An old schoolfriend recently moved to New York City to pursue her career in journalism. Unfortunately for her, I have lots of naive questions about her profession. Fortunately for me, she’s my friend and has to answer them without getting too irritated. Here are two examples.

1. Popularity-based quality metrics

How are journalists assessed by their editors, and how do business pressures shape the subject, content and presentation of what is published? More specifically, since the establishment of on-line newspapers and other news outlets, it’s possible to collect detailed information about which headlines attract readers and how long they spend reading the article. It’s easy to imagine that this would have consequences for quality metrics within the industry. But first let me establish the basic framework.

Here are two uncontroversial comments: 1. journalists want their work to be read; and 2. a major, and sometimes unique, source of revenue for news agencies is advertising (which is completely reliant on circulation). Now that there is abundant data on exactly how to maximise an article’s appeal to a particular audience, the cynics inside will cry that the public is essentially being fed its own excrement for the sake of journalists’ egos and/or news outlets’ bank accounts. Our generous side will, conversely, claim that journalists are not so immoral: that they are dedicated to a trade that they believe in and whose integrity they want to protect (it’s not as if they’re there for the lavish salaries).

Both these points of view are plausible to some extent (and relevant to more professions than just journalism). I expect most of you will be naturally attracted to the first one, especially since, even if the writers themselves are all noble, their bosses who have their eyes on the coffers might not be. And the second point of view is certainly Panglossian: exactly the same could be claimed of scientists, but from my vantage point within physics it’s clear that the principles of scientific methodology and objectivity are frequently (and in small doses, routinely) twisted as researchers vie for tenure-track positions or scarce grant money or more publications or an article in a prestigious journal. Regardless of a discipline’s august philosophical underpinnings, in the end it’s humans doing the work and they are frail and easily tricked by their own bullshit… Anyway, back to the story.

I’m not claiming to be the first to raise the possibility of bias in the press; though more popular / strident criticisms typically concern government or corporate interests actively corrupting the public’s principal information source (and there are many, many demonstrable instances of that happening). My angle isn’t new, but it is slightly less sinister – that internal pressures might lead to / sustain a narrowing of focus and diversity in the issues and conversations endorsed by the press.

Underlying all this is my expectation that, though a news outlet is a business, it simultaneously has a duty to inform the public about a breadth of issues and not just the tiny Venn intersection of the most popular ones. And I’m sure the overwhelming majority in the industry would agree without reservation. So my question is really this. How are the “public service” principles underlying journalism safeguarded against business pressures in practice? Are there mechanisms in place to forestall conflicts of interest like the ones I’ve been contemplating? To balance the integrity of a news organisation’s mission statement with the need to stay afloat in a competitive environment?

For example, it doesn’t seem such a far-fetched scenario that an editor tells their writers that they are allowed to write an article on some interesting but empirically unpopular topic only if they have garnered so many hits on a more popular article that month. Or perhaps salary raises are contingent upon the number of article views (and the attendant advertising revenues).

When I asked my journalist friend about this, she responded with a number of things (including a fierce exasperation at the incessant negativity about her profession). For one, the popularity-based evaluation systems I just envisaged are unthinkable and unthought in the world of respectable journalism (i.e. not click-bait sites). In her experience, journalists and editors do not equate popularity with quality, and are fully aware that a preoccupation with the former would be a disaster for the standing of their platform. Of course editors must try to increase readership, but the pressures that I was envisaging are as a rule not as corrupting as I feared.

I can see her point. There are many branches of journalism that are difficult and expensive to do, and are either not particularly popular or have a relatively low success rate of producing results: investigative journalism for instance. The model that I envisioned, if brought to its logical conclusion, would certainly have done away with these excesses. From her response I gather that the answer to my question (whether there are explicit institutional balances to protect journalists from having to prioritise page hits over breadth or quality) is no; but the culture of the profession and the prudence of editors prevents such distortions in reporting from becoming severe.

2. Wealth and class bias in all professions including journalism

Here is my premise. There are many professions for which the barriers to entry are high: perhaps there are many people vying for few jobs so the market is risky; perhaps they require a large investment of time or money; or perhaps it’s assumed that applicants for positions must first learn the ropes through months (or years!) of unpaid internships. All these forces will tend to exclude people from poorer backgrounds. It is so much easier to pursue jobs which offer risk, delayed returns or low initial pay if you have a financial safety net from family or partners or personal savings.

It would be interesting to see data on the variation of economic background by profession; and such studies certainly exist. Without reading them, though, I’d predict that law and medicine are dominated more than any other sector by the upper-middle class. (In the USA, entry to those fields requires additional, famously expensive degrees. In the UK the situation isn’t so extreme, but the time-lag between starting professional training and getting paid for working is still substantially larger than in other professions.)

