Agile Rules and Procedures: Just Do It

Written by Jamie Friday, 22 July 2011 11:11

In my last post I argued that priests and parasites were a side-effect of trade and specialisation.  In this post I will explain why.  

In 2005 I moved to one of the big consultancies, which was organised around a partnership model.  The partners sat on top of a pyramid with senior managers, managers, analysts, etc., all sat beneath them.  It wasn’t just that the people at the top enjoyed the power, the people at the bottom liked relinquishing it, too.  The submission of power to the hierarchy was therefore sadomasochistic - it was both mutual and enjoyable.  The army, university fraternities and certain software organisations, with their hierarchies of certified this and that, employ the same trick.  The result is conformity.

When a person’s power is derived from their position, we call it position power. A more benign form of power is expertise, which is neither sadistic or masochistic.  For example, we gladly submit to the knowledge of doctors and dentists.  Charles Handy has this to say:

Because this power source is the most socially acceptable it is also the most sought after and will bring out many spurious claimants.  Their bluff, if bluff it is, may well work, but if discovered the dethroned expert will find that she has created a credibility gap that may well contaminate future attempts to claim expertise for herself.  (Handy, p. 131)

This explains why priests and parasites arise from trade and specialisation.  It explains why there are so many conmen - i.e. agile experts - in our field but also in any other field (imagine a traveling ‘doctor’ selling snake oil).  It also explains why conmen will get found out - they create a ‘credibility gap’ that they eventually fall into.  

The Tail Wags the Dog
A common way to affect another person’s behaviour is to use rules and procedures.  We can bribe our children with rules for taking the car out: make sure you fill it up.  We can influence a whole team of developers with rules for committing code: if the build is broken, we have to stop everything and fix it.  The law influences the behaviour of millions of people with rules that forbid murder, rape, arson and theft.  Rule and procedures do not work out of the box, however, and the leader must have:

The perceived right to institute these rules and procedures;
The means and will to enforce them, i.e. the appropriate power base.  (Handy, p. 134)

In organisations that rely on position power, rules and procedures work well.  The leader in the hierarchy has, by definition, the right to institute.  He also has the power base, which is everyone below him.  My first agile transformation for the consultancy was easy: I came into work and said we are doing Extreme Programming.  Nobody argued.

Rules and procedures also work well if they are self-enforced by a team.  For example, the agile practices, as developed by Beck and Fowler, which would later become Extreme Programming, were a side-effect of the team.  The team were not a side effect of the rules.  In the same way, the law is a reflection of the human being’s need for safety.

Submitting to rules and procedures always means giving up personal freedom - the freedom to murder, the freedom to commit without tests - in exchange for group cohesion.  Sometimes people refuse to give up their personal freedom because they don’t agree that the cohesion is worth it, or they don’t agree that the rule is right.  The Poll Tax laws caused riots and contributed to Thatcher losing her position as Prime Minister.  The Importation Act of 1815, also known as the Corn Laws, were eventually rescinded in 1846.  The Volstead Act of 1919, which prohibited the sale of alcohol in the United States of America, failed utterly and it was left to gangsters like Al Capone to lead the rebellion.

Effects and their Causes
Why do some rules and procedures, which in hindsight look so backward, ever get tried?  Consider Ridley:

A calm home contains happy children; children who are hugged a lot are nice; children who are beaten a lot are hostile; an so on.  But this could be confusing cause and effect.  You could just as plausibly argue that happy children make a calm home; children who are nice get hugged a lot; children who are hostile get beaten a lot.  Old joke: Johnny comes from a broken home; I am not surprised - Johnny could break any home. (Ridley, p. 252)

This confusion of cause-and-effect combined with the desire to control is how we fall into the trap of rules and procedures.  Consider this reasoning:

all none-drinkers are moral
my husband doesn’t drink
therefore, my husband is moral.

From here, it can be argued that we can make all men moral by stopping them drinking.  How can we stop them drinking?  By introducing a set of rules and procedures - the Volstead Act.  We can work down the chain of reasoning, starting with the generalisation: all none-drinkers are moral.  Find a specific example: someone’s husband.  And then deduce a conclusion: the husband is moral.  We then work back up the chain of reasoning with more generalisations: we can make all men moral by stopping them drinking.  The problem, however, is that deductive reasoning relies on one indisputable fact.  But we know that it’s not true that all none-drinkers are moral, that’s why the rules for prohibition made no sense.

