Who could argue with the core of the argument here, built on the resuscitation of Henry Mintzberg’s ideas from a decade and half ago, that philanthropy should have clear goals, but maintain flexibility and openness in the ways that the goals are pursued? The main issue that I have with John Kania, Mark Kramer, and Patty Russell’s article is not related to the plausibility of their central argument, but rather to the missed opportunity to focus on what really ails strategic philanthropy.

To make the authors’ case for emergent strategy, the article creates a straw man of a hermetically sealed approach to philanthropic strategy that plods along toward goals, marching through causal models, committing to performance metrics and other relics of a trendy but technocratic approach to getting results. The authors protest effectively against rigidity and narrow causal approaches, but ultimately never get around to focusing adequately on the most difficult problem in strategic philanthropy: the enduring presence of vast amounts of “noise” or external confounding factors in almost all the work that philanthropy undertakes, from small local initiatives to broad international agendas. No matter how open funders are to adapting and adjusting, they must close some doors in order to take action. There simply is not enough money in philanthropy to take on all the dimensions of important social problems. Although the concept is appealing, one worries that emergent strategy would need to be huge and very expensive indeed in many philanthropic domains.

Whether the problem is simple, complicated, or complex, to use their language, there is always a deep question in philanthropy about where the boundaries of the problem are to be set, at least initially. Take the case of education. Underperforming schools could be improved with better teaching inside the schools, but the range of critical factors affecting school performance is much broader than teacher effectiveness. Research has long suggested that parental engagement is one of the most powerful determinants of academic achievement. Of course, many factors in turn shape the possibility of parental commitment, ranging from the ability of parents to find time to attend school meetings to conditions in the home that may shape the capacity of parents for deep engagement in their child’s education.

For a problem to be acted on, it must have some boundaries. As soon as philanthropy starts to work on part of a problem, it must accept that there are some parts of the problem that are not actionable and that noise around their approach will be present. Noise or confounding factors are widespread in social systems, especially in education. Beyond parental commitment, there are many other factors that shape what it is possible to achieve within schools, ranging from local public funding that affects school budgets, to bus schedules that may affect attendance, to changes in the local economy that may stress families and increase residential mobility. Working on one or more of these factors, even if they are closely connected to the goal at hand, does not obviate the fact that there is potentially far more explanatory power in the factors not addressed than in the one that is chosen for funding.

This is the enduring problem in philanthropy: much of what determines whether a goal is achieved or not is a function of factors well outside the funding control of the donor, no matter how broad, adaptive, and expansive the strategy becomes. Whether the approach is narrow or emergent, some factors left outside the system will be of real and significant consequence. The article never fully confronts this problem when it advocates instead a more open and sensing approach to strategy development.

The problem in strategic philanthropy thus comes down at its core to where the boundaries of a problem will be set and how many factors will be addressed in the quest to make an impact. Smart funders will start with the causes that are most proximate to the problem at hand. At the same time they will be aware that a huge amount of the change they seek to create is contingent on factors that are now and probably will long be outside their span of control. The most effective funders are constantly aware of the noise in the systems and attempt to internalize little by little some of the most significant externalities.

If emergent philanthropy helps foundations see the core problem of noise in complex systems and the weakness of simple causal models, then it can’t help but be useful. More helpful, however, would be a direct acknowledgement of the limits of philanthropic strategy in controlling externalities or adjusting to all the contingencies that can arise when a strategy is implemented. Emergent strategy offers a bit too much false hope.

In the end, it is interesting that the search for yet another banner for the field of strategic philanthropy to rally under has led us back 15 years to Mintzberg’s work in corporate strategy. This tells us just how enduring the problem of affecting change within philanthropy truly is. It just may be that the authors here who hope to shape strategy development in the world of philanthropy may have to do more than write an article to have a chance at getting to their goal. They may just have to embrace an emergent strategy themselves to reach this very complex and elusive objective.

Read the rest of the responses.