Discourse Matters: Designing better digital futures

A very similar version of this blog post originally appeared in Culture Digitally on June 5, 2015.

Words Matter. As I write this in June 2015, a United Nations committee in Bonn is occupied in the massive task of editing a document overviewing global climate change. The effort to reduce 90 pages into a short(er), sensible, and readable set of facts and positions is not just a matter of editing but a battle among thousands of stakeholders and political interests, dozens of languages, and competing ideas about what is real and therefore, what should or should not be done in response to this reality.

discoursematters

I think about this as I complete a visiting fellowship at Microsoft Research, where over a thousand researchers worldwide study complex world problems and focus on advancing state of the art computing. In such research environments the distance between one’s work and the design of the future can feel quite small. Here, I feel like our everyday conversations and playful interactions on whiteboards has the potential to actually impact what counts as the cutting edge and what might get designed at some future point.

But in less overtly “future making” contexts, our everyday talk still matters, in that words construct meanings, which over time and usage become taken for granted ways of thinking about the way the world works. These habits of thought, writ large, shape and delimit social action, organizations, and institutional structures.

In an era of web 2.0, networked sociality, constant connectivity, smart devices, and the internet of things (IoT), how does everyday talk shape our relationship to technology, or our relationships to each other? If the theory of social construction is really a thing, are we constructing the world we really want? Who gets to decide the shape of our future? More importantly, how does everyday talk construct, feed, or resist larger discourses?

rhetoric as world-making

From a discourse-centered perspective, rhetoric is not a label for politically loaded or bombastic communication practices, but rather, a consideration of how persuasion works. Reaching back to the most classic notions of rhetoric from ancient Greek philosopher Aristotle, persuasion involves a mix of logical, emotional, and ethical appeals, which have no necessary connection to anything that might be sensible, desirable, or good to anyone, much less a majority. Persuasion works whether or not we pay attention. Rhetoric can be a product of deliberation or effort, but it can also function without either.

When we represent the techno-human or socio-technical relation through words, images, these representations function rhetorically. World making is inherently discursive at some level. And if making is about changing, this process inevitably involves some effort to influence how people describe, define, respond to, or interact with/in actual contexts of lived experience.

I have three sisters, each involved as I am in world-making, if such a descriptive phrase can be applied to the everyday acts of inquiry that prompt change in socio-technical contexts. Cathy is an organic gardener who spends considerable time improving techniques for increasing her yield each year.  Louise is a project manager who designs new employee orientation programs for a large IT company. Julie is a biochemist who studies fish in high elevation waterways.

Perhaps they would not describe themselves as researchers, designers, or even makers. They’re busy carrying out their job or avocation. But if I think about what they’re doing from the perspective of world-making, they are all three, plus more. They are researchers, analyzing current phenomena. They are designers, building and testing prototypes for altering future behaviors. They are activists, putting time and energy into making changes that will influence future practices.

Their work is alternately physical and cognitive, applied for distinct purposes, targeted to very different types of stakeholders.  As they go about their everyday work and lives, they are engaged in larger conversations about what matters, what is real, or what should be changed.

Everyday talk is powerful not just because it has remarkable potential to persuade others to think and act differently, but also because it operates in such unremarkable ways. Most of us don’t recognize that we’re shaping social structures when we go about the business of everyday life. Sure, a single person’s actions can become globally notable, but most of the time, any small action such as a butterfly flapping its wings in Michigan is difficult to link to a tsunami halfway around the world. But whether or not direct causality can be identified, there is a tipping point where individual choices become generalized categories. Where a playful word choice becomes a standard term in the OED. Where habitual ways of talking become structured ways of thinking.

The power of discourse: Two examples

I mention two examples that illustrate the power of discourse to shape how we think about social media, our relationship to data, and our role in the larger political economies of internet related activities. These cases are selected because they cut across different domains of digital technological design and development. I develop these cases in more depth here and here.

‘Sharing’ versus ‘surfing’

The case of ‘sharing’ illustrates how a term for describing our use of technology (using, surfing, or sharing) can influence the way we think about the relationship between humans and their data, or the rights and responsibilities of various stakeholders involved in these activities. In this case, regulatory and policy frameworks have shifted the burden of responsibility from governmental or corporate entities to individuals. This may not be directly caused by the rise in the use of the term ‘sharing’ as the primary description of what happens in social media contexts, but this term certainly reinforces a particular framework that defines what happens online. When this term is adopted on a broad scale and taken for granted, it functions invisibly, at deep structures of meaning. It can seem natural to believe that when we decide to share information, we should accept responsibility for our action of sharing it in the first place.

It is easy to accept the burden for protecting our own privacy when we accept the idea that we are ‘sharing’ rather than doing something else. The following comment seems sensible within this structure of meaning: “If you didn’t want your information to be public, you shouldn’t have shared it in the first place.”  This explanation is naturalized, but is not the only way of seeing and describing this event. We could alternately say we place our personal information online like we might place our wallet on the table. When someone else steals it, we’d likely accuse the thief of wrongdoing rather than the innocent victim who trusted that their personal belongings would be safe.

