“But look, you found the notice, didn’t you?”

“Yes,” said Arthur. “yes I did. It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying ‘Beware of the Leopard.'”

–Douglas Adams, The Hitchhiker’s Guide to the Galaxy

Any apologia for Facebook’s recent behavioral study has to address one issue head on: that of informed consent. Informed consent is the bedrock upon which the ethics of all biomedical and behavioral research on humans rests. And it is where Facebook’s experiment went so wrong. Even the spirited defense by Meyer et al. concedes that the Facebook researchers did not obtain informed consent — before providing a shorthand explanation about why informed consent wasn’t necessary.

But this approach dodges a key criticism of the Facebook study. The researchers never argued that they didn’t need informed consent. They insisted that they had, in fact, obtained it from each and every one of their 689,000 subjects. In their view, any user who signed up for Facebook had tacitly agreed to be part of their experiment. After all, anyone with a Facebook account is bound by a “data use policy” which, the authors argued, gave the researchers carte blanche. Never mind that the data use policy is buried three clicks deep off of the main Facebook page, and that only a small fraction of users have likely read it, much less understood it. Never mind that, in fact, “research” was only added to the policy four months after the experiment began. Never mind that minors — who are not even capable of giving informed consent on their own — were probably included in the study.

This is not the standard of informed consent that scientists are expected to uphold — and required to by peer-reviewed journals such as the Proceedings of the National Academy of Sciences. That a Silicon-Valley company holds itself to a different ethical standard is not a justification for looking the other way when scientists breach the rules.

As to the secondary question about whether this research could ethically have been performed without informed consent, there is more room for disagreement. However, it is a fairly extraordinary claim that a study which involves an attempt to negatively influence moods — a harm, even if a relatively mild one — of hundreds of thousands of people, likely including tens or hundreds of thousands of minors, should not require explicit consent and oversight.

There’s no doubt that the ethical rules for human research are imperfect, inconsistently applied, and sometimes misapplied. Yet when experimenters feel entitled to potentially harm human beings without obtaining consent, when even in the midst of a public outcry fellow scientists refuse to condemn the practice for fear of a “chilling effect on valuable research” — this is when such rules prove how necessary they really are.