Book note – The Persuaders, by Anand Giridharadas

by Ingrid Robeyns on December 28, 2022

I recently listened to the new book by Anand Giridharadas, who is well-known for his previous book Winner Takes All. That book was about how (some of) the superrich are happy trying to contribute to some of the world’s problems, but never ask any questions related to why the world is so unequal as it is, what power and the workings of capitalism have to do with all of this, and whether their capitalist strategies are at all suited to address these problems. I thought that was a great book.

So I was looking forward to his new book. It is called The Persuaders. Winning Hearts and Minds in a Divided Age. It is a book about why we shouldn’t just give up on people who have political or social views that we find wrong, perhaps even horrible. The book presents a series of cases, the activists involved, and the techniques or strategies they use – interspersed with some insights from social psychology and other sciences on what works (and what doesn’t) to make people change their mind in a non-manipulative way.

My take-away from the book is that there is no point in believing you are right (or have the right policy, or the right analysis on what needs to happen on matter X), and believing the only thing that is needed for change is airing those views and that analysis. It’s just not enough. We need to actually spend time and effort to persuade others that this is the right analysis/policy/direction, and this persuasion cannot be merely cognitive; it requires understanding “where people are”, what makes them believe what they believe, and showing respect for them as a person at the outset. All of that requires listening, and being willing to engage in a genuine conversation, and finding out why people believe what they believe. Just believing I am right (and having all the arguments sorted out in my head) and airing my views, is not enough to also make a difference in the world, especially not in deeply divided societies. And, very importantly, trying to persuade others, and being willing to be persuaded, should be an essential part of any democracy. Thus, this book is also, at a deeper level, about what contemporary democracies need.

The Persuaders is about US-American activists and US-American politics. It is about how different groups of white activists and activists of colour overcame their differences to organise together the million women’s march; about a feminist expert on rape who decided not to ignore the call by a convicted rapist to help him stop raping others and who sets up a process to change him and his fellow inmates; about the differences in persuasive styles between Alexandria Ocasio-Cortez and Bernie Sanders; about door-to-door direct canvassing to gather support for legal change to give undocumented migrants a pathway to citizenship.

Although the book is very US-American, the strategies and techniques discussed, and the lessons that we can draw from the cases, are also applicable to other countries where some groups in society no longer talk to eachother – which, I guess, is almost everywhere. It also struck me, while listening, how the lessons drawn by the persuaders described in this book could be very useful for scholars and scientists who try to reach out to the public about their findings; it is just not enough to know what the problems are or what solutions are available. Technocratic appraoches have severe limits. Thinking about communication is just as important – and communication is not a one-way activity. Moreover, communication is also not a merely cognitive activity; it requires us to address our emotions, and it requires all parties to the communication being treated as persons that deserve respect (it is possible to respect a person as a person yet deeply disagree with what they think).

One issue that remains a question to me after reading this book, is when to decide that further listening and talking is not only pointless, but genuinely harmful. It can be harmful because the person you are trying to persuade is verbally violent, and you need to put up with that violence during your acts of persuasion, yet to no avail. I tend to be of the view that we should simply not tolerate verbal violence, and that it is not only permissible, but actually sometimes better, to block off communication with verbally violent people. Another form of harm is that the person you are trying to interact with is structurally using manipulative and propaganda techniques. For example, think of political leaders, but also some of their followers, who use proto-fascist techniques (some of which were beautifully explained by Umberto Eco in this essay in The New York Review of Books in 1995). Being open to be persuaded implies being willing to accept, at a meta-level, that there is a distinction between a true claim and a false claim, and that one should not deliberately claim things that one knows to be blatantly false.

Talking to such a person who no longer respect those basic rules of a reasonable and fair conversation might legitimise their discourse, and when done in a public arena might give them more air time to voice their lies, and hence, do more harm. Moreover, when my interlocutor is using verbal aggression, or manipulative or propaganda techniques, I can no longer simply follow the fair conversation-style that I want to use, since their strategy is a dominating one. I am not sure that in this case assuming that they are acting in good faith and that they are open to change their mind, is the right assumption. In the area of science denial (think of Covid or climate change), these are real worries. Does respect for democracy require us to always be willing to persuade and hence talk to others? Perhaps Giridharadas thinks that in the US progressives think too quickly and too often that those on the right are ‘unreasonable’ in the sense described above, and therefore mistakenly not even trying to change their hearts and minds. That might be the case (I do not know). Still, even if that were true, the question remains whether there are cases when it is best to ignore the person and not engage, because it will cause harm.



