentence: but the villains are doing their villainy, and that villainy isn’t just, they’re screwing us over in terms of money. It’s: they’re hoping that if we will blame people who are struggling to make ends meet, we’ll look the other way while they essentially fleece us all. That’s kind of the crux of it. That’s what I mean by narrating the dog whistle.

And then we seal the deal by making the call to action or demand that we’ve got …

RG: Right. And so, you were running a project around that in 2020? 

AS: Yeah. So, we had run a bunch of ads, and we had some early data, not just from in-channel testing, which is basically what we call anytime you’re doing a survey or a randomized control trial and the subjects that you are studying know that they’re being studied. Which is, by definition, an artificial environment that is necessary and helpful. But we had actually run a field experiment, so we were able to see how these ads were performing out in the world in terms of increasing early mail-in voting.

So, we had started that. We had a program going in 2020 in Wisconsin and Minnesota, and Dmitri saw that early data from the field experiment, got very excited about it, and very, very generously – and to his credit, I really want to say – very quickly, without a lot of the usual funder hemming and hawing and deliberation, resourced recreating that same program in Pennsylvania.

RG: He does seem like the kind of character that, if he believed in something, would just cut a check and be done with it. 

AS: Yeah. And that is refreshing.

I think that the rest of your question around my awareness of this kind of experimental thing, I don’t know the details of it, but it’s very much in alignment, I think, with the standard – and I don’t say this pejoratively – but the standard Silicon Valley ideate. You know, they have all of these catchphrases, like “fail fast to learn fast,” which does lend itself to this kind of quicker action, and this less deliberative kind of, “I will have deep thoughts for an entire year or longer about what might work. Oh, by the way, it’s already over, and nothing could work, because the election passed.”

RG: So, the first time that you came up in the conversation with Dmitri was around, defund the police. We can play that clip here:

DM: On crime, what are they going to say? And what they’re going to say is that there is a currently sitting member of the Democratic Party in the United States Congress who openly and expressly advocates for the end of funding to police forces. And there are quite a number of other Democrats who are in power now in administration positions and in Congress who don’t agree with that extreme position of, there should be no funding of the police. But, when asked if they believe in defund the police will give a complicated answer other than, “No, we should fund the police.”

Now, my friend Anat Shenker-Osorio, who’s another person that we have funded in the past, will tell you that if we say the words “fund the police,” that’s bad, because it’s increasing the salience of an issue. And if you run a poll, real time, and you ask people, “Hey, nothing else is going on in the world, it’s more than a year away from the next election, I’d like you to take a poll,” and you drop in the phrase, “fund the police,” out of nowhere, that will probably raise the salience of crime in that survey. And you will read the poll and you’ll say, “Ah, saying the words ‘fund the police,’ or attacking the defund the police movement from the center-left, attacking the left, those things actually reduce the poll results that you get in this survey a year and a half away from election day.”

RG: What would be your response? Is he accurately kind of capturing your take, that you shouldn’t say “fund the police?”

AS: He’s definitely capturing that part. I think that there’s a lot going on in that clip, so, can I kind of take it apart?

RG: Absolutely. Yeah.

AS: Okay. So, first, on the saliency point, one of the things that happens in research and, again, this is one of the perils of in-channel testing, and infield testing is much, much more expensive, and it requires doing those things after the fact, so you do have to test things. In survey and in RCT I do it as well, so I’m not saying I don’t, it’s just, you have to recognize what the limitations of it are.

If you have a survey, and you’re, basically, out of a clear blue sky, saying to people, “Hey, what do you think about defunding the police? Or, what do you think about funding the police?” You’re already introducing that idea to them in order to register their approval or disapproval of that idea, where they may or may not necessarily have that top of mind. To whit, I have been part of a consortium called The Research Collaborative, where we have been conducting qualitative research. So, not quant, but focus groups, two to four focus groups a week, every single week, since October of 2000. So, when you add all those people together, that is a pretty large number of people.

