These are some thoughts about what makes a good editorial. They are supposed to be the start of a discussion, which I hope we will continue via email. I also hope to start a similar discussion about the op-ed page and the Sunday section. Most of this is nothing new, and none of it is meant as criticism of the page now. And none of it is a hard-and-fast rule. No doubt many brilliant editorials violate all of these guidelines except the big one I start with. The essential quality of any piece of opinion journalism is intellectual honesty. Intellectual honesty is to an editorial or op-ed what factual accuracy is to a news story: the gold standard. Obviously editorials should be accurate, and news stories should be honest. But a dishonest argument rots an opinion piece even more than a factual error does. Intellectual honesty includes things like:• Sincerity (do you truly believe what you’re writing?)• Coherence (have you thought hard and thought it through?)• Consistency, or Sauce for the Goose (does this opinion fit with your opinions on other matters? are you prepared to apply the same logic in circumstances where it leads to a less agreeable conclusion?)• Skepticism (have you considered the counterarguments?) An unsigned editorial is a particular type of opinion piece. What is the point of this somewhat odd genre? To express the view of the newspaper, obviously. Except that the newspaper, in all its other pages, prides itself on having no view. So in a way, it makes even less sense for a newspaper to produce an official or semi-official opinion on the issues of the day than for, say, an auto company or a restaurant chain to do so. These businesses at least have no business interest in being thought unopinionated. And why should anyone care about an opinion expressed in an unsigned editorial? Many people don’t care. The editorial page is often one of the least-read parts of a newspaper. It is published to some degree as an indulgence of tradition. Our goal should be to make editorials more valuable to the reader and therefore to the company. A signed opinion piece can draw readers with its style or with its author’s reputation. It may be possible for an unsigned editorial to have similar appeal, but it certainly is harder. A better approach (or so it seems to me) is for editorials to offer something clearly different from signed opinion pieces. Something readers also cannot get from news stories and analysis. What is that something? Call it maybe “order in the universe.” The editorial page should be a place readers want to come to for a sense (inevitably false, but genuinely attempted) that someone has it all figured out. That despite the chaos of the news and the cacophony of the op-ed page, a general understanding of the way the world is and a coherent opinion of how it ought to be are both possible. Editorials have unsung advantages over news stories, and even so-called news “analysis”: they are allowed to state facts as a given, without the citations and quotations and counter-quotes required by the conventions of news writing. And, of course, they are not required to come to a screeching halt before tumbling into a conclusion. Even their brevity can be an advantage, if it forces concision. In short, a good editorial is one that explains and clarifies an issue in the news. It can, in fact it should, have an opinion. But ideally it also should give readers the understanding they need to disagree intelligently. Here are some of the specific elements of a good editorial: 1. A framework of analysis. You can’t always cover the big picture when you’re reacting to some specific controversy or event, but you can try to hint at it. And you should have it in mind whether you are able to discuss it or not. Over time, the editorial page should have a coherent worldview in which, eg, our view on driver’s licenses for illegal immigrants fits in with our position on immigration policy in general, which in turn fits with our views on trade, on the minimum wage, and so on. This is a matter of intellectual honesty and discipline, but it also is a marketing tool. The world is a puzzle, and we’ll show you how all the pieces fit together. 2. An intellectual problem. (“Intellectual” just means a problem that needs to be solved by reasoning. It needn’t be hoity-toity.) • The problem should arise from the news, or from some larger public concern. No doubt there are brilliant editorials awakening readers to issues that have been wrongly ignored. But the usual effect of trying to make the readers care about an issue in order to tell them what to think about it is the opposite of awakening. • Disagreement. If no one or almost no one disagrees, it also is probably not a good subject for an editorial. Of course events sometimes require commemoration but defy argument. Tragic natural disasters, for instance. But even when unavoidable, such editorials are likely to be boring. 3. An analysis and argument. • The argument should be intellectually honest, as per above. • It should be consistent with what we have said before on this issue and others. If it departs from previous editorials, this should be acknowledged and explained. There are only two possible explanations: circumstances have changed, or we have decided that we were wrong. If it is the second, the editorial should say so and say why.