P O S T E D B Y A L B E R T
This post was conceived as a companion piece to an earlier post on theories of change, titled How to Cause Things in Philanthropy.
1. Engineering without the math. A colleague of mine once pointed out, to my great satisfaction, that as grantmakers we are in fact social engineers. I still feel the glow of this flattering assessment. A PRI here, a dab of cash there, and soon the social machinery is purring contentedly, ending homelessness in Cincinnati or reversing global warming.
That’s the ideal picture, of course. The machines whose wheels we grease invariably have human constituents. These humans snag on one another or break down altogether. Sometimes the entire machine just gets up and walks away.
2. Building a better social intervention. At times like these we might cast a wistful eye at the board game Mousetrap which has delighted children and adults since it was first introduced in 1963.
During the course of this game, players take turns building an elaborate Rube Goldberg machine whose sole purpose is to trap little plastic mice. What a joy to set the mousetrap in motion and watch it work! You turn a crank that rotates a gear that rotates another gear that causes a plastic boot to swing up and kick over a bucket containing a small metal ball that then rolls down a set of stairs and into a chute where it’s channeled to a contraption that dislodges another ball from its high perch whence it rolls into a bathtub and falls through the drain hole onto one end of a teeter board causing a little plastic man at the other end of the board to fly backwards and into a barrel that triggers the descent of the mouse cage.
Those of us who struggle to get our grantmaking interventions to function properly can’t help but marvel at this beautifully choreographed march of causes and effects.
The game also has, I believe, some important lessons for grantmakers and others who work in the Third Sector.
The spectacular predictability of this elaborate mousetrap depends on the fact that interactions between inanimate objects—gears, balls, chutes, and the like—are governed by well-known laws. These laws express in mathematical language the regularities that physicists and others observe in the behavior of matter.
Laws governing human interactions, by contrast, are fiendishly difficult to come by. Unlike the creators of the Mousetrap game, we don’t often have the luxury of encasing our human subjects in plastic and flinging them off the ends of teeter boards or causing them to fall against one another like dominoes.
How then, as good social engineers, can we get our humans to behave in predictable ways? Fortunately, there are regularities of human behavior—stunning regularities—that grantmakers can learn to exploit.
3. The inner-life of cocktail-tropic fauna. In a previous post I considered how causal chains can become ridiculously complicated as they lead from our grantmaking intervention to our hoped-for outcomes. This thwarts our attempts to develop meaningful theories of change for our work.
Causal chains can also become rather brittle: all the stars may be in alignment for moving from some cause to a given effect, from one link of the chain to the next; all the felicity conditions might be satisfied; and yet, if the movement from cause to effect depends on the vicissitudes of human actors, we might be wise to put long odds on the outcome.
One of the challenging aspects of human actions is that they’re mediated by representations of the world. What does this mean? Shine a light near an amoeba and, without thinking, this phototropic creature will ooze its way towards the glow. Light up the sign for a bar and a cocktail-tropic human might move towards the entrance, but there’s no inevitability about it. If he does, he’s likely to pause at the threshold considering whether or not he likes the bartender, weighing his chances of meeting the man or woman of his dreams, estimating the likelihood that the lounge’s one television screen will be tuned to his favorite channel. He’ll consider his options for the evening. He’ll remember an article about beer not being as fattening as wine. In short, his beliefs and desires will interact in complicated ways to shape his resolve and ultimately bring him to a bar stool—or not.
In the case of the amoeba, the shining light activates various biochemical processes that produce the animal’s phototropic behavior. Biochemistry is also very much at work in the cocktail-tropic human, but it doesn’t determine human action without first creating or recreating various representations of the world inside the head of our subject. These representations—beliefs, desires, aversions, fears, memories—will either put a spring in the lounge-lizard’s steps or stop him in his tracks.
Most social interventions depend on linking and coordinating the activities of many human actors, each with his or her repertoire of beliefs and desires, each introducing some uncertainty into the unfolding path of events. Rather than act according to plans—our plans—these actors will be observing, remembering, desiring, resenting, and yes, perhaps even assessing whether or not they’ll cut off their own noses just to spite us.
4. You often get what you pay for. There are many cases in which desired outcomes do in fact arrive at the distal ends of long chains of coordinated human activity. Consider, for example, the writing and delivery of a letter, or the flying of 200 passengers from New York to Paris. These extraordinary accomplishments require the precise sequencing of thousands—perhaps tens of thousands—of human actions. But there are important differences that set these examples apart from those we typically encounter in nonprofit work where sought-after goals often elude us.
