Debate Policy

There are three reasons I’ll debate people.

  1. I want to. I judge it’s a good idea. It looks valuable to me. (See this explanation of how I allocate my attention.)
  2. You request a debate according to my debate policy, below.
  3. You use my Paths Forward Policy.

Why not just use my judgment and that’s it? That’s what other public intellectuals do. But that allows for bias and other errors. I may judge incorrectly. My debating and paths forward policies are safety mechanisms in case of incorrect judgment. They allow errors to be corrected so that I don’t stay wrong. For more info about the problem of bias, see Using Intellectual Processes to Combat Bias and the articles it links to.

If I don’t respond to something, I’m not taking it seriously, I express disinterest, I’m skeptical of the value of the topic, I’m skeptical of the quality of your knowledge, I’m not listening, or I’m otherwise wrong about anything, then you can request a debate according to this policy. The debate policy overrules my initial judgment.


  • Pick a name and stick to it for the whole debate.
  • Have a website under that name with 20+ articles, totaling 20,000+ words. Link it.
  • State that, in your honest and considered opinion, you have skill and knowledge about the topic and something important to say.
  • Give your word of honor that you will continue until we reach an agreement, you state a chain of 5+ impasses, or I state 3+ chained impasses.
  • Debates use asynchronous writing in English. We don’t schedule a time to both show up to debate. It’s not a real-time chat. Instead, we each write independently on our own schedule. Responding to messages within two days is standard but there isn’t a hard rule for timeline.
  • I normally discuss at my Critical Fallibilism Forum. Use it or suggest another forum with feature parity and easy exporting/mirroring.
  • I may decline a debate if I’m not interested in the topic. If I do, I forfeit the right to make claims about that topic contrary to your position that I refused to debate.
  • You may only request one debate at a time. And if a debate ends with chained impasses stated by me, you don’t get another debate unless you address the impasses or I decide to make an optional exception. There is no time limit to solve impasse problems, but don’t expect to ignore those problems with no solution.

To understand this policy, you’ll need to read my article explaining debates and impasse chains.

Exceptions may be available, especially for other types of content creators who don’t write online articles, but this policy is what I’m guaranteeing I’ll follow even if I have a low opinion of you. Anyone who wants to can meet these requirements.

Requiring 20 articles helps filter out unserious people who haven’t finished some intellectual projects. It also lets me review their past work to see what they’re like. And it prevents people from easily switching names and pretending to be a new person. Books, academic papers or other published writing are fine too. Videos or podcasts work for most purposes, but having a written debate with someone who doesn’t write is problematic, so I don’t guarantee to accept those. Articles or other past works must be at least a little bit relevant, e.g. by covering intellectual topics.

Post-Mortems and Second Debates

Suppose we debate an issue and you concede. Will I have a second debate with you? Yes, but not about any topic. I’ll want to address what went wrong on the first issue. Why did you make that mistake? What are the underlying errors and what further consequences do they have? I think it’s important to post-mortem errors. Errors are usually caused by a systematic problem that affects many things.

Low Requirements

My debate requirements are designed to be minimal so that anyone with something important to say (in their own opinion) can meet them. That means, if I’m wrong, anyone can correct me. I’m open to debate. The minimal requirements help filter out crap and protect my time, but still allow anything good to get my attention.

I’m offering real attention here: meet the requirements, request a debate, and I’m agreeing to spend time and effort on it to reach agreement or else I’ll state a 3+ impasse chain. (I’ll commonly do 5+ impasses but, to protect my time, I’m only guaranteeing 3.)

I’m making a promise here, which obligates me to spend time, which is open to the general public. That’s a big deal. The requirements are around the minimum I think can reasonably protect my time from abuse and wastes of time.

I’m putting my reputation on the line, and dealing with anyone from the public, because I value discussion and debate, and because I want it to be possible for me to receive external corrections. If I’m biased or making some other mistake, I want it to be possible for people to help me. Most intellectuals lack robust ways they can be corrected, and pretty much entirely prevent corrections from the general public. I’m trying to do better.

I also get a lot of informal, vague, usually-anonymous debate requests. When I make a judgment that it isn’t worth my time, sometimes they attack me as unwilling to debate. This policy will help deal with that problem. I can link them to it, or I can even say nothing because this information is prominently available. I think this policy, rather than being a burden, will help me save time while increasing the quality of discussions I participate in.

This debate policy itself can be criticized and debated if you see a problem with it. I’m highly interested in that. Also, if you have a rational debate policy of your own, I’d also be interested in seeing it and how it differs from mine, and I’d consider discussing using your policy.

World’s First Rational Debate Policy

I’m unaware of any public intellectual with a debate policy like this. This policy was developed as part of my efforts to understand and clarify what I do differently than other intellectuals. The key idea is being open to error corrections from the public rather than acting in a way where if I’m biased, mistaken or irrational, I’ll stay that way even though people knew better and were willing to help me.