• The argument should address the core of the problem, not some side issue—even if in practice the side issue could be determinative. In an editorial, there is no point in winning the argument on a technicality. The point is to help people understand. Editorials about small aspects of a larger problem should illuminate the larger problem. • “Polls show that people overwhelmingly support our position” is not an argument. The goal of an editorial is to affect public opinion, not to reflect it. In fact, if people overwhelmingly support the position we think is correct, there may be no need for an editorial. The work has been done. On the other hand, we will often want to take positions that are not supported by public opinion, and we don’t concede that this makes our argument invalid.• “Experts support our position” is also not an argument. Ditto organizations of almost any sort. If someone or some group has a persuasive point, or useful information,, there is nothing wrong with borrowing these and citing the source. But the unadorned fact that someone shares your conclusion rarely belongs in an editorial. Making a decision based on endorsements is a (sometimes necessary) substitute for thinking it through yourself. An editorial should think it through. . 4. A conclusion. The problem should be solved, at least to our satisfaction. • There should be a “should” or a call to action. • The call to action does not have to be directed necessarily to a person who actually reads the Los Angeles Times. But it should be aimed at someone within our general sphere of influence, who might plausibly take the action being recommended. (The Economist is always calling on the Taliban to adopt capitalist democracy, and so on. Avoid that.)• The following, or variants, do not count as conclusions:o “There are no easy answers” or “only time will tell” or the right answer “remains to be seen.” Any question with more than one easy answer probably shouldn’t be the subject of an editorial in the first place. o Delay, further study, or anything involving a panel of experts.o The parties involved should lay aside their differences, or negotiate in good faith• “Fundamental reform is needed” may sometimes be the correct conclusion, but should be used sparingly, because it’s anticlimactic.. Think about changing topics.. 5. Clarity over cleverness! Editorials are the voice of the group, or even the institution. And their goal is understanding and persuasion on difficult issues. So the structure should be straightforward, and the voice should be fairly consistent across different authors.. Don’t deny yourself a really good joke or pleasing turn of phrase. But show restraint. Some specific small tips:• It isn’t essential that every editorial be structured as problem/argument/conclusion, but that’s not a bad general model. • Avoid backing in to the subject with a discursive lead about the weather or the last time this regulatory body was enmeshed in controversy.• Avoid making your real message the “but” of a sentence that starts out saying the opposite. (Two bullets up, see an example of what not to do.)• Don’t assume that the readers are all familiar with the topic, but don’t assume either that they need to be made familiar with it in all its glorious detail. 6. Concision. o Try to avoid a lot of who-struck-John about how the committee voted and who might be replaced and how it used to work before the notorious 1954 reforms. Write about the substantive issue, not about the process.o Avoid color and human-interest. You need every line you’ve got to explain why drug-testing regulations are excessive, or inadequate. You do not have space for Mary Doe, 47, of San Luis Obispo, who took a pill and almost died, or was cured, or whatevero Use the editorial writer’s privilege: you don’t need to buttress every assertion with a quote from someone else. The premise of an editorial is that you, the writer, are the voice of authority. If the reader doesn’t trust your judgment more than that of Helene Knightly, chairwoman of Citizens for Cleaner Mailboxes or Professor Roger Jordan of the University of Manitoba Pocket Nuclear Bomb Lab, there’s no point. 6. Topics. We publish three (or so) editorials every day. Ask yourself; is this one of the three most important things a responsible citizen needs to be thinking about tomorrow? Or if not, is it one of the three most interesting things an active mind wants to be thinking about? If neither, why are we writing about it?• Unless it is something like the 9/11 Commission, the fact that a group has issued a report is rarely a good topic. Especially if all we have to say is that the report is interesting and/or we agree with it. • Ditto the meeting of some group, or the publication of an article in a trade journal.• Ditto legislation introduced that has no chance of passage.• Ditto the intermediate stages in some endless bureaucratic saga, • National and Local: try to have your cake and eat it too. Look for California topics with national implications, and national topics with a specific or disparate impact on California. 7. Counterarguments. Give the case for the other side, as objectively as possible, and explain why it’s wrong. This is not, or not just, a matter of fairness. It is to make your own case stronger. A piece that makes it seem as if no reasonable person could possibly disagree is unconvincing when the reader knows that there actually are reasonable people who do disagree. Or do you disagree?