Making the best of linear thinking, complexly: typologies for reframing “coordination”


–I come from a training with little good to say about calls for “more effective coordination.” When having nothing more to say but feeling compelled to recommend something, then comes the irrefragable “what we need is more effective coordination.” Who can be against effective coordination, period. Though sought without a tincture of what to do, step by step and in real time. Like gold in seawater, coordination is impressive, but pointing that out is of scant use.

I’m not the only one who hesitates reading further when getting to the bit where death and disaster are credited to “the lack of coordination.” Past reading has conditioned us to expect no discussion on whether this “lack” arose because the responders knew—or reasonably thought so at the time—that they were already undertaking priority activities as urgent. Nor can there be any expectation that reading more demonstrates how death and disaster would not have arisen had activities been coordinated the way of hindsight with all its “could,” “may,” “perhaps,” “might,” and that ever-ready “must!”

Of course, such detail is difficult to summon, but that it is so rarely attempted leaves us to wonder just whose inexperience is revealed—the responders criticized or the callers for more coordination.


–I think the above is still true as far as it goes, but leaving it at that is no longer good enough, if it ever was.

I propose coordination is best understood as the chief limiting factor in policymaking and management under conditions of high uncertainty, complexity, conflict and incompleteness. Calls for better coordination are unavoidable as long as pertinence of the uncertain, complex, incomplete and conflicted remains. As such, “better coordination” is not the aim, but the limiting factor on being reliable under these conditions. Coordination is not an explanation, or a cause, or a rationale of management and policymaking, but rather a test of the latter’s validity.

Coordination is limiting because no one knows how to do it, when it—or something like it—needs to be done. It is chief because, even when better management has been undertaken, pertinent issues remain causally unclear (uncertain), variably numerous and interconnected (complex), interrupted and unfinished (incomplete), and under great dispute (conflict), often by virtue of that uncertainty, complexity and incompleteness.

Calls for more coordination are unavoidable when the remaining uncertainty, complexity, incompleteness and conflict need to be reduced but are not at that point reducible: “Thus,” so it goes, “we must have better coordination.”

In this way, the call for thus-coordination is an empty signifier for our not recasting the issues to determine whether we can better tolerate their remaining uncertainty, complexity, incompleteness and conflict. That is: Recast the issue to see if the change in the amalgam of complex, uncertain, unfinished, and disputed is useful to you. (Think of it this way: The one who does the recasting brings a totalizing impulse to the exercise, not just to make sense of the parts and the whole, but also to narrativize and in that way fabricate useful interconnections.)

What, though, makes for “pertinence”? This is to ask: Pertinent with respect to what? In this case, it is with respect to public policymaking and management. Over our profession’s half-century of evolution, policy analysts and public managers have witnessed different amalgams of uncertainty, complexity, incompleteness and conflict arising and emerging out of really-existing but highly differentiated configurations of: cultures, socio-technical systems, organizations, types of unpredictability, performance modes, and specific cases we call “major issues of policymaking and management.” Let me unpack that.


–The great irony in taking complexity seriously is the usefulness of linear thinking in analyzing complexity. I have to quickly add: …when that linear thinking is in the form of multiple typologies considered together for analyzing uncertainty, incompleteness and conflict as well.

A two-by-two table or some such is easily criticized for simplifying a complex reality. This, though, misses what has always been the latent methodological function of typologies in the plural: to remind us that reality is indeed more complex than lines, boxes and lists can portray.

Multiple typologies are the norm in my profession, and to use them in sequence—one after another, different terms following upon different terms—is to render a major policy and management mess granular enough for differing implications to become visible. Multiple typologies are not the pieces that complete a picture puzzle; they make a puzzle detailed enough to see a different puzzle or puzzles already there.

