resign, and do not fail to do so

Beer tasting notes, part 1

This was brought to my mind by reading Yet Another person going on about being “disruptive”, but I quickly got a long way from there, as I started re-reading Stafford Beer’s “The Heart of Enterprise”, and was reminded of this lovely little box text:

(bonus picture of the tip of my thumb)

There is quite a lot to unpack here – in fact there is a lot to unpack in this whole chapter, so I thought I’d start a mini-series of notes on Beer. “Heart” is a very nice, very weird book – although Brain of the Firm is the definitive statement of his model, Heart is the one where he tries to make it practically relevant to actual managers. So much so that every chapter is followed by a section called “Later, in the bar”, where a group of fictional characters express their exasperation at all the abstract reasoning and ask what the hell the author is on about.

One thing I forgot to include in my guide to reading difficult books is that you often get the most interesting insights in the little asides, late on in the development of the idea, when the author feels the need to deal with conclusions and implications which look plainly wrong or crazy. (For example, I was reminded during the pandemic that Robert Nozick and almost every other major libertarian writer will always give you a footnote saying that obviously, everything they’ve said about state coercion is kind of all-bets-off when it comes to quarantine for infectious disease). The box above comes in chapter 14 of Heart, where Beer is finishing up the model and moving on to “Notes on Implementation”. It’s just about where he has to begin to deal in earnest with the fact that although the science of cybernetics is all about abstract systems and relations, actual enterprises have to be managed by actual managers. And so the box above (and the rest of the chapter) make a lot of references to “style”, which don’t appear anywhere else in the book, or at all in Brain.

Thankfully, “style” is one of the terms in Beer’s writings which is used in its ordinary language sense; he doesn’t bother with a formal definition and just says that:

“Draconian measures are sometimes necessary [if there is not much time to act before the crisis becomes a disaster] … such measures are, by definition, not benign. It follows, that a benign style, chosen only because we approve of benignity, renders the draconian measures ineffective. It results in confusion, and usually in what is called ‘lash-back”. To choose a draconian measure [restricts the available choices of style]”

So the big trade-off in dealing with an incipient crisis is between time and style. The “cost of time” is the risk that the incipient crisis will become an actual crisis. The “cost of style” is, basically, organisational damage.

The idea of this chapter is that many of the problems of actual organisations come about because the management system has to be two things at once. It has to be a viable cybernetic system (meaning: one that is capable of processing information from the environment and turning it into decisions sustainably). But it also has to be a functioning human system, in the sense of being a set of relationships that the people in the jobs can live with, so that they will actually communicate, notice things and act together.

The human system is based on a lot of shared understandings and promises; when you change the cybernetic system many of these have to be adjusted. (As Beer puts it, “as one watches the inexorable consequences of the new organisational design, as to which many guarantees have of course been given, it quickly becomes obvious that some luckless individual – despite his position of privilege and despite the guarantees – was born with a silver knife in his back”).

That’s what’s meant by “the stress of adopting an unnatural style”, although it’s also interesting to see that (again, in this chapter but almost nowhere else), Beer puts a lot of emphasis on the psychological strain on the individual decision maker. This is the point at which he seems to recognise that, particularly in a crisis situation, the most scarce resource is likely to be the manager’s own bandwidth and decision-making ability, and that doing something which is fundamentally not suited to your temperament is going to put extra demands on this.

So the Stafford Beer-approved crisis strategy is – find a list of actions which could potentially stabilise the situation in a useful time frame, then pick the one which causes the least organisational damage. Bearing in mind at both stages that the option you pick will determine the way you have to behave, and that this will have consequences for you, and for the organisation. If this process doesn’t identify any viable course of action, then move to a “war footing”, and consider tradeoffs which had previously been rejected, in the knowledge that you are going to have to spend time repairing fences afterwards; if there’s still no answer, you have to resign.

“I think that this model works, because I have used it myself many times. Frequently, the initial procedure leads to failure. In that case, it is necessary to “cancel all engagements” and to address the second procedure in a life-or-death spirit. That usually does work. It has failed me four times in my life. On each occasion I omitted to resign until later, because I simply could not believe that there was no way out. There was no way out, with hindsight. Thus I now believe in the reality of the model to its final conclusion, and will not ever be caught again”