Peter Dorman 12.28.22 at 7:54 pm

This reminds me of a lesson I learned as a teacher: you need to know what students bring to your class. I had learned this abstractly, but one day it hit me hard. I was grading an exam and came to an answer that just seemed weird. I could imagine why someone might not get the material they were being graded on, but I couldn’t figure out how they could have come to that particular answer. Then it dawned on me: the answer was an amalgam of the course material and the probable prior beliefs they carried around with them. What they wrote was internally contradictory, but it made sense as a mixture of old and new.

I realized from then on that my first step would always have to be listening, finding out what students already thought about the subject, whether they were conscious of it or not. But the next step was to engage that thinking. Of course it should be respectful, but it would have to be a sort of deconstruction to make way for a different understanding.

Political persuasion isn’t teaching, but I agree with Giridhardas that it starts with that listening. What follows really depends on how the process unfolds. Someone who is obstinate in their views, armored against any possibility of revising them, can’t be engaged. That’s true in the classroom as well as in “real life”. But the message of the book, as I understand it, is that we shouldn’t jump from the perception of cognitive difference, however emotionally fraught, to the assumption of immovability. You can only learn where engagement is possible by trying it out, and there has to be that moment of listening at the outset.

I agree with IR that it’s equally important to be aware of the ways dialogue can be corrupted and turned into a resource for exploitation. There really are sociopaths out there.


EB 12.28.22 at 10:47 pm

Starting “where people are at” is one of the iron rules of community organizing. Building a campaign that uses reasoning and rhetoric that make sense to people who haven’t yet been engaged is really important. This is why Defund the Police didn’t get anywhere on the ground in Illinois, despite great distrust of police departments. Black legislators got no calls from their constituents to shrink police budgets; they got many calls to change or abolish specific police behaviors, and did succeed in passing major legislation that accomplishes that.


Moz in Oz 12.29.22 at 4:37 am

Persuasion works but it comes late in the process from what I know. You have to identify the problem and come up with solutions first, then identify the people to persuade. Viz, persuading the pope that Russia should retreat from Ukraine is all very well but it doesn’t help anyone. Persuade Putin and you’ve achieved something very different.

The Teal campaign in Australia might be useful as an example. They didn’t win by persuading the 30-40% who vote Liberal to change their votes, they won by getting a critical mass of “climate change matters” people to vote on that basis, of whom some small fraction were former Liberal voters. Basically inverting the cliche “people care about climate change, but they won’t change their votes because of it”. With preferential voting they typically only needed 20-30% of the vote to win.

In managed democracy countries that sort of change is much, much harder – in the UK gerrymandering and other defects in the voting system mean that third parties find it very difficult to get seats in the lower house, while in the USA they’re effectively banned from state and federal government (the ban is effective even if it’s not explicit in legislation). You can’t fix that by small-scale activism, likely you need another revolution or internal war.

Which takes us back to… first work out what problem you’re trying to solve.


Moz in Oz 12.29.22 at 4:46 am

I tend to be of the view that we should simply not tolerate verbal violence, and that it is not only permissible, but actually sometimes better, to block off communication with verbally violent people.

It took me a long time to realise that some people genuinely disagreed with that position (rather than just mouthing enemy propaganda). I grew up with abusive parents so the idea that I could just stop having anything to do with them was appealing and obviously a good thing. But then some of the NVDA people I dealt with wanted to take the contrary position to Te Whiti levels, and I was just “nope, there’s no possible win for us from doing that”. Te Whiti “won” 100 years after he died… MLK might be a US analogue. Plus tactics have to change over time.

There’s also a jump from what works one on one, or in small groups, and what works when you’re out in public shouting over media and other pressure groups. It doesn’t matter how good your argument is when you’re in front of the White House contradicting Trump and Faux News, that just shows that you’re wrong and asking for a beating.

One political technique that absolutely does work is to have the lovely reasonable people asking politely for change, while outside an angry rabble are disrupting the normal course of business. We need both The Green Party (in democratic countries) and angry anarchists disrupting traffic or derailing oil pipelines or whatever the latest cause of outrage is. You never get what the rabble want and you should never expect to. Their demands have to be unreasonable to be effective. The USA coup attempt + traitors in the house is another example of that two-faced approach.


John Q 12.29.22 at 7:25 am

Following up Moz @3, there’s a division of labor here. The skills needed to work out what policy is needed aren’t, in general, the same as those needed to persuade some particular group of the merits of that policy. And there can be quite a few intermediate links, from experts who study a problem but don’t worry about policy, to policy-oriented academics to policy advocates to politicians to voters.

Of course, this presupposes a system in which people actually care about policy outcomes. If all you want to do is “pwn the libs”, you can short-cut most of these steps.