A still different frame might characterize personal information as an extension of the body or even a body part, rather than an object or possession. Within this definition, disconnecting information from the person would be tantamount to cutting off an arm. As with the definition of the wallet above, accountability for the action would likely be placed on the shoulders of the ‘attacker’ rather than the individual who lost a finger or ear.

‘Data’ and quantification of human experience

With the rise of big data, we have entered (or some would say returned to) an era of quantification. Here, the trend is to describe and conceptualize all human activity as data—discrete units of information that can be collected and analyzed. Such discourse collapses and reduces human experience. Dreams are equalized with body weight; personality is something that can be categorized with a similar statistical clarity as diabetes.

The trouble of using data as the baseline unit of information is that it presents an imaginary of experience that is both impoverished and oversimplified. This conceptualization is coincidental, of course, in that it coincides with the focus on computation as the preferred mode of analysis, which is predicated on the ability to collect massive quantities of digital information from multiple sources, which can only be measured through certain tools.

“Data” is a word choice, not an inevitable nomenclature. This choice has consequence from the micro to macro, from the cultural to the ontological. This is the case because we’ve transformed life into arbitrarily defined pieces, which replace the flow of lived experience with information bits. Computational analytics makes calculations based on these information bits. This matters, in that such datafication focuses attention on that which exists as data and ignores what is outside this configuration. Indeed, data has become a frame for that which is beyond argument because it always exists, no matter how it might be interpreted (a point well developed by many including Daniel Rosenberg in his essay Data before the fact).

We can see a possible outcome of such framing in the emerging science and practice of “predictive policing.” This rapidly growing strategy in large metropolitan cities is a powerful example of how computation of tiny variables in huge datasets can link individuals to illegal behaviors. The example grows somewhat terrifying when we realize these algorithms are used to predict what is likely to occur, rather than to simply calculate what has occurred. Such predictions are based on data compiled from local and national databases, focusing attention on only those elements of human behavior that have been captured in these data sets (for more on this, see the work of Sarah Brayne)

We could alternately conceptualize human experience as a river that we can only step in once, because it continually changes as it flows through time-space. In such a Heraclitian characterization, we might then focus more attention on the larger shape and ecology of the river rather than trying to capture the specificities of the moment when we stepped into it.

Likewise, describing behavior in terms of the chemical processes in the brain, or in terms of the encompassing political situation within which it occurs will focus our attention on different aspects of an individual’s behavior or the larger situation to which or within which this behavior responds. Each alternative discourse provokes different ways of seeing and making sense of a situation.

When we stop to think about it, we know these symbolic interactions matter. Gareth Morgan’s classic work about metaphors of organization emphasizes how the frames we use will generate distinctive perspectives and more importantly, distinctive structures for organizing social and workplace activities.  We might reverse engineer these structures to find a clash of rivaling symbols, only some of which survive to define the moment and create future history. Rhetorical theorist Kenneth Burke would talk about these symbolic frames as myths. In a 1935 speech to the American Writer’s Congress he notes that:

“myth” is the social tool for welding the sense of interrelationship by which [we] can work together for common social ends. In this sense, a myth that works well is as real as food, tools, and shelter are.

These myths do not just function ideologically in the present tense. As they are embedded in our everyday ways of thinking, they can become naturalized principles upon which we base models, prototypes, designs, and interfaces.

Designing better discourses

How might we design discourse to try to intervene in the shape of our future worlds? Of course, we can address this question as critical and engaged citizens. We are all researchers and designers involved in the everyday processes of world-making. Each, in our own way, are produsing the ethics that will shape our future.

This is a critical question for interaction and platform designers, software developers, and data scientists. In our academic endeavors, the impact of our efforts may or may not seem consequential on any grand scale. The outcome of our actions may have nothing to do with what we thought or desired from the outset. Surely, the butterfly neither intends nor desires to cause a tsunami.

butterfly effect comic
Image by J. L. Westover

Still, it’s worth thinking about. What impact do we have on the larger world? And should we be paying closer attention to how we’re ‘world-making’ as we engage in the mundane, the banal, the playful? When we consider the long future impact of our knowledge producing practices, or the way that technological experimentation is actualized, the answer is an obvious yes.  As Laura Watts notes in her work on future archeology:

futures are made and fixed in mundane social and material practice: in timetables, in corporate roadmaps, in designers’ drawings, in standards, in advertising, in conversations, in hope and despair, in imaginaries made flesh.

It is one step to notice these social construction processes. The challenge then shifts to one of considering how we might intervene in our own and others’ processes, anticipate future causality, turn a tide that is not yet apparent, and try to impact what we might become.

