As a writer I wear two hats. I am a staff writer for the New Yorker magazine, where I have been under contract more or less continuously since 1996. I also do public speaking, based on my second career as the author of two booksThe Tipping Point and Blink. Over the past four or five years, I have given talks to corporations, trade associations, conventions of one sort or another, colleges, think tanks, charitable groups, public lecture series and, on one occasion (arranged by my mother) my old high school. For some of those engagements, I have been paid. For those given to academic and charitable organizations, I generally have not.
Most of the time, these two hats complement each other. It was because of my work as a New Yorker writer that I was able to get a contract to write my books. It is because of my books that I have gotten speaking engagements, and it is, in part, because of my books and my speaking that some people have discovered me in the New Yorker. But I recognize that there is also the possibility that these two roles can come into conflict, as is always the case when someone has to serve two different constituencies. This note is an attempt to talk about that possibility, and to think through its implications. I will warn you that what follows is quite long. It is long because the question of potential conflicts of interest is a fraught and difficult subject for journalists, and I think it is worth taking seriously.
The majority of my time and effort is given to my job at the New Yorker. I have a contract with the New Yorker to produce a certain number of words every year (usually 40,000 to 50,000). I am forbidden, with certain small exceptions, to write for other magazines. The choice of what I write about is largely up to me, although I often take suggestions from editors at the magazine. I do not cover a specific subject area or beat, and so far the New Yorker has placed no restrictions on what I might write about.
Twice, since joining the New Yorker, I have taken an unpaid book leave. Over the past five or six years, I have also occasionally given paid speeches. This is not unusual. Writers of business booksor, indeed, general interest books of many kindshave become, in the past few years, regulars on the speaking circuit. A number of New Yorker writers are represented by speaking agents, and it is a safe bet that virtually every author on the New York Times non-fiction bestseller list also does paid lectures. The increasing overlap between the speaking and writing worlds is a function of two things. First, speaking has come to be seen as a integral part of promoting a book. Second, the tremendous growth in recent years in conventions has created a greater demand for public speakers. I am represented in my paid speaking by the Leigh Bureau. (www.leighbureau.com). They handle the negotiations over my fee, and the terms of the engagement, expenses, schedule and so forth.
Sometimes my speeches are drawn from my books, The Tipping Point and Blink. On other occasions, I talk about some idea or article that I'm working on, or come up with entirely new material for the occasion. I accept speaking engagements because I enjoy public speaking and traveling, because it is financially rewarding, because I realize that it's an important way of promoting my books, and because, most of all, I believe there is real value in the public presentation of ideas. For example, all three of the long case studies at the heart of Blink were previewed, while I was in the midst of writing my book, before various academic audiences. In every case, the comments of audience members helped me re-shape and sharpen my arguments. The written word, to my mind, improves to the extent that it comes to resemble the spoken wordand the discipline of having to present my work orally has helped to make me a clearer and more succinct writer.
The list of clients who have engaged my services is eclectic. To name just a few, in the past few months I have spoken to a hospital group in Las Vegas, a conference of arts and civic organizations in Michigan, a computer storage company convention in Texas, a group of entrepreneurs and local business leaders in Silicon Valley, and a national meeting on homelessness in New York. The first three of those were paid. The last two were not.
I realize that at some media organizations what I do as a speaker would be prohibited. I used to work at the Washington Post, and there were a strict set of ethical guidelines, governing potential conflicts of interest. We certainly couldn't accept paid speaking engagements. Nor were we allowed to accept gifts of any sort. (When the inevitable bottles and baskets of fruit and chocolates came from public relations agencies every Christmas, we would hand the goodies back to the startledand delightedmessenger who brought them.) When we took sources out to lunch, we had to pay. The editor of the Washington Post, Leonard Downie, even goes so far as decline to vote, because he thinks that casting a ballot will compromise his objectivity. This is what might be called the "hard" position on conflicts-of-interest. Once, when I was still at the Post, Downie admonished the staff against attending a big pro-choice rally that was being staged one weekend in Washington, D.C. His position was that abortion was a highly divisive issue; the Post was a paper committed to covering the news fairly and objectively; and that if Post reporters were known to be marching in an abortion rally it would undermine that mission. (Right around that time, I attended a seminar at a journalism school and was asked about the Downie rule. I saidin what I thought was an obvious jokethat the reason for the policy was practical not ideological; that so many Washington Post were pro-choice that if they didn't prohibit people from going to the rally there would be no one around to put out the next day's paper. I still think that's funny. My editors did not.)
