Field Guide

Change fails at the first and last steps

Lewin sketched change as three moves: loosen the current equilibrium, move, then stabilize the new one. Unless it is anchored, groups drift back.

Use it: A law is only the middle step. Unfreeze by making the case inside the institution; refreeze with replacement structures, clear roles and a scorecard.

Lewin, K. (1947). Frontiers in group dynamics. Human Relations, 1(1).

A mandate is one step of eight

Kotter watched more than 100 companies try to remake themselves. Only a few succeeded outright; most fell short, for recurring reasons: no felt urgency, no coalition to carry it, and victory declared before the change was rooted in the culture.

The only step an edict covers.

Use it: A law or directive removes an obstacle and stops there. Build the case and the coalition inside first, and anchor the replacement before calling it done.

Kotter, J. P. (1995). Leading change: Why transformation efforts fail. Harvard Business Review, 73(2).

Lower the resistance before you push harder

A situation holds steady when forces pushing for change balance forces holding it back. More push tends to raise resistance.

Use it: Remove one restraining force (a real role for affected staff, an answer for the accreditor) and the reform moves further than another mandate would.

Lewin, K. (1947). Frontiers in group dynamics. Human Relations, 1(1). Illustrative forces, not measured.

Orders hit the tip; culture lives below the waterline

Schein described culture in three layers: visible artifacts, the values people say they hold, and the unspoken assumptions underneath. The assumptions drive behavior.

Use it: Removing an office changes an artifact. If the assumptions stay, the artifacts grow back under new names.

Schein, E. H. (1990). Organizational culture. American Psychologist, 45(2); and Organizational Culture and Leadership.

Institutions build structures against anxiety

Studying a hospital nursing service, Menzies found many procedures protected nurses from anxiety rather than helping patients. These defenses felt necessary and resisted change.

Use it: Ask what anxiety a program relieves. That is the job a lawful replacement must do, or the old structure returns.

Menzies, I. E. P. (1960). A case-study in the functioning of social systems as a defence against anxiety. Human Relations, 13 (Tavistock retrospective).

Every institution has one job, and drift from it can be named

In the Tavistock tradition an institution is an open system defined by its primary task: what it must do to survive. Activity that serves members’ anxieties instead of the task is anti-task.

Use it: Test each program against the primary task: institutional neutrality in working terms, and a standard staff across the spectrum can accept.

Miller, E. J., and Rice, A. K. (1967). Systems of Organization. Tavistock.

Tight walls buy loyalty, and stop the group from learning

Every group needs a boundary. An over-bounded group makes it too thick: strong loyalty, rigid authority, purity rules, and outsiders treated as a threat. It rejects outside information for its source, so it cannot correct itself. An under-bounded group has no shared purpose, and members drift.

Use it: Outside arguments fail against an over-bounded group by definition. Bring information in through its own trusted members and stated values. In a captured office, open the boundary with cross-unit teams and outside review; storming the wall only thickens it. Check your own side’s walls too.

From Clayton Alderfer’s work on boundaries in organizations (late 1970s and 1980s), developed in group-relations practice.

A shared enemy holds the group together, so it must stay big

An anxious or divided group can push its doubts, rivalries and failures outward onto an enemy, which buys unity and spares it from facing its own problems. The enemy must stay frightening; if it shrinks, the infighting returns. A movement that rallies its base against one larger-than-life opponent runs this pattern, and many voters are now held more by fear of the other party than by liking their own.

Use it: Don’t play the bogeyman. Calm, ordinary, specific conduct makes the projected figure look exaggerated, and outrage confirms it. Turn the question back to the group’s own record: what have you built, apart from opposing him? When the enemy leaves the stage, expect the group to fracture, and welcome those who leave. Every movement runs this pattern, including ones we agree with.

Bion, W. R. (1961). Experiences in Groups. Tavistock. Volkan, V. (1985). The need to have enemies and allies. Political Psychology, 6(2). Abramowitz, A., and Webster, S. (2016). The rise of negative partisanship. Electoral Studies, 41.

About a quarter, committed, can flip a group’s norm

In online experiments with 194 people in 10 groups, committed minorities of 25% or more overturned an established convention; the largest that failed was 21%.

Use it: Reform needs a committed quarter of a department, not a majority. The same dynamic may help explain how small activist cadres reshaped institutions.

Centola, D., et al. (2018). Experimental evidence for tipping points in social convention. Science, 360. Lab conventions, not workplaces.

People who think they’re outnumbered go quiet, so the loud view looks bigger

Fear of isolation makes people who sense they are in the minority speak less, which makes their view seem rarer still. Apparent consensus can be far stronger than real consensus.

Use it: Anonymous staff surveys show the real distribution, and one visible, calm dissenter gives others permission to speak.

Noelle-Neumann, E. (1974). The spiral of silence: a theory of public opinion. Journal of Communication, 24(2).