Acknowledgments and references

Notably, the position I articulate here is not new or unique, but another variation on a long running theme of critical scholarship, which is well represented by members of the Social Media Collective. I am also indebted to a long list of feminist and critical scholarship.  This position statement is based on my recent interests and concerns about social media platform design, the role of self-learning algorithmic logics in digital culture infrastructures, and the ethical gaps emerging from rapid technological development. It derives from my previous work in digital identity, ethnographic inquiry of user interfaces and user perceptions, and recent work training participants to use auto-ethnographic and phenomenology techniques to build reflexive critiques of their lived experience in digital culture. There are, truly, too many sources and references to list here, but as a short list of what I directly mentioned:

Kenneth L. Burke. 1935. Revolutionary symbolism in America. Speech to the American Writer’s Congress, February 1935. Reprinted in The Legacy of Kenneth Burke. Herbert W. Simons and Trevor Melia (eds). Madison: U of Wisconsin Press, 1989. Retrieved 2 June 2015 from: http://parlormultimedia.com/burke/sites/default/files/Burke-Revolutionary.pdf

Annette N. Markham. Forthcoming. From using to sharing: A story of shifting fault lines in privacy and data protection narratives. In Digital Ethics (2nd ed). Baastian Vanaker, Donald Heider (eds). Peter Lang Press, New York. Final draft available in PDF here

Annette N. Markham. 2014. Undermining data: A critical examination of a core term in scientific inquiry. First Monday, 18(10).

Gareth Morgan. 1986. Images of Organization. Sage Publications, Thousand Oaks, CA.

Daniel Rosenberg. 2013. Data before the fact. In Raw data’ is an oxymoron. Lisa Gitelman (ed). Cambridge, Mass.: MIT Press, pp. 15–40.

Laura Watts. 2015. Future archeology: Re-animating innovation in the mobile telecoms industry. In Theories of the mobile internet: Materialities and imaginaries. Andrew Herman, Jan Hadlaw, Thom Swiss (Eds). Routledge Press,

The Facebook “It’s Not Our Fault” Study

Today in Science, members of the Facebook data science team released a provocative study about adult Facebook users in the US “who volunteer their ideological affiliation in their profile.” The study “quantified the extent to which individuals encounter comparatively more or less diverse” hard news “while interacting via Facebook’s algorithmically ranked News Feed.”*

  • The research found that the user’s click rate on hard news is affected by the positioning of the content on the page by the filtering algorithm. The same link placed at the top of the feed is about 10-15% more likely to get a click than a link at position #40 (figure S5).
  • The Facebook news feed curation algorithm, “based on many factors,” removes hard news from diverse sources that you are less likely to agree with but it does not remove the hard news that you are likely to agree with (S7). They call news from a source you are less likely to agree with “cross-cutting.”*
  • The study then found that the algorithm filters out 1 in 20 cross-cutting hard news stories that a self-identified conservative sees (or 5%) and 1 in 13 cross-cutting hard news stories that a self-identified liberal sees (8%).
  • Finally, the research then showed that “individuals’ choices about what to consume” further limits their “exposure to cross-cutting content.” Conservatives will click on only 17% a little less than 30% of cross-cutting hard news, while liberals will click 7% a little more than 20% (figure 3).

My interpretation in three sentences:

  1. We would expect that people who are given the choice of what news they want to read will select sources they tend to agree with–more choice leads to more selectivity and polarization in news sources.
  2. Increasing political polarization is normatively a bad thing.
  3. Selectivity and polarization are happening on Facebook, and the news feed curation algorithm acts to modestly accelerate selectivity and polarization.

I think this should not be hugely surprising. For example, what else would a good filter algorithm be doing other than filtering for what it thinks you will like?

But what’s really provocative about this research is the unusual framing. This may go down in history as the “it’s not our fault” study.

Facebook: It’s not our fault.

I carefully wrote the above based on my interpretation of the results. Now that I’ve got that off my chest, let me tell you about how the Facebook data science team interprets these results. To start, my assumption was that news polarization is bad.  But the end of the Facebook study says:

“we do not pass judgment on the normative value of cross-cutting exposure”

This is strange, because there is a wide consensus that exposure to diverse news sources is foundational to democracy. Scholarly research about social media has–almost universally–expressed concern about the dangers of increasing selectivity and polarization. But it may be that you do not want to say that polarization is bad when you have just found that your own product increases it. (Modestly.)

And the sources cited just after this quote sure do say that exposure to diverse news sources is important. But the Facebook authors write:

“though normative scholars often argue that exposure to a diverse ‘marketplace of ideas’ is key to a healthy democracy (25), a number of studies find that exposure to cross-cutting viewpoints is associated with lower levels of political participation (22, 26, 27).”

So the authors present reduced exposure to diverse news as a “could be good, could be bad” but that’s just not fair. It’s just “bad.” There is no gang of political scientists arguing against exposure to diverse news sources.**

The Facebook study says it is important because:

“our work suggests that individuals are exposed to more cross-cutting discourse in social media they would be under the digital reality envisioned by some

Why so defensive? If you look at what is cited here, this quote is saying that this study showed that Facebook is better than a speculative dystopian future.*** Yet the people referred to by this word “some” didn’t provide any sort of point estimates that were meant to allow specific comparisons. On the subject of comparisons, the study goes on to say that:

“we conclusively establish that…individual choices more than algorithms limit exposure to attitude-challenging content.”

compared to algorithmic ranking, individuals’ choices about what to consume had a stronger effect”

Alarm bells are ringing for me. The tobacco industry might once have funded a study that says that smoking is less dangerous than coal mining, but here we have a study about coal miners smoking. Probably while they are in the coal mine. What I mean to say is that there is no scenario in which “user choices” vs. “the algorithm” can be traded off, because they happen together (Fig. 3 [top]). Users select from what the algorithm already filtered for them. It is a sequence.**** I think the proper statement about these two things is that they’re both bad — they both increase polarization and selectivity. As I said above, the algorithm appears to modestly increase the selectivity of users.