And we frequently ask them, is anyone from base voters of various races, ages, configurations, genders, etc., and then, also, swing voters, same thing. Different kinds of configurations, over all of these years and groups. When you ask them, “What is your beef with Democrats?” I can tell you that none of them – and I really do mean none – the first idea that comes into their own head is, they want to defund the police.

Their beef with Democrats, I mean, I would allow you to guess, I’m sure that you can guess. That they’re slow, that they’re bad on the economy, that they spend too much money, a whole litany of things. And, in fact, one of the fun exercises we often engage in is, we ask them, “If you had to liken Democrats to an animal, what animal would you pick?” We do the same thing for Republicans. And, generally speaking, the animals that they give us for the Democrats is a snail, a sloth, a turtle, some permutation of a thing that doesn’t do much, and does it slowly.

RG: Ouch.

AS: So, people don’t volunteer that answer about Democrats. Now, where does this feeling and idea — and here, I think that that Dmitri rightly made this point. The thing that people think about Democrats doesn’t come out of what Democrats say. I frequently say, that would be a wonderful world, and I’d be on vacation in it, but it’s not our world.

The thing that people think about Democrats is made out of all sorts of impressionistic things out in the world. A combination of the media, but also what is said about Democrats.

So, this saliency question, that is what that point is. It’s sort of difficult in a survey to actually capture, how do people feel about this thing because, in order to capture it, you have to introduce the thing. Does that make sense?

RG: Yes. I guess his argument is, the thing is out there. Like, it’s a big thing, Republicans are talking about it, so you don’t have a choice.

AS: Yes, of course. When the thing is out there, you’re going to have to have a response to it. And so, I want to say two different things.

The first is, if you want to win the debate, you have to set the terms of the debate. As long as you are responding to their terms of debate, you are already losing. And so, the trouble with a defund the police / fund the police kind of back and forth is that, regardless of which one of those you are tossing out there, by definition, you’re making people think about the police, right? I mean, that is most of what either of those phrases are made out of.

When we think about what is going on with people’s feelings around crime, or quote-unquote, “law and order,” or policing, yes, they are concerned about those things. Absolutely. But the reason that they’re concerned about those things, the actual underlying psychological motivation and feeling, is a desire for safety.

What people actually want in their life is to feel safe, and they have been taught – and we see it, it’s very evident, it would be silly to deny it – they have been taught to make a connection between safety and police. That exists and is real. But the hunger and the desire is not for police, it is for safety.

And so, there is a different way to approach this message. It is not a sort of A or b, defund or fund. It is understanding what voters actually want, and coming up with a message – which I’m happy to share with you – that actually talks about that and delivers it.

Because if you attempt to say, “no, I want to fund the police,” then basically what you’ve done is you’ve said to them, “Okay, the way to think about this safety issue, the way to think about this thing that you desire, is how much police is there going to be?” And regardless of what you support or espouse or say, they are always going to pick robocop over mall security, if you tell them the way to think about this election is who is going to be toughest, because that is the Republican brand strength.

You’ve basically said, we agree to have your debate. We’re going to play soccer, we’re going to pick the side of the field where we’re staring into the sun and try to make goals that way. It doesn’t work.

RG: Before we get to your better framing of how to respond, this feels like a very 2023 conversation that is not taking into account 2020. In other words, what do we do with the fact that there were real material conditions that were producing energy around the idea of defund the police? Like, the people who projected that really did want to talk about the police. They wanted to talk about police violence in a way that would finally land, and would finally send home the message that enough is enough, that this needs to stop.

And I think one thing that frustrated people about Dmitri’s approach to politics was that he very much wants to separate politics and political messaging from the real world. What would be a better way to respond to the very real concerns that people have that are producing messages that, by now, we may agree aren’t the most effective in swing districts?

AS: I’m so glad that you asked that, because part of the entire purpose of the Race Class Narrative, RCN, is to recognize – and this is something that I say frequently – that the job of a good message is not to say what’s popular, the job of a good message is to make popular what we need said, and that is the north star and the fixed point. Otherwise, what is the purpose of politics if it is not the kind of ritualized practice of trying to improve people’s lives? Which I’ve been naive enough to believe is, in fact, the purpose of politics. It’s certainly my purpose.