Put aside for a moment the fact that mail delivery and air travel can fail in spectacular ways. Flying 200 passengers from New York to Paris is no doubt a complicated affair, but each step of the process, from the design and construction of the airplane to the training of the aircrew, operates according to principles that are well understood. By contrast, creating the political will to address climate change might be one of those “wicked problems” that will forever resist a comprehensive solution.
Then there’s the issue of buyer incentive, and here we need to face an uncomfortable truth: there are important social goods that nobody really wants to pay for. If you’re selling a product like an iPhone or a flight to Paris, you’ll find many willing investors and customers. If you’re raising venture capital to end racism in Mississippi you’ll likely need to go begging. When you do, you’ll find that what grantmakers and other donors drop into your hat will never come close to covering the cost of what it will take to accomplish your mission.
There’s a reason so many charities are incorporated as not-for-profits. The nonprofit status of organizations that provide services to the indigent reflects not a failure of entrepreneurial imagination or of will, but rather a clear-headed assessment of what people value enough to pay for freely. Lack of buyer interest means no working capital; it also means a social entrepreneur has little with which to incentivize her staff, cover program costs, or grow the capacity of her organization.
5. Small carrots, big sticks. The issue of an appropriate reward structure for humans whose actions we’re attempting to motivate is a vexing one—especially outside the sphere of well capitalized for-profit enterprises. How do you get complicated, extraordinary things to happen without paying people huge sums of money?
Certainly no catalog of the various ways to motivate human action would be complete without a consideration of intimidation and terror. We see that in North Korea, for example, they host an annual spectacle in which tens of thousands of citizens hold up colored placards at just the right time to create the likeness of Kim Jong-un, or of a missile pointed at San Francisco, or of some other great socialist achievement. These mass displays provide a striking model of coordinated human activity.
How does it work?
If you are one of the performers and you do well, you get an extra carrot in your annual ration of vegetables; if you don’t, the authorities stick you and your family in the North Korean gulag. This, I believe, is what’s known as the “carrot and stick” approach to human motivation.
These tough love tactics are clearly too radical for most nonprofit work. The ostentatious tableaux, however, might come in handy for foundation CEOs who, in lieu of publishing inspiring monographs of their pensées, can opt instead to communicate their key philanthropic insights with colorful, football-field-sized displays.
6. Greasy cheeseburger with a side of salad. And then there’s reason. We can’t bang our human subjects together like billiard balls; and it would be challenging to pay our way to good social outcomes in many key areas of nonprofit work. Fear and intimidation are out. Why not try an appeal to reason? In this domain at least we often observe arresting regularities. If forced to choose, more than nine times out of ten a person will strike his head with a kiwi fruit rather than with a hammer. This unprecedented degree of predictability provides strong evidence that we are in the presence of a rational actor.**
Without even thinking about it, we presume that the agents up and down our causal chains will be reasonable. This could be the Holy Grail we’re looking for. It could, that is, if human rationality were robust enough to overcome all the other stuff going on inside our heads, stuff that frequently messes up the calculus of homo rationalis.
Consider these glosses on the groundbreaking research of Daniel Kahneman and Amos Tversky, renowned for their studies of deviations from rationality:*
The set-up | What a rational actor would do | What actually happened |
“[S]ubjects were given brief personality descriptions of several individuals and asked to assess, for each description, the likelihood that it referred to an engineer or a lawyer. In one experimental condition, subjects were told that the descriptions were sampled randomly from a group of 70 engineers and 30 lawyers, and in a second experimental condition the proportion of the two professions was reversed.” | “A rational assessment of probabilities, … would involve some combination of prior probabilities (based on the relative percentage of engineers and lawyers) and current information (based on the description), leading to different probability estimates in the two experimental conditions because of differences in prior probabilities.” | “Subjects’ estimated probabilities were nearly identical, however, demonstrating the tendency to … neglect prior probabilities or base rates. Similar patterns have been identified in countless other experiments.” |
“[R]esearchers gave subjects the following description: ‘Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in antinuclear demonstrations.’ Subjects were asked to assess the likelihood that various statements about Linda were true, among them being that ‘Linda is a bank teller’ and ‘Linda is a bank teller and is active in the feminist movement.’” | “Given two events A and B, the probability of the conjunction … of A and B cannot exceed the individual probability of either A or B.” | “Over 85% of subjects believed it was more likely that Linda was both a bank teller and a feminist rather than just a bank teller, contrary to the laws of probability.” |
[People facing decisions over medical treatments are told in some cases that a particular treatment has a 90% survival rate; in similar cases they’re told that the same treatment has a 10% mortality rate.] | [The two options are logically equivalent so they should elicit the same response.] | “People … respond differently to a 90% survival rate than they do to a 10% mortality rate, although the two are logically equivalent. These framing effects are difficult to reconcile with rational choice ....” |
I remember another example from my graduate school days in which subjects were presented pairs of photographs, each depicting a possible meal, and asked to assess which of the two had the lower number of calories. In one set-up, picture A showed a greasy cheeseburger; picture B showed the same cheeseburger together with a side of salad. I leave it to the reader to guess the outcome of this particular experiment.
