Watch on YouTube

Beautiful. At the end of the day, despite our differences, we’re all the same, we’re all fruit!

BUT, but, but, but – as some undoubtedly under-qualified self-help influencer shared on socials once—

And here’s where we run into trouble… especially in politics.

Sorting the fruit

In this newsletter I talk a lot about intergroup theory. ‘Social categorisation’ is key to this theory. It refers to the tendency for humans to categorise one another into little identity boxes: woman, man, rich, poor, faux riche, apple, orange, cool, uncool etc. etc. This is a pretty lived experience for most of us. What’s much more interesting is why we categorise—

“A primary psychological function of social categorisation is creating meaning in an otherwise unstructured and chaotic world.” — Ellemers, N., & Scheepers, D. (2025).

That’s right. We loathe to admit it, but we’re all anxious little things who are absolutely desperate to make sense of our completely unexpected, unlikely, chaotic, and confusing existence.

You might recall, we are also absolutely desperate to 💅🏻 look good 💅🏻. This drives us not only to identify apples and oranges, but also to compare apples and oranges.2 Unlike many other social identities, in the case of politics, being an apple or an orange is a choice. So, we compare: which group (party) is better for my interests? Which looks, feels, behaves and thinks like me? Which increases my chance of getting what I want?

Social psychologists have also found that we also like to paint those within these categories with the same brush, as if they were cut from the same cloth… apples fallen not far from the tree 😉 …. Here’s Ellemers & Scheepers (2025) again:

“To facilitate meaning-making people tend to exaggerate differences between categories and underestimate differences within categories. This helps to perceive clear—and clearly distinctive—social categories, even if these group-based perceptions don’t always accurately render specific features of individual group members.”

Now, the tendency to over-simplify and ignore complexity is a deeply understandable, heartbeat, feeling, human thing. When we peel the onion, our eyes water. We’d have to admit that the world is indeed unstructured and chaotic and its occupants are also too complicated to categorise! And nothing makes sense anymore! And it’s scary!

In politics, however, it is also a deeply strategic thing to leverage simplicity and rigid stereotypes, to ignore similarities between groups, and to play up differences. That’s because fear, anger and resentment inspire stronger emotional engagement (votes!) and reinforce clearer group divisions.

While this is a crash hot political strategy, it’s not without consequence. Repeated exposure to hostile political messages hardens biases that persist long after an election is over. Overtime, these messages can shape people’s social lives, relationships and communities.

Preventing cross-contamination

If you’re spending much of your time holding negative opinions about your political out-group, and having that perspective validated by your trusted opinion leaders, you’re hardly going to seek out a conversation with “the greatest enemy since Iran”, are you?

Lucky for you: partisan sorting means that it is increasingly unlikely you’ll run into them in your spare time.

American politics increasingly permeates what were previously considered apolitical spaces. This means the food Americans eat, the things they buy, the shows they watch, the people they listen to, and the clothes they wear, are influenced, consciously or not, by their partisan identities.

All in all, opposed partisans are getting some space for each other. A reminder below, if you need it:

The majority of Americans report that their primary social networks are comprised of friends aligned with their party identity. Nearly 6 in 10 people in Democrats’ and Republicans’ social networks are also Democrats and Republicans. In fact the share of those with no friends with political opinions differing to one’s own is growing.

The same goes for marriages. 77% of Democrats and Republicans are married to fellow Republicans or Democrats. Surveys have also found 4 in 5 Americans would prefer not to date across the political aisle in a serious relationship.

Zooming out from the family home to American neighbourhoods, there too American partisans are distant from one another, with political geographic sorting into ‘red’ and ‘blue’ states on the rise since 1976.

Interestingly, recent research also shows partisans have also become clustered into like-minded communities within states (in fact it is at its highest since the Civil War), and that this is driven not only by economic and racial considerations, but also by a want to be connected and nearby those with whom one agrees politically.

One 2012 study looking at internal migration data across seven states, found, intentionally or not, when American partisans relocate, they tend to relocate into areas populated with co-partisans. A follow up study in 2015 also found that properties are evaluated more favourably if prospective buyers learn of their neighbourhood’s like-minded political views.

Heck— look at the extensive reporting of Democrats so devastated about the results of the 2024 US presidential election, that they decided to leave the country.3

The issue with this kind of geographic and social sorting on the basis of partisan identity is that it can all-the-more spur on more extreme political views and more acrimonious feelings toward the other party. In fact, studies show conversations between like-minded individuals cause their political views to become more extreme.

The political constitution of marriages and families is essential for belief formation. Assumptions about others and the word, prejudices, political values, priority policy issues are all transmitted, moulded and hardened at the family dinner table, through conversations with spouses, or as modelled by parents.

Changing a political mind is extremely difficult. More so, once political beliefs are tied to family, community and social group belonging, challenges to one’s worldview can feel less like intellectual disagreement and more like a personal attack. It’s a lot easier to continue believing that all conservative women are gun-slinging, bleach blonde, bible bashing, trad wives when you don’t hang out with any.

Upsetting the apple cart… with an orange

(Man! Really pushing this fruit metaphor to its absolute limit)

Our brains like simplicity. And many of us don’t have, or feel like we don’t have, the time or skills or gumption to get to terms with something more complex.

The problem is that without challenging oversimplifications— especially via the critical mediating role of accidental social contact in schools or the supermarket or our weekly game at the bowling alley— our often unkind views of the other side are rarely challenged.

I shared early on that Americans aren’t big fans of their country’s politics. And look… fair enough. But if something is to be done about it, we’re have to put our big boy pants on, get uncomfortable and spend some time with some other fruits.

Democracies don’t require consensus, nor do they depend on everyone changing their minds. But they do rely on citizens retaining their capacity to see political opponents as fellow people in a shared (messy, disgusting, confusing, terrible) political reality.

Democracies survive disagreement. What becomes far more dangerous is when political opponents cease to be viewed as people worthy of understanding.

A quick democracy-preserving tip!

It’s been a while since I’ve dished out some unsolicited advice, so here’s a quick tid-bit to take to your next dinner table:

Heard something you don’t agree with? Before jumping to ‘you’re wrong’, try: ‘tell me more about that’.4

Research on political polarisation and conflict reduction consistently find that people are less defensive and more open when they feel heard and understood. It also makes conversations more complex and less tribal, because people are forced to explain their reasoning rather than simply performing their partisan identity.