And so, I would agree with you, or with the listeners that you’re citing that brought this up, that we need to be able to talk about the things that are destroying people’s lives and, quite literally, killing them. And so, the way that we do that – and I will share the message in one second – is to also recognize that we need to be playing chess, not checkers.

And what I mean by that analogy is that, a defund the police message coming from activists is actually really critical and vital, because it provides us a left flank. It provides an opportunity for a democratic politician who is running for office to be able to say something that is also about accountability from the police, without literally saying “defund,” but still be understood.

So, the example that I would cite of that is, for example, in Minnesota, which is one of the states in which I work frequently. The rest of the states I work are more kind of classic-classic top-of-the-mind battlegrounds; Minnesota, a bit more purple, getting bluer every day. We ran a campaign in 2020, and then we used this messaging as well in the very, very difficult reelection campaign for Keith Ellison – who’s the Attorney General and, you know, a Black Muslim man – that said “fund our lives.”

Using the phrase “fund our lives,” because of the existence of the activist discourse on defund the police, that sort of rang to people. It’s kind of a call to defund the police messaging but, instead of making a negative demand, which is what defund the police is, we are making an affirmative demand. We’re talking about what it is we want to fund.

So, my first answer is that there’s a difference between designing messages for people who are running for office and the entire messaging terrain. And this is something that the right has understood a very long time ago. When Todd Akin famously talked about how it isn’t possible for a person to get pregnant from rape because if it’s a, quote, “legitimate” rape, the body has a way of shutting these things down. I don’t know if you remember that.

RG: Oh, yes.

AS: That egregious comment got him in trouble, for sure, but it also did what they always do, which is that they keep going to the right, keep going to the right, keep going to the right. So then, the other Republicans who quote “only oppose abortion in all cases, including rape or incest,” but at least will admit that rape is a thing that exists in the world, suddenly, they look progressive, because they just keep shifting things to the right.

Whereas we, on the left, are like, “Oh no, don’t say that. Oh no, don’t say that.” When a more strategic approach would be, “My esteemed dear friends here are understandably incredibly angry and upset around these horrible acts of violence by the people who have sworn an oath to protect and serve us. That’s why I’ve put forward this proposal for police accountability.” It lets you kind of bank-shot off of that, if that makes sense.

RG: Yes. And so, later on in his defund the police argument, he said this:

DM: In particular, defund the police as a national call and how that worked nationally, it’s hard to know how that played down in New York, etc. What I would say is that candidates that took an aggressive fund the police position in 2020 tended to win their races. Like Vicente Gonzalez, Abby Spanberger, etc.

Pretty much every one of those frontline members of Congress said, “I was getting hit at every town hall with defund the police. I was getting slammed with defund the police all over the place. I was only able to eke out a victory by aggressively articulating a fund the police position.” And that’s what they say. And it could be wrong. It doesn’t show up in Anat’s polls, but that’s what they say.

RG: Let’s say he is accurately articulating what those members of Congress believe — and I do think, whether they believe it or not, they do certainly say that a lot — what would you say back to them?

AS: First of all, I would say that we just saw in the ’22 cycle a whole lot of what I will shorthand as the “police love me” ads, right?

This is a kind of direct-to-camera sheriff or member of law enforcement who is speaking out on behalf of the candidate, and talking about how they support X-Y-Z candidate, because X-Y-Z candidate does indeed want to fund the police, and does indeed support them.

This ad was very popular. We saw many, many, many people do it. We saw Mandela Barnes in Wisconsin do it, we saw, of course, Tim Ryan in Ohio; we would expect that. We saw Mark Kelly in Arizona, we saw Stacey Abrams run it this time.

And what we see, when we actually look at how those ads test again in RCT, is that, compared with other ads that would say about taking our freedoms, which is the framework that we very much pushed and promoted. I.e., if you want to win the election, you’ve got to decide what the election’s going to be about, as much as you can, as often as you can.