The only reason I can think of that the study is framed this way is as a kind of alibi. Facebook is saying: It’s not our fault! You do it too!

Are we the 4%?

In my summary at the top of this post, I wrote that the study was about people “who volunteer their ideological affiliation in their profile.” But the study also describes itself by saying:

“we utilize a large, comprehensive dataset from Facebook.”

“we examined how 10.1 million U.S. Facebook users interact”

These statements may be factually correct but I found them to be misleading. At first, I read this quickly and I took this to mean that out of the at least 200 million Americans who have used Facebook, the researchers selected a “large” sample that was representative of Facebook users, although this would not be representative of the US population. The “limitations” section discusses the demographics of “Facebook’s users,” as would be the normal thing to do if they were sampled. There is no information about the selection procedure in the article itself.

Instead, after reading down in the appendices, I realized that “comprehensive” refers to the survey research concept: “complete,” meaning that this was a non-probability, non-representative sample that included everyone on the Facebook platform. But out of hundreds of millions, we ended up with a study of 10.1m because users were excluded unless they met these four criteria:

  1. “18 or older”
  2. “log in at least 4/7 days per week”
  3. “have interacted with at least one link shared on Facebook that we classified as hard news”
  4. “self-report their ideological affiliation” in a way that was “interpretable”

That #4 is very significant. Who reports their ideological affiliation on their profile?

add your political views

It turns out that only 9% of Facebook users do that. Of those that report an affiliation, only 46% reported an affiliation in a way that was “interpretable.” That means this is a study about the 4% of Facebook users unusual enough to want to tell people their political affiliation on the profile page. That is a rare behavior.

More important than the frequency, though, is the fact that this selection procedure confounds the findings. We would expect that a small minority who publicly identifies an interpretable political orientation to be very likely to behave quite differently than the average person with respect to consuming ideological political news.  The research claims just don’t stand up against the selection procedure.

But the study is at pains to argue that (italics mine):

“we conclusively establish that on average in the context of Facebook, individual choices more than algorithms limit exposure to attitude-challenging content.”

The italicized portion is incorrect because the appendices explain that this is actually a study of a specific, unusual group of Facebook users. The study is designed in such a way that the selection for inclusion in the study is related to the results. (“Conclusively” therefore also feels out of place.)

Algorithmium: A Natural Element?

Last year there was a tremendous controversy about Facebook’s manipulation of the news feed for research. In the fracas it was revealed by one of the controversial study’s co-authors that based on the feedback received after the event, many people didn’t realize that the Facebook news feed was filtered at all. We also recently presented research with similar findings.

I mention this because when the study states it is about selection of content, who does the selection is important. There is no sense in this study that a user who chooses something is fundamentally different from the algorithm hiding something from them. While in fact the the filtering algorithm is driven by user choices (among other things), users don’t understand the relationship that their choices have to the outcome.

not sure if i hate facebook or everyone i know
In other words, the article’s strange comparison between “individual’s choices” and “the algorithm,” should be read as “things I choose to do” vs. the effect of “a process Facebook has designed without my knowledge or understanding.” Again, they can’t be compared in the way the article proposes because they aren’t equivalent.

I struggled with the framing of the article because the research talks about “the algorithm” as though it were an element of nature, or a naturally occurring process like convection or mitosis. There is also no sense that it changes over time or that it could be changed intentionally to support a different scenario.*****

Facebook is a private corporation with a terrible public relations problem. It is periodically rated one of the least popular companies in existence. It is currently facing serious government investigations into illegal practices in many countries, some of which stem from the manipulation of its news feed algorithm. In this context, I have to say that it doesn’t seem wise for these Facebook researchers to have spun these data so hard in this direction, which I would summarize as: the algorithm is less selective and less polarizing. Particularly when the research finding in their own study is actually that the Facebook algorithm is modestly more selective and more polarizing than living your life without it.

Update: (6pm Eastern)

Wow, if you think I was critical have a look at these. It turns out I am the moderate one.

Eszter Hargittai from Northwestern posted on Crooked Timber that we should “stop being mesmerized by large numbers and go back to taking the fundamentals of social science seriously.” And (my favorite): “I thought Science was a serious peer-reviewed publication.”

Nathan Jurgenson from Maryland and Snapchat wrote on Cyborgology (“in a fury“) that Facebook is intentionally “evading” its own role in the production of the news feed. “Facebook cannot take its own role in news seriously.” He accuses the authors of using the “Big-N trick” to intentionally distract from methodological shortcomings. He tweeted that “we need to discuss how very poor corporate big data research gets fast tracked into being published.”

Zeynep Tufekci from UNC wrote on Medium that “I cannot remember a worse apples to oranges comparison” and that the key take-away from the study is actually the ordering effects of the algorithm (which I did not address in this post). “Newsfeed placement is a profoundly powerful gatekeeper for click-through rates.”

