Ethics and the Facebook Whistleblower
If you haven’t seen it yet, 60 Minutes interviewed Fances Haugen about what’s happening inside the social media giant Facebook. Ms. Haugen has also discuss this in The Wall Street Journal and appeared before congress to testify. I’m going to refer you to the links above for the details related to why she decided to become a whistleblower because I want to focus on evaluating the ethics of whistleblowing and if there was enough justification for her to take this action.
I teach engineering ethics in my day job and whistleblowing is always an interesting topic with the students. Many have no idea what it is until you describe it. Then they immediately say something like “oh, it’s when you rat out someone or tattle on the company.” After discussing it for a little while, most students start to understand that whistleblowing has its place in society and that there are good and bad reasons someone might choose to become a whistleblower.
The engineering ethics text that we use, Engineering Ethics: Concepts and Cases by Harris, Jr., et al., lays out two primary reasons why someone might become a whistleblower (pg. 86):
to prevent harm
to avoid being complicit
It is clear that Ms. Haugen sought to address #1 above. She’s explicitly states that she’s concerned about Facebook’s practices (e.g., their focus on increasing engagement through enticing anger, hate, and envy) and their knowing neglect to adjust the algorithm against these issues. Her position on #2 is less clear. Based on what I’ve learned, we can at best only make assumptions which we should avoid doing.
Focusing our analysis on preventing harm, Harris, Jr., et al. further detail Richard DeGeorge’s idea that whistleblowing can be morally permissible or obligatory. They provide DeGeorge’s criteria for these categories:
Permissible if:
- the harm caused is “serious and considerable”
- the whistleblower first reported things to their immediate supervisor
- after making no progress with the above, they tried every other internal option to resolve the issue.
Obligatory if:
- the whistleblower has documented evidence that their view is correct and the organization is in the wrong (note: the evidence must convince a “responsible, impartial observer”)
- the whistleblower can provide “strong evidence” that the act of whistleblowing will “prevent the threatened serious harm”
So were Ms. Haugen’s action permissible or obligatory according to these criteria?
Permissible Breakdown
Overall, there is just too much grey area here. I haven’t listened to everything Ms. Haugen stated or wrote, but it depends on what she tried to do within Facebook before she left the company. Let me explain.
Harm “serious and considerable”?
- I think it’s more than reasonable to say, yes. Facebook (and social media in general) has caused and is causing serious and considerable harm. Multiple studies and reports, independent of Haugen’s findings, have demonstrated this.
Reported to immediate supervisor?
- This is a grey area for me as I have not been able to review all of Ms. Haugen’s writing/speaking. She may have discussed this somewhere or in her congressional hearings, but unless she explicitly stated this we likely won’t know. (If you’re aware of this, please let me know.)
Exhausted all other internal reporting options?
- Again, this is hard to evaluate as an outsider and I haven’t been able to review everything. (Again, if you’re aware of this, please let me know.)
So were Ms. Haugen’s actions permissible (according to DeGeorge)?
> It depends on some unknowns that may or may not have taken place internally at Facebook.
Obligatory Breakdown
Again, my thinking here is mixed. Let’s break things down.
Did she have documented evidence that will convince a “responsible, impartial observer”?
- I think the answer to this is yes. The internal reports and the amount of material make her case here strong. I think the 60 Minutes piece, despite only discussing a small fraction of Ms. Haugen’s documentation, helped her make a strong case that she has evidence.
Can she provide strong evidence that the whistleblowing will “prevent the threatened serious harm”?
- Here, however, I’m less convinced. It comes down to the word “prevent.” Ms. Haugen states at the end of her 60 Minutes piece that she’s hoping something more can be done to regulate social media companies like Facebook. She states that she wants to help Facebook not harm them. First, as we’ve witnessed over the last few years, regulating social media companies (or the internet for that matter) is a highly contested issue. Are they platforms? Are the publishers? Who owns the content? Who is responsible for the content? There is clearly a line that can be crossed which causes content to be removed, but where exactly is that line? It’s a difficult issue to discuss and an even more difficult issue to get people to agree on. Second, while I think regulation might be helpful, I don’t think it will prevent the harm that social media imparts on society. This deserves more attention than I’m willing to give it in this post, but the primary issues driving the ills of social media are sin issues. They are not algorithm issues. They are human nature issues. These ills also did not start on social media. They’ve existed well before the internet and social media. Do outlets like Facebook, Instagram, Twitter, and Reddit (and their algorithms) allow this sin to be amplified? Yes, absolutely. Therefore, I would argue that regulation might help the issue, but it’s unlikely to prevent the issue. Moreover, many have discussed whether or not members of congress understand modern technology enough to develop realistic and relevant means of regulation.
So were Ms. Haugen’s actions obligatory (according to DeGeorge)?
> At this point, I would say no they were not. Primarily because I don’t think there is a convincing argument that her actions will prevent serious harm.
Conclusion
I’ve discussed that Ms. Haugen’s actions weren’t obligatory and there is too much grey area to determine if they were permissible.
This leads one to ask… Should she have spoken out?
Another question we could ask is… Is DeGeorge’s permissible/obligatory framework missing something important and, in turn, not an appropriate framework to analyze this situation?
I’ll intentionally leave these questions open for discussion and debate.