Curiosity is not agreement. “Tell me more about that” is a way of keeping the conversation focused on understanding the other, not being the one who is right.

‘Comparison is the thief of joy’ is just one aphorism whose truth I’ve actively resisted purely for the sake of being against aphorisms. (That’s probably why I still use LinkedIn—joylessly).

So, at the expense of my own internal consistency, I must admit: It is true. Comparison is undeniably the thief of joy… especially in US politics.

We all just want to look good.

In this series, I’ve talked about how group identity theory (thinking in terms of in-groups and out-groups) can help our understanding of political trends in the United States. A key mechanism of the theory that we haven’t yet discussed is the role of comparison.

In order to distinguish our social groups, we need to find points of difference: what is it that makes “us” different from “them”?

Social Identity Theorists say we practice this ‘intergroup differentiation’ in order to solidify our own sense of identity. We want to feel an affinity with members of our in-group, so we emphasise the ways in which we are aligned with our group and are misaligned with the alternative: “I belong with Group X because I behave, look, think like them and not like Group Y.”

Another reason we make these comparisons is for the purposes of evaluation. It’s not only about identifying oneself within a group. It is also about finding the reasons that being in the in-group is good (or at least looks good) for us.

You don’t have to go far into the topic of self-esteem to discover that we are all burdened by wafer-thin ones. We touched on this idea at the end of Part 4 where I concluded with a complex little takeaway: “we all just want to look good.”

As we discovered last time, when we compare and evaluate our groups, we instinctively have more positive evaluations of our own side. Why is this? Well:

However, the literature as I’ve seen it is murky on whether the reverse effect is true—that when we compare social groups, we are driven to create disparaging, negative evaluations of our out-groups. In fact, many studies have found that while there is almost always a compulsion to preferentially treat one’s in-group, the impulse for negative evaluations, ill-treatment or hostility toward out-groups is not guaranteed.1 Sure, it might help you feel good about your group by thinking your group is really good, but you don’t need to be a heckle the other side to get that feeling.

Or do you?

Enter: Politics

Partisan identities are funky social identities for a few reasons.2

One important reason is that political group identity, unlike many other social group identities, is ultimately a choice. Unlike biological sex or race, you’re not born with the inherent traits to signify whether you’ll grow up to be a Republican or a Democrat. You’re free to change your mind about your political identity whenever you like,3 and the defining traits of what it means to be “a Republican” or “a Democrat” can and have changed over time.

Another reason is that political identities also serve a fairly unique purpose: They exist to help individuals create coalitions to effectively compete for power.

Politics is ultimately a 💪🏻competition💪🏻 to determine “who gets what when and how”, as any political science student will tell you. Joining political coalitions and groups (most often: parties) gives individuals their best shot at influencing the political process, of having some say about “who gets what when and how.”4

Intergroup behaviour applied to the funky political context (of 💪🏻competition💪🏻), therefore, produces funky results.

Because politics is an inherently competitive context and social groupings serve a competitive purpose. Distinguishing between political groups therefore doesn’t just have the innocuous function of helping individuals to determine who’s who in the zoo (to identify what makes “us” different from “them”).5

Instead, partisan differentiation, is much more useful as a tool to help pursue political power.6

As a result, unlike comparing other social groups, intergroup differentiation in politics almost always inspires disparaging views of the opposition and out-group criticism. Pointing out the ways a party is superior to its opposition— has better ideas, is more trustworthy, and more interested in you, dear voter— is kind of the whole point.

In politics you’re much more likely to attract flies with manure (“they’re bad”) than honey (“we’re better, trust me”). In competitive elections especially, votes are corralled more successfully through fear and loathing (“vote for us because we’re not them”) rather than stirring up the sociological impulses of in-group favouritism (“vote for us because we’re brilliant”).7

This is especially true when your first priority is to ‘to turn out the vote’ because voting is not compulsory… *cough, cough* like in the United States.

Hurling insults at the other side is an effective way to make fence sitters pick your team by leveraging our innate sociological impulses to look good and associate with groups that boost our self-esteem.

💅🏻 **You want to look good**💅🏻, so you wouldn't want to associate with them, would you? Not when they’re corrupt and probably going to lose the next election… that would just make you look bad.

Be yourself; everyone else is taken.

Remember there are two functions of intergroup differentiation— one is group evaluation (which we just covered), the other is to solidify one’s own sense of identity.

It’s important to say, the identity-forming function of intergroup differentiation, that distinguishes one group from another, is not completely redundant here.

As much as us conflict-avoidant extroverts would love it, you can’t include absolutely everyone in your party because that would imply there is no disagreement on how things should be run…

There needs to be some level of discrimination and distinction to determine who fits in with the party and who doesn’t, and what the party stands for and against.

This identity-defining criteria of “who we are as a party” is usually clarified at big political flashpoints like in the process of an election where policy positions are articulated, campaign slogans are said too often, leaders and representatives are chosen, and certain social groups are targeted as the beneficiaries of policy proposals.

Other moments like when big speeches are delivered or consequential decisions are made, also help define a party. Think: when Democrat president Lyndon B. Johnson urged Congress to pass the Voting Rights Act in 1965, or Republicans chose Donald Trump to represent their party on a more populist platform in 2016.

Typically, only the most interested, motivated, involved political watchers with access to the most powerful movers and shakers shape this process of self-definition publicly.

So, when the party is defining what it means to belong to the party’s in-group, it’s ‘the party’ as defined by the most politically-switched on thought leaders. It’s the media that gives shape and words to these trends, its elected representatives, their campaign strategists and advisers, and maybe that small percentage of Americans who turn up to town halls, lobby, picket and protest, and vote in primary elections.

And what do we know about these utmost politically-switched on thought leaders…? They’re ideologically polarised. (See Part 1).

All for one and one for all.

We established earlier that partisan identity is unique because it’s ultimately a choice. Americans can choose to switch identities whenever they like. Logically, these big political flashpoints— speeches, upsets, new policies— should offer voters an opportunity to reflect on whether their preferences, worldviews and values (their self-identity) aligns with the identity of their associated party.