–The typologies in my own work come largely from sociology, political science and organization theory. In the most practical sense, you can begin with any typology, the entire point being there is no obvious macro, meso or micro start when it comes to reframing what is uncertain, complex, unfinished and disputed at the same time. The typologies I rely on include:

  • Different types of unpredictability, including measurable probabilities, unmeasurable uncertainties and unknown-unknowns (adapted from Andrew Stirling’s typology of incertitudes);
  • Different types of organizations, where production agencies for example differ significantly from coping agencies in terms of their observable/unobservable outputs and outcomes (J.Q. Wilson’s typology of agencies)
  • Different types of cases, e.g., “out there in reality” versus “the case emerges from your interaction with issues of concern” (Charles Ragin’s typology of cases)
  • Different types of large-scale technological systems whose centralized or decentralized operations vary as a result of component coupling and interactivity (Charles Perrow’s typology of high-risk technologies)
  • Different types of cultures for differentiating ways of life and policy orientations (the four cultures of Mary Douglas, Aaron Wildavsky and their colleagues)
  • Different performance modes—just-in-case, just-in-time, just-this-way and just-for-now—for the real-time high reliability management of large-scale socio-technical systems (a typology of Roe, Schulman and their colleagues)

No major issue emerges unchanged from the seriatim application of these typologies. More, all issues in this blog—inequality, poverty, war, climate change, pandemics, healthcare and more—merit application. Your typologies or like ones may be as fruitful in differentiating the amalgams of concern (e.g., “limiting factors” are treated as linear constraints in some models).

But in all this, remember the cardinal virtue of using typologies. It is to move you from the myriad types of contingent factors at work affecting a major policy and management issue—societal, political, economic, historical, cultural, legal, scientific, geographical, philosophical, governmental, psychological, neurological, technological, religious, or whatnot. It instead is to move you to the many criteria with which to identify and describe the factors that are pertinent. These reframing criteria are the dimensions (the horizontal and vertical gradients used in differentiating the cells) of each typology.


If the above is right, then calls for thus-coordination, rather than being conversation-stoppers, reflect that there is no single or final set of typologies—no last set of criteria—that resolves a still pertinent uncertainty, complexity, incompleteness and conflict. It is for that reason that my earlier dismissal of thus-coordination falls short and ended up exaggerated.

To call for more effective coordination is, in effect, to assume and to demand tolerance and forbearance. To put up with what you don’t like in terms of the uncertain, complex, incomplete and disputed is not easy.

All of this may seem like small beer, but to me it isn’t.

For it raises a question not yet posed. Elsewhere in this blog, I stress how important are the question, “What’s missing?,” and the admonition, “Be careful of what you wish for.” Put another way, it is to ask: Are we prepared to sacrifice one amalgam of uncertainty, complexity, incompleteness and conflict for another? Are we prepared to substitute the tolerance we are not altogether uncomfortable with for a forbearance about which the only thing we know is it too will be uncomfortable? In short, are we prepared to over-sample unpredictabilities and their drivers, given the fixation of others on the measurable and calculable? Note the we.

It remains too easy, in my opinion, for us to default to thus-coordination. Our costs of doing so need to be raised. One way is to insist that “we need more effective coordination” is to insist that “more effective” be a matter of comparing amalgams. Which do we prefer, the one for today’s major mess or the one that remains after having shown “we can reframe or recast that in the following way. . .”?

Principal sources

Perrow, C. (1984) 1999. Normal Accidents. Princeton, NJ: Princeton University Press

Ragin, C. 1992. “Introduction:  Cases of ‘What is a case?'” In What is a Case? (C. Ragin and H. Becker, Eds), Cambridge: Cambridge University Press

Roe, E., and P.R. Schulman 2008. High Reliability Management. Stanford, CA: Stanford University Press

Steinberg, L. (1972) 2007. Other Criteria. Chicago, IL: Chicago University Press

Stirling , A. 2010. “Keep It Complex!” Nature Comment, 23/ 30 December

Thompson, M., R. Ellis, and A. Wildavsky. 1990. Cultural Theory. Boulder, Colo.: Westview

Wilson, J. Q. 1989. Bureaucracy. New York: Basic Books

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s