I think there’s a major problem with the world and with intellectuals and debates in particular. The problem is their lack of publicly documented, structured policies for dealing with debate, criticism and questions. Publicly writing down the discussion rules is crucial for accountability and so potential discussion partners know in advance what to expect.

This problem is getting in the way of scientific and philosophical progress. People working in those fields are not rationally open to debate. They get things wrong and then stay wrong. They judge what discussions they want (with who, and in what circumstances, e.g. with short time limits and no intention of reaching a conclusion), and if they are biased or mistaken then there’s no way to fix it. Mistakes and biases are very common and should be expected, so lacking good ways for them to be corrected ruins everything.

There are many errors out there which someone knows about but which are very hard to fix anyway. If they were easier to fix, a lot more errors would be fixed. That would make science, economics, politics, psychology and other fields much more effective.

My policy is both a practical tool and a demonstration of how intellectuals should do things.

My debate policy prevents me from doing many intellectually bad things. It’s better not to bet my intellectual career on my ability to make good judgments and avoid bias. While I believe I’m good at that, no one is perfect, and there’s no reason to rely only on my own personal, private skill and integrity.

Other intellectuals need not have an identical policy, just a reasonable policy with robust anti-bias features. In particular, more famous people will need to protect their time more vigorously because so many people want to talk with them. Two of the main techniques for that are proxies (representatives who debate for you) and pre-written answers (you or your proxies can link stuff instead of rewriting it). My material on Paths Forward explains more about this. Another factor that reduces the time use is winning your debates quickly and clearly. And anyone creating a debate policy can come ask questions at my forum; that’s the kind of thing I’d like to help with.

Informal Discussion

Informal debate and discussion is fine. I do that when I want to. But I might stop responding without explanation. That means I didn’t judge it a good use of my time and attention (or I saved it for later). If I haven’t responded after three days, feel free to ask what’s going on or request a debate using my debate policy. If you think I should respond to something, but I don’t, this debate policy is the solution.

Informal discussions have no clear stakes or consequences. That’s OK. People can learn and do some parts of structured debate. But don’t try to hold me responsible for answering your arguments in an informal discussion. If you want me to take responsibility for answering arguments, use my debate policy.

The distinction between a requested debate and an informal chat is important. People try to have it both ways, avoid the responsibility of asking for a serious debate, but, at the same time, try to demand that I be responsible for answering all their questions and criticisms. Now, either you used my debate policy or you didn’t, which will bring greater clarity and make it easier for me to not respond to informal messages.

Informal discussion works best when people are being friendly, cooperative, have high opinions of each other, and see lots of value in the other guy and the conversation. Informal discussions also work well when there isn’t much concern about bias or dishonesty (easy stuff, non-problematic areas, low ambition goals).

The more you want to do something hard and important, the more worthwhile it is to request a debate and use documented, structured methods like debating to agreement or a chain of five impasses. The more it matters, or the more resources are being invested, the more value there is in making it hard for bias to ruin everything. Formal structures make it harder to be dishonest, fool yourself, be biased, etc.

A common, good reason for informal discussion is if you’re a beginner. An initial, exploratory discussion is often informal. If you’re feeling each other out, and finding out what the other guy thinks, and what you agree and disagree about, that’s a good time for informal discussion.

Informal discussion should be able to transition to formal discussion at any time (unless someone is a beginner and says they just aren’t ready). People often make it ambiguous whether they will or won’t transition, and won’t give straight answers about it, which is bad. A common cause of that ambiguity is that they don’t know what a rational, structured discussion is or how to do it (and they don’t admit that or ask about it). My article explaining debates and impasses will help with that.

With me, transitioning to an actual debate is dependent on my debate policy requirements being met. Policies should be written in advance and used longterm, not made up ad hoc in relation to a particular discussion. They should be general policies which are exposed to criticism, not excuses made up to ignore a specific individual.

I believe people (including myself) are too careless overall and ought to use well-structured interactions more. When there are problems, structured approaches are more effective per effort spent. Problems are common and underestimated.


I’m a public intellectual. I run a discussion forum to allow critical feedback and discussion. I study ideas, particularly philosophy, and try to figure out what’s true. I’m open to debate and discussion with the public. I make judgment calls about how to allocate my attention. I may make mistakes. I may be biased. If I don’t engage with an idea, and you think I should, you can informally ask about it, or you can use my debate policy. If my debate policy doesn’t solve the problem, as an extra backup, you can use my older Paths Forward Policy. These policies make my errors fixable. The errors of other intellectuals are largely unfixable because they don’t have policies like this.

I, Elliot Temple, first published this debate policy in 2019. I’ve made clarifying edits (most recently in 2022) but no significant changes.