After a long run as president of George Washington University, Stephen Joel Trachtenberg turned his thoughts to leaders whose time was cut short.

“Presidencies Derailed: Why University Leaders Fail and How to Prevent It,” published in August by Johns Hopkins University Press, takes on a subject of irresistible interest to those who track higher education.

Trachtenberg , who was GWU’s president from 1988 to 2007, wrote “Derailed” with Gerald B. Kauvar, a research professor at GWU, and E. Grady Bogue, a former chancellor of Louisiana State University.

In the book, they tell a story about a legendary University of California president, Clark Kerr, who observed that he left the UC presidency in 1967 the same way he had assumed it nine years earlier — “fired with enthusiasm.” It was a double-entendre, referring not only to Kerr’s zeal upon taking office but also to his “enthusiastic firing engineered by then-Gov. Ronald Reagan.”

Some critics maintained that Kerr had been too lenient toward student protesters at UC-Berkeley in the mid-1960s as the campus was seething with unrest. The Republican Reagan, a future U.S. president, preferred a harder line.

Being at odds with an ambitious governor is just one way for a university president to get ousted.

Trachtenberg and his ­co-authors find others: ethical lapses, poor interpersonal skills, governing-board troubles, inability to lead key constituencies and — an obvious one — failure to meet institutional objectives.

Much of their book focuses on examples in which university officials and fired presidents were guaranteed anonymity to obtain candid assessments. The authors define a derailment as the failure to finish the term of a first contract.

But two derailed presidents agreed to discuss their cases: William Frawley, fired from the public University of Mary Washington in 2007 after less than a year in the job, and Michael Garrison, who resigned from the public West Virginia University in 2008 soon after he began his second year as president.

Garrison, hired to be a change agent, ran into controversy over a football coach’s messy departure. Then he faced a governing board that Garrison said was “in disarray” following questions about ­record-keeping troubles with an MBA program for executives.

Frawley was axed after he suffered a “calamitous breakdown” that included two charges of driving under the influence of alcohol. Previously, he had been a dean under Trachtenberg at GWU.

Asked about the firing in an interview with Trachtenberg and Kauvar years later, Frawley said he believed it was important to be honest. He added:

“I think as a senior university administrator, you are forced into being functionally dishonest, saying empty positives in expected public academic discourse, all the while keeping the real truth of an institution — its crunches, impossibilities and realistic sense of itself and its future — behind villa walls. Over the course of twenty-five years, from chair to dean to president, I found myself becoming a robotic academic apologist and increasingly disillusioned with dishonesty. To do the job, I turned into a kind of ­anti-intellectual; the events that happened were a necessary part of me giving that up and getting genuine freedom, which I think I have now.”

Every derailed presidency has its own story. And many resignations, abrupt retirements or terminations do not qualify as derailments. Mysteries surround these situations because the key actors tend to clam up.

In December 2012, the University of the District of Columbia’s Board of Trustees fired President Allen L. Sessoms without cause, four years after he took office at the public school. Little was said afterward to explain the action.

In June, Joseph R. Urgo announced that he would step down as president of St. Mary’s College of Maryland when his three-year contract ended that month. The public liberal arts college had suffered two straight years of declining freshman enrollment.

In October, Sidney A. Ribeau abruptly announced his retirement after five years as president of private Howard University. He had a few months earlier signed a contract extension to serve through June 2015. Ribeau said at the time of the announcement that he was not forced out, but there were tensions last year between leaders of Howard’s governing board.

One of the most notable apparent derailments of the past two years was the ouster of University of Virginia President Teresa A. Sullivan in June 2012. That month, leaders of the U-Va. governing board engineered Sullivan’s resignation less than two years into her tenure — a power play that backfired. Eighteen days after the drama began, Sullivan was given her job back at the flagship public university. “A case study of a dysfunctional board,” Trachtenberg and his co-authors concluded.

Their recommendations for avoiding a derailment start with steps to ensure a successful presidential search, including a governing-board assessment, a strong search committee, clear priorities for characteristics of the next president and careful screening of candidates. Sometimes the process unravels, but no one will admit it. “When colleges, universities and systems set out to find a new president, they often begin and end in self-delusion,” the authors write.

They also advocate effective presidential transitions and advise leaders to “manage campus anxieties with care.”

The risks of failure are high. “When a derailment becomes apparent or known, the institution will be perceived as having failed in an important way,” the authors write. “Stories about why the president failed will abound — some true, some false, some both true and false.”