The same mistake is made when analysing software teams.

agile teams have high levels of trust (and write good code, etc.)
my team is trustworthy and writes good code
therefore, my team is agile.

From this deductive reasoning, we can argue that we can make all teams agile by making them trustworthy and having stand ups.  How can we do that?  By introducing a set of rules and procedures that will force them to be agile.  The power then is force, not expertise, and force is always met with aggression, passive or otherwise.

The Eel Uprising


Here is the Lindengracht, in 1886, which is not far from my house here in Amsterdam.  The rope crossing the canal has a live eel tied to it.  From a small, moving boat, participants had to stand up and grab the eel.  The winner was the person who could hang on until the creature’s head was pulled off.  To make it more difficult, the eel was covered in soap.  Of course, the real point of the game was not really to catch the eel, rather it was to laugh at all the participants falling into the canal.  Mak picks up the story:

Twenty Jordaan inhabitants each contributed half a guilder, a nice fat eel was bought, and there were still six guilders left over for the prize.  The police got wind of the affair, however, and untied the rope which had already been strung across the canal.  There was a small commotion, during which one policeman was beaten with an umbrella and another was thrown down into a cellar.  The police were not going to put up with this.  (Mak, p. 217)

Years earlier, eel grabbing had been outlawed, as it was cruel to the eel.  That rule, however, was not recognised by the citizens of Amsterdam and neither was the authority of the police.  The initial commotion caught the attention of a group of socialists, who were meeting nearby in the Volkspark.  In the spirit of universal suffrage, they came and fought on behalf of the eel grabbers.  Things calmed down but the next day the agitators were back and a full scale riot kicked-off.  The police were haled with missiles from behind a make shift barrier.  In retaliation, they shot into the crowd, killing a number of people and injuring dozens.  The rioters then turned all the lights off, creating confusion, before beating back forty policemen.  Eventually, the infantry were brought in and the crowds dispersed.

Within 48 hours of the eel been taken down by the policemen, 25 people were dead.  As for the eel, its remains resurfaced in 1913 at an auction, were it was advertised as the ‘one true eel of the Eel Uprising’.  It went for one guilder and seventy-five cents.

*

If the mild mannered ancestors of my adopted hometown were willing to riot over an eel, what chance does the agile consultant have?  If, like the police that day on the Lindengracht, the consultant is spoiling the fun, and if he does not have a legitimate power base or means to police, then he doesn’t have a chance.  Many agile initiatives fail not because of any problems with agile per se, but with the way the priests, the non-expert experts, resort to rules and procedures.  This is because:

• The agile rules and procedures are an effect, not a cause, of good software teams.  The Chrysler team made XP, XP did not make them.
• Rules and procedures are often a mask for force, a form of power that is hostile.  People won’t tolerate this, they will fight back.
• Force is no substitute for expertise just as a conman is no substitute for the real thing.

Conclusion
The enforcement of rules and procedures, the bastion of all priests, often has more to do with the urge to scratch an emotional itch than it does real change.  The priest is always fearful, and conquers his terrors by conquering other people.  If I can paraphrase Onfray,

[rules and procedures give the priests] the means to bolster their faith, for they find in it the material essential for reinforcing their own need for mental help.  Just as psychoanalysts often treat others in order to avoid questioning themselves too closely about their own weaknesses, so the vicars of monotheist gods foist their vision of the world on the faithful - and day by day their own convictions become more secure.

The Protestants who pushed for the Volstead Act were driven by the childish urge to control, to perfect, and in doing so their own convictions became more secure.  The same logic can be found in the insecure, agile expert, whose own logic is derived from a need to control but not always a need to improve.  This is what leads to rules and procedures, but they are often so transparent that failure is inevitable.

Further Reading
The final part of this series is here.

I was inspired by, and borrowed a lot of my classifications from Handy’s Understanding Organisations.  I read about the eel uprising in Mak’s Amsterdam a Brief Life of the City.  Ridley covers cause-and-effect in Nature via Nurture.