Update: (5/10)

A comment helpfully pointed out that I used the wrong percentages in my fourth point when summarizing the piece. Fixed it, with changes marked.

Update: (5/15)

It’s now one week since the Science study. This post has now been cited/linked in The New York Times, Fortune, Time, Wired, Ars Technica, Fast Company, Engaget, and maybe even a few more. I am still getting emails. The conversation has fixated on the <4% sample, often saying something like: "So, Facebook said this was a study about cars, but it was actually only about blue cars.” That’s fine, but the other point in my post is about what is being claimed at all, no matter the sample.

I thought my “coal mine” metaphor about the algorithm would work but it has not always worked. So I’ve clamped my Webcam to my desk lamp and recorded a four-minute video to explain it again, this time with a drawing.******

If the coal mine metaphor failed me, what would be a better metaphor? I’m not sure. Suggestions?

 

 

Notes:

* Diversity in hard news, in their study, would be a self-identified liberal who receives a story from FoxNews.com, or a self-identified conservative who receives one from the HuffingtonPost.com, where the stories are about “national news, politics, [or] world affairs.” In more precise terms, for each user “cross-cutting content” was defined as stories that are more likely to be shared by partisans who do not have the same self-identified ideological affiliation that you do.

** I don’t want to make this even more nitpicky, so I’ll put this in a footnote. The paper’s citations to Mutz and Huckfeldt et al. to mean that “exposure to cross-cutting viewpoints is associated with lower levels of political participation” is just bizarre. I hope it is a typo. These authors don’t advocate against exposure to cross-cutting viewpoints.

*** Perhaps this could be a new Facebook motto used in advertising: “Facebook: Better than one speculative dystopian future!”

**** In fact, algorithm and user form a coupled system of at least two feedback loops. But that’s not helpful to measure “amount” in the way the study wants to, so I’ll just tuck it away down here.

***** Facebook is behind the algorithm but they are trying to peer-review research about it without disclosing how it works — which is a key part of the study. There is also no way to reproduce the research (or do a second study on a primary phenomenon under study, the algorithm) without access to the Facebook platform.

****** In this video, I intentionally conflate (1) the number of posts filtered and (2) the magnitude of the bias of the filtering. I did so because the difficulty with the comparison works the same way for both, and I was trying to make the example simpler. Thanks to Cedric Langbort for pointing out that “baseline error” is the clearest way of explaining this.

(This was cross-posted to multicast and Wired.)

What does the Facebook experiment teach us?

I’m intrigued by the reaction that has unfolded around the Facebook “emotion contagion” study. (If you aren’t familiar with this, read this primer.) As others have pointed out, the practice of A/B testing content is quite common. And Facebook has a long history of experimenting on how it can influence people’s attitudes and practices, even in the realm of research. An earlier study showed that Facebook decisions could shape voters’ practices. But why is it that *this* study has sparked a firestorm?

In asking people about this, I’ve been given two dominant reasons:

  1. People’s emotional well-being is sacred.
  2. Research is different than marketing practices.

I don’t find either of these responses satisfying.

The Consequences of Facebook’s Experiment

Facebook’s research team is not truly independent of product. They have a license to do research and publish it, provided that it contributes to the positive development of the company. If Facebook knew that this research would spark the negative PR backlash, they never would’ve allowed it to go forward or be published. I can only imagine the ugliness of the fight inside the company now, but I’m confident that PR is demanding silence from researchers.

I do believe that the research was intended to be helpful to Facebook. So what was the intended positive contribution of this study? I get the sense from Adam Kramer’s comments that the goal was to determine if content sentiment could affect people’s emotional response after being on Facebook. In other words, given that Facebook wants to keep people on Facebook, if people came away from Facebook feeling sadder, presumably they’d not want to come back to Facebook again. Thus, it’s in Facebook’s better interest to leave people feeling happier. And this study suggests that the sentiment of the content influences this. This suggests that one applied take-away for product is to downplay negative content. Presumably this is better for users and better for Facebook.

We can debate all day long as to whether or not this is what that study actually shows, but let’s work with this for a second. Let’s say that pre-study Facebook showed 1 negative post for every 3 positive and now, because of this study, Facebook shows 1 negative post for every 10 positive ones. If that’s the case, was the one week treatment worth the outcome for longer term content exposure? Who gets to make that decision?

Folks keep talking about all of the potential harm that could’ve happened by the study – the possibility of suicides, the mental health consequences. But what about the potential harm of negative content on Facebook more generally? Even if we believe that there were subtle negative costs to those who received the treatment, the ongoing costs of negative content on Facebook every week other than that 1 week experiment must be more costly. How then do we account for positive benefits to users if Facebook increased positive treatments en masse as a result of this study? Of course, the problem is that Facebook is a black box. We don’t know what they did with this study. The only thing we know is what is published in PNAS and that ain’t much.