BUT THIS RARELY HAPPENS.

Voters, even when they might not agree with the direction the party is heading, rarely switch sides. They’d rather just not vote.

There are a few reasons for this. As we know from Part 3 partisan sorting means that one’s partisan identity is more sentimental and highly valued. To defect or to question one’s allegiance or alignment with their party then risk social acceptance—and an identity crisis (!!)—on a number of levels.

We also know, from last time, that more often than not when the party moves its positions on something, it’s partisans will follow suit.8 We discussed how most partisans are ‘party identity-first, beliefs-second’ actors: most voters tend to mould their preferences to suit whatever the party—their identity group—is doing, saying and thinking.

That’s voters. Elected representatives face a next-level conundrum.

In a different much more public and consequential way, elected representatives run into trouble with this ‘fall in line or fall apart’ dilemma all the time. It is hard to take a principled position on a particular issue that reflects your true personal beliefs and not the beliefs of the party, particularly when its make-or-break for your re-election.

In recent times, this has proved a real issue for Republicans who don’t ‘fall in line’ with Trump’s new direction for the party. For example, July 2025 brought a fresh slate of Republican retirements from lawmakers who have sought greater independence from Trump and his new direction for the Republican party. These retirements follow the various retirements and defeats of other GOP lawmakers like Jeff Flake, Liz Cheney and Adam Kinzinger who had broken ties the President (the de facto GOP leader) in the last four years.

The reason for this, summarised perfectly by one former comms director of the National Republican Congressional Committee is that “in a post-Trump world, 100% loyalty is the party litmus test.”

It’s not just a Trump/ Republican thing though. Democrats too have more incentive to ‘fall in line’ and present themselves as a unified coalition than they used to. If you’re a devout Catholic Democrat that isn’t historically super on board with codifying abortion rights, you’re likely to change your position to suit the new party positions than risk isolating the new mainstream of your party or (God forbid) being associated with those be woman-hating, revisionist Republicans! This was Joe Biden’s transformation, btw.

But I’m also thinking about in 2023 when Democrat Senators voted with Biden 99% of the time, and Democrat House members who voted with Biden 93% of the time; recent criticisms of ‘group think’ around Dem’s consistent support for Joe Biden as Democrat candidate in 2024 despite his outrageous lack of popularity; and Democrats’ pretty-unified backing of several COVID-19 measures like lockdowns, social distancing and mask mandates that contradicted other party interests.

Since at least 2017, but possibly even before that, what we’ve seen is greater consolidation around the “party-line” and a drastically lessened tolerance for anything that diverges from the mainstream party position— both in elected officials and in public support.

One man’s trash is another man’s treasure.

At the same time, the party line for both parties has increasingly been defined in opposition of one another— “what they (my in-group) said”, “if they’re for it, I’m against it”, “If Trump said it, must be bad” etc. etc.

Polling on a number of issues shows Democrats and Republicans positioned in a near mirror image of one another. If Democrats care, Republicans don’t; if Republicans think something is a good idea Democrat’s don’t.

Previously shared understandings between the parties of what, above all, it means to be “an American”, or a “good citizen”, have all but disappeared. The scope of ideological overlap between Democrats and Republicans has seriously diminished, as has the perception of the partisans’ ability to agree on anything, including basic facts.

There are various consequences of this. Among several others, it raises the stakes of losing. To lose an election to the outgroup, or allow any concessions on policy or perspective, now means the realisation of everything you don’t want and don’t value…. Because “if they’re for it, I’m against it.” This is one of the reasons for the apocalyptic feelings and hectic levels of anxiety about the outcome of the 2024 election.

Another consequence is that it incentivises playing up threats of ‘the other’ and disparaging one’s political rival. As we discussed, playing up your own side’s brilliance doesn’t necessarily convert to votes on Election Day, while fear and loathing of the other side are often much better lighter fuel.

And yet another consequence is that partisan identity (values, interests, attitudes, traits, affiliations) is therefore also, so often, defined in opposition to ‘the other’. Because of partisan sorting, political identity permeates everyday life. Americans signal their political identities in the food they eat, the things they buy, the shows they watch, the people they listen to, the clothes they wear, and the events they say yes to.

The two political groups are becoming more and more distinct and opposite in character. And what happens when Democrats go voguing in gay bars, while Republicans go boot scooting…. their partisans become more socially distant! Cross-pollination with diverse others is less and less incidental, and of even greater consequence—it’s less tolerated.

TL;DR

At least some of the state of the United States today stems from, I think, the very human tendency to make comparisons: Most people feel they need to embody the attitudes, style and behaviour of their social in-group to belong and to feel as though they 💅🏻 look good 💅🏻; and to validate their choice, to really feel as though they belong and look good, there is a very natural desire to distinguish the alternative choice9 as looking less good.

In the political context— where 💪🏻competition💪🏻10 is the name of the game and partisanship is a choice— this is all turned up to 11. It’s not only that the alternative party is a sub-optimal choice, it’s that their constituents are morally bankrupt and their ideas will send the country into ruin.

Seems like the kind of people I’d want to get some distance from… and more on that idea next time.

The suggestion to keep politics away from the catalogue of acceptable dinnertime conversation is an understandable one. But it is much less helpful than many think.

Welcome back to my series Bringing politics back to the dinner table— the series where I look at how politics came to be so stomach churning, and what we might be able to do about it.

I like to think of these newsletters as little political analysis Hor d’Oeuvres. They’re beautifully constructed little morsels that aren’t quite satisfying unless you sample each and every one…1

Subscribe now

I know there’s been a little delay in the service, so I’ll do some of the heavy lifting upfront with a few reminders about where we’ve been and where we’re heading.

We’re looking to the United States to better understand what drives political division. This sets an introduction for some key political science & social psychology terms that we’ll use to analyse Australian politics down the line.

Last time, we explored the increasing salience of Americans’ political (party) identity in their sense of self-identity.

Interestingly, very, very few Americans say that their political party identity is important in describing who they are. They are much more likely to cite their family status, religion, gender and age as being important to their identity. And fair enough.

