On Liberal Media Bias
A few years ago, there was a mini-uproar about a supposedly “racist” computer webcam that failed to detect the faces of black folks and others with darker skin tones. As the video below demonstrates, there was definitely something very odd about the way the new HP webcam functioned:
I learned about this via a blog I frequented at the time which focused primarily on race and pop culture. The commentariat was up in arms about the “racist” computer and racist people at HP. Many argued that this was evidence of deliberate racism and that HP did not want black people to buy its computers.
Hogwash.
A computer cannot be racist. It is not sentient. It does what its programmers and/or users tell it to do, nothing more and nothing less. So, no, neither the computer nor the webcam were racist. Nor is it the case that whoever made or built the cameras was a racist who didn’t want people to enjoy HP’s products. It would make zero sense for a company as large and with as big a public profile as HP to deliberately discriminate against a significant portion of the market.
But, still, something was up, right? I mean, that video doesn’t appear to be doctored and other videos and stories popped up about similar difficulties. Maybe HP, its employees, and their products weren’t racist, but you still had a bunch of dark-skinned people unable to take advantage of one of their newest products.
The most likely explanation is two-fold:
- The facial recognition technology did read different skin colors differently and, depending on the available light, might struggle to pick up darker skin tones.
- There were few or no darker-skinned people in the rooms that these products were being developed and tested, meaning the problem did not become evident until after roll-out.
So the products did have issues with race in part because of broader issues with race present at HP. But it would be a bridge too far to conclude that HP or its employees were racist, that the outcome was intentional, or that it was part of a broader, concerted effort to deny computers to black folks. What really happened is that the bulk of HP’s developers and testers were lighter-skinned and through a collective and unintentional bias, they rolled out a product that seemed explicitly intended to discriminate.
What does any of this have to do with liberal media bias? WELL, CLEARLY THE FACT THAT MOST OF YOU NEVER HEARD OF THIS HORRID STORY OF RACIST ROBOTS MEANS THE MEDIA IS REALLY BIASED TOWARDS THE RIGHTWING!!!
Wait. No. That’s not it. That’s the opposite of it.
This has to do with liberal media bias because I think the causes of what we tend to describe as liberal media bias work similar to the causes of HP’s “racist” computers. Media cannot be biased; not anymore than a webcam can be racist. The people who create and distribute media… they can certainly be biased. But I’m still not convinced that the broader slant we see in the news is the result of a deliberate, calculated, and orchestrated attempt to further particular goals and stymie others. Instead, what is likely at play is that the rooms where most decisions about media are decided are full of liberals. Maybe not exclusively so, but largely. (I believe Jaybird linked to an article demonstrating as much, which I unfortunately cannot find at this time.) So, if you fill a room with mostly liberals, it is natural that the majority of their decisions are going to be biased toward liberalism. Filled a dozen rooms with mostly liberals and suddenly it seems as if our media is awash in liberalism, carefully selected by a cabal of leftist media moguls who filter down only approved stories to their underlings.
Do I think, on the whole, the media is biased towards the left? Yes. Do I think there is a widespread conspiracy afoot? No.
And just as HP would be best served to seek ways to diversify their product development and testing teams, the mainstream media would be best served to diversify the political leanings of their decision-makers.
************************
It is worth pointing out that I didn’t follow the HP story super closely, so it is possible that there was something else to the ultimate outcome. My reporting on the facts might be less-than-perfect, but as an analogy, my interpretation of those events seems to parallel what I think is going on in the media today.
Do you think that there is such a thing as “institutional racism”? The idea that the forms and structures and physical layout and practices and governing rules of, say, a court or a college, might have the net effect of putting a disfavored minority at a functional disadvantage as compared to a member of a more privileged class?Report
Were I one of the greatest beneficiaries to “institutional racism”, I would probably argue that there was no conspiracy, how there isn’t a group of people using racial slurs, and how when something like Serious For Real Racism shows up, everyone tramples everyone else like they’re at a Who concert in their haste to get to the microphone to denounce it.Report
Burt-
If there question is simply, “Do you think that there is such a thing as “institutional racism”, yes or no?” I would answer, “Yes.”
If given room to elaborate, I would argue that most things that we would label as “institutional racism” are often not a part of a broader racist conspiracy. But if you tip the scales just enough in each action, over enough actions, the impact becomes massive.
So, in somewhat of a response to Jaybird as well, I am one of the greatest beneficiaries of “institutional racism” but also one of it’s most fervent opponents (as in, I believe it exists and oppose it; I am not an opponent of the idea of “institutional racism”). But I don’t think that a bunch of white guys are sitting around smoking cigars and wondering how they can get black people to smoke more crack.Report
So if institutional racism can exist, the next question is, can it exist in the media?Report
Yes.
And I think I see what you’re getting at. If I am right, I anticipate my next response being that racism, particularly the forms that tend to be “institutional” are neither liberal nor conservative. But I might be way off on where you’re going. Please proceed!Report
Not quite.
If the media (or more precisely a particular media outlet, e.g., NYT or Fox News) can become tainted by institutional racism, then can it not also be tainted by other kinds of institutional bias?