Of course, if Facebook did make the content that users see more positive, should we simply be happy? What would it mean that you’re more likely to see announcements from your friends when they are celebrating a new child or a fun night on the town, but less likely to see their posts when they’re offering depressive missives or angsting over a relationship in shambles? If Alice is happier when she is oblivious to Bob’s pain because Facebook chooses to keep that from her, are we willing to sacrifice Bob’s need for support and validation? This is a hard ethical choice at the crux of any decision of what content to show when you’re making choices. And the reality is that Facebook is making these choices every day without oversight, transparency, or informed consent.

Algorithmic Manipulation of Attention and Emotions

Facebook actively alters the content you see. Most people focus on the practice of marketing, but most of what Facebook’s algorithms do involve curating content to provide you with what they think you want to see. Facebook algorithmically determines which of your friends’ posts you see. They don’t do this for marketing reasons. They do this because they want you to want to come back to the site day after day. They want you to be happy. They don’t want you to be overwhelmed. Their everyday algorithms are meant to manipulate your emotions. What factors go into this? We don’t know.

Facebook is not alone in algorithmically predicting what content you wish to see. Any recommendation system or curatorial system is prioritizing some content over others. But let’s compare what we glean from this study with standard practice. Most sites, from major news media to social media, have some algorithm that shows you the content that people click on the most. This is what drives media entities to produce listicals, flashy headlines, and car crash news stories. What do you think garners more traffic – a detailed analysis of what’s happening in Syria or 29 pictures of the cutest members of the animal kingdom? Part of what media learned long ago is that fear and salacious gossip sell papers. 4chan taught us that grotesque imagery and cute kittens work too. What this means online is that stories about child abductions, dangerous islands filled with snakes, and celebrity sex tape scandals are often the most clicked on, retweeted, favorited, etc. So an entire industry has emerged to produce crappy click bait content under the banner of “news.”

Guess what? When people are surrounded by fear-mongering news media, they get anxious. They fear the wrong things. Moral panics emerge. And yet, we as a society believe that it’s totally acceptable for news media – and its click bait brethren – to manipulate people’s emotions through the headlines they produce and the content they cover. And we generally accept that algorithmic curators are perfectly well within their right to prioritize that heavily clicked content over others, regardless of the psychological toll on individuals or the society. What makes their practice different? (Other than the fact that the media wouldn’t hold itself accountable for its own manipulative practices…)

Somehow, shrugging our shoulders and saying that we promoted content because it was popular is acceptable because those actors don’t voice that their intention is to manipulate your emotions so that you keep viewing their reporting and advertisements. And it’s also acceptable to manipulate people for advertising because that’s just business. But when researchers admit that they’re trying to learn if they can manipulate people’s emotions, they’re shunned. What this suggests is that the practice is acceptable, but admitting the intention and being transparent about the process is not.

But Research is Different!!

As this debate has unfolded, whenever people point out that these business practices are commonplace, folks respond by highlighting that research or science is different. What unfolds is a high-browed notion about the purity of research and its exclusive claims on ethical standards.

Do I think that we need to have a serious conversation about informed consent? Absolutely. Do I think that we need to have a serious conversation about the ethical decisions companies make with user data? Absolutely. But I do not believe that this conversation should ever apply just to that which is categorized under “research.” Nor do I believe that academe is necessarily providing a golden standard.

Academe has many problems that need to be accounted for. Researchers are incentivized to figure out how to get through IRBs rather than to think critically and collectively about the ethics of their research protocols. IRBs are incentivized to protect the university rather than truly work out an ethical framework for these issues. Journals relish corporate datasets even when replicability is impossible. And for that matter, even in a post-paper era, journals have ridiculous word count limits that demotivate researchers from spelling out all of the gory details of their methods. But there are also broader structural issues. Academe is so stupidly competitive and peer review is so much of a game that researchers have little incentive to share their studies-in-progress with their peers for true feedback and critique. And the status games of academe reward those who get access to private coffers of data while prompting those who don’t to chastise those who do. And there’s generally no incentive for corporates to play nice with researchers unless it helps their prestige, hiring opportunities, or product.

IRBs are an abysmal mechanism for actually accounting for ethics in research. By and large, they’re structured to make certain that the university will not be liable. Ethics aren’t a checklist. Nor are they a universal. Navigating ethics involves a process of working through the benefits and costs of a research act and making a conscientious decision about how to move forward. Reasonable people differ on what they think is ethical. And disciplines have different standards for how to navigate ethics. But we’ve trained an entire generation of scholars that ethics equals “that which gets past the IRB” which is a travesty. We need researchers to systematically think about how their practices alter the world in ways that benefit and harm people. We need ethics to not just be tacked on, but to be an integral part of how *everyone* thinks about what they study, build, and do.

There’s a lot of research that has serious consequences on the people who are part of the study. I think about the work that some of my colleagues do with child victims of sexual abuse. Getting children to talk about these awful experiences can be quite psychologically tolling. Yet, better understanding what they experienced has huge benefits for society. So we make our trade-offs and we do research that can have consequences. But what warms my heart is how my colleagues work hard to help those children by providing counseling immediately following the interview (and, in some cases, follow-up counseling). They think long and hard about each question they ask, and how they go about asking it. And yet most IRBs wouldn’t let them do this work because no university wants to touch anything that involves kids and sexual abuse. Doing research involves trade-offs and finding an ethical path forward requires effort and risk.

