This is a revised version of a Friendly Atheist post.

Whether the secular movement flourishes will depend on how well it can carry on a dialogue with its religious friends and foes. It’s through conversation that we will change our public image, negotiate political gains, and form alliances on specific issues. It’s conversation that will determine whether our numbers expand.

But the stakes are drastically higher than that. In an increasingly interdependent world, our ability as human beings to resolve disputes verbally is the only abiding safeguard against violence, against polarization, against seeing informed democracy degenerate into shouting matches.

Why, then, are we so averse to talking to those with whom we disagree? Why do dialogues fail? Why are we so rarely persuaded? If we can understand why we’re so bad at resolving our differences, maybe we can do a little to change that fact.

Greta Christina noted at the Secular Student Alliance Annual Conference that “arguing about religion is not a waste of time.” Although debaters themselves may be bewilderingly obstinate in the heat of battle, onlookers remain surprisingly receptive to new ideas. This suggests that the best way to promote atheism is to argue before a large audience.

However, there will always be cases where we need to get a point across to someone directly. Most interactions between theists and nontheists will be in small groups, or one-on-one. And it is these direct chats that are ideal for reaching out to those individuals who are least informed about atheists, least inclined to waste time on the Internet perusing atheist blogs or YouTube debates. Aside from the occasional prime-time atheological sound bite, person-to-person discussion will be what tends to define our image and plant the earliest seeds of doubt. So the question retains its urgency. What makes discussion break down? Judging by the debates I’ve seen and participated in, there are two main culprits.

I. Our discussions aren’t collaborations.

We see debate as an opportunity to defend ourselves, attack another position, fight for dominance and power and respect. We see it as something either I win or you win, not as something both sides succeed or fail in together. Our discussions are antagonistic because we enjoy being right, we take pride in the strength of our reasoning — and we feel shame and dismay when we are proven wrong.

Why don’t we feel the happy excitement of a new discovery when someone corrects a mistake of ours? Because the discussion has been framed as a competition, not as a mutual pursuit of deeper understanding. It’s not enough to make overtures of camaraderie; even exchanges between the best of friends can become bitter squabbles if either side becomes too invested in who is right, overshadowing what is right. A healthy discussion should feel like trading recipes or researching a common interest; each side should keenly (or casually) desire to understand the other, to learn and not just to teach. We have plenty to learn from the religious; if nothing else, we have plenty to learn from them about other religious people, and how to better reach out to them and find common cause. This sort of cheerful shared curiosity must drive discussions. A religious exchange motivated only by the evangelical desire to banish ignorance, however well-meant, is doomed to failure.

Remember: It’s not fun to be wrong. Always put yourself in the other person’s shoes. Sam Harris suggests that our brains process falsehoods with an experience akin to disgust. It’s a delight to encounter beliefs you agree with. It’s a pleasure to hold beliefs which seem to make sense of your experiences. At the same time, it’s not easy living with contradictions and lacunae; cognitive dissonance is unpleasant. A discussion of deeply held beliefs is more like an affective rodeo than like trading indifferent data points.

What’s the take-away? Be nuanced. Be moderate. ‘Nuanced’ doesn’t mean ‘complicated’. Express your views clearly and concisely, but follow up a negative comment with a positive comment, to mitigate the inevitable emotional sting while leaving the intellectual content intact. And ‘moderate’ doesn’t mean ‘wishy-washy’ on points of substance. You will come across as moderate if you are willing to make concessions, ask sincere questions, compliment the other side, and admit your own shortcomings, even if these sugar-coated asides are irrelevant to your central argument, and even if the argument itself is a radical one. A very little friendliness and good humor goes a very long way. Indeed, just coming across as a nice person tends to do a lot more to attract skeptics and allies than even the most devastating logic. And, of course, it leads to way better conversations.