(This is to parse out bias by a particular reporter from bias in the media structure itself.)Report
Oh. Yes. That was sort of my whole point. Did it read otherwise? I think the bias is there. I don’t think that is indicative of deliberate bias on any individual’s part or of a collective effort towards bias.Report
I read it otherwise. Perhaps I just needed to rephrase it in my own terms to digest the point.Report
Please elaborate… it appears I might have been less-than-clear and welcome an opportunity to clarify.Report
You conclude:
From this, I take away “Liberal people are going to pick other people like themselves (liberals) to help them make their decisions and they’re going to collectively decide things in liberal ways.” That to me conveys a sense that we’re talking about a series of intentional acts. What it seems to me you were getting at was the notion that these kinds of biases are the result of unconscious factors, which is different than decisions made intentionally.Report
Gotcha.
I think it is sort of a both/and.
Liberals may be more prone to select other liberals to fill their newsroom not because they have the word “LIBERAL” on their resume but because, hey, that guys says all the right (read: liberal) things!
It also might be that a series of intentional decisions might be seen as independent but, taken collectively, demonstrate a broader collective bias.
Sure, the FIRST guy you hired was a liberal. And the SECOND. And the THIRD. But you didn’t MEAN to hire all liberals. It just sort of worked out that way.
Etc.
Am I more clear? I wish I could not speak bad.Report
Yup. Just look at CMU. It’s a decidedly unfriendly environment for people in wheelchairs, with floors that you can race chairs on (all the way down to glass doors at the bottom…).Report
I think however much the media is biased to the “left” is anchored by the fact that those controlling the pursestrings and what actually gets on the air is biased to the “right” and the fact that most Beltway journalists are easily fooled by somebody with some charisma. Anybody who actually remembers the 2000 Election and the media’s dogpile on ever single Gore statement and still thinks the media is tilted to the left gets a laugh from me.
I think the media is biased toward conflict and sensationalism in most cases (see the “Romney comeback!” stories everybody came out with 7 seconds after the debate was over, not that I’m claiming Obama actually won the debate) and on the issues, they’re solidly in the neoliberal consensus when it comes to economics (shared sacrifice! by which, we mean cuts in Social Security and Medicare and lower the rates for high earners!) and probably more to the left than the average person on social issues.Report
So somehow on piddling, obviously silly stories like the one you describe, the pervasive liberalness of the media shines brightly through. Yet when it comes to issues that actually mean something they generally parrot the government line like it’s inherently the truth.
I guess we’re working on different definitions of “liberal” here. It looks statist & oriented towards an upper-class friendly status-quo to me…
Now, that isn’t to say it’s conservative either. Pretty much no prominent media is conservative. Even Fox News isn’t conservative — they’re capital R Republican, with philosophical consistency not even on the radar.Report
This is a salient point. Perhaps the bias is towards Democrats and away from Republicans. Would that seem more accurate?Report
Given the composition of Sunday talk shows and the like in terms of guests, I don’t think that’s accurate either.
There is a substantial bias toward upper middle-class (or upper class) white men and their concerns.
See: Jim Lehrer’s debate topic choices.Report
Do upper middle-class white men skew D or R? I’d guess D but don’t know for sure. If you add in the natural urban bias that networks based in urban areas have, I think you’re still going to come out skewing D no matter what way you slice it.
I should also add that I’m not necessarily saying this is a wonderful thing or a horrible thing… it’s just a thing.Report
Well, whites skew Republican, men skew Republican, and people with high incomes skew Republican. I’m not 100% sure that upper-middle-class white men skew Republican, but I can’t imagine that they don’t.Report
Do people with high-incomes skew Republican? I suppose it depends on how we define “high”.Report
Obama had the top and the bottom locked up, but white uppermiddle men tend to trend Republican. less so in places with more income equality, mind.Report
No, it doesn’t depend. Just, “yes.” Obama won a certain subset of the upper class.Report
If we define upper class as “more than $200k a year” that subset can actually be called “most.”
There are other ways to define upper class, but that goes to Kazzy’s point about definitions, doesn’t it?Report
Upper middle-class, slightly older people tend to be white, tend to vote, and tend to have a lot of disposable income.
This makes them an attractive audience for both politicians and anyone who sells ads for a living.Report
This is what I have observed- that journalists tend to be well educated, well compensated members of the professional class, with almost zero personal identification with the concerns or culture of the working class.
Its not conscious, or malign, its just that when you are a TV pundit, magazine correspondent, or newpaper writer you tend to write stories based on your personal perspective. And that perspective is one of that is attuned to urban social culture, and upper middle class financial concerns.
So we get endless discussions of the budget deficit, and calls for austerity which will never be borne by any of the people who write about them.Report
Bingo. Being more in favor of gay marriage than the average American is proof the media is biased liberally and we need affirmative action for conservatives in the media, but being largely in favor of cuts to Medicare and Social Security is kind of swept away.Report
That’s a bit of a stretch. Plenty of good reporters come from blue collar backgrounds. Thing is, to be a journalist of any standing, it really does help to be a college graduate. Blue collar people aren’t usually college grads, which means they didn’t take any journalism courses. So at best, qualified journalists are pale blue collar.
Furthermore, who buys the journalism product? Blue collar people don’t buy newspapers for the intrepid in-depth reportage from Bungabungastan, where we’re given backgrounders on the Bungese and how they’re adapting to current events. Blue Collar types watch news on the Teevee and all they want to see is the bang-bang. The Bungese don’t enter into the picture, the American public wants columns of smoke and FLIR video from the helo as they waste these moving dots on the ground like it was some fucking video game.