The problem is, as we discussed in Part 3, most of the other non-political identities on that list— gender, religion, region, age— have each become increasingly intertwined with politics.

Electoral data over the past forty or so years shows several ‘non-political’ social identity groups (e.g. black Americans, evangelical Christians, non-college educated young men) are fairly reliable constituents, affiliates or voter blocs that favour either one of the two major parties. This is a trend called ‘partisan sorting’. 2

These identities not only help inform how individuals vote, but they also help inform how the two major parties determine their priorities, messaging, agenda and interests.

It’s a mutually informing process that has resulted in the two parties being made up of constituents representing two distinct clusters of grouped social identities that have relatively similar views, social demographics and life experiences.

Back in Part 2, we identified that it is not political polarisation, but affective polarisation that lies at the heart of American’s political divisiveness.

Affective polarisation refers to evidence that Americans are increasingly hostile and disparaging about members of the opposite party (out group hostility), while maintaining rosy opinions about their own party (in group favouritism).

Partisan sorting and its related effects has helped affective polarisation to thrive since the 1980s. There are a number of reasons for this. We’re going to keep the teensy-weensy hor d’oeuvres flowing and go through each of the consequences in separate newsletters.

🔥🔥HOT TIP INCOMING 🔥🔥 It’s has so far been and will continue to be a very unpleasant amuse bouche so, for a special bonus I’m going to share my premiere hot tip to help you overcome some tension when political disagreement takes a seat at your dinner table.

Alright then, Bon Appetit.

Partisan sorting promotes in-group biases and in-group loyalty

Social Identity Theory3 suggests when we have a shared identity or group categorisation, we evaluate those on our side more positively than ‘the others’ for no logical reason.

This ingroup biasing effect is true even when our social categories are arbitrary and random (e.g. in a social experiment where participants are given “Category A” vs. “Category B”).

Perhaps expectedly then, the effects of ingroup bias are much greater in the context of partisan sorting.4 This is because partisan sorting boosts the strength and sentimentality of our political identity.

In sorted political groups, we are more likely to experience a greater sense of belonging and connection because we share similar characteristics, values, interests, opinions and life experiences to the other people in our group. It’s easier to feel like we fit in, that we are surrounded by people who are like us and understand us.

These strong in-group feelings inspire a number of weird and irrational political behaviours.5

For example, American partisans approve and positively evaluate candidates and/or policies labelled ‘Republican’ if they’re Republican and ‘Democrat’ if they’re Democrat, regardless of whether that candidate or policy aligns to their actual, declared political ideology or their party’s known stance on a particular issue.6

This suggests specific political issues are peripheral to Americans’ voting, whereas their partisan identity and group affiliation is paramount.7 Or, as the allegedly British saying goes, “I would vote for a pig if it was my party’s pig”.8

Strong partisan identity and its associated cognitive biases might help us understand why President Trump enjoyed enduring support from Republicans despite frequent scandals (two impeachments, four indictments, the E. Jean Carroll trial which found him liable for sexual assault, and a criminal conviction) in 2023-24, even though 2 in 5 of them said such dramas and criminals convictions would be enough to curtail their support.9

Republicans were merely backing their party’s pig.

This little piggy went to market, so I went to market too

At the top of this newsletter, I alluded that the centrality of political identity in one’s self-concept is not a conscious choice for most people; they don’t consciously consider their political identity as essential to who they are. Afterall, only 16 per cent openly acknowledged it in the survey I mentioned at the top.

This makes sense. A foundational assumption of political science is that the majority of the public are politically checked out, they don’t follow politics closely or care about it that much.

If people profess that their politics is not important to who they are, how come the American electorate is increasingly affectively polarised? How come broad swathes of the public are willing to say disparaging things about their political opponents?

I think some of the answer to this lies in partisan sorting.

Most Americans are not ideologically motivated to get involved in politics. The ‘folk theory of democracy’ where citizens read newspapers and debate details of policy with their friends and use this deep contemplation to form their political attitudes— has largely been debunked as a fallacy.10 As has the ‘belief-first’ model of political engagement, where political actors first form their political beliefs and then get involved in the political process.

Partisan sorting makes politics increasingly inescapable in other social settings. It also means that most Americans’ entrée into politics is not on the basis of their political beliefs or opinions, but on the basis of association with their other social groups. There are lot more groups-first beliefs-second thinkers in the mix.11

Being brought into politics via non-political social groups means most partisans are likely to have a low base of detailed political understanding, or coherent or ‘true’ opinions about policies or politicians.12

Without truly held or principled political beliefs, the politically blissed-out are also less likely to have detailed nuanced views about politics and more likely to compensate for this knowledge gap with over-simplified mental shortcuts (hermeneutics).

They think more along the lines of “other party = bad, my party = good” than “some members of the other party are okay when it comes to this particular issue, but on most other issues my party is better”.

Group-first, beliefs-second thinkers are also much more likely to go along with whatever everyone else in the group is doing and to use their group’s attitudes to determine their own.13 It’s not laziness that prompts us to act in this way, but our very human hardwiring to seek out connections, to find belonging and to be accepted.

I might not be a card-carrying Democrat, but if I’m hanging around a lot of them—likely by virtue of other social groups: my female-author-only book club, my gay friends, my fine arts degree— I’m likely to mould my political beliefs to match theirs, especially if I don’t care too much for politics or have my own ideas. That way I can fit in and be accepted in other social settings that are important to me, and I don’t watch the news, or know any better, and I like those people, so they must be right, right?15

This little piggy cried wee, wee, wee, “fake news”, wee all the way home

Here's the problem:

You’d think these ‘non-political political’ people, brought into the political fold by their other social identities, will not be as biased as avid political watchers. But it is not so. They are similarly, if not more so, subject to in-group preferencing proclivities, especially when it comes to the manufacturing of political beliefs.

Political ideologues, those with well-informed and truly held beliefs about what is right and wrong in politics, are more likely to think critically about whether or not a certain politician, policy or platform truly aligns with their political values.

It was, after all, deeply engaged Republicans with strongly-held traditionally-conservative political ideologies who waned in their support for President Trump (someone broadly recognised to not represent those values) a short-time after he began his first term.16

On the other hand, the ‘non-political political’ folk simply do not have a rounded enough base of political knowledge to ‘know any better’ than to go along with what the group thinks.