Those “police love me” ads — that’s my shorthand for them — they didn’t do particularly well. And the reason for that is because they’re not that exciting to the base. It’s also not true that they’re completely anathema to the base, it’s a lie to think that they are. They’re anathema to activists, but the base is not quite where I would love them to be, and not quite where I think activists frequently think that they are. This is the hindrance of actually conducting public opinion research, right? You’re subjected to public opinion. It’s not always fun.

But they’re not motivating to the base. And, in fact, there is evidence that they are demobilizing. And, to swing voters, they just kind of feel, to use the technical terminology, like bullshit. And the reason that they feel like bullshit is because, how are you going to undo in one 30-second ad, even if you spend a lot of your TV money on it, the rest of what is being said about you? 

And, you know, clearly, Dmitri’s argument is, that’s why you’ve got to say “fund, fund, fund, fund, fund.” We don’t see it in the data. And the same holds for a giant study that was done by Justice Research Group through many, many, many iterations of testing lots of messages and huge samples. What they actually found, and I can actually read it to you if you want, is that the fund the police message did not perform the best. Partly because, like I said, it tips off swing voters’ bullshit meter, it just doesn’t feel credible. And it potentially has a demobilizing effect on the part of the base that is most likely to kind of drop off.

So, the message that did the best, and this was across demographics, specifically with battleground state voters. And I’m not saying that this message is exactly what I would say, but it goes: “Democrats say you should support candidate” – so, the name of the candidate – because, name of candidate, believes in investing in crime prevention and community police officers who will walk the beat, who will know the neighborhood and restore trust and safety. Let’s not abandon our streets or choose between safety and equal justice. Let’s come together to protect our communities, restore trust, and hold law enforcement accountable. We should all agree the answer is not to defund the police, it’s to fund the police with the resources and training they need to protect our communities.”

So, there is the fund the police sort of call out at the end. In subsequent retesting of that, it was found that that framework of police accountability works, even if you don’t use that line.

A second message that did the best with our low propensity – i.e., the voters that are our greatest flight risk – is the, we know what keeps us safe message, which is what I was alluding to before, sounds like this: “Democrats say you should support name of candidate because name of candidate knows what keeps us safe. It’s living in communities where we look out for our neighbors, where we have the great schools, necessary services, and affordable healthcare we need to get and stay well, overcome challenges, and care for our family Candidate believes that no matter our zip code, background, or color, we all deserve to live in a place where we are safe, and will deliver proven approaches to safety that the majority of us favor.”

So, basically a message that works, a landing spot, is this idea of, again, the shared value, that we know what keeps us safe, and a call out to what that requires, including having the people who are sworn to serve and protect us act in our interests, work to prevent crimes, and treat us all as equals. That is across, you know, dozens and dozens and dozens of these experiments what we actually find to work, when you factor in both turnout and swing voters.

And a lot of these studies – and I don’t know the details of what Dmitri is citing — but there’s a tendency to overlook this mobilization question, which is justified by the fact that it’s really hard to measure mobilization. People lie about it, right? We ask them, does this make you want to vote? And they say, yes. It’s like, the, “I will pay you tomorrow” of public opinion research. People are always saying yes, and it’s not true.

But the fact that it’s hard to measure mobilization doesn’t justify that you don’t even look at it and don’t try. Because, otherwise, how are you dealing with the fact that a lot of our problems in a lot of states is our side just not turning out?

[Deconstructed mid-show theme music.]

RG: One of the other controversial things that he said during the interview was around Roe v. Wade, and he said it would be a mistake to run on codifying Roe v. Wade. And that came within his broader claim that you really shouldn’t run on anything, that you can’t inspire people to come out to vote by saying that you’re going to do good things, or pointing to the good things that you’ve done. That the only way to actually motivate people to come out is scaring them about bad things that are going to happen, or things that you have that are going to be taken away from you.

