The ‘Thou Shalt Nots’ of Systems Change

Enthusiasm for systems change is not new, and a broader historical perspective will help systems change enthusiasts learn from the past what NOT to do: replicate the ineffective mindsets and engineering approaches that have defined so much of the systems change work in our sectors.

By Christian Seelos, Sara Farley & Amanda L. Rose Jan. 14, 2021

(Illustration by iStock/kimberrywood)

We are currently witnessing a new wave of systems enthusiasm among philanthropic and development organizations eager to be identified as system leaders, with a host of implementing organizations and development partners aligning around reinvigorated calls for “systems change.” But systems change is not new: since the beginning of the last century, disciplines ranging from biology to psychology adopted system perspectives to, as Magnus Ramage and Karen Shipp put in their historical reflection on systems thinkers, “make sense of the complexity of the world [by looking] at it in terms of wholes and relationships rather than splitting it down into its parts and looking at each in isolation.”

In 1956, Ross Ashby wrote that the new systems science of Cybernetics “offers the hope of providing the essential methods by which to attack the ills—psychological, social, economic—which at present are defeating us by their intrinsic complexity.” Utopia seemed to be within mankind’s grasp, as Ashby and other cyberneticians expanded this discipline beyond machines, applying systems thinking to everything from organizations and medicine to the “reengineering” of entire countries. And in 1966, inspired by the new possibilities of space travel that emerged from systems science, Senator Gaylord Nelson introduced his vision of “A Space Age Trajectory To The Great Society” by asking:

“Mr. President, why can not the same [scientists and engineers] who can figure out a way to put a man in space figure out a way to keep him out of jail? Why cannot the engineers who can move a rocket to Mars figure out a way to move people through our cities and across the country without the honors of modern traffic and the concrete desert of our highway systems? Why cannot the scientists who can cleanse instruments to spend germ free years in space devise a method to end the present pollution of air and water here on earth?”

Within a few years, system analysis contracts created a multi-billion dollar industry driven, as Ida Hoos put it, by the “mythology” of Nelson’s proposal and an “aura of precision lent by a plethora of formulas, charts, and diagrams.” Hoos argued that this new mantra of engineering systems change was “exploited by a myriad of vested business and professional interests, and fostered by government officials eager to be identified with and to take advantage of advanced concepts of management science.”

But does this perspective inform effective action? After all, despite ambitious intentions and great efforts, the utopian Great Society did not emerge. Indeed, the convergence of social problems like COVID-19, systemic racism, food insecurity, and education disparities only further confound our efforts to achieve positive social outcomes. Have the supply of social problems increased and exceeded the supply of otherwise effective system solutions? Is system thinking a naïve mythology rather than a framework for bringing about effective solutions? Are we implementing the principles of systems thinking in ineffective ways?

We recently hosted a “Frank Conversation about Systems” with senior decision-makers from the philanthropic and development sectors to explore these questions. And although the conversation found no consensus on the first two, the group identified a long list of pathological behaviors upheld by numerous organizations deemed incompatible with a system perspective, what we’ve come to call the “Thou Shalt Nots” of systems thinking. For those committed to adopting system perspectives in their work and in their organizations, we offer the following account of our “frank conversation” as a reality check intended to ground decisions in a realistic assessment of the opportunities and potential stumbling blocks.

We’ve organized these stumbling blocks into four key areas: Processes, Cognition and Attitudes, Values, and Roles and Success Criteria.

Processes

Embedding the processes to “work in systems” within our organizations is difficult with so much inertia behind standard engineering approaches to problem solving and breaking the linear grip on planning and reporting is especially challenging in many quarters. Our participants noted increasing frequency of reporting and inadequate resources to maintain partnerships as two major challenges. But perhaps more problematically, a host of organizations confessed to the ambiguity or even resistance to letting others lead, which is required to pursue systems change strategies. Other key process challenges cited include:

Reliance on simple quantitative metrics for reporting that camouflage the messiness and nonlinearity of systems change coupled with a hunger for rapid reporting that is out of pace with early system change efforts
Difficulty cultivating ease among staff with zooming out and in as a reflexive part of operating in systems
Resistance to adapt strategies and plans despite emerging important information

Cognition and Attitudes

Many organizations pointed to the obsession with speed, action, and results as an attitude that can undermine a system mindset: impatience and bias for action legitimizes shortcutting the less clear, obvious, or attributable actions of organizations that might otherwise contribute to system change. Simplification of reality comes at the detriment of possibility when it comes to system change strategy. Other challenges noted by participants include:

Assumptions of knowing how to change systems ex-ante rather than exploring possibilities iteratively
The fascination and embrace of system-thinking toolsets, which may disguise a deeper unwillingness to change mindsets and attitudes (while the toolsets may be ill equipped to encourage mindset shift or to reveal how difficult it really is to do so)
A never-ending hamster wheel of pilot projects (“Pilotitis”) coupled with weak downstream interfaces/handoffs to advance successful projects toward broader system impact at scale
Preordained solution sets, which are then sought through grants, projects, and contracts (as opposed to empowering frontline staff to navigate systems by their judgement in real time)
A lack of humility and introspection to enable us to examine how we contribute to the problems we seek to solve. We are not simply funders of solutions!

