How to talk across political and value divides

Rarely through a single conversation, but yes — over time and with the right approach. **Broockman & Kalla (2016)** found that **deep canvassing** — non-judgmental, personal conversations lasting around 10 minutes — reduced transphobia measurably and durably, even weeks later. The operative word is _deep_: superficial debate reinforces existing views. What moves people is feeling genuinely heard and then encountering a question or story they cannot easily dismiss. One conversation plants a seed; a relationship is the soil it grows in.

**Deep canvassing** is a conversation technique developed by political organizers and studied by **Broockman & Kalla (2016)**. Instead of presenting arguments, the canvasser asks the other person to share a personal story related to the topic, listens without judgment, and then shares their own story. The result is a **genuine emotional exchange** rather than a debate. The prejudice reduction observed in the study was roughly equivalent to the attitude change seen over 25 years of cultural shift — achieved in about 10 minutes. The mechanism mirrors what **Allport (1954)** identified in the contact hypothesis: real, equal-footing contact dissolves the abstract 'enemy' into a specific human being.

Because argument signals threat, and **threat triggers defensiveness**, not reflection. Justin Lee's work in *Talking Across the Divide* makes the point clearly: the goal is not to be right but to be heard — and strategic dialogue creates the receptivity that argument destroys. When someone feels attacked, they protect their identity by doubling down. When they feel respected, they can afford to examine their own views. The same information lands completely differently depending on whether it arrives as a challenge or a discovery. Argument gives you the satisfaction of saying it; dialogue gives you a chance of it actually landing.

Two moves help most. First, **challenge one branch, not the whole tree**: instead of attacking the underlying worldview, find one specific claim that is factually testable and focus there — Lee calls this separating the branch from the trunk. Second, **present corrective information as your own discovery**, not their error: 'I used to think that too, and then I came across...' rather than 'that's not true.' The second framing removes the sting because it doesn't require them to admit they were wrong. They can update quietly, which is the only way most people actually update.

**Allport's contact hypothesis (1954)** is the finding that prejudice between groups decreases when members of those groups have direct, personal, equal-status contact. It was originally tested on racial and ethnic divisions, but the principle extends: the more someone sees their political 'opponent' as a specific, three-dimensional human being with comprehensible reasons for their views, the harder it is to sustain cartoon-villain stereotypes. **Broockman & Kalla's (2016)** deep canvassing research is effectively the contact hypothesis in practice — the conversations worked precisely because they replaced abstract group-think with personal story.

Treat them as the **understandable protagonist of their own story** — a frame from Justin Lee's *Talking Across the Divide*. Their beliefs didn't arrive from nowhere; they came from experiences, communities, and identities that made those beliefs feel not just plausible but obligatory. Asking 'how did you come to think that?' is more useful than asking 'can you defend that?' The first is curiosity; the second is a challenge. You do not have to agree with the journey to understand it, and understanding it is the only platform from which real dialogue is possible. Our guide on [how to disagree without damaging the relationship](/en/blog/disagree-without-damaging-the-relationship) covers the mechanics in more detail.

Then arguing evidence will not help much, because evidence isn't what the belief is doing. **Boghossian & Lindsay** identify this as the key diagnostic: when a belief functions primarily as a **moral or social credential** — signalling loyalty to a tribe, affirming a community identity — presenting contradicting facts feels like an attack on the person, not a correction of a claim. The conversation that can help is one that acknowledges the underlying value ('I can see this is about fairness for you') before quietly separating the value from the specific policy conclusion. Never ask someone to abandon their tribe; ask them to consider whether this particular position actually serves the value they care about.

Echo chambers do more than reinforce existing beliefs — they make the _other side_ seem more extreme than it actually is. Justin Lee argues that **echo chambers cause polarization** directly, not merely through confirmation bias: when you only hear your own side's characterization of the opposition, you form a distorted picture of what that opposition actually believes. That distortion makes conversation feel pointless before it starts. Breaking the echo — even slightly, through one trusted relationship across the divide — recalibrates your mental model of who 'they' are. This is part of why the relationship itself matters as much as any single conversation.

**Synthesize identities rather than forcing one side to yield** — a frame from Daniel Shapiro's *Negotiating the Nonnegotiable*. The goal is not agreement on policy; it is a shared relationship that contains the disagreement without being defined by it. Practically: agree on which topics are off-limits for now, make explicit what you value about the relationship beyond politics, and refuse to let any single conversation become the verdict on the whole person. Our guide on [how to have a difficult conversation](/en/blog/how-to-have-a-difficult-conversation) covers how to open those conversations without triggering a shutdown.

Because **cultural legacy shapes communication defaults invisibly**, as Malcolm Gladwell documents in *Outliers*. What reads as respectful directness in one context reads as aggressive confrontation in another; what reads as thoughtful restraint in one culture reads as evasive dishonesty in another. Before you interpret someone's conversational style as bad faith, it is worth asking whether their style was simply shaped by different norms. This doesn't mean all styles are equally productive — but it does mean that matching your approach to theirs, rather than insisting on your own defaults, is often the fastest path to actual contact.