Lyotard once said information was made to be sold. If journalism reflects the upper middle class financial concerns, that’s because the working class doesn’t buy information. They will, however, buy bang-bang.Report
Whenever I sweep my hand grandly over an entire category, I include a few of the innocent with the guilty.
I suppose I should narrow it to say that the journalists who set the national agenda and decide What It Is We Are Talking About [tend]to be upper middle class white collars
I have never seen a panel on cable news networks discussing the issues which included anyone who was on public assistance.
Whenever the talk is of military matters, they are sure to engage teams of retired generals and defense industry lobbyists (but I repeat myself).
Yet the topic of public assistance and its effect on The Poor is regularly discussed by people who wouldn’t know SNAP from crackle and pop.
No one thinks this is odd; which itself betrays a bias.Report
I remember reading somewhere that Henry Rollins once tried to get a truly democratic movie show on the air. His idea was that he’d get a famous director or actor or something and a regular schmoe from a regular job (a construction worker, a short order cook, a waitress) and they’d all watch a movie together and discuss it.
The director he had on the show talked about the blocking of the various scenes… and the schmoe ranked the hotness of the actresses in the movie and expressed that he wished that one in particular had removed her blouse.
The show didn’t get off the ground.Report
That’s hilarious. If they had a third guy to make fun of both of them that show would’ve been a hit!Report
Well then, the movie failed at getting the schmoe’s attention beyond the boobs of the actresses. Considering the audience consists of more schmoes than fellow directors, the movie in general failed, assuming the schmoe was representative.
Not that there is anything wrong with observing such, only in just getting that from what is supposed to have a plot.Report
Oh, truer words were never spoken. For all this hand wringing and moaning about the Condition of the Poor and Downtrodden, it’s amazing how their voices are never heard.
When it comes to matters military, we’ll never hear the voices of the enlisted men, unless it’s some faceless, voiceless dude behind a pair of Oakleys, hunkered down and firing. Or hugging his beloved kiddies and/or Significant Other. There are exceptions: PBS Frontline is one. Mr. Romney has some ideas about how to deal with that situation. Oh yes.
Information is only of value when it can be sold. Knowledge without any sale value is useless to these fucks. Never mind that we spend billions on these problems, benefitting only the Usual Suspects, the aforementioned defense industry lobbyists and many others, building large office complexes in the environs of Babylon-upon-the-Potomac, the better to suck at the mighty teats of that monstrous sow, the Department of Defense. Mr. Romney has a plan for them, too, somewhat different than his plans for PBS.
The meek shall inherit nothing. That’s Frank Zappa.Report
This is interesting, and apropos.
http://www.rochester.edu/news/show.php?id=4622
The famous Marshmallow Study is reconsidered. The original study was used as a way of explaining how poor people make their own poverty by being unable to delay gratification. The new one demonstrates that sometimes instant gratification is in fact the wisest course, given their lived experience.
I can’t comment with expertise on this new study; but what occurs to me is how the conversation about poverty is most commonly done by people who themselves are not poor, have never been poor, and most importantly, have no possible frame of reference for the daily lived experience of poverty and what motivates their behavior.Report
Obligatory.Report
Do I think, on the whole, the media is biased towards the left? Yes. Do I think there is a widespread conspiracy afoot? No.
I think this is fair. It’s not a conspiracy, it’s just that over time, generations of (mostly-unconscious) self-selection and wanting to be with others ‘like you’ (be it racially, or religiously, or ideologically, or whatever), and you end up with a fairly ideologically-homogenous media population.
Then a million little (mostly-unconscious) editorial choices made with regard to framing, word choice, and tone, accumulate to produce biased results.
Possibly relevant anecdote.Report
Damnit. That basically sums it up much more neatly and succinctly. From now on, you write my posts.
I think the self-selection piece is important, as well.Report
RE: self-selection – I suspect that it we were to start over somehow, in a few generations we’d be right back where we are. Because I would guess that in your initial population, you might get higher proportions of applicants who want to go into journalism who believe for example, ‘I will be a champion of the poor’, and a lower proportion who believe, for example, ‘the poor we shall always have with us’ and are focused instead on other issues than poverty.
And maybe in the end, this is to the good. But it seems acknowledging it is still worthwhile.Report
Can I acknowledge it, but not really think it makes a big difference in the long term of what the media covers? I mean, yes, Republican’s on this site and others are pissed the Nightly News isn’t Solyndra and Fast & Furious wall to wall, but there were plenty of things during Bush’s term that DailyKos thought the nightly news should cover daily as well.
I think the fact that more and more rich people with their levers on power seem to be sociopaths committed only to enlarging their own treasure at the cost of the American people is a more worrying trend than whether 80 or 85% of reporters voted for Obama, but that’s just me.Report
Jesse-
That is a fine position to take. This post was, as the title suggests, on liberal media bias… no more, no less. I thought of it while reading a comment Burt made on Tom’s recent post and with the story of HP fresh in my mind after telling it recently to Zazzy. It all just sort of came together… a bit of a lightbulb moment.Report
I just found it interesting that you think the “mainstream media would be best served to diversify the political leanings of their decision-makers”, but you don’t seem as worried about the political leanings of other decision makers in society. Or at least, you were worried enough about liberal bias in media to write a post about it, but not about conservative bias in other parts of life.Report
I apologize for not being all things to all people.Report
If the media are the gatekeepers, and the gatekeepers are biased (if unintentionally/unconsciously) – how can that not matter? This is an ‘input’ to our Democratic process – we vote, based on what we know, and what we know is biased (BIBO: bias in, bias out). You don’t see this as a risk?
