Topsight > What is topsight?

A while ago, I posted my review of David Gelernter's book Mirror Worlds, an influential book that is about developing software to help us better understand complex systems. Published in 1991, this book was influential in a number of ways. Specifically, it introduced the notion of topsight.

As I wrote in my review of the book,

Gelernter argues that we often have trouble getting to the big picture, understanding the entire system. Instead, he says, we get mired in the details, something that he calls ant-vision. “Ant-vision is humanity’s usual fate; but seeing the whole is every thinking person’s aspiration. If you accomplish it, you have acquired something I call topsight.” Topsight—the overall understanding of the big picture—is something that we must "pursue avidly and continuously, and achieve gradually." It's a systemic understanding, a way of seeing the whole (see p.11; 30; 42; 51).

And I added:

topsight, like insight, comes gradually; don't confuse it with the model itself, understand it as something that the model make it possible to achieve.

As our systems become more complex, topsight becomes more critical to achieve. But it also becomes harder to achieve. That's certainly true in the complex systems that Gelernter wants to model, such as cities and nuclear power plants. But it's also true in the complex, overlapping, polysemous sociotechnical systems in which we work. How do we make sense of an organization in which several specialties overlap, an organization that uses off-the-shelf software and texts from different domains, an organization that has to work with multiple sets of rules? How do we tie together second-by-second operations, minute-by-minute tasks, and year-by-year activities, using them to yield a more complete understanding of the organization? How do we figure out how information flows through organizations, where it gets stuck, and how it becomes unstuck? How do we gain topsight—which, as Gelernter says, must be achieved gradually, like insight—when even a simple organization can develop unwritten rules, hold contrasting objectives, and be enmeshed with other stakeholders?

The term topsight, in fact, points to something that I've been studying since 1997 and teaching since 2000: how to investigate, analyze, diagnose, and model the ways that organizations circulate information. I've discussed parts of this work in my two books, in my many publications, and in seminars.


But Gelernter says something else about topsight. It shouldn't just be for a few people. If you want to drive smart, informed changes, you have to make sure that everyone has at least a chance to develop topsight.  He imagined this happening via broadly accessible models of complex systems. I imagine it happening by bringing methods out of academia—out of books, publications, and seminars—and making them accessible to the people in these systems, to consultants, and to undergraduates.


So for the last several months, I've been working on a project meant to pull my topsight-related work together and make it more broadly accessible. In the next few days, I'll be discussing that project—and topsight—on this blog. Stay tuned.