Unstirring your coffee (more viscous fluids, actually)

Here is an article in the New York Times about unstirring viscous liquids. Actually, while coffee is used as an analogy, it's not close to viscous enough to match what's been done. The researchers including nonlinear dynamics and fluid mechanics Jerry Gollub as well as physicists who would probably be considered more traditional. (Gollub overlaps enough with my community that he's one of us. He does a bunch of pattern formation and other such things.) Anyway, the researchers took two concentric cylinders filled with very viscous stuff (and particles with the same density so that they floated in the liquid) and turned one of them and turned it back. They apparently found a sharp transition in terms of when they could recover their initial configuration and when they couldn't. I'll say it can be expressed analytically as a phase transition, but I didn't check if there's a model that actually puts it in that framework. I'm sure somebody will put such a model forth soon enough if it hasn't happened yet. Of course, this is a nonequilibrium phenomenon, and going right to bifurcations in the usual pattern formation perspective would be more useful than just a phase transition model.

The article was shorter than I would have liked---it feels like it stops abruptly and could have used more detail---and the title of the article is exceptionally bad for what's actually contained therein.