Values

What values and principles do organizations bring to their systems change work? Participants acknowledged the central importance of values for local ownership, trust, democratized power structures, and inclusion of diverse voices in changing systems. However, they also voiced concern that popular “systems change” verbiage could be taken up by organizations without actually shifting away from the common business values of urgency, measurement, and control; “systems change” can become just another term of art rather than a vehicle for doing development differently. And when it comes to the values that drive organizational priorities and culture, participants agreed that leadership matters. Bottom up efforts to shift organizational values towards more participatory, inclusive, locally-led systems change initiatives will only go so far without top-down support.

Other values-related challenges raised by the group include:

The ethical justifications for changing systems, such as who decides on the direction and contours of the change being pursued
Power imbalances that often characterize donor-partner relationships, due to a fundamental disparity in resources
Reliance on codified scholarship and technically trained experts above traditional knowledge, local voices, youth, etc.

Roles and Success Criteria

Closely linked to the question of values are issues related to roles (i.e., Who gets to decide?) and success criteria (i.e., How is success measured?). Participants acknowledged challenges associated with shifting power and voice to those who do not have it now, pointing to the need to better clarify roles along change processes. Participants also raised but did not answer a difficult question: What, indeed, is the ethical and appropriate role for an external actor to play in driving systems change vis-à-vis system intrinsic actors? Generally, participants acknowledged a disconnect between the values inherent in systems change approaches and the oft-used practice of defining what success looks like ex ante. Participants expressed a desire to move toward celebrating the process by which change occurs (Is it inclusive? Is it locally led and supported?) instead of just what is achieved.

Additional challenges the group highlighted include:

Need to engage external system actors as catalysts or supporters of systems change, but not as drivers or dominant voices/actors in local change processes
Letting go of the desire to measure attribution and contribution in a systems context where many factors go into whether change manifests or not

Thou Shalt Not …

The “systems turn” in philanthropic and development sectors can be made more productive by focusing less on implementing new bold initiatives than on ceasing to operate from the ineffective mindsets and engineering approaches that define much of the work in our sectors, as one of the authors has recently argued. In this context, “Thou Shalt Nots” can help organizations reduce pathological behavior and generate space for a genuine system perspective. And while we may not yet have a consensus about what working from a genuine system perspective should look like, contrasting “traditional” and “systems” approaches can inform how we engage with problems, realities, and communities.

To structure our “Frank Conversation about Systems,” we offered the following stylized characteristics of the two contrasting approaches:

Traditional: Engineering Change

Well-defined: Narrow, focused, precise, often follows a linear path.
Observable: Objective information about key parts is deemed sufficient to define problematic situations, expecting predictable future trajectories.
Prioritizes solutions: Best practices built on superior/technical (Western) knowledge, an expectation of meeting defined success criteria, and “treating” the poor.
Technical solutions: Implementable solutions to be controlled, quantified, and managed, creating predictable change.
Result-oriented: Implementation plans are input designed by experts/consultants with fixed timeline and defined end states (outcomes).
Accepting risk (even w/o being aware) by making big bets: Accepting probabilities that assumptions, strategies, and plans may be wrong. Being ready to move on after failures.
Monitoring and Evaluation (M&E): Establishing upward accountability and reporting, and proving causal contribution to legitimize efforts.

Systems: Embracing Context

Broad: Diffuse, messy, and ambiguous, entailing multiple pathways.
Hidden: Validates the existence of the “things in-between” and acknowledges subjective realities, hidden power relations, and complex historical trajectories.
Creates possibilities: Adapting to socio-political context, locally-smart approaches, and expectation of learning and identifying and building on local knowledge.
Adaptive explorations: Evolving interventions that recognize change manifests in myriad cultural/cognitive/ behavioral shifts, that will mostly defy any single institution’s control.
Vision-oriented: No defined timeline nor pre-defined end-states, with shared visions guiding ongoing learning, and adapting co-owned by local “beneficiaries” and sponsors of change.
Managing risk by making smaller bets: Shaping probabilities by discovering what is locally possible, desirable, and how it can be scaled. Preventing failure and/or enable healing to continue.
Monitoring, Evaluation, Research, and Learning (MERL): Providing input to improve local decisions, sustain meaningful progress, and refine donor assumptions about social context.

To enable progress on this thinking, we are reaching out to you for sharing your ideas for your own “Thou Shalt Not” list and for ways of overcoming the barriers that prevent our organizations from adopting systems perspectives effectively. Please post your thoughts and ideas in the public comments section below this article. We hope that as a broader community we can find ways to support those who believe complex challenges do not outmatch systems approaches. Rather, it is up to all of us to rise to the task of mindset shift and organizational culture change. No easy feat indeed.

ORIGINAL Article