I would love to get out of politics if that is actually true. But, just a clear-eyed look at the data and the public opinion research, what’s your assessment of that claim?

AS: I mean, there are so many ways into that question. One rejoinder I would offer is that Martin Luther King didn’t get famous for saying, “I have a complaint,” nor “I have a multi-bulleted list of policy proposals,” right? He presented a dream in, arguably, some of the most harrowing, horrifying, terrible conditions.

We can look at example after example after example, and what we need to remember is that, in any election, there are actually three people running. There is your person, their person, and stay at home. And stay at home literally has the home team advantage, because, guess what? People are already at home.

So, our task and our job isn’t merely to get people to vote against their person, it’s to get them to vote in the first place. And it turns out, I think, if you know anything about US politics, that that’s a really challenging thing to do, and it’s a thing that is being made deliberately more challenging by these MAGA Republicans who are absolutely determined to pick their voters. Because they know that voters, broadly speaking, are getting less and less and less likely to pick them. Our side, staying at home is pretty much the anchor that weighs us down, and we can see that in election after election after election.

For example, in 2022, while pundits widely reported that, you know, it wasn’t turnout, it actually wasn’t turnout, that is just because they do not know how to look under the hood. In point of fact, in the 15 states that we won, turnout equaled 2018, which was historic. And the fact that it equaled 2018 is even more historic, because that was turnout in an incumbency. In the 35 states where we did not do that hot, turnout was as predicted, and our side stayed home.

And so, in point of fact, we see over and over and over again – we see it in ‘22, we see it in ‘18, we see it in ’20 – that, in order to win, we’ve got to mobilize our side. So, this question of, we can just say the other side are a bunch of bastards or whatever messaging you’re going to use, there’s a piece of that that I absolutely agree with, and that is that we do need a clear, coherent articulation of the villains. 

But what we have found in our research, and it’s the reason why we came to this Protect Our Freedoms framework in 2022, that we pushed out through ads, and through scripts, and through canvas, and texting and so on, and we got a fair number of candidates to actually adopt, in battleground states like Pennsylvania, to some extent Wisconsin, definitely Michigan, Arizona, etc. What we found is that drawing a contrast between an antagonist, which are MAGA Republicans that want to take away your freedoms, which is a phrase I use very deliberately; “take away” activates loss aversion. Freedoms is an incredibly important and central and resonant concept. It is a core value for Americans across demographics and ideologies.

And voters – notice I say voters as the protagonists of the story, not Democrats, so it is, antagonist, MAGA Republicans, protagonist, voters – voters who need to turn out to protect our freedoms. That was essentially the story that we found to be most effective.

But, in that, you do have to give people a reason to show up because, otherwise, they will simply just stay at home.

RG: Yeah, and his argument was that, instead of running on codifying Roe, you should actually run on a promise that you will make sure that children who are the victims of rape have access to abortion. Which, first of all, is practically difficult, because if all other abortion services are banned, you’re going to have abortion clinics across the state shut down. So, it would be legal in fact, but actually not available in practice. Like, there literally wouldn’t be anywhere to go and get a procedure.

But putting that aside, what was your reaction to the idea that you should pick something that crazy that the other side is against, and kind of drive that as the wedge? Rather than something broader like, we’re going to restore this right that was taken from you?

AS: Generally speaking, I like to tell people, don’t take your policy out in public, it’s unseemly. This is a more kind of generalized messaging lesson. The policy is not the message. So, I am not a big fan of messages that explicitly name any kind of policy, because what you want to talk about is not the name of the policy. You can do that in sort of boring back rooms, that’s totally fine. But, as far as what you’re actually selling to voters, you want to sell the brownie, not the recipe. Meaning, you want to sell what the policy is going to deliver.

So, the way that I would talk about it is, Democrats, or voters, need to turn out to: protect our freedoms; our freedom to decide whether and when we have children; freedom for our kids to learn the truth of our past; freedom to earn a good living and be able to retire in dignity. I’m giving you lots of different options. Freedom to be ourselves, love who we love, and create the families that we desire.