Most of the time, the manufacture of political beliefs is fairly harmless. As Paul Bloom illustrates,

“If I have the wrong theory of how to make scrambled eggs, they will come out too dry: if I have the wrong everyday morality, I will hurt those I love. But suppose I think the leader of the opposing party has sex with pigs or has thoroughly botched the arms deal with Iran. Unless I’m a member of a tiny powerful community, my beliefs will have no effect on the world. This is certainly true as well for my views about the flat tax, global warming, and evolution. They don’t have to be grounded in truth, because the truth value doesn’t have any effect on my life...”

Bloom touches on an interesting idea: the truth often doesn’t yield much social reward, whereas conformity and group think (as a means of group acceptance) does.

In this study, researchers Schaffner and Luks asked nearly 700 Americans to compare a photo of Trump’s 2017 inauguration crowd to that of Obama’s in 2009. They asked a simple question: ‘which photo has more people in it?’

Obviously and factually, Obama’s inauguration had a larger crowd. Yet nearly 15 per cent of Trump supporters (= 218 people) said Trump’s had larger crowds! A clear deliberate misreporting of fact and also of their true beliefs!

Public opinion polls time and again show that partisans are often unable to agree upon the facts, like this one. Some political socio-psychologists say the divergence of the two parties over what is considered to be factual is because political beliefs are socially expressive beliefs. They are not about right/ wrong or about truth or about fact.

In isolation, what Paul Bloom says might be true— fibbing about one’s political beliefs or failing to profess the real truth may not have major effects on one’s life, or on the world.

But what we’re seeing in the United States is a trend where ingroup acceptance requires the dismissal of accepted facts, speaking disrespectfully about other people, and in extreme cases committing acts of violence.

On a large scale, we can already see that the trading of truth for group acceptance has had some funky, dare I say democratically corrosive, outcomes for America’s politics.

This article in The Atlantic, which analyses the demographics of January 6 rioters, helps to illustrate this point:17 It doesn’t take being an ideological radical to take radical political action. All it takes it the very human drive to have connection, to feel a part of something, to belong.18

Being a prized pig: We all just want to look good.

To sum up, I’ll give a plain and simple mention of the psychological driver behind our in-group favouring action— we just want to look good.

Because one’s partisan identity is (increasingly) important to their self-concept, as we discussed last time, people are motivated to perceive their ingroups favourably to maintain or enhance their group-esteem, and their self-esteem by association.

To admit our strongly endorsed political leader or policy idea was wrong vicariously puts egg on our face. It’s not great for the self-esteem. (This might help explain why a stubborn two-thirds of Republicans believed the 2020 presidential election was stolen from Donald Trump despite extensive evidence to the contrary… maybe they just didn’t want to admit they were wrong?).

So, here’s: 🔥🔥MY FIRST DIVISION MITIGATING TIP OF THE SERIES🔥🔥

If you’re ever up against a political opponent with whom you cannot agree about anything, including basic facts, your best chance of cutting through to them is not to prove to them that you are right.

Firstly, we’ve learned it’s not about right or wrong, it’s about group loyalty, acceptance and belonging. So, the disagreement isn’t really about fact or correctness, it’s about group attachments.

Secondly, for them to admit you’re right would make them look bad, and we all just want to look good, so you’ll end up at a stalemate.

I can’t know your relationship or how the conversation had unfurled so far, but I can say that your best approach will be to let them have the win.

Disgusting, I know. But if you can manage to assure yourself that you still look good, even in conceding, you’ll be okay. And the world is unlikely to explode if this one person cannot change their mind.

In fact, the world is all the less likely to explode if they have thought-divergent friends like you and cross-cutting social linkages in their life, people and groups that make them feel loved and accepted outside of their political group.

If you want to do your democratic, liberal, truth-promoting duty, you’re better off holding onto that person than to show them how wrong they are.

At the heart of the politics-is-sport analogy is the acknowledgment that team (or group) dynamics are key to the game. There are opposing sides, spectators have their loyalties, don their colours, cheer them on and boo the other team.

Politics and sport are undergirded by Social Identity Theory 101. Our groups, sporting and political, are expressive sources of self-identity. They give us a sense of belonging, a sense of ‘our people’, an ‘us’ that is distinguishable from a ‘them’.

Viewing partisanship as a form of social identity blast through the political sciences with The American Voter in 1960 when Angus Campbell and his co-authors pedalled a preposterous idea:

Most American voters, they said, are inattentive to politics and so have trouble connecting the dots between their interests and the promises of the two major parties. Instead, they found voters cast their ballots on the basis of their partisan identity.

In this respect, partisans act more like sports fans (governed by group loyalties) than investment bankers (governed by strategic, rational thinking).

Expectedly, talking about 🥴FEELINGS🥴 rather than 🫡 RATIONALITY🫡 in politics really got under the skin of the American political science community which, at that time, overwhelmingly preferred to view political science as a “science”.

Yet, this view has come to be more accepted because it has proved useful to understand partisans’ frequently irrational, self-interest-diminishing, and sometimes outrageous, political behaviour. Like, for example:

• Children, who do not vote, tend to adopt a partisan identity very early in life. Here’s a fascinating study that shows as many as a quarter of American six year olds prefer one party over another. It also shows young children have affective-laden views (views that favour co-partisans and disparage those from the other party) when they’re asked to talk about or draw parties and their leaders.

• Even when Americans experience meaningful changes to their economic and social their circumstances, the kind of changes that should recalibrate their political leanings, most Americans’ partisan identity is stable and consistent throughout their lives.

• Partisan identity is even consistent when partisans accuse their own party of poor performance and weak leadership, or when they change their stance on core party issues.

• Despite significant evidence to the contrary, more than two-thirds of Republicans still hold Donald Trump was cheated out of a 2020 presidential election victory. Increasing numbers of those who think the election was stolen also admit there is ‘no solid evidence’ for their belief.

Strange, irrational behaviour.

For our purposes, viewing partisanship as an expressive group identity is especially useful to understand the trends of affective polarisation discussed in Part Two.