I have a great deal of respect for the hard position. I think that it makes a difference in a town as ideologically divided as Washington, that the head of the dominant paper sits publicly and firmly on the fence. That said, many reporters at the Postincluding me, at the timedid not agree with Downie's abortion rally ban, and nor did they (or me) follow his lead on not voting. A writer is a professional, and part of what it means to be a professional is that one is assumed to be capable of separating one's personal from one's professional life. A doctor who misdiagnoses you is not allowed to say, in his defense, that he had a fight with his wife that morning; nor can a doctor who has a personal aversion to homosexuality refuse to treat a gay patient. We rely, as patients, on the fact that a doctor can create a reasonably clear boundary between the various domains of his life. I think we can reasonably have the same faith in journalists.
I'm also not sure how much sense the hard position has for a writer in my position. One of the big reasons the Post has such strict rules on accepting gifts and paying for meals is that reporters cover beats. They have a specific, narrow field that they follow closely, and the intimacy that arises between reporter and source inevitably creates the potential for a problem. When I covered the health bureaucracy for the Washington Post, there were senior officials in the government that I talked to nearly every day. I talked to some sources more than I talked to my family. That's why newspapers are understandably so nervous about the potential for bias, because friendship can easily muddy the waters of objectivity. At the New Yorker, however, I don't have a beat. I write about everything under the sun. Most of the time, when I interview someone for a story, I never talk to them again. More to the point, the rules at the Washington Post were intended to protect the paper's status as an objective source of news. The perceptionand factof objectivity is central to the mission of a modern newspaper. At the New Yorker, I'm under no obligation to be objective. I'm only under the obligation to be fairand the difference between fairness and objectivity is considerable. The test of a newspaper article is that when a reader finishes reading it, he or she has no idea where the writer stands on the issues under discussion. That's objectivity. With fairness, the bar is a little lower. It is perfectly permissibleeven advisablethat a reader of a New Yorker article know where the writer stands on the issue under discussion. It is important only that we be fair: that we accurately and appropriately represent the ideas at hand. During the election, many of the New Yorker's political editorials were written by Hendrik Hertzberg. No reader could have any doubts about where Hertzberg's sympathies lay. He is a Democrat. He is friends with many Democrats. He was a speechwriter for Jimmy Carter. He consistently and powerfully argued against the Republican orthodoxy. Hertzberg could never cover the White House beat for the Washington Post. But he is a brilliant political writer for the New Yorker because he manageseven when his sympathies lie with one side of the argumentto be scrupulously fair. He does not misrepresent his intellectual opponents. He meetsand confrontsthem on their own terms. That's the truest test of a polemicist.
There's another complication with the "hard position", and that has to do with the difference between working for a media organization and writing a book. At the Washington Post, I was not responsible for the economic success of the newspaper. I was completely free to write whatever I choseso long as it was fairly reportedeven if that article resulted in, for example, lost advertising revenue. The same is true, in this context, of the New Yorker. The New Yorker has made a great deal of money running ads from automobile companies, in particular ads for SUVs. In January of last year, I wrote an article that was highly critical of SUVsand of people who bought themand no one from the business side of the magazine so much as said a word to me about what impact that might have on the magazine's bottom line. That is as it should be. We have a wall between the business and editorial sides of the magazine.
Writing a book, though, is a little more complicated, because that wall doesn't exist. A writer is required by his publisher not just to write the book but also to promote it, to wear both a business hat and an editorial hat, and that changes the writer's role substantially. In my upcoming book tour for Blink, I will do dozens of talks and readings at bookstores and companies and lecture series and civic groups, and doing that kind of promotion is considered part of the job. As a Washington Post reporter, if a company like Microsoft came to me and asked me to speak to them, Leonard Downie would have, quite properly, required me to say no. After all, as a reporter for the Post, I wrote about technology. That would create the appearance and perhaps even potential for a conflict. But what is one of the stops on my book tour? The Microsoft campus, because they have a series in which they bring in authors they think their employees might find interesting. And if I said no to that, my publisher would quite properly be annoyed. The only reason they published my book, after all, is that they want to sell lots and lots of copies of it, and they wouldn't be terribly impressed by any ethical qualms I might have about who I should and shouldn't promote my book to. The Downie principle is a rule that works for newspapers. It doesn't necessarily work for book publishing.
