For many or most of our projects a small on site team of a handful of people was enough. At this scale, everyone talks to everyone else directly and organization questions are trivially solved. Now starting to think about some larger adventures, and correspondingly we need to consider some issues about the complexity of larger teams. How they are integrated effectively into a free roaming yet coherent pack. A pack is just another name for a group of entities that share a task.

Two things about scaling up teams were clear for a long time already. One is the necessity of local autonomy for any single actor providing motivation, responsibility and accountability. On a theoretical basis, it is also clear that any significant scaling of any system of relevance requires a hierarchical integration of a base cellular design. Single players, small teams, teams of teams to form larger teams and so on all the way to the lead.

A major discovery in 2020 came from reading “small is beautiful” [1] by way of chapter 16 titled “Towards a Theory of Large-Scale Organisation”. In that chapter first the local autonomy principle is decomposed into four constituent principles. This is complemented by the fifth part of the theory, the “middle axiom”. The middle axiom was new to me in explicit form, makes total sense, and provides an actionable policy for integrating large organizations without violating any of the other constraints. From that middle axiom comes the phrase

Good government is always government by exception.

So, in summary the question is how to achieve hierarchical self-organization for missions that require a large scope. How to develop a large organization and stay agile like a small pack, both at the same time.

1/ Achieve smallness within large organisation

2/ Balance order and chaos. Balance structuring orderliness and creative chaos. Both is needed at the same time, always, to make any headway.

3/ Organize by these five principles: 3/1/ subsidiarity, 3/2/ vindication, 3/3/ identification, 3/4/ motivation, 3/5/ the middle axiom.

They are explanded in more detail right below with their full name in the original chapter, my idiosyncratic names, and selected clippings from the original text. Lets dive in …

1 The Principle of Subsidiary Function

(local-autonomy, local-knowledge)

The higher level must not absorb the functions of the lower one, on the assumption that, being higher, it will automatically be wiser and fulfil them more efficiently. Loyalty can grow only from the smaller units to the larger (and higher) ones, not the other way round - and loyalty is an essential element in the health of any organisation.

2 The Principle of Vindication

(no-single-best-average, respect-the-outliers, support-the-living-artist, social-contract-explorers-we)

To vindicate means: to defend against reproach or accusation: to prove to be true and valid; to justify; to uphold; so this principle describes very well one of the most important duties of the central authority towards the lower formations. Good government is always government by exception. Except for exceptional cases, the subsidiary unit must be defended against reproach and upheld. This means that the exception must be sufficiently clearly defined, so that the quasi-firm is able to know without doubt whether or not it is performing satisfactorily.

3 The Principle of Identification

(self-accountability, closing-the-information-loop-on-self-actions, no confounding, local accountability)

Each subsidiary unit or quasi-firm must have both a profit and loss account and a balance sheet. From the point of view of orderliness a profit and loss statement is quite sufficient, since from this one can know whether or not the unit is contributing financially to the organisation. But for the entrepreneur, a balance sheet is essential, even if it is used only for internal purposes.

4 The Principle of Motivation

(motivation is everything, self-motivation, self-story, lawful/meaningful, ikigai)

It is a trite and obvious truism that people act in accordance with their motives. All the same, for a large organisation, with its bureaucracies, its remote and impersonal controls, its many abstract rules and regulations, and above all the relative incomprehensibility that stems from its very size, motivation is the central problem. At the top, the management has no problem of motivation, but going down the scale, the problem becomes increasingly acute, This is not the place to go into the details of this vast and difficult subject.

5 The Principle of the Middle Axiom

(decentralization, global-integration, exception-friendliness, impact-statistic, action-consequences)

Top management in a large organisation inevitably occupies a very difficult position. It carries responsibility for everything that happens, or fails to happen, throughout the organisation, although it is far removed from the actual scene of events. It can deal with many well-established functions by means of directives, rules and regulations. But what about new developments, new creative ideas? What about progress, the entrepreneurial activity par excellence?

We come back to our starting point: all real human problems arise from the antinomy of order and freedom, Antinomy means a contradiction between two laws; a conflict of authority: opposition between laws or principles that appear to be founded equally in reason.

The centre can easily look after order; it is not so easy to look after freedom and creativity.

What is required is something in between a middle axiom, an order from above which is yet not quite an

Example of UK coal board: issue updated and small set of rules / self-questions to go through before starting a new project. Why do you do it the way you do? Did you check for A and B? OK.

Another middle axiom can be found in the device of Impact Statistics. Normally, statistics are collected for the benefit of the collector, who needs -or thinks he needs - certain quantitative information. Impact statistics have a different purpose, namely to make the supplier of the statistic, a responsible person at the lower formation, aware of certain facts which he might otherwise overlook. This device has been successfully used in the coal industry, particularly in the field of safety.

Notes and references

  • [1] E. F. Schumacher, Die Rückkehr zum menschlichen Maß: Alternativen für Wirtschaft und Technik, “Small is Beautiful”, Rowohlt, 1977. Full text
  • [2] E. F. Schumacher, Small Is Beautiful: A Study of Economics As If People Mattered, 1973. Full text
  • [3] Geoffrey West, Scale: the laws of life and death of organisms, cities and organizations, 2017.
  • [4] Christopher Alexander, A pattern language, 1968. (work towards regions smaller than 10M)
  • [5] unchecked McChrystal: Team of teams

Technical notes

timestamps: 2021-02-07, 2021-01-27, 2020-11-26, 2020-11-24

subtags: philosophy books jetpack feed schumacher small beautiful scaling research hacker culture cyberspace internet books


tag cloud

robotics AI music books research psychology intelligence feed ethology computational startups sound jcl audio brain organization motivation models micro management jetpack funding dsp testing test synthesis sonfication smp scope risk principles musician motion mapping language gt fail exploration evolution epistemology digital decision datadriven computing complexity algorithms aesthetics wayfinding visualization tools theory temporal sustainability stuff sonic-art sonic-ambience society signal-processing self score robots robot-learning robot python pxp priors predictive policies philosophy perception organization-of-behavior open-world open-culture neuroscience networking network navigation movies minecraft midi measures math locomotion linux learning kpi internet init health hacker growth grounding graphical generative gaming games explanation event-representation embedding economy discrete development definitions cyberspace culture creativity computer-music computer compmus cognition business birds biology bio-inspiration android agents action