For one thing, a key tenant of Social Identity Theory, and social psychology writ large, is the idea when individuals are categorised into groups, we evaluate those on our side more positively than ‘the other side’ and discriminate against the ‘others’ for no logical reason.

Click below if you can’t remember what ‘affective polarisation’ is…

Individuals are a group soup

While American politics is often thought to be fuelled by the dynamic of the Republican-aligned group vs. the Democrat-aligned group, it’s not so simple.

Individuals are a group soup. We each contain lots of interesting nibbly bits and pieces from our complex and various social identities and group attachments.

We are each a jumble of different sources of self-identity, from the characteristics we were born with (race, sex, social class) to life experiences that shape us along the way (education, geography, religious groups etc.).

For allegedly ‘non-political’ identities, these other social identities are woven into American all the time!

This is a complaint for those who don’t like ‘woke identity politics’. The inundation of identity in American politics, they say, has created a highly flammable, emotional environment, unfocused on the ‘real issues’.

Being in an era of identity politics means we spend less time talking about ‘traditional political’ matters like bureaucratic process and unemployment, and more time talking about emotionally-charged issues like Black Lives Matter, gender-neutral bathrooms and trigger warnings.

This might mathematically make sense: More identities = more competing identity group dynamics = more emotions, upsets, loyalties, passions.

There is also evidence that shows political disagreements on identity-based issues like migration, religion and LGBTQ+ rights fuel more hostility towards political opponents than disagreements over economic issues.

Yet, to suggest identity politics is the reason for America’s more recent affective polarisation problem implies that identity otherwise had very little to do with ‘old school’ American politics.

This is a curious concept for a nation whose most significant political developments have had ‘non-political’ social identities at the centre— the Civil War, women’s suffrage, literally any kind of rights and reform movement.

Putting the woes of identity politics to the side for a sec, when election time rolls around, individuals’ identity-group soupiness is one of the reasons rationally determining who to vote for can be so difficult.

Some suggest, to make political decisions, individuals form an ‘identity stack’ which houses all their group identities (e.g. “white”, “female”, “doctor”, “childless”, “marathon runner”, “Walmart shopper”1^, “Virginian”, “Gemini”2). And then, when it’s time to get political, the identities that they hold most dear climb to the top of that stack and help them determine which party best represents the aggregate of their in-group interests.

While this all sounds fine, as the authors of The American Voter would argue, the idea that voters spend time actively calculating their total group-based interests is unrealistic and inaccurate. To do this sum would be a time-consuming challenge that very few voters have the political knowledge, interest or resources to do at all, let alone accurately.

Personally-meaningful ‘non-political’ social identities are certainly politically formidable identities. This is well-known.

It’s one of the reasons Hillary Clinton ran such a female-forward 2016 presidential campaign; the assumption that the ‘woman vote’ would be cast for the first US female president because of shared in-group interests between American women.

Donald Trump, aware of the potency of shared identity in politics, accused Kamala Harris had suddenly “turned black and now she wants to be known as black” during the 2024 presidential race. The comments were clearly dog-whistling. But, they also illustrate this idea that non-political identity group attachments are politically powerful— that by sharing a racial identity with 1 in 5 Americans, Harris was a shoo-in to capture Black and/or Asian voters.

Yet, politics is also not so straight-cut that shared social group identity coverts to votes. More black voters voted for Joe Bidenin 2020 than for Kamala Harris in 2024. More white women voted for Trump than Clinton in 2016. Also, there simply aren’t enough middle-aged wealthy, white businesswomen from Georgia, or a young, queer, church-going black men from New York to account for everyone’s individual group-soupy constitution.

Rather than other social identities, partisan identity is consistently the strongest, most salient identity in politics.3

Unlike the explanation of ‘identity politics’, the reason for affective polarisation and America’s more emotional politics is not because there are too many identities groups swimming around in the American politics group soup.

It’s because, over the last half century, partisan identity has increasingly fallen into alignment, not competition, with Americans’ other social identities. This is referred to as ‘partisan sorting’.

American politics is not group soupy, it’s sorted

You have likely heard Black Americans always vote Democrat, or rural Whites and evangelical Christians love Trump, or that “the Latino vote is in play this year”. Identity wasn’t always aligned with partisanship in this way, nor were the two major parties so distinct in their political worldview or in their social demography.

In the mid-twentieth century, constituents of the Republicans and Democrats overlapped so much that a study in the 1950 edition American Political Science Review advocated for more polarisation in American politics. They said, basically, ‘no one knows who to vote for because they can’t tell the difference between the two major parties and what they stand for.’4

Things changed in the 1960s, when Democrats’ leadership on the passage of the Civil Rights Act (1964) and the Voting Rights Act (1965) prompted a party realignment and sharper partisan differences.

Most Black Americans and social progressives then began identifying as Democrats, where just two decades earlier ‘the black vote’ was split evenly between the two parties. At the same time, those opposed to racially progressive policies, mostly in Southern states, left their Democrat-identities behind and became Republicans.

Over the next two decades American politics is all about women’s lib and women’s reproductive rights5; what to do about God6; environmental protection, nuclear power, nuclear families, and the Vietnam War.

These all become major (nuclear!) political wedges and created greater ideological and demographic differences between the two parties.

Those who wanted social reforms found alignment with the more socially progressive proclivities of the Democrat party. Those with more conservative worldviews, who preferred maintaining the status quo, preferred to vote for Republicans. (This is called partisan-ideological sorting).

Finally, the two party’s political worldviews are starting to look different. The authors of that 1950s paper begging for greater distinction between the two parties would have beamed!

Skipping ahead to the 1990s and into the 21^st century, there’s the internet, the proliferation of cable news broadcasters, the rise of social media, economic shocks, not to mention 9/11, climate movements and Covid-19.

All of which deepened the cultural and economic sorting of America’s various identity groups into reliant bases of the two major parties.

Partisan sorting means politics needs more helmets

Without looking at me, if I say I am a Democrat, you have a much better chance now opposed to fifty years ago to guess where I might live, what news I consume, whether or not I attend religious services, several of my policy preferences and likely even my race or gender.

That’s the result of partisan sorting.