There is a second, very different, approach to the conflict of interest question, and it has been best articulated by Michael Kinsley, the editorial page editor of the Los Angeles Times, who formerly edited Harper's, the New Republic, and the online journal Slate. While at Slate, Kinsley inaugurated a practice, in sharp contrast to Downie, of having all his writers publicly declare their political preferences. Here is Kinsley's memo on the subject. It dates from the 2000 U.S. presidential election. It's worth reading in full:
Like many lunatic ideas, Leonard Downie's has a certain inner logic: If opinions are corrupting, just don't have opinions. Downie, executive editor of the Washington Post, is well known for believing thatin the service of objectivitya journalist in his position should not vote. Writing on the Post op-ed page a couple weeks ago, Downie went even further. He said he does not even allow himself the luxury of deciding whom he would vote for if he was into that sort of thing.
Many journalists (including me) find this excessive. Journalists are still citizens, with the rights and duties of citizenship. Journalists are also people, for the most part, and people naturally develop opinions about the world around them. This is not a capacity you can turn on and off like a switch. The critical faculties that make for a good journalist probably make purging yourself of all relevant opinion even less plausible. Downie is certainly right that there is no point in not voting officially if you're voting mentally. But in concluding that he therefore shouldn't even vote mentally, he is buying into the fallacy that having an opinion is the same as having a bias.
What's the difference? Bias is a failure to suppress your opinions or (if opinion is in your job description) to state and defend your opinions openly. Like avoiding opinions, avoiding all bias is probably impossible. Among other difficulties, objectivity is not a huge safety zone. It is a narrow path between bias on one side and bottomless relativism on the other. Journalists are not supposed to be neutral between fact and falsehood or about certain basic shared values. We may state baldly that two plus two is four and may assume without supporting evidence that democracy is a good thing. But beyond that, the fog of disagreement sets in.
So perfect objectivity is not just unachievable but indefinable. That doesn't make it a false ideal. Avoiding bias is a more reasonable aspiration than avoiding opinion itself. If you reject the Downie Solution, though, you'd better have an alternative way to demonstrate your lack of bias. Fortunately, the burgeoning field of journalistic ethics has an all-purpose alternative solution for almost all ethical dilemmas. It is disclosure. Let your readers know that your great-aunt's ex-husband owns 10 shares of AT&T, and they can decide for themselves whether this biases your coverage of the telecommunications industry.
Why shouldn't the same logic apply to politics? If you're not going to refrain from voting, why not let your readers know how you voted so they can judge your objectivity for themselves? If you're asking them to trust you despite your political opinions, shouldn't they know what those opinions are? If you believe you do an adequate job of preventing your opinions from curdling into bias, what are you afraid of?
In this spirit (and to fill in the eerie voting-day news gap between the end of the campaign and the beginning of the election returns), Slate invited its entire staff to declare how they plan to vote and to briefly explain why. The exercise was voluntary, but most Slatesters joined in. You can read the results here. And Press Box launches a crusade here to get other journalists to follow our example.
One result is no surprise: Slate's staff is voting overwhelmingly for Al Gore. Fear of confirming conservative suspicions about the liberal predisposition of the media is probably the main reason other journalists will resist following our lead.
No doubt it is true that most journalists vote Democratic, just as most business executives (including most media owners) vote Republican, though neither tendency is as pronounced as their respective critics believe. This is a natural result of the sort of people who are attracted to various careers. It is not the product of any conspiracy. There is no Liberal Central Committee drafting young liberals into journalism against their will or blackballing young conservatives. And there is nothing that can be done to change this disparity, unless conservative press critics would like to see the media institute a political quota system, favoring conservatives over better-qualified liberals (affirmative action for opponents of affirmative action).