I would further suspect that this can lead to feedback loops, amplifying ‘distortion’ – we vote based on what we know, and then we are ‘rewarded’ with ever-more-positive knowledge of the results of our votes. Over time, and absent correction, it would seem we could be led far astray.
This should not be read as a defense of the Republican party, which is largely off its rocker anyway. But it still seems salient.Report
Again, I don’t think the media is as liberal biased as you seem to be (on a scale of -10 being Stalin and 10 being Fascisttown, I’d go with it being a solid -.5 or -1), but I also don’t think it affects their story decisions to a large scale. Do they maybe decide that the story about the single mother losing her house due to a bank screwing her over is more important for a 1 minute fluff piece than a story about a guy shooting a thief in his own home? Maybe. I think it really depends on whether there’s video or not.
OTOH, on the big stuff, I think the media is biased toward conflict, sensationalism, and a heavy dose of Beltway Parochialism. Again, an actually biased media committed to winning victories of leftism wouldn’t have leaped on every single misstatement of Gore’s during the 2000 Election, wouldn’t have treated the Swift Boat Veterans as worth covering, and wouldn’t have elevated John McCain to SuperMaverick status.
I do think that maybe they got a little carried away during the early stages of the Obama campaign, but anybody who thinks the press has been overly friendly to Obama during his Presidency is living in their own Roger Ailes-built bubble.Report
No, I don’t necessarily think they are far-left as a whole, just somewhat to the left of the average American (and again, maybe that is to the good). But if the feedback mechanism does exist, a small variance can end up far greater over time.
I hope that their bias toward sensationalism and conflict (and competition – even the basket case that is Fox, for all the criticism they get and deserve) is a sufficient corrective.
We don’t need any one source to be ‘fair and balanced’ IMO, so long as the ecosystem as a whole mostly is, and voters are taking wide enough media samples.Report
And like I said, I think if you added up every reporter in every single newspaper in America, you’d get a population a tick or two to the left of the average American.
But on the other hand, I think, among the 50 or so reporters who actually can affect public perception via what they write and say on TV, they believe in the neoliberal consensus on economics, are probably a little more hawkish than the average American, and a little more liberal on social policy than the average American. The most “far-left” reporter or columnist who gets regular TV time is probably Paul Krugman and even he believes in free trade. 🙂Report
That’s because he’s an economist, we all believe in free trade right across the ideological spectrum. The only real exception I can think of is Joseph Stiglitz, and his middling scepticism places him way outside the mainstream of the profession.Report
That was my point. Even somebody on the “far left” of acceptable opinion in the US is actually a pretty mainstream economist, even if you may disagree with him on his opinions on debt spending and such.Report
A serious question: if one of your objections to media coverage of economics is that it is too deferential to the consensus of economists, how is this qualitatively different from conservative objections that media coverage of, say, global warming is too deferential to the consensus of global warming scientists?Report
I think it is telling though that Krugman is basically the most liberal academic we see on TV or regular column space in the US.
Unlike say in the UK where Eric Hobsawm was considered a historian mainstream enough that even his ideological nemesii like Niall Ferguson were lavish in their praise.Report
I think the problem is not that they claim the media is too deferential to the GW consensus (i.e. just that they wish that skeptics got more attention), but that they want the media to act as if there isn’t a consensus, or as if, if there is one, it’s a conspiracy or illegitimate in some other way.Report
…But if that *is* the leftist argument about economics coverage (or if it’s not the denialit argumkent about climate science coverege – but it is), then yeah, not much.Report
The only real exception I can think of is Joseph Stiglitz, and his middling scepticism places him way outside the mainstream of the profession.
But given the centrality of the role, how do you explain his scepticism away?Report
“the role he played…”Report
Jesse:
Krugman is a left-wing economist, the average economist is very slightly left of centre in the US, and Krugman is more to the left than that. But economists don’t fall neatly into the standard ideological groupings, which makes sense given that we know things non-economists don’t. Even the left of the profession overwhelmingly supports free trade, and that should tell you something.Report
Stillwater:
I don’t think there’s much to explain, Stigliz is an eminent economist (his Nobel is well-earned) but he didn’t make nay major contributions in trade theory, his Nobel work was on collective action problems.
Furthermore Stiglitz is not a Naomi Klein style anti-globalist, he’s just less enthusiastic than the average economist, I still think he’s more in favour of liberalising trade than the average non-economist.
And even on an issue like trade where the consensus is near-universal there will always be outliers, such is the nature of a random distribution.Report
James K: I wasn’t referring to Stiglitz academic work but rather his role as chief economist at the World Bank, a role which gave him a rather unique perspective on the effects of free trade policies.Report
A perspective on the “free trade” policies you get in practice maybe, where agriculture is excluded as a matter of course. As it is the majority of “free trade” deals the US signs are little more than an imposition of American Copyright laws on other countries.
There is a real basis for criticising how trade agreements work in practice, but that’s largely because most of them don’t really liberalise trade very much.Report
Jesse,
Even just a half- or full-point towards liberalism is enough to identify a skew that is both real and meaningful.Report
Well, IMOHO, almost all of that half a point is on social issues, not economic or foreign policy issues, so I’m perfectly OK with the elite media being more friendly to gays than the average population, just like it was more friendly to blacks than the average population for a long time.
