This week and next, Slate is publishing three exclusive excerpts from The Terror Presidency: Law and Judgment Inside the Bush Administrationby Jack L. Goldsmith. Goldsmith served as head of the Office of Legal Counsel from October 2003 to July of 2004. It was his job to advise the president on the legal boundaries of executive power. But when Goldsmith began to review the legal work of his predecessors, he became concerned that part of the legal framework limiting the conduct of the military and intelligence agencies in the war on terror was profoundly flawed. Attempts to fix those mistakes led to a legal upheaval in the war on terror, a series of clashes with David Addington, Vice President Dick Cheney's then-legal adviser, and ultimately to Goldsmith's resignation less than 10 months later. Today's selection reflects some of those early clashes.

A hastily arranged meeting in late February 2004 was typical of a dozen or so meetings I attended (and, I suspect, of many meetings before I arrived at OLC). Gonzales asked a group of lawyers to convene in his office to assess the implications of the Supreme Court's announcement that it would review a lower court's approval of the government's detention, without charge or trial, of Yaser Hamdi. Paul Clement, the deputy to Solicitor General Ted Olson and the best Supreme Court litigator of his generation, said the court's action was bad news. He explained that although we had solid legal arguments, the Supreme Court might not accept traditional wartime detention in the seemingly indefinite and ill-defined war on terrorism.

"Why don't we just go to Congress and get it to sign off on the whole detention program?" I asked, explaining that the Supreme Court would have a much harder time striking down a wartime detention program that had Congress' explicit support. Clement concurred, as did John Bellinger, Condoleezza Rice's legal adviser, and Department of Defense general counsel Jim Haynes. Those men had made this argument before. They had always been shot down, just as I was about to be.

Watch Dahlia Lithwick interview Jack Goldsmith.

"Why are you trying to give away the president's power?" Addington responded. He believed that the very act of asking for Congress' help would imply, contrary to the White House line, that the president needed legislative approval and could not act on his own. The president's power would diminish, Addington thought, if Congress declined its support once asked, especially if it tried to restrict presidential power in some way. Congress had balked, during the month after 9/11, at giving the president everything he had asked for in the congressional authorization to use force and the Patriot Act. Things would only be worse in 2004 and beyond, Addington believed.

Addington would always ask two simple questions whenever someone proposed that the White House work with Congress to clear away a legal restriction or to get the legislature on board: "Do we have the legal power to do it ourselves?" (meaning on the president's sole authority), and "Might Congress limit our options in ways that jeopardize American lives?" In the Hamdi meeting and in many others, everyone agreed that the president had the lawful authority to detain enemy soldiers during wartime. We also could not deny that going to Congress might limit the president's power at the margins, or that the limitation might conceivably cause us to release a detainee or fail to get information that resulted in another attack.

Of course there was an obvious counterargument: The relentlessly unilateral approach in a novel war might lead us to lose in the Supreme Court in a way that tied the president's hands much tighter than Congress would, thus jeopardizing American lives (if one thought this way) even more. This argument had at least a little traction. Whenever the Supreme Court threatened to review one of the administration's terrorism policies, Paul Clement was able to eke out small concessions from the White House. On the day of the Hamdi meeting, for example, the White House agreed to push harder on the efforts already in progress to establish, as a matter of executive discretion, more formal procedural protections for detainees like the ones Haynes and his subordinates had crafted over a year earlier.

Relatively small steps of this sort would, over several years, eventually add up to nontrivial protections for the GTMO detainees. Addington never liked these developments. But at least they were decisions by the executive alone, and thus were consistent with his fundamental stance of nonaccommodation toward Congress. They were also consistent with his relentlessly short-term perspective. Addington awoke every morning primed to do battle with the terrorists and focused on preventing an attack that day or that week, and not on what might happen next year or beyond. "We'll deal with that when and if it is necessary to do so," he would say when someone raised the possibility of difficulties that might result later from being overly aggressive now.

Addington once expressed his general attitude toward accommodation when he said, "We're going to push and push and push until some larger force makes us stop." He and, I presumed, his boss viewed power as the absence of constraint. These men believed that the president would be best equipped to identify and defeat the uncertain, shifting, and lethal new enemy by eliminating all hurdles to the exercise of his power. They had no sense of trading constraint for power. It seemed never to occur to them that it might be possible to increase the president's strength and effectiveness by accepting small limits on his prerogatives in order to secure more significant support from Congress, the courts, or allies. They believed cooperation and compromise signaled weakness and emboldened the enemies of America and the executive branch. When it came to terrorism, they viewed every encounter outside the innermost core of most trusted advisers as a zero-sum game that if they didn't win they would necessarily lose.

* *

Addington's hard-line nonaccommodation stance always prevailed when the lawyers met to discuss legal policy issues in Alberto Gonzales' office. During these meetings, Gonzales himself would sit quietly in his wing chair, occasionally asking questions but mostly listening as the querulous Addington did battle with whomever was seeking to "go soft." It was Gonzales' responsibility to determine what to advise the president after the lawyers had kicked the legal policy matters around. But I only knew him to disagree with Addington once, on an issue I cannot discuss, and on that issue the president overruled Gonzales and sided with the Addington position.

"When history looks back, I want to be in the class of people who did the right thing, the sensible thing, and not necessarily the fashionable thing, the thing that met the aesthetic of the moment," Douglas Feith told Jeffrey Goldberg in The New Yorker in 2005, referring to his controversial decisions in the Pentagon. This was precisely David Addington's attitude to hard legal decisions about terrorism. The vice president's influential lawyer had unshakable beliefs about executive power, about the correctness of his legal positions, about the nature and severity of the terrorist threat, and about what was necessary to defeat the threat. He thought that others had less information than the White House did about the threat, and were motivated by concerns other than what would best protect the American people. He believed that faced with tough decisions, true leadership required him to do what was right, even, and especially, if it was unpopular. As the complaints grew and grew, as the pressure to change course increased, Addington became more and more insistent that the administration was doing the right thing, and he stuck to his guns with an ever-firmer grip. Many times when Addington faced enormous resistance, I thought to myself, "He'll have to back down now." But he rarely did. And he never did on something he thought was important unless an immediate and unavoidable disaster would result. I grudgingly admired Addington's perverse integrity, even when I thought his judgments were crazy.