Butfor the millionth time!an opinion is not a bias! The fact that reporters tend to be liberal says nothing one way or another about their tendency to be biased. It does suggest that when political bias does creep in, it is more likely to tilt liberal than conservative. But there are so many other pressures and prejudices built into the newsincluding occasional overcompensation for fear of appearing biasedthat raw political bias plays a fairly small role. And any liberal bias in reporting is more than counterbalanced by the conservative tilt of the commentariat. Or so I believe.
Of course it is not easy to persuade folks of this, and many will never believe it. No doubt it is easier just to keep your political opinions secret and imply that you don't have any. But that absurdity or dishonesty itself undermines your credibility. Or it ought to.
There are two powerful ideas here. The first is about the virtues of transparency. Kinsley believes that its more harmful to pretend you don't have a particular opinion or conflictand leave outsiders guessing and speculating about where your "true" allegiances liethan it is to simply put your potential conflicts on the table. The second of Kinsley's notions, thoughthe idea that there is a real distinction between a bias and an opinionis a bit more complex, and I think it's worth digging into it in more detail. On many occasions, I think, we use the words "bias" and "opinion" interchangeably. But Kinsley is right. They are very different.
Bias, for example, is global. Opinion is particular. If I'm a film critic, and I hated the latest Werner Herzog movie, that's an opinion. That's why we read film critics. If I'm a film critic, though, and I hate all German movies, that's a bias. A bias is different in degree from an opinion. In the work of a critic who holds opinions, furthermore, there may be no discernable pattern: she may like this Western and dislike that Western, or like "Taxi Driver" and "Raging Bull," but dislike "Gangs of New York" and "Casino." Biases have a consistency that opinions do not. Thirdly, biases are predictive while opinions are reactive. The problem with the critic who hates all German films is that she's made up her mind about a class of movies before she's seen themthe effect of her bias is to close her mind. The critic who merely holds opinions, by contrast, is at least making an attempt to go into the latest German film with an open mind. When the great New Yorker film critic Pauline Kael was asked why she retired, she said that it was so she would never have to watch another Oliver Stone film. That was a quipbut it had a ring of truth. Kael recognized that a critic who rules out an entire category of film, a priori, can no longer be an effective critic.
I think we can all agree that biases are a problem, particularly for a journalist. Writers with biases are predictable in the worst way and, more than that, they are dishonest. They pretend to have given thought to a subject, when all they've done is apply a fairly rigid set of preconceptions. For a writer to have an opinion, on the other hand, is a wonderful thing. The ability to form opinions is a sign of engagement with the world. And, like Michael Kinsley, I believe that the process of journalism is immeasurably improved when writers are open about their opinions. So let me start by making a few disclosures. You should know that in my early twenties, I was very conservative politically. I had a picture of Ronald Reagan on my wall in college. (Yes, I was that much of a geek). Today, if I could vote (and I can't, because I'm Canadian) I would vote Democrat. I am pro-choice and in favor of gay marriage. I believe in God. I think the war in Iraq is a terrible mistake. I am a big believer in free trade. I think, on balance, taxes in Americaparticularly for rich peopleought to be higher, not lower. I think smoking is a terrible problem and that cigarette manufacturers ought to be subjected to every possible social and political sanction. But I think that filing product liability lawsuits against cigarette manufacturers is absurd. I am opposed to the Death Penalty. I hate SUVs. I think many CEOs are overpaid. I think there is too much sex and violence on television. Of perhaps most relevance to my writingbecause I write so much about health careI have recently come to think that the United States needs to adopt a Canadian-style single-payer, government funded, universal health care system.
Do any of these opinions rise to the level of bias? I don't think so. They don't cohere in a single identifiable ideology. And they aren't predictive, in the sense that they lead me inexorably towards writing in a pro-God, pro-Democratic, pro-choice, pro-gay marriage, pro-free trade, pro-higher taxes, anti-Iraq war kind of way. If you look at my articles, in fact, you'll see that I rarely even write about the subject areas where I have the most strongly held opinions. What's more, when I interview or profile people I don't ask them for their opinions on these same subjectsso there's very little chance for any conflict or agreement in our attitudes to become an issue. I should also say that, by the time you read this, any number of the opinions I've stated above may well have changed. That's another important difference between biases and opinions. Biases are pretty stable. Opinions come and go.