Or, to be blunt, do you think it was a problem there weren’t more people in favor of segregation working for the New York Times in the 60’s?Report
I don’t think there necessarily is a “problem”. And I also think the social issues in the news are largely what drive people’s perceptions of it.Report
And again, to be blunt, I don’t really give a damn if a straight white Christian male thinks the media is too friendly to minority group x, z, or c.
I was kind of making a joke, but I do seriously think, that if the same society existed in 1964 or so, there’d be articles from VSP in the New Republic and Atlantic about how liberals really need to listen to Strom Thurmond’s points on race-mixing and how the media really is too unfriendly to people who just don’t want black people in their schools.Report
Jesse-
Is anyone making that argument?
First off, I’m liberal. More liberal than most folks here.
Second off, the reason that I advocate diversifying their political beliefs is not because I think equal shrift should be given to climate change deniers. Far from it. But if the media is as I believe (and I wish JB or whomever would chime in with that link from earlier) and most of the folks involved are liberal, it risks making the same mistake that HP did, wherein they completely miss something very important because there was a lack of differing perspectives in the room.
If the “liberal” take on an issue is the correct one, that should be reported. Likewise if it is the “conservative” take that is correct. I’m not advocating less accurate reporting. I’m advocating a better system of checks through a broader range of viewpoints.
HP wouldn’t have rolled out “racist” computers if they had more dark-skinned people in the room when they were developing them.Report
Considering the biggest “miss” the “liberal media” has made in the past few years was a massive conservative-leaning miss (ie. Iraq War), I could make the argument that the problem with the media is that it’s full of liberals scared to be called by liberals so they pull the false equivalence “he said/she said” game out to cover their own ass.
For instance, I don’t even see the story you posted as an evidence of liberal media bias. I see it as a bias toward sensationalism – “HP CAMERA DOESN’T SEE BLACK PEOPLE” gets hits.
But again, I don’t see the liberal bias in the media that you do. I wish the media was as liberally biased as people thought it was.Report
For the record, I wasn’t holding up the HP story as indicative of any type of bias one way or another. I was paralleling the “way things ended up the way they did there” with “the way things ended up the way they did in the media”.
I read about the HP story on that blog. I don’t know where else it was covered. I didn’t hear anyone insist that the story deserved more or less airtime than it received.Report
+1000. It becomes the central problem of our age.Report
I agree with that. But that in and of itself isn’t a form of bias. Bias is a relational property, one that relies on differences, and in this case it’s a difference in ideological perspective. One thing conservatives have been able to do – and I say this with reluctant praise, from a tactical pov – is to politicize everything in society. There are no neutral topics. There is no such thing as “objective” reporting. And it manifests, it seems to me, in the following way: a news report which fails to embrace the conservative perspective is by definition evidence of anti-conservative bias. A new report which fails to mention the conservative view of things is by definition evidence of anti-conservative bias.
Psychologically and politically, this type of view of the news and media generally may make perfect sense. But it suffers from circularity, I think. It’s bordering on tautological. Any news story which fails to present the conservative view of things is anti-conservative, where “anti-conservative” is defined as a news story which fails to present the conservative view.
That’s not to say there isn’t actual evidence which might establish institutional bias in the media against conservatives. But until the conservative can define what a neutral, or objective, news story would actually be without circularly defining it in terms of the conservative view, then the empirical case can’t be made.Report
One thing conservatives have been able to do – and I say this with reluctant praise, from a tactical pov – is to politicize everything in society.
In the 80’s, there was this “everything is political” thing going on and the people saying this sort of thing were from non-mainstream left positions. (The feminist criticism of science provides a particularly glaring example.)
Conservatives may be doing it well… but they weren’t the first to do it. They’re reacting.Report
There’s a difference between a random Women’s Studies group in Cambridge believing it and a national political party believing something. But, I know, a pox on both houses!Report
As much as I love denigrating Women Studies groups (and doubly so for ones in Cambridge), that wasn’t what I was doing here.
I’d say that the Conservatives are picking up post-modern criticism tools that the Liberals have stopped using (and, in some cases, haven’t used for a while).
How’s this? The Liberal Media, through its pervasive bias, is attempting to change the internalized assumptions of the culture at large in such a way that will preclude certain options from even being seen. We’re not talking about them hammering people over the head, we’re talking about them gently placing filters in front of people so that when they encounter something that doesn’t fit a preconceived notion, their automatic response is confusion or anger and to attack the other position as if it were saying something that it were not and thus reinforce the world view that they mistakenly see as their own.Report
Conservatives may be doing it well… but they weren’t the first to do it. They’re reacting.
