We talk as if people carry tidy, worked-out political belief systems. Most do not. Underneath the labels, political life runs more on identity, group loyalty, and who is doing the talking than on a coherent set of reasoned positions, and the strongest force in it now may be how much each side dislikes the other.
Let's begin with what ideology is meant to be: an organised, reasonably consistent system of beliefs about how society should be arranged, the kind of thing where knowing one of your positions lets someone predict the rest. The foundational finding in political psychology is that most ordinary people do not have one. In a landmark study, Philip Converse showed that the mass public's political attitudes have low constraint (positions that ideology says should go together often do not) and low stability (the same person, asked the same question two years apart, frequently gives an unrelated answer). He called many of these answers "nonattitudes", responses generated on the spot rather than drawn from a settled view (Converse, 1964).
That claim has been fought over for sixty years, and the fight is instructive. Christopher Achen argued the wobble was mostly an artefact: survey questions are vague, and once you model that measurement error, ordinary people look far more coherent than Converse made them seem (Achen, 1975). It is a real and partial rescue. But the most thorough modern reassessment lands closer to Converse: across decades of data, most citizens are, in the authors' phrase, "innocent of ideology", organising their politics by group and party rather than by abstract principle (Kinder and Kalmoe, 2017).
So if it is not coherent doctrine, what is people's politics made of? Several things, none of them the tidy ideology of the pundit. First, left-right does capture something real. It is a meaningful psychological dimension, and where people place themselves on it correlates with dispositions like openness, threat-sensitivity, and a preference for order (Jost, 2006). Second, the label and the content come apart. Many people who call themselves conservative hold operationally liberal positions on specific policies, because the self-label is a piece of cultural identity, not a summary of their views (Ellis and Stimson, 2012). Third, and most tellingly, identity often swamps content outright. In a striking set of experiments, people's support for a welfare policy was driven almost entirely by which party was said to back it, overriding both the policy's actual details and the person's own stated principles, and they confidently denied being influenced while assuming their opponents would be (Cohen, 2003).
Put those together and the modern picture comes into focus. The central feature of contemporary politics may not be ideological distance at all but affective polarisation: partisans increasingly dislike, even loathe, the other side, and this animosity is only loosely and perhaps artificially tied to actual policy disagreement (Iyengar, Sood and Lelkes, 2012). People have not necessarily moved further apart on the issues. They have come to hate each other's teams. That is a story about identity and feeling, the territory of L3-07, far more than about belief systems.
The disputes here are live and they matter for how confidently anyone should talk about "ideology".
The first is how coherent the public really is, the Converse-versus-Achen question. It has never been fully settled, because it turns on a hard methodological problem: telling genuine instability of opinion apart from sloppy measurement of a stable one. The weight of modern evidence favours the low-coherence view (Kinder and Kalmoe, 2017), but the honest position is that some of the apparent incoherence is real and some is the questionnaire's fault, and reasonable researchers still draw the line in different places.
The second is the psychology of left and right. The best-known claim is that conservatism functions as "motivated social cognition", linked to needs for certainty, order, and the management of threat, supported by a large meta-analysis (Jost et al., 2003). The correlations are real, but they are mostly modest to moderate, and their interpretation is genuinely contested. Critics note that the research focused asymmetrically on explaining the right, as if conservatism were the condition needing a psychological diagnosis, and that rigidity and intolerance can be found at both political extremes. The finding is best read as "these dispositions lean one way on average", not as a clean account of why people are conservative, and certainly not as a verdict on anyone's character.
The third is how much of the polarisation is real disagreement versus tribal feeling. One camp argues Americans' actual policy views have pulled to the extremes; another argues most people remain fairly centrist and what has grown is sorting and animosity, not issue extremity (Iyengar, Sood and Lelkes, 2012). The evidence leans toward affect outrunning ideology, but how much of each, and whether the same holds outside the United States, is unresolved.
Almost all of this evidence is American, drawn from a two-party system and decades of one country's election studies. In multiparty democracies the picture of identity, sorting, and polarisation looks different, and the very concept of a single left-right line travels poorly. The findings are also about populations on average, not individuals: that the public shows weak ideological constraint does not mean no one is a principled, consistent thinker, only that such people are rarer than the political conversation assumes.
Is rising hostility between political camps driven by genuine value conflict or mostly by identity and the media that feeds it? Does the low-coherence finding hold in multiparty systems, or is it an artefact of two-party politics? And can the dispositional tilt of left and right be separated cleanly from the cultures and identities people are raised into?
The usable and uncomfortable core: your audience is less ideologically coherent than it looks, identity and group loyalty move people more than policy argument, and the strongest force in the room is often how the two sides feel about each other. This is powerful knowledge and it sits right on the line this series keeps drawing, because the same insight that lets you communicate across a divide also lets you deepen it. Stoking tribal animosity works, and it is the move that corrodes the shared ground a society and a market both need.
A brand that wades into politics should know what it is actually touching. Audiences rarely reason from a worked-out ideology; they respond to identity cues and to which tribe a message seems to belong to, and the same statement will be read as "us" or "them" before its content is weighed (Cohen, 2003). The honest implication is humility about what a political stance does: it is received as a tribal signal, not as an argument, and it will be loved and loathed on those terms regardless of how carefully you reasoned it. If a stance flows from a value the company genuinely holds and will stand behind, that can be worth the cost. Taking one to chase a segment, while assuming people will weigh your reasoning, misreads how politics actually lands and tends to anger as many as it wins.
This is the most powerful and the most dangerous use in the whole series. The evidence says the reliable way to move people is through identity and affect, and that negative campaigning, messages that paint the other side as contemptible, dependably hardens partisan animosity (Iyengar, Sood and Lelkes, 2012). It works. It also degrades the thing being competed over: affective polarisation eats the common ground democracy runs on, and a campaign that wins by teaching people to loathe their neighbours has changed the country it then has to govern. Two cautions, as elsewhere. First, mobilising belonging and pride does not require manufacturing an enemy, and the choice to stoke hatred is a choice with a democratic cost, not a technical necessity. Second, the research on left-right dispositions is not a licence to diagnose your opponents as fearful authoritarians; the effects are modest, the science is contested, and treating half the electorate as psychologically defective is both unsupported and self-defeating.
Two facts shape the work. Because the public is more cue-taker than ideologue, trusted messengers and framing matter more than the abstract merits of a policy: who delivers a message and which identity it speaks to can decide its reception more than its content (Cohen, 2003). And because affective polarisation erodes the shared reality that governing depends on, public communication should avoid feeding the us-them spiral and lean on cross-cutting, superordinate identities, the believable common "we" that lowers the temperature. The same evidence that shows how division is manufactured is the manual for cooling it.
Three habits. First, hold your own "ideology" a little more loosely: a good deal of it is identity and cue-taking too, and the experiments find we are blind to this in ourselves while seeing it plainly in opponents (Cohen, 2003). Second, when a position instantly appeals or repels you, pause to ask whether you are judging the policy or the team that proposed it. Third, watch how good it feels to despise the other side, because that feeling is the most manipulable thing in your political life, and the people who want your vote or your outrage know it.
.png)