But regardless of the numbers, there are a few areas where I’d expect the societal effects of bias to have the most important ramifications: personal lawyers, public legislators, and journalists. All of these are tasked to some extent with protecting and educating the public despite the prevailing power structures which frequently marginalise minorities and the poor. It is especially important to have experienced advocates for the unenfrachised in the areas where they stand to lose the most.

Turning now to journalism specifically, these days job applications are often considered only after a good deal of unpaid labour, so from the foregoing reasoning we might expect journalists to be predominantly middle-class. Of course this doesn’t mean that they only serve the middle class – there is a huge variety both in what is written and the in the audiences it is written for. I’m merely suggesting that non-representative perspectives may become standard, entailing certain expectations about and attitudes toward society, and tacit embedded assumptions about how the world works.

To rephrase this a little more explicitly, whatever the objectives and culture of the publication in question, if mid-to-upper-middle-class writers and editors dominate the landscape, this will affect what gets written. Contrast this with a scenario where there are working-class journalists writing for middle-class readers, and also working-class journalists writing for working-class readers – I imagine the tone and approach and even content of those articles would be different on the whole. If everything is written from a position of privilege, the corpus is consequently written from a position of ignorance.

An analogy. Imagine that, by chance, journalists were exclusively male. Of course they would still publish material for women to read; and much of it would be very good by all standards. I can’t help feeling, however, that something would be lost.

But so far, that’s all this is: feeling and speculation. I e-mailed my journalist friend about it and she came back with some interesting comments (edited and shuffled slightly for clarity).

What you’ve been talking about has definitely become an issue in the UK – it’s become far more “private school” this generation. People are angry about all the unpaid internships and nepotism because previously it wasn’t like that at all and was very diverse. Indeed, many of the very best journalists had no university education and a working class background (e.g. Harold Evans, Rebekah Brooks).

In the US it’s worse on the whole; for example most journalists have been to university. But within my publication there’s a major focus on diversity, so it’s hard for me to judge.

Of course the demographics of writers affects what’s written and is an issue. But I think this pales in comparison to how the demographics of readers affects what’s written.

Conclusion: yes, there is a bias of backgrounds within the profession, but in her estimation there isn’t much evidence that it distorts the news. I’m making a storm in a teacup.

I find her answer comforting: she’s a smart and self-aware lass and more clued-in than I am when it comes to workplace sociology (indeed, when it comes to most things). However, the broader issues of diversity and access to different professions, with their commensurate social justice ramifications, are still at large. I only have to look around my physics department to see that we have a problem with gender balance.


Roc 30 November 2015, 09.42

Great piece. The bias of the media, and the failure of education to teach people (including the journalists themselves) to detect this, is a candidate for human problem number one in terms of creating avoidable suffering.

Setting aside the important social mobility and justice issues, I wonder if the middle-class bias will have any additional filtering effect after the already powerful pre-existing biases have done their bit :

The cost constraints which mean that most ‘quality’ press content (from memory 73% in one Cardiff University study) consists of uncritically recycled government and corporate press releases.

The requirement in commercial media not to alienate the advertisers and in some to maximise profit.

Have you seen Chomsky’s "Manufacturing Consent"? He has an interesting take on the problem.

On the gender imbalance in the physics department, this is part of a massive and general issue of whether you are aiming at equality of outcome or equality of opportunity.

To me it goes without saying that it is unacceptable to discriminate on grounds of more or less artificial and oppression-purposed categories like race or gender because it means giving up on the ideal of the individual and returning to a group conception of humanity, collective guilt, etc.

But to deliver equality of outcome, that’s what you have to do. (This is an empirical question, but my hypothesis would be that if there is any gender bias in the physics department, it is directed at recruiting more female physicists, i.e. securing equality of outcome.)

To me it is valid to discriminate on grounds of behaviour or ability or enthusiasm provided your criteria are functionally related to the enterprise and not artificially constructed to favour a privileged group.

In reality, only by abandoning the artificial concept of gender or race for the purposes of your own analysis, and looking at people’s behaviour (which may include believing in gender or race) can you really understand what is happening. These essentially statistical concepts, even if they are held in good faith rather than to propagate malicious stereotypes, do not properly correspond (though they may contribute) to reality and so they hinder understanding.
Cato 30 November 2015, 17.14

Thanks for the comments: very thought-provoking. I mentioned in passing that this was only one source of bias in the news media, and I totally agree it isn't the most important one -- merely one that I hadn't heard before. The issue with press releases is certainly interesting and it's already on my list to investigate; so far it's been hard to reconcile those accusations with other testimonies about the culture of journalism. We must be aware that slamming the press is a popular pastime and they will come under fire for imagined, as well as real, biases. The opportunity for misdeeds doesn't constitute misdeeds themselves, just an institutional loophole that we should be aware of and perhaps try to plug. I plan to write more about the press in the coming months.

