The best fertilizer on any farm is the footsteps of the owner.
Given the sheer coincidence, I figured it must be a good book. The book is different from many other non-fiction books I’ve reviewed recently in that it’s an academic strain. The points laid out in the book are compelling but it can be difficult to wade through, especially if you’re casually reading.
Nonetheless, the book holds several great stories that illuminate the dangers of incorrect measurements and good intentions. The crux of the book can be summed up as this: The coarseness in measurement of a large actor (a state in this case) can preclude the desired outcome.
This is made more dangerous as these states become more authoritarian than not, and as things are done in the name of high-modernism (or an unfaltering confidence in science and technology). Central planning is the poster child of this disaster waiting to happen.
This is a book review of Seeing like a State, along with some ruminations on why this book in particular was a common topic at Deconstruct.
Why measure at all?
Standards are a relatively new invention of history. Seeing like a State details the days of many different fiefdoms having slightly different measurements for things like sacks of grain.
Each different town or kingdom had just a slightly different way of measuring. These methods also had quite a bit of slack in them: Were the bags of grain flattened off, heaping, or given enough room for the sack to close?
In due time, standard measurements proliferated, mostly for reasons of easy taxation. I guess our ancestors were tired of fighting, so they settled on spaces instead of tabs.
Contrast that to today: You go to a grocery store and order meat by the pound or gram. The scale that it’s measured on is certified by The Division of Measurement Standards. The United Kingdom takes it one step further and regulates what consitutes a shot of alcohol.
These standards originally came about for reasons of taxation, but a certain fatigue must also play a part. As the world shrinks, people seek common languages for insignificant details. Arguing over what a “sack of grain” eventually became an unproductive argument.
We are what we measure
There’s a meme in startup culture that, “You are what you measure.”
Whether it’s a Key Performance Indicator or a Key Result, these metrics can be valuable motivators or terrible distractions. We’ve got a name for the distractions, too: a Vanity Metric.
While an entire lack of measurement is an organizational smell, so too is a focus on the wrong measurements. Take for example, the phenomenon of clickbait. Enticing and dishonest headlines driving clicks are an over-optimization for a single metric: the pageview.
Seeing like a State covers clickbait before the internet, courtesy of some Germans in the late 1700’s. Scientific foresty in German forests was an optimization of the forest. Eliminate all external variables and optimize for the yield of wood each year. The singular purpose of each forest is also reflected in its ordiliness: only one species of tree would be planted in perfect rows.
These forests achieved a heightened yield their first few harvests. The plan was working! Over time, the lack of diversity in each forest made it extremely succeptible to failure. The monoculture only needed one enemy before it was toast.
Furthermore, this failure was only observed after everyone who had engineered these forests had died.
In the German case, the negative biological and ultimately commercial consequences of the stripped-down forest became painfully obvious only after the second rotation of conifers had been planted. “It took about one century for them the negative consequences to show up clearly.”
What seeds do we sow today that will be maligned tomorrow, or that will tank our organization? We could die seeing all of the metrics of fame and success, only to have that to be shown as a Bad Idea™ by history.
We would not argue that measuring the yield of a forest or a pageview is wrong, but the danger comes when that becomes a singular focus. These are important, objective results that serve as as commodities or leading indicators.
Give a mouse a metric
From the high-level, unchecked metrics can be at odds with the health of the patient, the company.
Can metrics work then as a motivating force? If you give someone a metric to hang their professional progression on, surely they would appreciate the simplicity and focus?
The state officials and agrarian reformers reasoned that, once given a consolidated, private plot, the peasant would suddenly want to get rich and would organize his household into an efficient workforce and take up scientific agriculture.
I’ve heard legend of programmers that were paid by the number of lines of code that they wrote. That is a bit of an urban myth from what I can see. Still, these dynamics can still play out at a young company.
If a CEO sets a target for his nascent sales team to make 100 phone calls per day to leads, it’s easy to see how that may go wrong. Even by measuring something more results-oriented like dollars in the door can prioritize the short-term over the long-term. If those customers acquired on unsustainable contracts, our forest will surely die after the first harvest.
In the 1970’s, this was the case for Ujamaa villages in Tanzania. These villages were supposed to be a high-modern transformation of the country’s farming economy. Citizens were relocated into more “efficient” (read: more legible) villages with the promise of better farming practices, access to better equipment, and more. These villages turned out to be a logistical disaster, but it was driven in large part by the incentives.
The government of Tanzania incentivized local officials to relocate as many villagers as possible. These officials were not measured by the health or success of each village, but by the sheer numbers of bodies shuffled.
With their careers on the line, these officials persisted moving citizens from their existing villages to Ujamaa villages and were awarded as such.
The result was farming villages situated entirely along highways or with other entirely illogical layouts. The villages were an objective failure. The officials made local optimizations at the expense of the country.
A poor metric given to middle management can be just as bad as the wrong metric at the organizational level.
An organic mass
If people find the new arrangement, however efficient in principle, to be hostile to their dignity, their plans, and their tastes, they can make it an inefficient arrangement.
Redesigning the lines and boxes in an organizational chart is simpler than changing how that organization in fact operates.