It’s far too easy to say “informed consent” and then not take responsibility for the costs of the research process, just as it’s far too easy to point to an IRB as proof of ethical thought. For any study that involves manipulation – common in economics, psychology, and other social science disciplines – people are only so informed about what they’re getting themselves into. You may think that you know what you’re consenting to, but do you? And then there are studies like discrimination audit studies in which we purposefully don’t inform people that they’re part of a study. So what are the right trade-offs? When is it OK to eschew consent altogether? What does it mean to truly be informed? When it being informed not enough? These aren’t easy questions and there aren’t easy answers.

I’m not necessarily saying that Facebook made the right trade-offs with this study, but I think that the scholarly reaction of research is only acceptable with IRB plus informed consent is disingenuous. Of course, a huge part of what’s at stake has to do with the fact that what counts as a contract legally is not the same as consent. Most people haven’t consented to all of Facebook’s terms of service. They’ve agreed to a contract because they feel as though they have no other choice. And this really upsets people.

A Different Theory

The more I read people’s reactions to this study, the more that I’ve started to think that the outrage has nothing to do with the study at all. There is a growing amount of negative sentiment towards Facebook and other companies that collect and use data about people. In short, there’s anger at the practice of big data. This paper provided ammunition for people’s anger because it’s so hard to talk about harm in the abstract.

For better or worse, people imagine that Facebook is offered by a benevolent dictator, that the site is there to enable people to better connect with others. In some senses, this is true. But Facebook is also a company. And a public company for that matter. It has to find ways to become more profitable with each passing quarter. This means that it designs its algorithms not just to market to you directly but to convince you to keep coming back over and over again. People have an abstract notion of how that operates, but they don’t really know, or even want to know. They just want the hot dog to taste good. Whether it’s couched as research or operations, people don’t want to think that they’re being manipulated. So when they find out what soylent green is made of, they’re outraged. This study isn’t really what’s at stake. What’s at stake is the underlying dynamic of how Facebook runs its business, operates its system, and makes decisions that have nothing to do with how its users want Facebook to operate. It’s not about research. It’s a question of power.

I get the anger. I personally loathe Facebook and I have for a long time, even as I appreciate and study its importance in people’s lives. But on a personal level, I hate the fact that Facebook thinks it’s better than me at deciding which of my friends’ posts I should see. I hate that I have no meaningful mechanism of control on the site. And I am painfully aware of how my sporadic use of the site has confused their algorithms so much that what I see in my newsfeed is complete garbage. And I resent the fact that because I barely use the site, the only way that I could actually get a message out to friends is to pay to have it posted. My minimal use has made me an algorithmic pariah and if I weren’t technologically savvy enough to know better, I would feel as though I’ve been shunned by my friends rather than simply deemed unworthy by an algorithm. I also refuse to play the game to make myself look good before the altar of the algorithm. And every time I’m forced to deal with Facebook, I can’t help but resent its manipulations.

There’s also a lot that I dislike about the company and its practices. At the same time, I’m glad that they’ve started working with researchers and started publishing their findings. I think that we need more transparency in the algorithmic work done by these kinds of systems and their willingness to publish has been one of the few ways that we’ve gleaned insight into what’s going on. Of course, I also suspect that the angry reaction from this study will prompt them to clamp down on allowing researchers to be remotely public. My gut says that they will naively respond to this situation as though the practice of research is what makes them vulnerable rather than their practices as a company as a whole. Beyond what this means for researchers, I’m concerned about what increased silence will mean for a public who has no clue of what’s being done with their data, who will think that no new report of terrible misdeeds means that Facebook has stopped manipulating data.

Information companies aren’t the same as pharmaceuticals. They don’t need to do clinical trials before they put a product on the market. They can psychologically manipulate their users all they want without being remotely public about exactly what they’re doing. And as the public, we can only guess what the black box is doing.

There’s a lot that needs reformed here. We need to figure out how to have a meaningful conversation about corporate ethics, regardless of whether it’s couched as research or not. But it’s not so simple as saying that a lack of a corporate IRB or a lack of a golden standard “informed consent” means that a practice is unethical. Almost all manipulations that take place by these companies occur without either one of these. And they go unchecked because they aren’t published or public.

Ethical oversight isn’t easy and I don’t have a quick and dirty solution to how it should be implemented. But I do have a few ideas. For starters, I’d like to see any company that manipulates user data create an ethics board. Not an IRB that approves research studies, but an ethics board that has visibility into all proprietary algorithms that could affect users. For public companies, this could be done through the ethics committee of the Board of Directors. But rather than simply consisting of board members, I think that it should consist of scholars and users. I also think that there needs to be a mechanism for whistleblowing regarding ethics from within companies because I’ve found that many employees of companies like Facebook are quite concerned by certain algorithmic decisions, but feel as though there’s no path to responsibly report concerns without going fully public. This wouldn’t solve all of the problems, nor am I convinced that most companies would do so voluntarily, but it is certainly something to consider. More than anything, I want to see users have the ability to meaningfully influence what’s being done with their data and I’d love to see a way for their voices to be represented in these processes.

