Here’s a scary thought: “Google has been paying university professors for favorable policy papers.” The basic idea, put forth by the Google Transparency Project (GTP), is that Google is paying millions of dollars to academics in order to “buy” research that is favorable to their policy positions. Their report is called “Google Academics, Inc.” Catchy, I guess. It plays into dominant narratives that tech companies are evil. But it also plays into an increasingly troubling narrative that science can’t be trusted. In an attempt to discredit Google, this project’s report also implies that academics can be easily bought off. That research is for sale to the highest bidder. It is an easy leap from this concept to thinking that a political party is paying for climate change research, for example — and this is a narrative that is dangerous right now.
To be clear, the issue of research funding not being disclosed is a big deal. And it definitely happens. The Wall Street Journal article based in part on GTP’s data seems to have uncovered some problematic individual cases. And we do already know that this is an issue in the pharmaceutical industry, for example. But the good news is, this is something that must not happen very often in the tech industry. Because despite claims of “hundreds” of Google-funded, policy-influencing papers, in a database of 330 articles, 7 of them are mine, and mine isn’t the only questionable case.
One of the first papers listed in this database is titled Remixers’ Understandings of Fair Use Online. It is a research paper based on an interview study, where I talked to online content creators about their understandings of fair use, and how they make decisions about what they can and can’t do when it comes to creating and sharing remixes on online platforms. You might be wondering how this paper is favorable to Google’s policy positions. I am, too! If anything, the paper is critical of YouTube, since their policies and practices around fair use have chilling effects for content creation.
You might also be wondering why Google funded this work. Well, they didn’t. The inclusion of my papers in the database is based on the fact that I received a Google Policy Fellowship when I was a PhD student, in 2011. This fellowship paid for my living expenses while I had an unpaid internship at Creative Commons. The amount was $7,500 which sadly was not quite enough for me to break even for the summer (because rent in the bay area!), but without it, I definitely wouldn’t have been able to take that internship at all. And even though I never did any work for Google, and the work I did at Creative Commons that summer was completely unrelated to any of my research, that fellowship was apparently enough to taint every subsequent paper I have published that has a whiff of policy bent.
I also think it’s very important to note that most of my research included in this database was generously funded by the National Science Foundation — which is acknowledged in the papers themselves.
In other words, Google did not fund this research, even “indirectly.” And yet, according to GTP, I should credit Google in every paper I publish for the rest of my career that has anything to do with policy. (They seem to have done a keyword search for “copyright” among my publications.)
I’m highlighting my particular case because it is just one of some really questionable inclusions in the database. GTP wrote a blog post responding to criticism of their methodology, including my case. They note that myself and other policy fellows “raise valid points about whether a small Google stipend for a policy fellowship should constitute ‘indirect’ Google funding for subsequent research projects that support Google’s policy positions.”
(I’m not even going to speculate how they think my computer science research papers “support Google’s policy positions” but I have a feeling that they did not actually read the papers. Especially since my 200-page dissertation is one of those listed.)
However, according to GTP, I should have disclosed my fellowship in all of these published papers, because the sole purpose of these fellowships is to “influence their thinking and ultimately their future scholarship” and to buy “good will” from the recipient. They’re suggesting not just that Google can fund positive research, but that they can buy a researcher for their entire career.
I can’t help but find it slightly insulting to suggest that because of a small amount of money received in grad school, any research someone conducts for the rest of their career could be biased towards Google. (Even, apparently, the research that has nothing to do with Google!) I suppose there’s a slippery slope argument here, but by this logic, every paper that academics write should disclose any company they ever had an internship with, any company that ever gave them funding even if it is entirely unrelated to their current research, and any company that funds centers they are a part of or conferences they attend. (And at the far end of the spectrum, we should potentially also be keeping track of t-shirts and socks given away at various events.)
Computer science papers typically have page limits. Trust me, this would be a really big problem.
Typical practice is to disclose funding sources for research in an acknowledgment section of a paper — which is precisely what we do. And we also list all funding sources on our CVs. Mine lists my policy fellowship. But because that fellowship did not fund any of my research, I don’t mention it in my papers — not anymore than I mention the donated PS4 I won for a coding competition in any of my papers. I also don’t list the NSF funding for research that wasn’t funded by that grant. Funding is a good thing for us — not something that we’re trying to hide.
I suppose there are some who would argue that academics should not take corporate funding at all, even eschewing free t-shirts. But unfortunately, with government funding rates (e.g., NSF, NIH) dropping, this becomes increasingly difficult. And to be clear, for those who are unfamiliar with academic funding, in the sciences at least it is necessary. External funding pays for PhD students, equipment, and study expenses. (What it is not doing is going into a professor’s pocket.)
Maybe I’m naive and Google really was trying to influence my thinking for the rest of my career (in my case, it definitely didn’t work), but particularly when it comes to funding computer science research, I think that they have a vested interest in supporting good work and good people. And in the case of grad students, in investing in talent in hopes they might want to come work there someday. So yes, maybe there is something shady happening here. But I can at least tell you that it wasn’t happening with me. And if Google or anyone else does give me money for my research, I will happily crow about it from rooftops and acknowledge it in all my papers. It also wouldn’t keep me from continuing to criticize YouTube’s copyright practices.
Though my point in this post isn’t to argue whether or not I should personally be in this database, but to point out that the data included is ambiguous at worst. And if the data had to be padded with cases like mine, then that suggests that the real problem is probably a pretty small one. (And unfortunately, any real point they might have made about a real problem has been obscured by messy data. Headlines like “Google pays for favorable research!” that cites a couple of real cases and then says “and there are hundreds more!” while pointing to this data is misleading, at best.)
So I encourage people to think about this problem — the real problem of potential academic bias — using real, rigorous methods. Because with the current public climate with respect to science, this kind of witch hunt is irresponsible.
(Addendum 1: I waited a bit to write this, as I was hoping for some clarification from GTP. If they respond, I will update this post, but here is what I asked them.)
(Addendum 2: There are also questions about GTP’s funding and the motivation for this project, though I don’t know enough about that to speak to it. The Wired article has some speculation.)
(Update: If you were curious about Google’s response, they sent a gif to TechCrunch that basically covers the rebuttals from folks like me.)