John Q 12.29.22 at 7:35 am

The range of difference that can be bridged by persuasion is also limited. No one is going to persuade me that Trump is a stable genius, and nothing I say is likely to convince a Trump fan otherwise. I tried a while back to think about the range of people with whom I could usefully engage. Apart from the gamut spanned by CT (as DD once put it “from social democrat to democratic socialist) that includes liberaltarians, non-dogmatic Marxists, and recovering soft neoliberals. But I long since learned the folly of treating purveyors of rightwing talking points as if they actually had something to say.

Again following Moz, we don’t need to persuade everybody, just a sufficiently large majority.


Ingrid Robeyns 12.29.22 at 9:48 am

Moz – yes, you are right about what you say in @3 – and John about the division of labour. Giridharadas’s book starts from the assumption that people have worked out what they think need to be done, and then looks at the second half.
Moz: what does NDVA stand for? (I have an assumption/guess but rather want to ask you).
On JOhn’s point @6 – that we don’t need to persuade everbody: yes, that’s right, and it was one of the things that is discussed in the book. In my own country, it would make a huge difference if the left-leaning parties (who for many years have lost to a coalition of right-wing parties), would start with persuading those who stopped voting or those who haven’t voted. That could make all the difference. There is no need (and probably little point) in trying to convince those who currently vote right-wing that they should vote differently.


Moz in Oz 12.29.22 at 8:59 pm

Sorry, non-violent direct action. Capitalised it’s a variously formalised system with proponents ranging from “don’t hit cops” to Quakers. Some proponents fall into the “anything other than respectful engagement is violence” trap, and did so decades before microaggressions were even thought of :)

I agree with your other points, and one of the common sources of tension in activist groups is the question of who we are trying to influence. Critical Mass, for example, is generally a fun bike ride (it’s not a protest!) that usually targets cyclists and wanna-be cyclists as much as it targets politicians or other usual suspects. It’s not trying to persuade committed motorists that there’s any benefit to them, that’s for the polite and reasonable crowd… even though that lack of persuasive effort violently offends some motorists.

I feel the need to quibble with John Q, we rarely need to persuade a majority, or even a near-majority. Just swinging enough voters, or influencing key decisions, is usually enough. A 5% swing at an election is huge…


John Q 12.29.22 at 11:35 pm

Australia is fortunate in having both compulsory voting and a mix of instant runoff and PR systems. Mostly introduced by conservative governments, who saw partisan advantage at the time, but currently more beneficial to the centre-left and left.


conchis 12.30.22 at 3:37 am

Just a quick note to say thanks for the recommendation: have just started reading it, and is very good so far.


ozajh 12.30.22 at 7:09 am


“Defund the Police” is so bad as a slogan that I honestly wonder whether it might have been deliberately coined (or, perhaps more likely, emphasised) by the authoritarians. “Reform the Police” would have been so much better, in the sense of being harder to rationally oppose.

I feel the same way, for different reasons, about “Black Lives Matter”. It’s almost too easy to derail the conversation by spouting “XXX Lives Matter” for XXX = White/Blue/All/etc. and complaining about BLM exclusivity. “Black Lives Matter Too” would have been harder to attack, although no doubt Fox et al would have found a way.


Ingrid Robeyns 12.30.22 at 10:53 am

Moz – thanks – that was what I thought NVDA stands for, but just wanted to be sure.

conchis – glad you like it and happy reading!


EB 12.30.22 at 3:29 pm


Black Lives Matter isn’t even a goal or a program. It’s a value, a very legitimate value, but it does not point in the direction of any specific action that could be taken to uphold that value. Of course, the general public rightly read between the lines and realized that BLM was a reaction to police killings of Black people, mostly unjustified, often of men who were unarmed. But the name itself is so vague as to provide cover for almost any reform or program. And in fact it has been used to endorse everythoing from community-based food banks to (rarely) violoent protests.


TM 01.05.23 at 8:35 am

JQ 5: “The skills needed to work out what policy is needed aren’t, in general, the same as those needed to persuade some particular group of the merits of that policy”

And we should add that the people who are good at the first set of skills are rarely good at the second, but often aren’t aware of this deficiency.


TM 01.05.23 at 8:43 am

Regarding who to try to persuade: in the US, we know from the outcome of referendums (e. g. minimum wage, abortion rights) that there is a slice of the electorate who vote Republican despite disagreeing with their actual policies. We analogous groups exists in other countries too but referendums, where they are held, provide direct evidence. This is the group that I would try to persuade if I had the skills to be a good persuader.

Comments on this entry are closed.