It’s talking about what the outcome is going to be. Going into policy details, first of all, people just absolutely glaze over. You would be astounded at the number of people who actually are still unaware of what Roe v. Wade, let alone Dobbs, actually is, even as they do know that something massive happened on abortion. 

Getting into the details of protecting the ability of a child to be able to have an abortion unleashes what people in the reproductive rights and justice community rightly call the “good abortion, bad abortion dynamic.” Where some abortions are just fine because it’s circumstances beyond your control, or it’s kind of these most horrific cases. And, aside from the practicality, which you very, very rightly lifted up and, in fact, we live in a country, even pre-Dobbs, where you have the purported right to an abortion but, in many, many places it’s inaccessible. And, thanks to the Hyde Amendment, you couldn’t pay for it anyway unless you have money from somewhere.

Basically, what you’re doing is, you’re falling into this slippery slope of conceding that abortion is actually a pretty terrible thing, but we need it in these super extreme cases, in order to avoid this much more terrible thing. I mean, I cannot tell you enough how much, again, we’re talking about two to four focus groups a week, every single week, for the last, you know, it feels like 7,000 years of my life, but in fact has only been since late 2020. I mean, we did focus groups before, but not in this kind of deliberate, all the time, over and over kind of way.

I cannot tell you the number of times that in a group of, for example, white swing men in the outskirts of Green Bay, Wisconsin, who have spent 45 minutes trashing Democrats and talking about how terrible and horrible they are, once we say the word “abortion,” a switch flips, and they say, “You know, at the end of the day, I’m not going to have anyone tell me and my wife what’s going to happen to our family and the decisions that we –” I mean, it really is like a lightning bolt, even for men.

Or another group that I remember really, really vividly in Texas, again, swing voters, trash talking Democrats, all the crap, all the crap, all the crap. And then suddenly saying, when we just introduced the topic of abortion, saying, “Money goes up, money goes down. I can learn to live in a budget. What I cannot learn to do is live without my freedoms.” Direct quote. Over and over again.

And so, what this abortion thing has done is it has made real, it has catalyzed the recognition that MAGA really is coming for your freedoms, and this is the first one that they’ve come for, and they’re coming for the rest. And it’s the creation of that loss aversion, and this sort of repugnancy to what feels corporeal – meaning in your body, your lived experience – that has mobilized, and activated, and turned people out in what should have been by historical precedent and punditry a red wave blood bath against us, and wasn’t.

RG: I noticed, though, the beginning of your answer, when you were talking about the messaging around the issue, you didn’t actually use the word abortion. “Choice of when to have children, nobody tells my family what to do.” I’m curious if that was deliberate, or am I picking up on something that wasn’t really there?

In other words, is there some consensus among messaging people that using the word abortion is not ideal? I would think, maybe, yes, it is. But then, on the other hand, you were saying that in the focus group when abortion came up, it flipped a switch for people.

So, if it is kind of a stigmatized word in messaging, why?

AS: So, there has been a historical feeling among mainstream Democrats that it’s not a good word to use. And, in fact, it sort of used to be widely referred to as, quote, “the A word.” “Don’t say the A word, the A word isn’t helpful.”

And even up until, I’m sure you remember, when the Dobbs decision came down, there were sort of two weeks of kind of weird, uncomfortable silence from the White House. Nancy Pelosi quite famously read a poem. It was sort of like WTF? Are you going to respond, are you going to react? And there really wasn’t a lot of response and reaction, and there really was this feeling.

And it was said over and over, and folks in the repro movement, they would tear their hair out — I would argue, rightly so – that people were like, “Oh, don’t campaign on abortion, don’t campaign on abortion. It’s not a winning issue, it’s not a winning issue.” And folks that had been at this, and are at this for decades, have been like, “You’re wrong, you’re wrong, it is a winning issue.” And, again, as I keep saying, if you want to win the debate, you have to set the terms of debate.

