No Villain in the Room

I spoke with a business leader recently who is preparing to convene a group of “senior executive types" who are “used to being right and heard.” As peers from different organizations, the familiar hierarchical dynamics don’t apply. Nevertheless, as the leader noted, “part of having a lot of respect for each other means maybe a natural aversion to conflict or contradiction of one another.” Consequently, the meeting host is concerned that the participants might self-censor their divergent perspectives, leading to a conversation resting on incomplete situational awareness.

We reflexively blame silence on power because power gives us a villain to confront. But research indicates the most common causes of self-censorship are varied and subtle.

Hierarchical power dynamics can and do suppress information sharing. In one research study, “85% of respondents said they had been in situations where they felt unable to raise an issue to a supervisor even though they felt the issue was important1.” However, the most common reason for their silence wasn’t fear of retaliation or punishment. Rather, it was fear of being labeled negatively and of damaging valued relationships.

Even so, differential status raises the stakes of friction and may automatically trigger silence. Implicit voice theories posit that humans may be hardwired by evolution to be vigilant and self-protective around higher-status individuals2. That is, there are survival benefits to avoiding offense to those with greater power. Consequently, in a hierarchical organization, employees may self-censor regardless of how approachable or open their supervisors actually are.

Even in the absence of an obvious hierarchy, information can be withheld because silence is not simply the absence of voice; it’s an active behavior with distinct motivations. Understanding silence, therefore, requires understanding underlying motivations3.

• Acquiescent silence is the withholding of information based on resignation and is motivated by a sense of helplessness, that speaking up won’t make a difference.

• Defensive silence is the withholding of information based on fear. It’s motivated by self-protection. This is the silence exacerbated by power differentials.

• Prosocial silence is the withholding of information for altruistic motives, such as protecting colleagues or organizational interests.

So, silence among peers can have one or more inherently ambiguous motivations, since it doesn’t provide overt clues.

Complicating the narrative of cause and effect are instances in which information sharing is suppressed even when there is neither a power differential nor a motivated silence. The meta-analysis of hidden-profile research, covering dozens of studies and thousands of groups, shows that shared information is mathematically more likely to surface4. In hidden-profile experiments, group members each hold unique information critical to the best decision, but because shared information is statistically more likely to be mentioned, the unique insights often never surface. It’s silence without the silencer. Nobody chose to stay silent; there was simply no opportunity to give voice to an unshared perspective.

We can design interactions that make courage unnecessary, mitigate impacts on social capital, and reduce the friction from time pressure. Nevertheless, design has limits:

• Good conversational design probably can’t overcome a sense of futility. If people stay silent because they expect nothing to change, anonymous input doesn’t mean they’re likely to share their thoughts. After all, why share when input is likely to be ignored? The core problem is behavioral rather than structural.

• Sometimes, friction is productive. Real-time disagreement is how groups learn to grapple with tension. Engineer out all the friction, and that capability might atrophy.

• Some silence may be wisdom, not failure. The prosocial withholding of information can help keep a team intact. Surface everything, and you also surface what was better left unsaid (a colleague’s private frustration, a half-formed doubt that would only sow confusion).

The withholding of information probably isn’t due to the lack of courage to speak truth to power. We can’t make people brave enough to defeat a sampling artifact. When we presume that silence is caused by power differentials, we create a villain and a prescription for bravery. But the question usually isn’t, “How do we make people brave enough to speak?” It’s, “What lets the right things get said without anyone needing to be brave?”