II. Our discussions aren’t specific.

When we speak of persuading people about atheism, we aren’t really speaking about some isolated bit of theology. Atheism here is code for a very broad and complex world-view, rich in methodological and theoretical commitments. This is our long-term strength, because it provides something with which to fill the epistemic void left by deconversion. But it’s our short-term weakness, because it makes our discussions too all-or-nothing. It forces us to demolish a towering world-view in one fell swoop, when we’d be better off chipping away slowly at the foundations.

We are at our strongest when we can debate particular, relatively weakly held claims. This allows us to show off the power, the richness, the appeal of scientific and philosophical reasoning — without drowning out that appeal in the backlash of immediate outrage. Why leap to debate God when you can sharpen Ockham’s razor first on ghosts, or homeopathy, or climate change denialism? In this way you can teach the intellectual methods motivating atheism, which are in any case far more important and life-saving than atheism itself. If the methods manage to take root, they will do more to eat away at dogma from within than any argument made by another ever could.

Sticking to specifics makes it easier to convince the other side of some particular claim; and even if the issue is a trivial one, there is much value simply in the act of learning to inquire skeptically and revise one’s views. Moreover, it is on these innumerable factoids, far more than on deep and unshakable moral convictions, that theists and atheists disagree.

The same, surprisingly, is true of American liberals and conservatives. If a discussion were had on interpreting some specific data or theory, the dialogue could advance and both groups could come away better educated. Because the debate is instead halted at incredibly broad topics — we don’t debate some claim about abortion, we debate abortion itself — no progress is made. Instead, both sides fall into the well-rehearsed rituals of their cherished established beliefs, camouflaging a mass of negotiable factual disagreements as a monolithic dispute of irreconcilable values. This is how sides in a dispute fossilize into factions. There are indeed real conflicts over values — but these are as dust compared to the mountains of cost-benefit analyses, empirical generalizations, and causal interpretations on which the two sides would first diverge. When the discussion stays in vague, well-trodden territory, we do nothing but go in circles.

How does this work in practice? Paul Veyne, in “Did the Greeks Believe in Their Myths?”, quotes the missionary Évariste Huc’s account of Tibet:

We had adopted a completely historical mode of instruction, taking care to exclude anything that suggested argument and the split of contention; proper names and very precise dates made much more of an impression on them than the most logical reasoning. When they knew the names Jesus, Jerusalem, and Pontius Pilate and the date 4000 years after Creation, they no longer doubted the mystery of the Redemption and the preaching of the Gospel. Furthermore, we never noticed that mysteries or miracles gave them the slightest difficulty. We are convinced that it is through teaching and not the method of argument that one can work efficaciously toward the conversion of the infidel.

Set aside the manipulative evangelism and notice the lesson in psychology. Even the best arguments tend to fail when they’re pitted against the deepest convictions of a competing religion, cemented by habit and guarded by stereotyped, mantra-like counterarguments. Non-argumentative, factual accounts, on the other hand, slip through the cracks quite easily. This is not simply because they are framed as indisputable facts, nor because they are too idiosyncratic and exotic to brook easy retort. It is because they are friendlier, less confrontational. They invite listening and learning, rather than intellectual combat.

Such a technique, of course, can easily be abused. It merely replaces one authority with another. We want to encourage productive and dynamic dialogues, not just a one-sided soliloquy. Yet if we want the communication without the rancor, we must make argument its own reward. It must be a happy act aimed at real discovery and mutual enrichment.

If our only goal were to make everyone believe the same thing we believe, we’d be better off relying on the rhetorical power of facts and figures and jargon. But orthodoxy, even scientific orthodoxy, isn’t our goal. Our goal is a world of open-minded critical thinkers, of people who have made a habit of questioning, and of seeking, and of imaginatively advancing the human discussion in science and in politics. However you envision secularism’s end-game, no path is possible in the absence of civil and productive dialogues between people with radically different world-views.

This is not to say that such dialogue is easy. It is to say that we have no choice. We have to talk.