Let's talk, then, about my speeches, since that's clearly the area the question of biases and opinions is most fraught. Does the fact that I am periodically paid by company X or organization Y to deliver a speech about the Tipping Point or Blink affect, in some way, the content or the direction of my writing? I think the answer to that has to be yes. For example, I have in my book Blink an entire chapter devoted to an attack on market researchin particular the heavy reliance in corporate America on focus groups. I got interested in that question because I once spoke about the Tipping Point at a music industry meeting, and at a cocktail party afterwards got into a long conversation with a industry executive, who told me how he thought pop music was being ruined by radio's adherence to market research. Would I have developed that view of market research had I never gone to that conference? I don't know. But probably not. When you give a speech to a particular group, you expose yourself to the opinions of that group, and that exposure cannot help but affect the way you think. Similarly, I once gave a talk at the Yale University psychology department, for which I was paid an (small) honorarium. Afterwards, I spent the better part of a day chatting with various psychologists on the Yale faculty. Out of those conversations have come a number of ideas and thoughts that have made it into my writing. Here I feel quite confident in saying that if I had never gone to Yale, I would not been exposed to those ideas. Yale paid me to come to New Haven, and I left New Haven excited about the psychological research being conducted at Yale. I think, in fact, that the people who invited me to Yale would readily admit that this was their intention. Why else invite a journalist to campus, and set up a day of interviews with him, if not to influence him in some way?
I'm not sure there's anything wrong with this kind of influence, though. I didn't develop a pro-Yale biasan enduring, predictive, monolithic predisposition towards Yale. I developed a set of opinions shaped by my Yale experiences. As it happens, those opinions were favorable to Yale. But opinions are fickle things, and the opposite could just as easily be true. I could have been so unimpressed by the people I met at Yale that I could have been indifferent or even hostile to their work. For instance, I've given lots of paid talks, over the years, to medical groups of one sort or anothergroups of doctors or insurers or hospitals. Virtually all of those groupsat least on an institutional basisare firmly opposed to a Canadian-style universal health insurance. Yet it is over that same period that I have changed my mind and become a strong advocate of the Canadian system. The effect of meetings and talking to all those people in the U.S. health care world has been to convince me of the intellectual bankruptcy of their position. Like I said, the effect of influences on opinions is not a one way street.
Now, you might say that that's a rare example, and in most cases the fact of being paid by someone predisposes you to be sympathetic to their cause. This assumption, in fact, is one of the cornerstones of the Downie school: money is the fastest route to journalistic corruption. But is it really? In my experience, personal friendships are far more serious source of bias. For example, the CEO of Research in Motion, the Canadian firm that makes the Blackberry, is a man named Jim Balsillie, who was a friend of mine in college. Furthermore, RIM is closely linked to the University of Waterloo, an institution where my father taught for 30 years and to which I have profound emotional ties. Have I ever received a dime from RIM? No. Can I be objective on the subject of the Blackberry? I really doubt it. Yet I'm quite sure that if I wrote an article on hand-held communication devices and came out heavily in favor of the Blackberry, I'd get in far more trouble if I failed to disclose any financial dealings I'd had with the company than if I failed to disclose my sentimental attachments to RIM and Waterloo. At a place like the Washington Post, in fact, having given a speech to RIM would disqualify me from covering the telecommunications industry. But growing up in Waterloo would not. Isn't that a bit strange?
Let me go further. I don't think that the health insurance examplewhere I ended up taking an opinion opposite to the interests of the people I got paid to speak tois all that rare. In fact, it has happened on several occasions. On behalf of the business side of the New Yorker, I have repeatedly given talks or presentations to representatives of companies that advertise with the magazine. For some of those presentations, I have been paid. And on a number of occasions, those groups have included people from the U.S. automobile industry. Has that biased me in favor of the Big Three? Well, no. As I've stated, last January I wrote an article bitterly attacking the SUV, which has been the cornerstone of the financial success of Ford, General Motors and Chrysler over the past ten years. Giving a speech does not buy my allegiance to the interests of my audience. Why? Because giving a paid speech to a group for an hour is simply not enough to create a bias in that group's favor. It's a very different sort of transaction. I'm not invited to speak to those medical groups because I promise to agree with their position on health care, and I'm not invited to speak to groups from Detroit because I promise to agree with their position on SUVs. In fact, my position on health insurance or SUVs never comes up. I'm invited because those audiences want to hear about my work. Financial ties are in danger of being corrupting when they are ties, when they are, in some way, permanent and when resources and influence and information move equally in both directions. There is no question, for instance, that I am hopelessly biased in favor of The New Yorker. I've been writing for them for ten years. They've been paying my salary for ten years. The people who work on the magazine have become some of my closest friends. If there's some deep dark scandal buried beneath the surface of the New Yorker, it's probably not a good idea to rely on me to ferret it out.