Partisan identity has become a “mega identity”– an identity fortified by other social identities.7

The consequences of this are several. We’ll go through these in Part Four. For now, what you need to know is— partisan sorting strengthens partisan identity.

There is extensive evidence to show that cross-cutting social and ideological groups decrease social tensions, while aligned social, ideological and partisan identities ramp up affective polarisation. This trend has been observed since the 1960s.8

The theory goes that belonging to social groups that do not align with one’s partisan identity—you’re a transgender Republican or a gun-toting Democrat, for instance— means you are more tolerant of partisan outgroups and less biased. Your cross-cutting identities enable you to see positives, similarities, and kinship in both sides.

Those with unaligned social and political identities have a more nuanced, grey-coloured understanding of politics.

The reverse is true. When group identities are aligned— say for example you’re a black, secular Democrat or a conservative, evangelical Tea Party Republican —emotions are heightened, cognitive biases are motivated and intergroup differences are exaggerated.

When parties are sorted, politics is black and white: either you’re with us or against us, you’re right or you’re wrong, you’re agreeable or you’re evil.

Sound familiar? A bit like today’s political conversations, perhaps?

We know from the example of sport, where there are strong team loyalties, there is little chance the game will be watched passively. It gets passionate. The highs are higher, the lows are lower. The victories are more important.

Herein lies the increased sensitivity in US politics. Partisan sorting has increased as the strength of partisan identities and so increased the stakes of the game.

Perhaps politics should involve more helmets.

Or maybe not.

In this edition, we’ll be continuing our case study in the United States to get some more context for the nature of political division du jour. So, before you sink this one down, I highly recommend you go back and read Part 1 of this series.

Read Part 1

If you absolutely cannot be bothered, here are the most important bits:

• ‘Political polarisation’ — the movement of political attitudes to greater extremes— is the in vogue explanation for America’s searing hot political division.

• But, the extent to which the American public hold truly ideologically polarised, let alone ideologically consistent, views is questionable at best.

• There is growing evidence that suggests ‘everyday’ Americans think they are more polarised on political issues than they are in reality. This includes attitudes on a number of thorny issues like gun rights, abortion and social security.

• The issue in the United States is not necessarily that opinion-opposed Americans are taking on more extreme political views (à la political polarisation). Rather, as we’re about to see, Americans like their political opponents a whole lot less than they did before.

• So, to get to the heart of America’s political division, we’re going to have to talk about our **feelings**.1

Part Two: We’re going to have to talk about our feelings

The emotional cousin of political polarisation – affective polarisation – can help us understand why America’s political climate appears so acrimonious while most Americans hold unexpectedly moderate political views.2

Affective polarisation looks at the gap between in-group favouritism (positive, warm views toward one’s own political party) and out-group animus (negative, cold views about ‘the other’ party). It’s often detected using a “feelings thermometer” and a few other behavioural studies methods.

The American National Election Studies (ANES) has asked about registered Democrats’ and Republicans’ feelings toward one another since 1976. Their data shows that Americans’ warm, positive feelings toward their political opponents have markedly cooled-off over the last 40 years.

For Democrats, warm feelings toward Republicans dropped 27.7° between 1976 and 2020. Republicans’ warmth toward Democrats dropped 29.6° over the same period.3 At the same time, the data shows Americans’ view of their co-partisans has remained consistently rose-coloured over the same period.

What we see then is a growing gap between in-group favouritism and out-group animus— a trend of affective polarisation.

The feelings thermometer affirms several other notable trends that demonstrate Americans are becoming more actively hostile toward their political opponents.

• Between 2016-2022, growing shares of Republicans and Democrats said members of their opposing party are more immoral, dishonest, close-minded, unintelligent and lazy than most Americans.4

• About 2 in 5 Americans have reported being willing to use dehumanising or demonising language about members of the other party.

• Seventy-five percent of Republicans who falsely believe Donald Trump won the 2020 presidential election think that all Democrats are ‘downright evil’.

• And nearly 1 in 4 Americans think political violence may be justified to “save” the country.

Feelings…? That’s not evidence!

If self-reported feelings aren’t your thing (“feelings…? that’s not evidence!” and all that),5 we can talk about the various behavioural economics and psychological studies of partisan bias. But, spoiler alert, these all show a similar result to the feelings thermometer— we tend to give preferential treatment to those on our political team and to be more sceptical, if not more antagonistic, towards those who are not.

• Leading Stanford researchers Iyengar and Westwood devised a mock-college selection committee who had to choose a scholarship winner between two students with similar academic credentials, but different party IDs. About 80% of the time, the mock committee members picked the applicant matching their own party identity. This was true even when the out-group candidate had significantly higher academic grades.

• Other behavioural economics experiments reveal people are willing to accept lower compensation when their employer shares their political affiliation; and are more likely to interview partisan-aligned job applicants.

• In a 2022 psychology study, Democrat and Republican identifying participants were asked to judge what emotion a series of ambiguously happy/angry/surprised-looking faces were expressing. Respondents misjudged those from the opposite party as more angry more often.6

• I get that you get it, but just one more! A 2016 study of online dating behaviours found that people view potential dating partners more favourably and a more likely to interact with them online when they share their political characteristics. Cute.

The findings about in-group favouritism and out-group animus are not unique to the United States, or to political identities alone. They mirror several social science studies investigating identity group-based prejudices like racism, sexism, ageism… basically all the -isms.

This research finds we more readily like those who are like us; and we don’t as easily warm to those who aren’t like us. And we often can’t help it. It’s biological or something.7

Yet, while there are biological presets that make us less lovey-dovey with our political rivals, what is uniquely compelling about out-group animus in US politics is that it has actively souring in recent decades.

This sort heightened out-group animus is not seen in the in-group/out-group behaviour of other social identities, which appear mostly stable over time. As Iyengar and Westwood write in the American Journal of Political Science (July 2015, p.690):

“Unlike race, gender, and other social divides where group-related attitudes and behaviours are constrained by social norms, there are no corresponding pressures to temper disapproval of political opponents. If anything, the rhetoric and actions of political leaders demonstrate that hostility directed at the opposition is acceptable, even appropriate.”

Why can’t we just get along?

