interests. It must allow the skeptical to resolve their suspicions, not simply provide facts: relying on the testimony of “experts” is clearly insufficient (e.g., Kash et al., 1976; Hoos, 1978). The system must be prepared to receive communications, not just disseminate them. Receiving communications identifies where credibility is a problem, points to unforeseen informational needs, and helps officials to refine their own perceptions about what is actually happening in response to an emergency.
Given such a formidable challenge, it makes sense not to attempt to gather every bit of detailed information, but instead to ensure that communication lines are kept open. In a crisis situation, unusual kinds of information may be needed quickly. For example, there may be a need to know what legal authority state and local officials have, which vacant buildings are available for shelter, or whether school buses can be used to supplement mass transit systems. The requirements for such information—and the facts themselves—will usually be highly localized. It would be futile to try to maintain a flow of timely and correct information to meet so many localized needs. More appropriate is a system that focuses centralized attention on a limited number of information functions and that assures their credibility. It is important for the system to be an effective emergency communication system as distinct from an information system.
Meeting information needs in an emergency through a decentralized, pluralistic structure increases the importance of communication among individuals and among local organizations. Consequently, attention must be given to such prosaic matters as maintaining up-to-date lists of correct names and telephone numbers of key people. It is usually easy to identify ahead of time who will have a particular type of information, even if it is difficult to gather the information and keep it up to date. Also, because some communication links may be congested or broken in an emergency, the need for localized information suggests a need to build some redundancy into the emergency communication system to be used.
The more severe a future energy emergency is, the more certainly will the federal government play a major role in meeting emergency needs. This fact creates a real dilemma for the allocation of emergency responsibilities among levels of government and the rest of society. To provide any in-