Let's look at another example. This is a harder case. And I bring it up because it recently was written about publicly (www.slate.com/id/2110572). The article was called "High Prices" You can find it on the archive page of this website. It made the following argument.
Have I given paid speeches to companies or industries mentioned or affected by that article? Yes I have. As I stated earlier, I have given my Tipping Point talk to groups of doctors, hospitals, insurers, as well as Pharmacy Benefit Managers and groups funded by the National Institutes of Health. More specifically, I have on several occasions over the past four years given paid speeches on the Tipping Point to pharmaceutical companies. So did that create a bias in favor of the pharmaceutical industry? Well, the evidence is for all to see in the article itself, and I think the answer is pretty clear. For instance, I said that consumers would be better off ignoring Astra Zeneca's marketing campaign for its heartburn drug Nexium and just buy cheaper generic or over-the-counter versions of that drug, or similar drugs that category. I also suggested that most consumers would be better off skipping Pfizer's heavily marketed and highly expensive drug Celebrex, and sticking to ibuprofen. Pfizer's Celebrex and Astra Zeneca's Nexium are two of the most profitable drugs in the world. "High Prices" advocated systematic generic and theraputic substitutiona policy that would cut the sales of Pfizer and Astra Zeneca, and everyone else in the brand-name pharmaceutical business, by tens of billions of dollars. It advocated the use of third-parties, like pharmacy benefit managers, to more rigorously control physician prescribing practices. Finally, I said I was in favor of a Canadian-style single payer system. And if we had that kind of system, of course, the U.S. government could do what the Canadian government does now, which is use the enormous bargaining power of a universal health system to demand steep discounts from the industry. You can see why I wasn't terribly concerned about the possibility that someone would think, as a result of my speaking, that I had a pro-drug company bias. (I should point out that, to my great puzzlement, I ended up being accused by some readers of exactly that.)
But that doesn't entirely resolve the issue. Suppose I had wanted to write something overwhelmingly positive about the brand name prescription drug industry? What if I learned, while at some speaking engagement, say, that the company was on the verge of curing cancer? Now things start to get more complicated. Clearly as a journalist I'd want to write about that fact, and you as a reader would want to read it. And the fact that I got however many thousand dollars to talk about the Tipping Point would be dwarfed in importance by what I'd learned. Still, if I wanted to say something nice about that firm, I'd have to make my previous arrangement with those firms plain. I would have had to prove that my encounter with them created only an opinion about what to write aboutnot a bias. I would have to make very sure that my discussion of the issue at hand was fair, that I accurately represented both sides of the issue even if I favored one side over the other.
All of that creates a fairly substantial burden. In a case other than a cure for cancerwhere what I learned was less consequentialI might prefer to take the easy way out and not write the article at all. And then where would we be? Well, we'd have a case where the hat I wear as Tipping Point and Blink author and speaker would affect the direction and content of my writing for the New Yorker. Faced with the possibility of suspicion of conflict of interest, I would be censoring myself.
I should stress that this is a hypothetical example. I don't think that this kind of situation is likely to come very often. In most of the articles I write, there is no real potential for a perception of bias. Nonetheless, I'm not willing to dismiss this final consideration either. It's a real issue, and one of the reasons I wanted to work though this subject at such length (and I applaud those who are still with me) is that I wanted to put this smallbut genuineproblem out in the open.
So what to do? I'm open to suggestions. But let me start by enumerating a series of principles, to which I pledge to adhere, that I think make the possibility of opinion turning into bias and fairness into unfairness less likely.
I don't claim that these four principles entirely resolve the questions raised by the two hats that I wear. But I think that they are a start.
December 10, 2004