Jared Diamond spoke at the Commonwealth Club last month. I have always been a huge fan of his Pulitzer prize winning book “Guns, Germs, and Steel” as well as “Collapse” and other works. This talk introduced his new book “The World Until Yesterday: What Can We Learn from Traditional Societies?” The answer: more than you think.
The talk is very interesting but only one story he told was really relevant for this blog. He said this:
“Most of us have probably read in high school about the medieval peasants in Europe having a strip system where a peasant did not have one block of land, and peasants would have 7 strips scattered in 7 different places. We might say how stupid that is because the peasant has to walk between 7 different strips. Why not just amalgamate those strips?“
“There was an anthropologist doing a study in Peru, where these potato farmers in Peru scattered their strips; typically 15 different strips all of them tiny. An agricultural expert came in and said ‘How mind-blowingly stupid these peasants are. They should just trade and amalgamate their land.’ “
“But it turns out that if you have 15 strips in different directions, the risk of an insect blight, or people stealing your stuff, or a frost, varies in different directions. So if you have 15 strips, what the peasant cares about is that no matter what happens their potato yield will never drop to the point that they will starve. If you put all your potatoes in one big block, it might save walking back and forth, an in 50 years your yield might be higher, but in 17 of those 50 years you would have starved to death. If your strips are scattered in different directions, then you can virtually guarantee that every year at least some of those strips will produce.“
The question we should ask ourselves is why would the agricultural expert find this to be mind-blowingly stupid? (And why would I think that myself quite frankly before the risk was explained?)
The reason of course brings us back to the “Enlightenment Bias” that all phenomenon can be explained at the core by fairly simple rules. Thus a field is a field is a field. Farming is an action that you do in a field, and nothing else matters. It is this desire to reduce activity to the simplest form, and then perfect that, that blinds us to the risk.
Ken Burns documentary “The Dust Bowl” portrays an infamous example of just this sort of thing. The dust bowl was caused by a huge mono-culture: everyone was growing wheat. When things started going bad, it started going bad everywhere. Farming in that part of America was negated for almost a decade. We now know that large scale mono-cultures are risky. It is far better to grow a mix of things, because if one crop is wiped out, you can fall back on the rest that is not effected.
Yet, we have not learned this lesson yet in the automation of business processes. The ideal is always “Find the one, best practice, and get everyone to do it.” It sounds good in abstraction, just like it the idea to amalgamate the fields sounds good. It sounds good because of the Enlightenment Bias.
In my last post, How Technology causes Fragility, I talked about how mono-cultures caused by technology can make an organization weaker. What stuck me about Jared Diamond’s statement is that not only is this true, but ancient cultures had figured this out the hard way.
It also strikes me that our goal should not be that of finding the best process and creating uniformity. Efficiency is not uniformity. Instead, don’t worry about enforcing a best practice, but instead attempt only to identify and eliminate “worst practices”. A variety of approaches should be encouraged, because if one of those approaches turns sour, you can fall back on the others.
Thinking Matters 
Hi Keith, a great story from one of my favorite authors that resonates with the subject of naive intervention:
In the three posts series I have pointed out that more execution options and safety redundancies are seen by bean counters as inefficiencies. Process standardization creates a monoculture AND a lack of knowledge how to improve.
Anyone surprised that I have been arguing against BPM being the silver bullet?
Thanks, Max
Of course, those of us who grew up around farming wouldn’t be surprised why they’d want strips instead. Farming communities have ever been communities – that pull together or work out collectively beneficial arrangements like that.
Pingback: Requisite Inefficiency | strategic structures
Keith,
Fully agree with “don’t worry about enforcing a best practice”. But I’m not sure that “instead attempt only to identify and eliminate “worst practices” is always a good advice. Often some redundancies are seen as “worst practice” while they are vital for survival. I tried to elaborate on this and related consideration here: http://strategicstructures.com/?p=597.
Thank you for the inspiration!
Ivo
Excellent post in your blog! You touch on Dave Snowden’s work particularly apt, and his “safe fail” as opposed to “fail safe” is an important philosophical approach — which I hope to cover in my talk at BPM Next.
While variety is important, one does not sit by and watch bad practices continue. There is still a “process improvement” going on. Ideally you base this on measured statistics of real world differences, and not on a conclusion about how a model would predict. Like the immune system, you attempt to recognize bad practices and get rid of them.
I used the term “eliminate” and that has a couple meanings: one is to completely and finally eradicate something, and you argue in your post that might not be optimal. It is a very good point. However what I really meant was “stopping the common practice of a bad practice”. In other words, it is still OK to try and improve your regular practice by identifying and avoiding bad practice. Just don’t attempt to isolate a single best practice, and as you point out in your post, don’t 100% eliminate what might today be considered a bad practice. Both points lead to the same conclusion: some variation is important.
My next post will be about how we use models incorrectly to attempt to identify good and bad practice.
Hi Keith, I want to elaborate on how to identify best practices.
I like the comparison to the immune system but the human immune system has no designed best practice or measures statistically. It is trained by exposure. “Adaptive (or acquired) immunity creates immunological memory after an initial response to a specific pathogen, leading to an enhanced response to subsequent encounters with that same pathogen. This process of acquired immunity is the basis of vaccination.”
What we propose with ACM is an immunization against bad practices. There are no statistical measurements that are necessary. They can be used, but in principle a performer or customer has a bad experience (getting the flu) and then adapts to avoid it next time. Avoiding the experience by overly sanitizing reduces the ability of the immune system to respond. Process models kill the ability to adapt, but they have to be redesigned. It is like giving someone Penicillin everytime they have temperature just to be safe.
It is a lot about gradual change and adaptation. Big change is hard on everyone, but gradual change we deal with much more relaxed. Variation is nothing else than gradual change. The exposure immunizes the process and increases the ability to deal with more and different changes.