So, to return to our original purpose— to use the United States as a case study to understand why talking about politics at dinner feels so yuck— we’ve nailed one big thing down:

It is not political polarisation (divergent ideas and opinions) that is solely responsible for the suffocating air of political division in the United States.

Americans increasingly don’t like members of the other party, they don’t trust the other party to govern well, and they don’t want to cooperate with the other party. And this is a major contributor to the fraught, hostile and acrimonious feeling attached to discussion of politics in the United States.

But why is it now that Americans like their opponents less?

In the next few newsletters, I’ll present some related political trends that might be driving both political polarisation and affective polarisation in the United States. Then, we’ll look at what’s been happening in Australia.

It’s probably a good time to make sure you’re subscribed.

Subscribe now

The advice to not talk about politics is a prudish way to prevent social discomfort because it’s obvious that disagreement is commonplace and that it is very awkward. It’s important – as this and this and this and this social scientist will tell you – but it’s very, very awkward. And it’s very, very hard to do well.

To avoid talking about politics is a delightful little pearl of wisdom. It might just save your friendships1 and the civility of your dinner party.

Yet, steering clear of politics is pitifully unrealistic. In 2025, we have hundreds of searing headlines rolling in from the United States to remind us­ of how quickly small talk about the weather can turn Trumpy. And, no matter what we say or do, politics, virtue signalling, baiting, cancelling, offence-taking, dog-whistling and general insensitivity is seemingly more and more a part of our everyday conversation.

In this series of newsletters, I’m going to make the case that avoiding the social awkwardness of political disagreement is bad advice. I’m also going offer you some research-backed tips to help you safely invite politics back to your dinner table.

To get there, first we’ve got to do some scene (or table?) setting…

Part 1-4 investigates the gold standard of uncomfortable, disruptive political division— the United States. We’re going to look at the rising heat of politics in the US and how this came to be. I’m also going to introduce some useful theoretical concepts to help us understand these trends and, later on, compare them to our situation in Australia.

Sound good? You might want to make sure you’re subscribed.

Subscribe now

Part One: A case study of political division in the United States

US politics provides not only life-saving small talk but it is also a great starting point to understand why we seem to be disagreeing so much, over so many things, all the time.

The US political environment is in a word “divisive”.2 The fancy among us use the term ‘polarised’ which describes the movement of political attitudes toward more extreme positions.

Data visualisation sourced: Pew Research Centre.

In recent years, ‘polarisation’ has become the explain-all buzzword of news headlines and commentaries dissecting the latest intractable political feud among American voters or between warring congress members on the Hill. It was even crowned Merriam-Webster’s ‘word of the year’ in 2024.

Polarisation makes sense as an explanation for increased feelings of division and disagreement in America. Indeed, the growing polarisation among members of the US House and Senate since the 1990s is a fairly well-known, well-researched and reasonably uncontested affair.

With more people taking more extreme ideological positions on political and social issues, of course there is less of a meeting in the middle, less common ground, less bipartisanship, more disagreement.

Polarisation is kind of a red herring

While political polarisation is a compelling idea, as far as understanding why there appears to be so much disagreement and division in the United States, it’s kind of a red herring.

Americans think they are more ideologically polarised on political issues than they are in reality. There are a few reasons for this.

For one thing, warring politicians, protestors, lobbyists, and political organisers – the ones with the meatiest, most outrageous or most stalwart opinions – have an easier time getting attention in the media. Your slightly politically-checked-out fence sitters do not.

This means the discussion of politics and the display of political opinion in the media typically snapshots the polarised ends of the political spectrum, not in the persuadable, agreeable middle.

It is up these more polarised ends, among politicians and the most politically engaged (about 1 in 5 of Americans),3 that political opinions are the most calcified and the most polarised.

Meanwhile, ‘everyday’ Republican and Democrat constituents are also, of course, divided on politics. Naturally. It’s reasonable to think busy, attention-poor, mostly indifferent Americans will take their opinion cues from the most readily available sources (i.e. the media and their political leaders) and so follow in their increasingly polarised footsteps.

Yet, the extent to which the middle ground between Americans has actually, fatally eroded is less clear. And the extent to which the public truly hold ideologically polarised, let alone ideologically consistent views, is questionable at best.

There is growing evidence that suggests most Americans actually agree on several important political and social issues.

If you need a mood boost, here’s a CBS News article showing 50 surprising things Americans actually agree on including agreement on thought-to-be contentious ideas like support for social welfare programs, that abortions should be legal in most or all cases, and that gas-powered cars contribute to climate change.

Looking to polarisation for the answers to America’s divisions is a bit misleading, but greater still, being hyperaware of polarisation makes it almost self-fulfilling.

Research has also shown that anticipating polarisation – spurred on by talking about it to a wit’s end in the 24hr news cycle – fuels more inaccurate and exaggerated perceptions about the views, policy positions and preferences of ‘the other side’.

There’s also a bunch of in-the-weeds political science stuff that suggests the public opinion polling relied upon to detect polarisation is prone to various errors.

• For example, the ones with strongest political opinions are unlikely to select ‘Don’t Know’ or ‘Undecided’ on a public opinion survey measuring attitudes toward a policy or issue.

• Also, survey respondents can feel compelled to tick one of the available answers (to make up an opinion on the spot) rather than admit they don’t know or don’t have an opinion on a political view.4

• Some respondents might also give the answer they think is the most virtuous or the most expected of them, a ‘preference falsification’, rather than the answer closest to their actual opinion.5

But our purpose here is not to pour through the decades-long debate about whether mass public opinion polls are trustworthy.

The point is — the number of ‘reasonable’ ‘everyday’ ‘moderates’ associated with one’s rival party are often greatly underestimated. As is the number of those in the opposing political party with whom one might agree.

I’m not about to gaslight Americans and say their experience of rising political division is only in their heads. It’s not.

But the issue here isn’t that opinion-opposed are Americans taking on more extreme political views (à la political polarisation).

The issue is that Americans like each other a whole lot less; they’re assuming the worst of each other, misunderstanding each other, and distancing themselves from one another.

The problem of America’s political division is largely emotional.

So, for a more accurate discussion of US politics, we need to investigate Americans’ ✨🥴 FEELINGS 🥴✨, not opinions.

Share Dinner Table Politics

Subscribe now