As for the question of equality, this requires deeper discussion, but I think one can frame it not only in terms of opportunity and outcome, but also in terms of idealism and pragmatism. Though your comments as an idealist are very very important and should be at the forefront of people's morals, pragmatic approaches should be at the forefront of people's actions. As social animals, group allegiance is fundamental to human interaction. We are not intelligent enough to overcome this. In any case, there is currently no equality of opportunity for e.g. women in science or black people in the US judicial system, even if all the legal documents we can find were to attest otherwise.
Roc 30 November 2015, 13.01

As I think you say, gender and race blind equality of opportunity has not been and needs to be delivered - but there are still many questions, such as whether to deliver equality of opportunity you correct for someone having intelligence, a supportive family, a good university, suffering institutional discrimination etc. Much it goes back to reforming poor education, which in Britain is one of the greatest human rights abuses. Anyway, that is where I would focus my effort, above all perhaps in detecting covert mechanisms to thwart equality of opportunity like the one you talk about in journalism.

Maybe part of my point is it is either/or. To have equality of outcome you must have discrimination and anecdotally the discrimination transfers resources to the more privileged of one more or less arbitrarily defined group from the less privileged of another, i.e. it is regressive at the level of individuals. But you could argue for equality of outcome on grounds of defusing unrest.

I feel even the most basic presuppositions in this debate are questionable so you have to go far back in terms of defining morality and justice and identity, both assigned by others and chosen by yourself.

And on the journalism front there is almost a problem by definition - to be a journalist you have to go to university and once you have gone to university you are now middle class in many of your interests and allegiances. I share your distrust of the media establishment’s impartiality and self knowledge but maybe this becomes less important as the media are disrupted.
Cato 05 December 2015, 03.25

All very interesting points, and I definitely agree with your focus on education (though I probably got that form you in the first place!). The comment about regressive policies is interesting, and definitely something of which to be aware, but that seems to me to be a problem of implementation rather than philosophy. In any case, the accounting will always be questionable -- how much value does one put on the very fact of being white? In the US especially, it is of huge value -- financially, culturally, for jobs, for law enforcement, ... -- and this fact of "social capital" must be included in deciding about regressiveness.

I don't know where to draw the line on how much discrimination is too much. But I do feel that when there's a clear history of oppression which is directly linked to deprivation and inequality today, reparations should not be out of the question on the basis of inverse racism. It's levelling the playing field. Improved education is only one element in our response to blatant iniquity -- I don't think any of this is either/or unless one is purely ideological (and as you know I am mistrustful of ideology) -- we need role models and we need people who understand the issues in positions of power, and that will not happen in any meaningful way if we rely on "trickle-up". It's not absurd to claim that the existing power structure of middle class white men (crudely) cannot fix the grievances / win the trust of minority or under-represented groups who have spent the last twelve millennia taking orders from exactly those people.
Roc 05 December 2015, 07.38

All that makes total sense and I hope my position isn’t irritating. It may be we agree and I would love to discover exactly where we disagree if so - it might be at the deep level of abstract morality etc.

I think I’m coming from my UK experience where I see favouritism for privileged people who manage to spin themselves as underdogs, and none for the real underdogs, because access to favouritism seems usually to be an expression of power, hidden or otherwise. It could be the position of the real underdogs is compromised by this (this is my hypothesis) but it may be irrelevant to them and simply a transfer of resources between privileged groups, and it could be it indirectly helps the real underdogs by eroding stereotypes etc. And probably a mixture of all three over the whole population of underdogs.
Cato 06 December 2015, 22.13

All three scenarios are interesting. Of course we'd have to talk about a particular example to say anything concrete. My instinct is that in the best-case scenario (i.e. not in reality but not prohibited by the laws of societal interaction): 1. isn't so relevant because the conners are a minority, albeit one with which people can be whipped into fury. For instance benefit fraud is a real thing, but tax avoidance / aggressive lobbying by big companies is also real and "costs" the government much much more. 2. Even if resources are given to the top echelon of underdogs, it is still progressive. And though initiatives organised by central governments are more likely to fail at redistribution (or be compromised from the outset), local governments and NGOs might in general have a better chance at effective targeting. There are a lot of intelligent compassionate people. 3. This I absolutely agree with: eroding stereotypes is important, but it is also crucial to have spokespeople from marginalised groups who can bring focus to the issues they face, and thereby break the cycle of marginalisation.