Mētis is typically translated into English as “cunning” or “cunning intelligence.” While not wrong, this translation fails to do justice to the range of knowledge and skills represented by mētis. Broadly understood, mētis represents a wide array of practical skills and acquired intelligence in responding to a constantly changing natural and human environment.
James Scott, the author of Seeing like a State, offers the counter-point to state-level thinking that often leads to a quagmire. It is to trust the mētis.
Although mētis sounds like an area on the inside of your elbow or a dish from Epic Meal Time, it is actually a Greek word without a direct English translation.
Whereas a high-modernism might be best described as book-smarts, mētis is street-smarts. It is the knowledge that one can only learn from experience. There is big difference between reading programming books and working within a software organization for a decade.
In many ways, mētis is a large part of software development. Mētis and the superstitions we develop are very similar. We all have weird incantations of environment variables, method structures, and more, not because we studied in a book but because that is the hound that has bit us the least.
Although science is in the name of the degree that many acquire before entering the discipline, there is still plenty of voodoo within the day-to-day of the profession.
A path forward
Comparing German agriculture in the 1700’s or 1970’s villagization in Tanzania to young companies might seem like a bit of a stretch. Such a comparison could be entirely due to your author’s inability to interpret the world outside of his own experiences. That’s probably true.
But reading the recommendations of the author about how to avoid large-scale, state-driven catastrophes may as well have been plucked from a book on running effective software development organizations:
One might, on the basis of experience, derive a few rules of thumb that, if observed, could make development planning less prone to disaster.
- Take small steps
- Favor reversibility
- Plan on surprises
- Plan on human inventiveness
Were this written for software developers, the list maintains an uncanny resemblance:
- Release work incrementally
- Allow for easy rollback
- Plan on scope creep
- Trust your team
Some of the best software projects I’ve had the pleasure to work explicitly or implicitly embraced these four rules and then some. First principles thinking led our projects, and we did not succumb to crude metrics.
Metrics and even processes to some extent can overly dummy-proof the operation of a company.1
The point is simply that high-modernist designs for life and production tend to diminish the skills, agility, initiative, and morale of their intended beneficiaries. The bring about a mild form of this institutional neurosis.
As we build and operate these organizations, we need to remember that humans are at their core. Attempting to organize efforts aesthetically will likely result in failure. For small companies, macro-level metrics are not the only tool of measurement. Rather than throwing things over the wall, we can always walk over to a desk or start a video call.
Although companies have all the aspects of a scientific subject, they are also organisms. We can never know how each input will affect its well-being.
The two hardest things in Computer Science are: People, and convincing others that "People" is the hardest thing in Computer Science.— Brad Grzesiak (@listrophy) June 17, 2017
The most tragic episodes of state-initiated social engineering originate in a pernicious cominbation of four elements. All four are necessary for a full-fledged disaster. The first element is the administrative ordering of nature and society—the transformative state simplifications described above. The second element is what I call a high-modernist ideology. It is best conceived as a strong, one might even say muscle-bound, version of the self-confidence about scientific and technical progress, the expansion of production, the growing satisfaction of human needs, the master of nature (including human nature), and, above all, the rational design of social order commensurate with the scientific understanding of natural laws.
The third element is an authoritarian state that is willing and able to use the full weight of its coercive power to bring these high-modernist designs into being.
A fourth element is closely linked to the third: a prostrate civil society that lacks the capacity to resist these plans.
The utopian, immanent, and continually frustrated goal of the modern state is to reduce the chaotic, disorderly, constantly changing social reality beneath it to something more closely resembling the administrative grid of its observations.
The farm, unlike tha plan, was not a hypothecated, generic, abstract farm but an unpredictable, complex, and particular farm, with its own unique combination of soils, social structure, administrative culture, weather, political strictures, machinery, roads, and the work skills and habits of its employees.
It sometimes seems to me that if I could persuade everyone to say “systematize” each time he wanted to say “liberate” and to say “mobilization” every time he wanted to say “reform” or “progress” I would not have to write long books about government-peasant interaction in Russia —George Yaney, The Urge to Mobilize
A modern population must live in communities with a certain physical layout—not just villages, but proper villages.
Most section and village authorities were content to go through the motions when it came to communal cultivation. And they were reluctant to impose fines on their neighbors who neglected the labor rules in order to tend to their all-important private plots.
The believe in large farms, monocropping, “proper” villages, tractor-plowed fields, and collective or communal farming was an aesthetic conviction undergirded by a conviction that this was the way in which the world was headed—a teleology.
Returning once again to the case of polyculture, we can see why agronomists might have scientific as well as aesthetic and institutional grounds for opposing polycropping. Complex forms of intercropping introduce too many variables into simultaneous play to offer much chance of unambiguous experimental proof of causal relations.
It a work-to rule action (the French call it grève du zèle), employees begin doing their jobs by meticulously observing every one of the rules and regulations and performing only the duties state in their job descriptions.
More than thirty-five years ago, in recognition of the refractory complexity of ambitious social policy, Charles Lindblom coined the memorable expression “the science of muddling through.”
Late colonial modernizers sometimes wielded their power ruthlessly in transforming a population that they took to be backward and greatly in need of instruction.
[Story of the Trabant wheeler-dealer who helped a factory survive by driving around Germany and bartering for parts.]