I’m glad that this study has prompted an intense debate among scholars and the public, but I fear that it’s turned into a simplistic attack on Facebook over this particular study rather than a nuanced debate over how we create meaningful ethical oversight in research and practice. The lines between research and practice are always blurred and information companies like Facebook make this increasingly salient. No one benefits by drawing lines in the sand. We need to address the problem more holistically. And, in the meantime, we need to hold companies accountable for how they manipulate people across the board, regardless of whether or not it’s couched as research. If we focus too much on this study, we’ll lose track of the broader issues at stake.

eyes on the street or creepy surveillance?

This summer, with NSA scandal after NSA scandal, the public has (thankfully) started to wake up to issues of privacy, surveillance, and monitoring. We are living in a data world and there are serious questions to ask and contend with. But part of what makes this data world messy is that it’s not so easy as to say that all monitoring is always bad. Over the last week, I’ve been asked by a bunch of folks to comment on the report that a California school district hired an online monitoring firm to watch its students. This is a great example of a situation that is complicated.

The media coverage focuses on how the posts that they are monitoring are public, suggesting that this excuses their actions because “no privacy is violated.” We should all know by now that this is a terrible justification. Just because teens’ content is publicly accessible does not mean that it is intended for universal audiences nor does it mean that the onlooker understands what they see. (Alice Marwick and I discuss youth privacy dynamics in detail in “Social Privacy in Networked Publics”.) But I want to caution against jumping to the opposite conclusion because these cases aren’t as simple as they might seem.

Consider Tess’ story. In 2007, she and her friend killed her mother. The media reported it as “girl with MySpace kills mother” so I decided to investigate the case. For 1.5 years, she documented on a public MySpace her struggles with her mother’s alcoholism and abuse, her attempts to run away, her efforts to seek help. When I reached out to her friends after she was arrested, I learned that they had reported their concerns to the school but no one did anything. Later, I learned that the school didn’t investigate because MySpace was blocked on campus so they couldn’t see what she had posted. And although the school had notified social services out of concern, they didn’t have enough evidence to move forward. What became clear in this incident – and many others that I tracked – is that there are plenty of youth crying out for help online on a daily basis. Youth who could really benefit from the fact that their material is visible and someone is paying attention.

Many youth cry out for help through social media. Publicly, often very publicly. Sometimes for an intended audience. Sometimes as a call to the wind for anyone who might be paying attention. I’ve read far too many suicide notes and abuse stories to believe that privacy is the only frame viable here. One of the most heartbreaking was from a girl who was commercially sexually exploited by her middle class father. She had gone to her school who had helped her go to the police; the police refused to help. She published every detail on Twitter about exactly what he had done to her and all of the people who failed to help her. The next day she died by suicide.  In my research, I’ve run across too many troubled youth to count. I’ve spent many a long night trying to help teens I encounter connect with services that can help them.

So here’s the question that underlies any discussion of monitoring: how do we leverage the visibility of online content to see and hear youth in a healthy way? How do we use the technologies that we have to protect them rather than focusing on punishing them?  We shouldn’t ignore youth who are using social media to voice their pain in the hopes that someone who cares might stumble across their pleas.

Urban theorist Jane Jacobs used to argue that the safest societies are those where there are “eyes on the street.” What she meant by this was that healthy communities looked out for each other, were attentive to when others were hurting, and were generally present when things went haywire. How do we create eyes on the digital street? How do we do so in a way that’s not creepy?  When is proactive monitoring valuable for making a difference in teens’ lives?  How do we make sure that these same tools aren’t abused for more malicious purposes?

What matters is who is doing the looking and for what purposes. When the looking is done by police, the frame is punitive. But when the looking is done by caring, concerned, compassionate people – even authority figures like social workers – the outcome can be quite different. However well-intended, law enforcement’s role is to uphold the law and people perceive their presence as oppressive even when they’re trying to help. And, sadly, when law enforcement is involved, it’s all too likely that someone will find something wrong. And then we end up with the kinds of surveillance that punishes.

If there’s infrastructure put into place for people to look out for youth who are in deep trouble, I’m all for it. But the intention behind the looking matters the most. When you’re looking for kids who are in trouble in order to help them, you look for cries for help that are public. If you’re looking to punish, you’ll misinterpret content, take what’s intended to be private and publicly punish, and otherwise abuse youth in a new way.

Unfortunately, what worries me is that systems that are put into place to help often get used to punish. There is often a slippery slope where the designers and implementers never intended for it to be used that way. But once it’s there….

So here’s my question to you. How can we leverage technology to provide an additional safety net for youth who are struggling without causing undue harm? We need to create a society where people are willing to check in on each other without abusing the power of visibility. We need more eyes on the street in the Jacbos-ian sense, not in the surveillance state sense. Finding this balance won’t be easy but I think that it behooves us to not jump to extremes. So what’s the path forward?

(I discuss this issue in more detail in my upcoming book “It’s Complicated: The Social Lives of Networked Teens.”  You can pre-order the book now!)