Last decade, NATO went ahead with the expansion of its ranks without the deepening, or streamlining of decision-making, that was needed to make it work.

It paid a high price.

In the 1990s, deepening was avoided because the alliance's rhetoric of doing everything "by consensus" was sacrosanct. To be sure, there were a number of genuine practices, some quite flexible, under the consensus label.

But no one dared challenge this rhetoric, which lay in wait to attack any real improvement in, or adequate use of, existing practices.

This year, the NATO secretary-general, George Robertson, has at last challenged this thinking. He has insistently raised the issue of enhancing decision-making procedures so that the newly expanded alliance can better manage itself.

Robertson is finding support, interestingly, from circles that had promoted widening NATO without deepening. Ron Asmus, a scholar at the Council on Foreign Relations and a former U.S. diplomat, has said NATO must re-examine its decision-making procedures fearlessly, without deference to bureaucratic prejudices. Former U.S. Secretary of State Madeleine Albright has written that the practice of decision-by-consensus "be revisited as NATO membership reaches into the middle 20s." For those experienced in Atlantic matters, this is an amazing step forward.

People are in a hurry to propose solutions, many of which are half-baked. Yet on this matter, even a half-baked solution is probably better than none at all. The official decision-making mechanism within NATO is flexible. If the sacred cows are pushed aside and the path opened to using this built-in flexibility, the necessary fine-tuning could be done later.

In the 1990s, in the absence of improved decision-making procedures, there was no space for considering Russian membership in the alliance. Nor was there much space for cooperation with Russia. Instead, there was a persistent fear that Moscow might disrupt the existing NATO consensus and divide the allies if it was allowed to talk freely with them.

Based on this fear, it was decided that NATO must not talk with Russia in the Permanent Joint Council until NATO had first reached its own common position separately, thus ensuring the council's ineffectiveness. The same fear has been used to weaken the brand-new NATO-Russia Council as well, by introducing a new "protective" device: that any single NATO member may, at any time, veto further discussion with Russia on any issue.

This is "protection" run amok. The only thing not protected in this is the ability of NATO to hold cumulative discussions with Russia. Or the vital interest of the United States in closer cooperation with Russia in the war on terror. No matter how much the West needs Russia's cooperation, Hungary would be able to veto the discussion. Or Poland. Or, pretty soon, Estonia.

When NATO admitted its three newest members in 1999, the space for consensus shrank. With a big-bang expansion of membership due in November, the space for consensus and decision will shrink further, unless decision-making is made more flexible.

Many Russian analysts console themselves with the thought that this will reduce NATO to an ineffective talking shop, but they're deluding themselves. A bloated NATO may not be able to agree on much, but it can agree to be anti-Russian.

The aspiring new members all have, understandably, deep historical resentments against Russia. Once NATO expands to the Baltics, it may end up that just about the only consensus left is on anti-Russian postures.

This would be a disaster for Russia and a disaster for Western security as well. Deepening, or adding options on decision-making approaches is NATO's only way out.

The basic methods for solving these problems were thought through in the 1990s by the Committee on Eastern Europe and Russia in NATO, of which I am member. Ours was the first Western nongovernmental organization to advocate NATO expansion. It has always been for expansion in a form that would not be "mechanistic" (as the Putin government aptly describes the current plan), but would undertake adequate reforms as well as a deepening of NATO to make it work.

The committee was set up in early 1992 by a small band of Atlanticists who had been talking about a post-communist reconfiguration of NATO since the mid-1980s. I had the honor of being one of the initiators, along with Igor Khripunov, then at the Russian Embassy in Washington, the late Bernard Yoh and Alan Lee Williams, head of the British Atlantic Council.

Later in 1992, David Abshire, former U.S. ambassador to NATO and a friend, though not a member, of CEERN, succinctly laid out the three main options: consensus minus one or two, weighted voting, and procedures for suspension of misbehaving members.

In a 1995 report released by former CIA director William Colby, our group analyzed the main options for decision-making and made its recommendations. (The report, along with a more exhaustive one from 1994, can be found on several sites, including an independent one, www.fas.org/man/nato/ ceern.)

Unfortunately, the plan of expansion that NATO adopted seven years ago neglected this issue and left it for the future. The options remain the same today, the problems having only grown worse in the meantime.

The basic point of the 1994-95 reports was that there is plenty of space for improving decision-making in NATO, even without a new treaty. The bulk of decisions do not impinge on national sovereignty, nor do they involve life-or-death matters for the member countries – except in the sense that the failure to take a strong enough common decision could have life-or-death consequences.

The North Atlantic Treaty deliberately left it to the North Atlantic Council to set its own procedures for making decisions, in the belief that a unanimity requirement was neither necessary nor desirable. The main options are:

(a) consensus minus one or two;

(b) weighted voting (with each country getting a realistic weight, probably population based, and a two-thirds or three-quarters majority required for a decision);

(c) suspension of a member that is persistently disruptive.

NATO could continue to use consensus as the normal operating procedure, yet adopt these procedures as options in reserve, for use if excessive obstructionism is making consensus unworkable. This would suffice to set the more obstreperous members on notice that they had better hurry to agree on a consensus because they could not blackmail NATO by holding out.

It would be a conservative reform, not undermining the consensus culture of NATO but upgrading it, by enabling it to function more smoothly. With an option of weighted voting among its procedures for dealing with obstruction, NATO could proceed to a "big bang" expansion without fear, and without drawing new dividing lines in Europe.

Thus came the CEERN proposal in 1995 for a "big bang" expansion. It was to have been a truly big bang, with sufficient reform to make the widening workable for NATO. It was to have been a bang in the interest of the whole communist, the Western countries and all the post-communist countries, not just the interests of a few small states in-between at the expense of the others. Yet the latter would be the result of a narrowed version of the "big bang" that was discussed in 1999-2000.

NATO itself has now turned to a variation on the big bang and is trying to figure out how to make it work. Its leadership has finally raised the underlying question –the widening-deepening question. There is a good answer to that question. It can be found in the original version of the "big bang."

Ira Straus is Fulbright professor at the Moscow State Institute of International Relations and U.S. coordinator of the Committee on Eastern Europe and Russia in NATO, a nongovernmental organization.