Lo and behold, now in this post Dobbsian World, most people are suddenly awakened to the fact that, oh, wow, it turns out that in an arena in which one of our biggest problems — and this is really the tragedy of watching this many focus groups. The idea of both sides is really, really thoroughly entrenched among voters. And this idea that, you know, yes, the Republicans are bad, and this action was bad and that action was bad, and so on. But that’s politics as usual, and politicians are bad, and they’re corrupt, and that’s just how it is.

The vilification of government project by the right has, I would say, been very successful, and it makes it much harder for people to sort of single out and say, “Oh no, but that’s like a new kind of bad,” right? That’s not just like, politicians-are-bad kind of bad. Dobbs and January 6th are two crystallized moments that really do that.

And so, to your question, which I did not forget, it’s not on my part a deliberate refusal to talk about abortion. We definitely talk about abortion – do not call it the A word, call it abortion. But the framework that I was employing, not when I was directly quoting focus group participants, which is, you know, them using their own words, right?

RG: Right.

AS: The freedoms framework is really important, and so, the reason why I use the phrase, “our freedom to decide whether and when we have kids,” is because that is a way to, if you’ll let me invent a word, “freedomize” the abortion question, so that we can link it to the book bans, so we can link it to the economic assaults, so that we can have a narrative that kind of everything fits into. 

RG: We talked on this podcast to a woman who was on the ground in Kansas working on that referendum, and listeners to this show will remember that she forecasted a landslide victory, and she was correct. And she talked about how resistance and skepticism of the vaccine mandates in Kansas was very successfully, actually, translated into messaging to defend abortion rights in that contest. Which I found surprising at the time, but it did seem to work, that there were a lot of people who didn’t like vaccine mandates, and then they came back to them with, well, if you don’t like that, if you believe in that freedom, then why shouldn’t a woman have the freedom to choose whatever healthcare decision she wants? And they’re like, “You know what? Fine. I guess I’m for abortion rights.”

Did you follow the Kansas fight? Do you have a sense on what enabled them to come from behind and win in a blowout there?

AS: Yeah. I did a very, very early conversation with the folks who ended up leading that campaign, and at that time I just talked to them about this freedom idea, because I’ve been on this freedom train for a very long time. Because we find it to be really, really effective, not just in, quote, “social issues,” but also in economic issues.

I think that they did an extraordinarily good job of – as we would say in Spanish, no tener pelos en la lengua – not biting their tongue, but rather making it really, really clear that what’s on the line, what’s at stake, and framing it around this concept of freedom. The anecdote or the information that you shared about connecting back to this kind of visceral idea that people had just had about feeling of intervention or forced coercion around something else that has to do with the body is really interesting. I wasn’t aware of that.

But the other framework on abortion – and this doesn’t apply in Kansas, but very, very much applied in Michigan – that we’ve seen be really effective, and this actually emerges out of work that I was fortunate enough to be able to do in Ireland, when we repealed the law against abortion there in 2018. And, to a certain degree, also informed by that same kind of victory in Argentina, which happened at the end of 2020, is a framework that I call, “someone you love.” And, basically, what it says is, someone you love may struggle with a pregnancy someday, someone you love may need an abortion. And what will they do then?

It’s basically taking the storytelling approach that is so central and vital, and has been the work of Black, women-led reproductive justice movements across this country. And taking that single-person-narrated story of abortion, and making it not just about past examples when, you know, the narrator has had this experience, or needed this health service, but also getting people to reflect on the idea of the future, and what it is going to mean when someone in their orbit – whether it be themselves personally or, you know, a sister, a coworker, a friend, a neighbor, whomever, which also applies to men – is going to need an abortion. And then what are they going to do?

And that someone-you-love approach is something that Gretchen Whitmer used a lot.

RG: Anything else that jumped out at you in the interview with Dmitri that I didn’t get to, that you wanted to hit?

AS: Yeah. What I want to say – and, in a way, this is partially a reaffirmation of some of what he said, with which I agree and, in some ways, perhaps, a difference that we share between us – is that, first of all, politics isn’t solitaire. And so, we need to recognize that people don’t just hear from us, they hear from the other side. And our message needs to act not just as a motivation, but also as a rebuttal to what the other side is saying.