Agreed. I think this is another instance where conservatives see the successes of liberals, identify the argument underpinning those successes, and turn them into their own weapons. Whereas the feminist critique of the status quo challenged the idea if neutrality in interpersonal social norms especially as they relate to women, conservatives are challenging the idea of neutrality in news reporting, academia, science, history, etc. It’s devilishly clever.Report
I should add – or maybe I shouldn’t! – that Corey Robin talks quite a bit about this very phenomenon in The Reactionary Mind.Report
I wish that Robin was more anthropological in his approach (as opposed to feeling like he was explaining “here’s what’s the matter with Kansas!”) but that’s a paltry complaint given the sins I had feared the book would commit.Report
See, the treatment of Saul Alinsky as some Guru of leftism when in reality, Rush Limbaugh and Newt Gingrich built their careers on Alinsky-style tactics.Report
I think there’s a difference between those two instances. The way conservatives (and to an extent, though perhaps a lesser one) have politicized everything is to suggest that everything is either liberal or conservative. The way certain left-wing groups (feminists, e.g.) did it in the 80s and 90s was more to suggest that everything had political implications, not that everything was either liberal or conservative. In essence, what’s happening these days, nad we see it all the time with Tom’s comments and posts, is the view that everything is either Fox or MSNBC.Report
I have relatives for whom saying “MERRY CHRISTMAS!” at Target (assuming it’s between Thanksgiving and, oh, New Year’s) has become a revolutionary act.
It’s difficult for me to not have flashbacks to Foucault.Report
That pendulum swings both ways.Report
That’s what the Republicans heard, so they banned pendulums from their convention.Report
You should read Foucault again, then, if you think that comes out of him. (If you get that out of the “theory” people of the 80s, though, then you’re probably right.)
Though it’s never bad to be reminded of Foucault.Report
This comment just reminded me of the Foucault-Chomsky debates.
Thisis a good summary.Report
Still, I remember the first time I watched that. I thought two things: “Wow, Foucault looks like Uncle Fester,” and, “Wow, I didn’t know two people could talk past each other that well for that long.”
The 5 second summary is awesome, though.Report
I thought it was the assertions of power relationships thing. People who don’t have to assert their power vs. people who do vs. people who know better than to try, etc.Report
Ah, then I misunderstood why it reminded you of Foucault (but I now see why it reminded Still of the Foucault-Chomsky debates, given that I misunderstood you).
Still, always good to be reminded of Foucault.
And coincidentally, I was just thinking that the whole “liberal media bias” thing reminds me of Chomsky.Report
Hmmm. I agree that it is hard to quantify. But I don’t see why that changes what we can reasonably surmise, and at least try to factor it into our thinking.
Take as given the ’80 percent of journalists vote Obama’ figure (no idea if this figure is correct, it’s just a thought experiment, so replace Obama with ‘candidate/party X’ if you want).
Various studies of human behavior, show that we will go to great lengths to rationalize or justify our assumptions, actions and beliefs, even unconsciously. We tell ourselves stories, that help us make sense of the world and ourselves, in a light that makes us feel justified or comforted. When we are doing this rationalization to others, we tend to do this via the spoken or written word.
Is it then such a leap to assume that a population which leans 80% one way ideologically/politically, and *speaks and writes for a living*, will tend to make a great many unconscious choices in their speech or writing that attempt to justify and further their own beliefs and ends?
There is nothing nefarious, or conspiracy-minded here.
Just a population, and observed human behavior.Report
Glyph, I agree with some of that, especially the part about it being rational to assume certain things. I’m pointing in a different direction. You wrote:
But I don’t see why that changes what we can reasonably surmise, and at least try to factor it into our thinking.
I think it does change what we can reasonably surmise, and in the following way: if conservatives are defining “liberal media bias” as “the failure to present the conservative view of events”, then we can’t surmise anything about the media. What we can surmise – because it’s pretty obvious – is that the conservative is adopting a tautological definition of liberal bias which begs the question in their favor. On this definition, the media is biased against the conservative if it presents a story which is either objectively pro-liberal, or ideologically neutral. Anything short of pro-conservative constitutes evidence of an anti-conservative bias.
Does the conservative have a definition of objective, ideologically neutral news reporting? It seems to me we cannot assume that they do, since the standard by which conservatives measure bias is the failure of media to present the conservative view. There is no neutrality for them.Report
Btw, if you’re following one line of reasoning here, you might think the problem is easily resolved: lets settle on a definition of ideological neutrality and conduct some tests, some experiments, some analysis of actual news reporting. But who would a conservative trust to conduct this test? Not just any old person in academia, that’s for sure…Report
Agreed, I get where you are going. My point was more about logical inferences, and I think they can be made from knowledge of the voting patterns, if those patterns can be established to consistently run one way or another (and irrespective of which direction they run).
That is, if 80% of journos vote R, I would expect, at minimum, a slight bias toward R in the reporting.
This in turn feeds back into Kazzy’s point, I think – if the journo population were more ideologically-balanced (not neutral, which is slightly different), then their output should also be more ideologically-balanced.Report
Sure, from one pov having more conservatives in media would balance things out, at least in terms of voices. But that gets to the deeper question here, which is why (supposing it’s true) 80% of media people are liberals who vote D? Is it because liberals in media are consciously trying to indoctrinate the populace with liberal ideology? (The cabal theory!) Is it because they’re unconsciously selecting workmates with shared political beliefs? (How does this get off the ground? Were media firms created by and stocked with radically partisan “seed” liberals back in the day, a core group which initiated this self-selection cycle and in virtue of which it now persists?) Is it perhaps due to non-partisan factors, like basic curiosity about the world, and ability to see different pov, a desire to lead a “non-traditional” lifestyle? (Who knows?)Report
Are journalist jobs more likely to hire people with journalism degrees? Are college educated people more likely to be liberal on culturally sensitive issues when compared to 50-year old “some college” types who live out in the sticks?Report
Does Fox hire people with J school degrees? If so does make them a Fox News a liberal bastion. How about the WSJ? What biases do people with Econ or Business degrees tend to have? What would that say about the biases of the Econ and Business mediaReport
I remember when there wasn’t a Fox News. The “is media biased?” debate was a lot different. “What about the WSJ? What about Firing Line?”Report
Jaybird, I know you know this, but it bears mentioning: just because people say there’s a media bias doesn’t make it so. Even if they’ve been saying it for a long time. I mean, most of the conservative memes of the last thirty years were articulated almost verbatim during the civil war era. The surface rhetoric doesn’t change. The underlying arguments do, tho.
