The psychology of it

Every rule in the method trades supervision for trust. This page sets those rules against what psychology already knows about motivation, groups and being watched. It is our reading of the research, not part of the manifesto, and it is an argument, not a proof: nobody has measured this method.

Trust and freedom

The method starts from a belief: people reach their full potential when they are granted trust and freedom. One well-studied version of that idea is self-determination theory. It finds that people are more motivated when three needs are met: autonomy, acting by your own choice; competence, being effective at what you do; and relatedness, belonging to a group that matters to you. At work, environments that support autonomy go with more self-driven motivation and better well-being, and, on complex work, with better performance (Gagné and Deci, 2005).

Three needs, and the rules that serve them

Autonomy

Acting by your own choice

  • Work is pulled, never pushed
  • Full freedom inside the trunk: the how belongs to the owner

Competence

Being effective at what you do

  • Invariants and acceptance tests say clearly what done is
  • Done is defined by the people who will meet it

Relatedness

Belonging to a group that matters to you

  • Importance is decided together, at the weekly
  • The team still meets, daily and weekly

Needs from self-determination theory, as applied to work by Gagné and Deci (2005).

Being watched

Every document a member works in is visible to every other member, and everyone sees the same view: no view exists that only a lead can open. What the method refuses is recording the person, and the research is on the side of that line. A 2023 meta-analysis of 94 samples covering 23,461 workers found that electronic monitoring of performance did not improve performance, and that it went with more stress however it was done. Where monitoring was transparent, meaning people knew what was recorded and why, and where it was less invasive, workers’ attitudes were better (Ravid and colleagues, 2023).

The map is not a monitor of people. It shows the documents each member works in, the same view to everyone, with no read receipts, no activity trail and no “last seen editing”.

What a group does to effort

In the 1880s the French agricultural engineer Max Ringelmann measured how hard men pulled on a rope, alone and in groups. Alone, a man pulled 85.3 kilograms on average. In groups of seven, each pulled 65.0, and in a group of fourteen, 61.4 (Kravitz and Martin, 1986). The more people pulled, the less each one pulled.

The figure usually quoted comes from a table in the same article, and Ringelmann did not say which tasks it came from. In 1927 the psychologist Walther Moede presented it as rope pulling, and later writers repeated that. It falls in a straight line, and faster than the rope trials did, but it shows the shape plainly:

Ringelmann's table: more workers, less from each

Each worker gave less as the group grew

Output per worker, as a share of working alone

0%25%50%75%100%1100%234567849%

The eighth worker added nothing

Total output, in multiples of one worker alone

0123412345673.9283.92

Show the numbers
WorkersOutput per worker, as a share of working aloneTotal output, in multiples of one worker alone
1100%1.00
293%1.86
385%2.55
477%3.08
570%3.50
663%3.78
756%3.92
849%3.92

Ringelmann (1913), as reported by Kravitz and Martin (1986, Table 2). Ringelmann did not say which tasks these figures came from; the rope label comes from Moede (1927), and later writers repeated it. His own rope trials curved instead: 85.3 kg per man alone, 65.0 kg in groups of seven, 61.4 kg in a group of fourteen.

Two more details are usually told wrong. Ringelmann mostly blamed coordination. He read the loss as “the lack of simultaneity” of the men’s efforts: a group cannot pull at the same instant in the same direction. The other cause appears only in passing, in his story of prisoners at a mill, each man “trusting in his neighbor to furnish the desired effort”. It took Ingham and colleagues (1974) to separate the two, by blindfolding people and letting them believe others were pulling behind them. Nobody was. There was nothing to coordinate with, and effort dropped anyway. That residue is motivation, and it is the part that crosses into work where there is no rope. And the drop is not a straight line. In the later studies most of the loss arrives with the first few people and then flattens, and Ringelmann’s own rope trials point the same way: from seven to fourteen, each man lost little more. A team does not escape it by staying small: four people are already past the steepest part.

Psychologists call it social loafing. A review of 78 studies (Karau and Williams, 1993) found it robust across tasks and populations, and found what shrinks it. Set against the method’s rules, most of that list is already there, and one item is given up on purpose:

What reduces social loafing, set against the method
Social loafing shrinks whenIn the method
each person's contribution can be identifiedA trunk has exactly one responsible person, and the name is in its header.
the contribution is unique rather than redundantTrunks are cut at the problem's natural seam, and two people who want the same trunk take it to the meeting instead of quietly sharing it.
the work matters to the person doing itNobody is assigned work, and taking a trunk means writing down why you want it.
the group is one its members valueThe prerequisite: a team that decided on purpose that the method suits it. No rule can install that.
each person's output can be evaluated against a standardGiven up on purpose. The method does not evaluate people, and it accepts that cost.

Conditions from the review by Karau and Williams (1993). Evaluation, among the strongest of them in that review, is the one the method does not use.

The manifesto does not cite this research; the comparison is ours, made afterwards. A fit like this suggests the method is not naive about groups. It is not evidence that the method works.