And I think that he would probably agree with that, I just think that the approach of doing that by presenting ourselves as essentially the B-minus version of our opposition, right? They say they want to secure the borders, we say we want to secure the borders. They say they want to fund the police, we say we want to fund the police. That’s not actually a rejoinder. That’s not actually a, “Hey, you don’t want Pepsi, you want Coke.” It’s basically saying, “Hey, no, Pepsi is good, and we are also Pepsi.” That’s very confusing for people, it’s demobilizing for our base, and it doesn’t draw that contrast that we actually require with swing voters. And so, I just want to really hit that point hard.

The other point that I really want to hit hard is, it’s really challenging to break a signal through the noise. Getting a message out into the world, let alone getting it repeated so it can actually be heard, is one of the toughest things that we have. Even if we land on some kind of perfect message that we have focus-grouped, and RCT’d, and surveyed, and we are absolutely certain that this 100-word paragraph is the greatest 100 words that have ever been compiled together. If the base won’t repeat that, if they won’t wear the equivalent of the red MAGA hat, that means that the middle isn’t going to hear it.

Because, even if they see your one ad that is perfectly crafted that one time, they are also seeing eight billion other ads, and seeing flyers, and hearing other things. And so, if you don’t attend to what your base actually believes and is willing to repeat, you can’t persuade the middle. If your words don’t spread, they don’t work. That’s the underappreciated piece of this.

RG: Reed Hoffman doesn’t have enough money to get those paid ads in front of enough people for it to matter. You have to get people to do the work for you. And, to do the work, they’ve got to be a little bit motivated. Is that a good way of putting it?

AS: Yeah. I mean, all of the money in the world does not actually create saturation. And things that people see in a TV ad are far less convincing to them than social proof.

Social proof is what I sometimes call the middle school theory of messaging. It’s the fact that people believe the thing they think people like them believe. And so, for example, when it was socially sanctioned, and even sort of widely okay to think, you know, gay men should not be able to get married, right? This was not that long ago. Or lesbian women should not be able to get married. It’s anathema.

RG: Or marijuana should be illegal.

AS: Yeah, marijuana should be illegal. You know, there’s all sorts of social attitudes that have changed really, really rapidly that we’ve witnessed. And part of that change is this kind of broad idea — let’s take marriage equality — of pivoting away from, “this is a really contentious issue and people feel conflicted about it,” to essentially messaging from inevitability, and saying “love is love.” Instead of saying, you know, “This is our right, and we should have this right, and this is about being equal, and this is about –” The kind of taking your policy out in public. And, instead, claiming the moral high ground, and having a message that a person in line at the grocery store might actually repeat to someone else.

Unless you’re there, you’re not getting saturation. If you’re not getting repetition, saturation, you are not getting social proof, and you cannot move the needle. 

RG: Well, Anat, thank you so much for joining us. Really, really appreciate it. 

AS: Thank you.

[Deconstructed end-show theme music.]

RG: That was Anat Shenker-Osorio, and that’s our show. Anat is the host of the Words to Win By podcast and a Principal of ASO Communications, where she examines why certain messages falter where others deliver.

Deconstructed is a production of The Intercept. Our producer is José Olivares. Our supervising producer is Laura Flynn. The show is mixed by William Stanton. This episode was transcribed by Leonardo Fairman. Our theme music was composed by Bart Warshaw. Roger Hodge is The Intercept’s editor-in-chief, and I’m Ryan Grim, DC Bureau Chief of The Intercept.

If you’d like to support our work, go to theintercept.com/give. If you haven’t already, please subscribe to the show so you can hear it every week, and please go and leave us a rating or review, it helps people find the show.

If you want to give us additional feedback, email us at podcasts@theintercept.com or at Ryan.Grim@theintercept.com.

Thanks so much, and I’ll see you soon.

Join The Conversation