These are refined studies, but the evidence for our persistent irrationality is often in plain sight. Ten years after 9/11, 41 percent of Americans still believed that “Saddam Hussein’s regime was directly involved in financing, planning or carrying out the terrorist attacks,” and according to a recent Pew study, 33 percent of all Americans reject evolution saying that “humans and other living things have existed in their present form since the beginning of time.” Some of us still avoid stepping on cracks or walking under ladders.
Other cases elicit more pity than derision. Nonprofit staffers who work with people in crisis, for example, know that these individuals may not always make rational choices; they will not consistently desire more rather than less good for themselves. Unfortunately, there are powerful forces that mess with our ability to make wise decisions for ourselves and our families. Even when we’re not in crisis, reason will often yield to more pressing demands on our psychic machinery.
Finally, we need to consider that we can have perfectly rational actors operating in a flawed scheme to address some social challenge. The scheme might be flawed because our analysis is unsound, or because our strategies and tactics are defective, or simply because perfectly rational actors might have reasonable grounds for refusing to cooperate with our boneheaded plans.
7. The paranoid madman. What do we do in the face of these challenges? Dear reader, if you’re a grantmaker, have you ever had a brutally honest conversation with a disaffected grantee, with one of those nonprofit leaders you’ve appropriated into your strategy for making the world a better place? If you have, then you probably have some inkling of why it’s so difficult to do things with humans. People don’t like to be played. They don’t like to be managed. They prefer to be the authors of their lives and the work they undertake.
And yet so many exemplars of “strategic philanthropy” are top-down affairs. We grantmakers, myself included, act as arrogant elites, drawing arrows and triangles on the whiteboards of our well-appointed conference rooms with no one around to challenge our flawed thinking. We strut about like giant roosters puffing out our breast feathers and clucking incoherently about “disruption” and “theories of change.” We look foolish to everyone except ourselves and those even more foolish than we are.
What’s missing here is a certain kind of sensibility rooted in the imperative that we treat people as ends not as means. This sensibility has as one of its fruits the goodwill people feel towards those who approach them in humility, asking for help rather than offering to save them from themselves. With this sensibility it becomes immediately possible for us to do things with humans rather to humans. Our causal chains become less brittle. A sense of commonality sustains our social change efforts through their rough patches.
Beyond this sensibility, what’s needed most, in my view, is a profound distrust of those who would straightjacket the work of civil society, imposing on it the apparatus of technocracy, framing human behavior as a problem that requires management by social technicians. “What if history was a gambler, instead of a force in a laboratory experiment?” asks Ralph Ellison in his Invisible Man. “What if history was not a reasonable citizen, but a madman full of paranoid guile?” To admit these as possibilities is to forge our causal chains one tentative link at a time, in solidarity with others, quickly reaching the point where we’re left staring into the dark of an unseen future, rejoicing in our anticipation of what’s possible rather than dreading what we cannot foresee and manage.
-----
* All of this material is excerpted from “Daniel Kahneman: Judgment, Decision, and Rationality” by Jack S. Levy, Rutgers University, available for download here. I’ve arranged excerpts from his text into table form.
** The rational actors of some economic and social theories figure prominently in the work of grantmakers who promote consideration of the “triple bottom line,” the economic, social, and ecological benefits produced by for-profit enterprises. I’ve been informed by private communication that researchers at the University of Schmerz am Überhogen are very close to developing the quadruple bottom line. In a landmark paper appearing in the Mathematikerzeitungsbringengelächterjournal, Alfred Stüber has proven that n-tuple bottom lines are impossible for any prime number n greater than or equal to 95,651.
Comments
You can follow this conversation by subscribing to the comment feed for this post.