Conservatives, almost by definition, are fighting a rearguard action against progress. They’re fighting against reality of progress. That’s why Colbert can accurately say that reality has a liberal bias. I don’t say that as a knock on them, btw. I think that’s just descriptive. And there are aspects to modernity I’m uncomfortable with, so I understand the underlying sentiment to some degree.Report
Is there anything that could possibly, in theory, demonstrate a bias on the part of the media?Report
That’s an excellent question. The right one, I think. It could be demonstrated only if there was agreement on what constitutes ideological neutrality, it seems to me. I think we’re way past that, tho. Don’t you?Report
It used to be easy enough to quote Kipling:
I keep six honest serving-men
(They taught me all I knew);
Their names are What and Why and When
And How and Where and Who.
We know that they aren’t good enough, anymore.
If we listen to “Democracy Now!” and their coverage of an Occupy Wall Street rally in Seattle and listen to Rush Limbaugh’s coverage of the exact same rally, what do we know? Well, we know that the stuff that they agree on probably happened.
So I suppose ideological neutrality would look like a story where you walked away knowing what probably happened.Report
What you say makes sense That it’s the same kind of thing or justifies the wholly conscious, planned bias of Fox or the Washington Times is what I dispute.Report
I would leave Fox News and MSNBC out since I do think it is more deliberate there. I don’t read the Wash Times.Report
As someone who has professional engineering experience in pattern recognition technology (both speech recognition and visual object recognition)…
Pattern recognition systems perform best when they have been trained and tested across a HUGE set of data. The models that are constructed for speech recognition systems need to be trained with thousands of hours of speech from thousands of different speakers of a language with different pitch, timbre, and accents. The same thing with facial recognition — they should be trained with thousands of samples of video with lots of different people with different facial characteristics (male, female, bald, short hair, long hair, bearded, mustachioed, asian, indian, white, black…). If HP trained and tested their system with data obtained from just their development and test personnel (which could very well be the case), they’re guilty of engineering malpractice.
Obviously, their training/test data wasn’t a rich enough set. Wherever they obtained the data from, it was clearly substandard. I’d be curious to know where the data came from before I started thinking about “what it all means” regarding race, bias, and all that.Report
Thank you, Shannon. I am woefully ignorant of the details you ask for and apologize for not being able to offer them.
Regarding “what it all means”… I’m not sure it necessarily means anything except that SOMETHING went wrong with how HP developed their software with regards to how it responds to different skin color.
This is all making me think of the blind men and the elephant or whatever that little story is…Report
I do AI and vision systems work. Here’s an ancient problem, one which might help you along with what Shannon’s (correctly) pointing out.
The military set out to write a vision system which could hunt for tanks hiding in a treeline. It’s the very best place for a tank to hide, it can see everything and humans have trouble spotting it. But machine vision systems can be applied to this problem.
So they trained this system. At first, it had trouble, but once it was helped along just a bit, as all these expert systems must, it did a fine job.
So they put this critter in a plane and started flying around, trying to spot tanks they knew were hiding out there in the weeds. Couldn’t find anything. You see, the vision system had been trained on sunny days and this was a cloudy day.
Photographing persons of colour is harder than it looks. It’s a contrast problem. You need extra light to create distinction. You need to use an incident light meter and take a reading right at the cheek, then comp down the exposure so the highlights still work. When I was doing vision systems for cell phone testers, lighting was a far more difficult problem than actual pattern recognition. We built the robot so it would lift the phone into a black box where we could control the light and keep glare from blinding the vision system.Report
Thanks, Blaise. There do indeed seem to be technical differences when it comes to analyzing different skin tones (I’m reluctant to say it is necessarily harder to do darker skin because I think the difficulty is largely a function of your starting point). The technical limitations are real. The greater issue with HP was that they somehow failed to notice their existence.Report
I think it’s simply very likely that, well, HP’s system was built and tested by fairer skinned folks.
It’s the simplest explanation and generally the correct one. It’s also a very IT explanation, as “It works fine for me” (ie: the expert who designed it) or “I find it fun, if you don’t there’s something wrong with you” (the gaming guru who built his dream game, only to find it had a small audience).
One of the bigger issues with designing websites, GUI’s, or any sort of user interaction system is stepping outside your head as “the guy who wrote it” or “yourself as user” and trying to see how other people will approach it.
Personally, I start with imagining my dad using the system. You wouldn’t believe how much more user friendly it gets after I make a pass under the Dad-filter. 🙂Report
“I think it’s simply very likely that, well, HP’s system was built and tested by fairer skinned folks.”
Exactly. Which is probably fine the vast majority of the time. But, sometimes, it’s not.
Likewise, I think most of the decisions in the media are made my liberals. Which is fine and dandy most of the time. But, most of the time, it’s not.Report
Problem is, it is tougher to photograph black people. That’s because, well, they’re black. Photographing anything black is tough. Black won’t reflect light. Planck’s Black Body problem.
