Crisis-Related Time and Information Pressures

3. Factors That Make Conflict Intractable

This introductory article was written by ChatGPT at the direction of Heidi Burgess, who reviewed, edited, and approved the final content.
For more information

Crises create a dangerous combination of urgency, uncertainty, fear, and incomplete information. Decisions must often be made before leaders have had time to collect all the relevant facts, consult all the right people, understand the full context, or evaluate all the possible consequences. Yet waiting too long can also be disastrous. When lives, security, democratic institutions, or social stability are at stake, decision makers are forced to act on the basis of what they know — or think they know — at the time.

This problem is especially severe in complex systems, where causes and effects are hard to trace and where interventions may produce surprising, delayed, or unintended consequences. In such situations, there may be no single “right answer” waiting to be discovered by better analysis. Instead, leaders must make provisional judgments, test assumptions, monitor results, and adjust quickly as new information becomes available. This is difficult even under ideal conditions. It becomes much harder when information is fragmented across different agencies, organizations, political camps, media systems, or expert communities.

Crisis-related time and information pressures can therefore make conflict much more destructive. People may seize on the first plausible explanation, blame a convenient enemy, rely on familiar but outdated procedures, suppress dissenting warnings, or act on information that is wrong, biased, or incomplete. Decision makers may also be pushed by public pressure, media demands, partisan incentives, or bureaucratic procedures that are poorly suited to the emergency at hand. Under these conditions, actions intended to solve the crisis can unintentionally make it worse.

Many well-known failures illustrate this danger. After Hurricane Katrina, official reviews found that responders lacked accurate, real-time situational awareness and that slow, confusing procedures delayed needed assistance. The 9/11 Commission found that existing hijacking-response protocols assumed that officials would have time to work through normal chains of command and that aircraft would remain identifiable — assumptions that did not fit the actual attack. The Challenger disaster similarly showed how high-stakes organizations can make catastrophic decisions when warnings are not fully communicated and decision makers rely on incomplete or misleading information.

More conflict-related examples include Israel's Yom Kippur War of 1973, which was a case of warning signs filtered through fixed assumptions. Brookings’ “The Fog of Certainty” argues that Israel’s failure to recognize and respond to key warning indicators before the 1973 Arab-Israeli War shows the danger of relying on unchallenged basic assumptions. The US State Department also produced a memo warning of imminent war, but that memo was ignored as well.

The October 7, 2023 attack is likely another example. There most certainly were warning signs that Hamas was preparing for such an attack, but some of those warning signs and plans were noticed, but were discounted, misinterpreted, compartmentalized, or not acted upon because they conflicted with prevailing assumptions. A Reuters report on the Israeli military inquiry says the IDF underestimated Hamas’s capabilities, believed Hamas was not seeking large-scale conflict, relied heavily on intelligence and defensive systems, and had focused attention on other fronts. Other accounts point to Israel’s belief that Hamas was deterred or pacified, overreliance on the border fence and surveillance technology, diversion of resources to other threats, and internal political turmoil as parts of the larger failure pattern.

The U.S. Cuban Missile Crisis of 1962 is another conflict-related example. Here, many things were done right — indeed, we would not be here writing this had they not been, as nuclear war between the U.S. and the U.S.S.R. was only narrowly averted. Michael Dobbs’ USIP report is useful because it challenges the tidy “rational crisis management” story. He argues that the real danger came not only from Kennedy and Khrushchev’s choices, but from unpredictable events neither leader fully controlled. The report says Kennedy was often only dimly aware of events elsewhere and that chance incidents—such as an aircraft going astray, a soldier losing his temper, or a spy misidentifying something—can redirect history during war and crisis.

A final example is the U.S.'s decision to go to war with Iraq in 2002 because leaders believed that Iraq had and was making more weapons of mass destruction. The 2005 WMD Commission concluded that U.S. intelligence seriously misjudged Iraq’s alleged nuclear program, failed to consider Iraq’s political and social context, failed to keep an open mind about new data, and did not sufficiently foster dissenting views.

The lesson from all these examples is not that leaders should avoid acting until they know everything. In crises, that is usually impossible. The lesson is that societies need decision-making systems that are designed for uncertainty: diverse information channels, trusted experts, clear communication, dissent-friendly deliberation, rapid feedback, pre-planned emergency procedures, and a willingness to revise course as better information emerges. In constructive conflict terms, the challenge is to act quickly enough to meet urgent needs, but carefully enough to avoid escalating the conflict, deepening mistrust, or locking in mistakes that could have been corrected.

—————

This page was created by ChatGPT in response to this prompt. It was then reviewed, edited, supplemented and approved by Heidi Burgess. More information about how and why we are using AI in this way, and about the growing number of ways in which Beyond Intractability is using ChatGPT, Claude and other AI systems to generate content and build out the BI system, is available on our BI/AI Overview Page.