The group is not only a tax

There is an opposite effect too. In the 1920s Otto Köhler had members of a Berlin rowing club curl a heavy bar until they were exhausted, alone and then in groups of two or three sharing one bar. A group could last only as long as its weakest member, yet groups often outlasted that member’s time alone, most of all when the difference in strength was moderate. Erich Witte brought the finding back in 1989, and it is now known as the Köhler effect: the weaker member of a group can work harder in company than alone (Hertel, Kerr and Messé, 2000).

We think the everyday version is familiar to anyone who has worked next to somebody visibly flying through their own work: it raises what you believe a week can contain. As we read the method, it spends its whole budget on autonomy, and autonomy is not solitude. What it removes is supervision, not the company of other people.

Numbers people invent for each other

Without a stated number of hours, colleagues guess each other’s commitment, and the guesses are rarely kind. A known bias points the same way: the fundamental attribution error, the habit of explaining other people’s behaviour by their character rather than their circumstances (Ross, 1977). A colleague who delivers less looks lazy, when they may simply have four hours a week to give.

Declared hours replace the guess with a fact the person wrote themselves. It is not a target and nothing is measured against it. It only lets everyone read the rest of the picture correctly.

Estimates that turn into promises

People underestimate how long their own tasks will take, even when they know that similar tasks ran late before. This is the planning fallacy (Buehler, Griffin and Ross, 1994). In many teams the estimate then becomes a target, and Goodhart’s law, in Marilyn Strathern’s wording, names what happens next: “When a measure becomes a target, it ceases to be a good measure” (Strathern, 1997). The method removes estimates, story points and velocity altogether. What remains is a person’s own declaration of hours, and a definition of done written by the people who will meet it.

Letting go of work without shame

The more something has cost, the harder it is to abandon, even when it should be abandoned. This is the sunk cost effect (Arkes and Blumer, 1985). Building two versions on purpose decides in advance that one of them will be dropped, before anyone has a reason to defend it. And the rule that a half-finished version is not trash exists for the person whose version lost: their finished parts are still compared with the version that shipped, and still count.

Safe to put it back

Releasing a trunk is announced, never hidden, and it is explicitly not a failure. That matters because teams learn more when their members feel safe taking interpersonal risks, such as admitting a mistake or saying that they are stuck (Edmondson, 1999). A method that punished giving work back would teach people to hold on to it silently, and in the manifesto’s own words, a header that has not changed in six weeks is the one thing the method genuinely cannot read for you.

The limit, stated honestly

Rope-pulling is not software, and an hour in a laboratory is not a quarter in a company. Nothing on this page measures this method. It shows only that the method’s rules sit on the right side of known failures, and it names what would have to be defended if somebody proposed dropping one of them.

Sources

  • Arkes, H. R., and Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35(1), 124–140. doi.org/10.1016/0749-5978(85)90049-4
  • Buehler, R., Griffin, D., and Ross, M. (1994). Exploring the “planning fallacy”: Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. doi.org/10.1037/0022-3514.67.3.366
  • Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. doi.org/10.2307/2666999
  • Gagné, M., and Deci, E. L. (2005). Self-determination theory and work motivation. Journal of Organizational Behavior, 26(4), 331–362. doi.org/10.1002/job.322
  • Hertel, G., Kerr, N. L., and Messé, L. A. (2000). Motivation gains in performance groups: Paradigmatic and theoretical developments on the Köhler effect. Journal of Personality and Social Psychology, 79(4), 580–601. doi.org/10.1037/0022-3514.79.4.580
  • Ingham, A. G., Levinger, G., Graves, J., and Peckham, V. (1974). The Ringelmann effect: Studies of group size and group performance. Journal of Experimental Social Psychology, 10(4), 371–384. doi.org/10.1016/0022-1031(74)90033-X
  • Karau, S. J., and Williams, K. D. (1993). Social loafing: A meta-analytic review and theoretical integration. Journal of Personality and Social Psychology, 65(4), 681–706. doi.org/10.1037/0022-3514.65.4.681
  • Kravitz, D. A., and Martin, B. (1986). Ringelmann rediscovered: The original article. Journal of Personality and Social Psychology, 50(5), 936–941. doi.org/10.1037/0022-3514.50.5.936
  • Ravid, D. M., White, J. C., Tomczak, D. L., Miles, A. F., and Behrend, T. S. (2023). A meta-analysis of the effects of electronic performance monitoring on work outcomes. Personnel Psychology, 76(1), 5–40. doi.org/10.1111/peps.12514
  • Ringelmann, M. (1913). Recherches sur les moteurs animés: Travail de l’homme. Annales de l’Institut National Agronomique, 2nd series, 12, 1–40.
  • Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution process. Advances in Experimental Social Psychology, 10, 173–220. doi.org/10.1016/S0065-2601(08)60357-3
  • Strathern, M. (1997). “Improving ratings”: Audit in the British university system. European Review, 5(3), 305–321.
  • Witte, E. H. (1989). Köhler rediscovered: The anti-Ringelmann effect. European Journal of Social Psychology, 19(2), 147–154. doi.org/10.1002/ejsp.2420190206