Every artist has known this since the Renaissance. Da Vinci:
To any white body receiving the light from the sun, or the air, the shadows will be of a bluish cast. Black is like a broken vessel, which is deprived of the capacity to contain anything. Report
So there are indeed specific technical limitations/difficulties when it comes to photographing dark skinned folks. Thanks for the technical know-how, Blaise.
Now if only there was some way for HP to have known that…Report
I’d like you to look at this famous image taken by Gordon Parks of the Fontanelle family. Look at the lighting. Notice how the father’s face has gone completely black. Look at the other faces and their orientation to the light.Report
I do image processing as well–biometrics and sometimes face recognition specifically. As BlaiseP pointed out, it’s a hard problem due in large part to contrast issues. It’s compounded by the fact that most of the easily implemented algorithms work in black and white, so contrast is really all you have left for extracting features.
The other issue is training sets. You need lots of data to train the system, and that usually comes in the form of commercial or research databases or lots of pictures of people you get around the office. That means that it’s either whatever some government or university was able to get or, as the tech industry tends to have handy, a bunch of white and south or east Asian people. Mostly men to boot.
Biometrics is rife with these challenges. We often end up having frank discussions that might be…uncomfortable…in most of the office environments. You can overcome the issues, but you need test data. My bread and butter is iris recognition. For a long time, one of the largest test databases was 100% Asian. You can imagine how limiting that might be if your project doesn’t have the money to set up shop somewhere and pay people for pictures of their eyes.Report
Anybody doing that sort of work should be aware of the issues of their training and test databases from the get go.
It’s like, you know, “step one” on the checklist. You don’t start data mining, or training neural nets or heuristics, or anything that involves large data sets without first taking a good, long hard look at the data and figuring out how clean it is, how comprehensive it is, and what you need to do to make it work for you.Report
It really is remarkable they let that product go to market. And if you look at the video, the guy isn’t even THAT dark skinned! Clearly major errors in their “data mining” or whatever it’s called… :-pReport
Believe me, we’re aware of it. But it’s still a problem for a variety of reasons, not least of which is the fact that computer vision on a low CPU budget is hella hard. I’m pretty sure that’s the technical term for it.
If I put my face tracking hat on for a moment and look at this stream, I think that his problem is not so much being black as it is being black and backlight simultaneously. Probably the worst case scenario for a consumer grade webcam. Since it’s a consumer webcam designed to run in real time, it’s probably an ultra lightweight algorithm that doesn’t hold up under anything but the best lighting configurations.Report
“…a cabal of leftist media moguls”
Could you name this cabal and provide evidence that they are, in fact, leftist?Report
Morzer-
I’m not arguing that a cabal exists. I’m arguing that if enough groups are just every-so-slightly liberal, the ultimate result of that can make it APPEAR that a cabal exists, even if the cause is much more mundane.
Think of it this way…
Imagine two teams playing a single elimination tournament. If the teams are even, it is a 50-50 proposition. If one team has a 5% edge, it is a 55-45 proposition.
If you are playing a best-of-three tournament, suddenly the better team wins not 55% of the time, but 58% of the time.
The further out you go, the bigger the disparity gets. But no individual unit is so greatly skewed. It just seems that way when you look at the final scoreboard.Report
Cabal ? Journ0-list. Andrew Sullivan, case closed:
Etc.
http://www.theatlantic.com/daily-dish/archive/2010/07/the-corruption-of-journo-list/184567/Report
The mainstream media would be best served to diversify the political leanings of their decision-makers.
Actually, I’d argue that the mainstream media (when it comes to news reporting) would be best served to diversify *all* of their biases. Well, they’re stuck with one; you want to hire people that have good speaking English.
Find top notch reporters in Indonesia, England, Russia, Brazil, China, Pakistan, Uganda, Malaysia, Australia, and hire them.
Bring them here.
Pay them lots of money to report, and give the reporters a good chunk of editorial leeway in deciding what to report on.
I cannot imagine how this wouldn’t turn out a product that is ten thousand times better than what we have now. This is what CNN should do, if it wants to compete with MSNBC and Fox. Do actual world reporting, and world perspective on local reporting.Report
Unfortunately, I’d think it’d be a better network, but still get worse ratings than whatever network was showing a special on a Pretty White Woman that was missing.Report
Dude, there’s plenty of Pretty Exotic Women that you can throw up against Pretty White Women.
A multiracial woman with a tinge of a British accent? That’s money, baby.Report
I believe this network already exists. It’s called BBC International and Al-Jazeera. (To some extent CNN International used to be like this, too)Report
Well, American’s won’t watch either of those, so yeah.Report
Both al-Jazeera and BBC have government sponsorship. Sure wish we had something like that here in the Land of the Free.
Oh, wait, we do. Sorta. It’s called al-Hurra. And Americans are forbidden to watch it. Well, you can if you’re clever and can pick it off from some feed in the Middle East.Report
“Do I think, on the whole, the media is biased towards the left? Yes. ”
Why don’t you come back to this space-time continuum? Then read
What Liberal Media? The Truth About Bias and the News by Eric Alterman.Report
The Nation’s Eric Alterman.
After that, you can read “he wasn’t particularly abrasive” by Christopher Hitchens.Report
There is a bias. It’s a bias towards “statism”. Few give the opposing view.Report