On August 24, 1992, Hurricane Andrew slammed into southern Dade County, Florida. What followed was tragic, costly, and remarkable many respects. In this article, we focus on one multi-faceted aspect of Andrew's aftermath - the policy aspect, political and administrative. For this aspect was every bit as remarkable as all the others. The event threw a spotlight on a largely ignored policy subsystem - emergency management - and on the complicated and interesting political and administrative relationships that comprise that subsystem, and the context for that policy or, as some might say, nonpolicy (Wamsley and Milward, 1984). This article uses that spotlight to sharpen our understanding of some of the pathologies of our governmental institutions and processes general and, more specifically, to chart the destabilization and change that is taking place the emergency management subsystem. Something that is bringing about the first significant changes the subsystem since the Federal Emergency Management Agency (FEMA) was formed 14 years ago.(1) The subtitle of this article, Escalating a Quagmire, is borrowed from a paper about America's increasing involvement Viet Nam delivered by Daniel Ellsberg to a convention of the American Political Science Association the late 1960s, apparently before he became infamous for obtaining and leaking the Pentagon Papers (Ellsberg, 1967). While the escalating involvement and expenditures of America emergency management will not have consequences as tragic and traumatic as the Viet Nam War, they are nonetheless significant and worthy of our concern. And most assuredly, the emergency management policy subsystem and its political-administrative problems constitute a quagmire. The Escalation(2) Requests for presidential disaster declarations and declarations of disaster have grown steadily since 1988 and have steadily set record highs since 1990. Fiscal year 1993 established yet another record with over 70 requests and over 50 declarations. No one can say with certainty that this trend is irreversible. In FY 1984, declarations peaked at 42 and then fell steadily until 1988, before beginning the steady climb to the current record high. Today, from the local level to the White House, those involved emergency management widely believe that the trend will continue inexorably upward, and not because of nature's whims. It is generally believed by people throughout the subsystem that disasters are being both nationalized and politicized and that involvement is growing because of a combination of advancing technology and human-created forces and dynamics (Wamsley, Interviews, 1992-1993; Wallace, 1994). It is also recognized that some states have contributed more than others to this escalation. As one FEMA executive opined, in Texas they want a declaration every time a cow pisses on a flat rock (Wamsley and Schroeder, Interviews, 1995). The CNN Syndrome and the Media Presidency The first of these forces might be called the CNN Syndrome or the Camcorder Policy Process. In a few decades, we have gone from conveying news via an occasional action photograph backed by postevent printed and spoken words to instantaneous, live-action images and words our living rooms, the Oval Office, and even the White House Situation Room. Can there be a local or even a regional disaster under these conditions? The news media have become increasingly intrusive and influential the emergency management subsystem. At the same time, the media have increasingly reflected (and created?) the growing public pessimism regarding governmental capabilities (Lipset, 1987; Mitchell, 1987; Cloud, 1989; Washington Post, 1995). In other words, problems are being defined as national problems at the same time the national government is being portrayed as more and more inept. This has dramatically changed the context for the policy subsystem (NAPA, 1993; 18). News media have always played a role politics and influenced policy, but something more is at work today, something of a qualitative change. …