russian roulette

navigation, see below for text
frontline online

interview: bruce blair

Dr. Blair is a Senior Fellow in Foreign Policy Studies at the Brookings Institute. His academic research concentrates on the nuclear command and control of U.S. and Russian strategic weapons. He is a former Colonel in U.S. Strategic Command and served as a Minuteman launch officer.

The Norwegian incident. How seriously did the Russian military take the sighting of whatever this blip was?

Any unidentified missile launch from the area of the Norwegian or Barents Sea always is treated seriously by the Russian military, because that's an area in which U.S. Trident submarines are known to patrol, on occasion, and this was the weapon system that most threatens the Russian military planner. Those weapons have pinpoint accuracy, high yield, and short flight times to Moscow. So a missile launch from the area of the Norwegian Sea would be automatically a potential threat, a threat of decapitation of the top political and military leadership in Russia. That is the sort of canonical worst case scenario in the minds of the Russian military. It was by definition, I think, a serious event, because of the short flight time. And because of the fact that Russia relies, today, very heavily on quick launch of its strategic forces, because both the Russian command system and their nuclear forces are, today, quite vulnerable....

[The Russians] are supposed to detect an attack, assess it, and reach a decision on retaliation, all within a span of 10 minutes.  That's an extraordinarily short period of time for a rational decision process to run its course. Under the procedures for deciding whether to retaliate to an apparent attack, Russia has established, in their command posts, a ten minute deadline, which means in principle they're supposed to detect an attack, assess it, and reach a decision on retaliation, all within a span of ten minutes. That's an extraordinarily short period of time for a rational decision process to run its course. And when false information gets fed into a highly tense deliberation like that, it is potentially a formula for disaster. The Norwegian incident was less dangerous, because the political context was calm, generally. But if that event had occurred in a slightly different context, if there had been a domestic crisis in Russia at the time, if there had been some international crisis, then it would have been far more dangerous. The fact that confusion reigned for several minutes, perhaps as many as eight, into the ten minute deadline for reaching a decision, is, in and of itself, to me, cause for alarm about the danger that this event posed to the world. Now, if the Russians had interpreted this event as being truly a life and death threat to them, I believe that their preliminary alert messages that they sent during the period of deliberation would have been more serious than they were. They did issue some orders to move to increased combat readiness, orders that went to the Strategic Forces. And that is an indication that they were treating it seriously, they were taking precautionary steps at the period when they were uncertain as to whether this attack was real or not. But if they had been convinced that the event represented a threat to Moscow by a US Trident missile submarine they would have issued orders that would have flushed or dispersed their mobile land based rockets out of their garages, orders that would have led the bomber crews to meet up warheads to their bombers, and steps like this, that weren't taken....

As the operators in the early warning centers watched the blip on the screen, it split into sections. What happened there? Do those sections in any way resemble what would have happened with a Trident missile?

...I believe, though I don't know for sure, that what they observed from the radar returns was a missile that had the characteristics of a Trident ballistic missile that can carry eight or ten warheads. I believe that the trajectory, in the initial minutes of the detection, was quite uncertain. It could have been a missile that would land on Russian territory. The velocity, the number of stages of the missile..., its trajectory, its path, looked very similar to a Trident missile. And when the stages separated, and part of the missile fell aside, it certainly would have resembled a Trident missile, and it's very understandable that radar operators would conclude that this issue was above their pay grade, and needed to be evaluated by a higher authority. So, what I believe they did, since they had never seen a scientific rocket with these characteristics, and had seen Trident missile characteristics that were very similar, that they passed the buck...first to the Early Warning Center, outside of Moscow, and also to the General Staff...war-time command post at Chekhov, south of Moscow, and they also were unable to make a firm determination as to the identity of and nature of this event, and immediately activated, from the General Staff War Room the famous nuclear briefcase carried by President Yeltsin and the Defense Minister and Chief of the General Staff, and initiated an emergency conference, just like the conference the United States would hold under similar circumstances. [This] immediately brought President Yeltsin into consultation with his top nuclear advisers to receive information about the event that is displayed on a panel inside the nuclear briefcase. And to hear, through an emergency telecommunications conference, over the phone, the assessment of his military, as to whether this was an attack and what we should do about it. And I think that that was, no doubt an extraordinarily high adrenaline phase of this process.

Has it ever happened before, in Russian military history, that the nuclear briefcases have been activated like that?

The irony of the event is that this, to my knowledge, neither Russia nor the Soviet Union had ever gone into an emergency drill of this sort with the nuclear briefcase activated by a false alarm....

Was there a realistic possibility that, with discussions, they could have concluded that it was an American or a Western missile and there was a genuine threat? Was retaliation ever in the cards?

Well, the problem today is that Russia relies very heavily on the quick launch of its forces, so there's very little time to...ensure that the information is totally reliable. So under the circumstances, I think it's fair to say that there was a risk. How large is hard to say, but there was definitely a risk of a mistaken launch triggered by the Norwegian scientific rocket. If the event had occurred in a different context, that risk would have gone up. It could have been very high, in the right situation, or, I should say, the wrong situation. How are you to know how close one comes to a catastrophic human mistake, under the circumstances of such enormous pressure? I believe that there's an inherent risk in the operation of these nuclear arsenals, with thousands of warheads poised on missiles, ready for firing at a moment's notice and governed by a doctrine or a strategy of quick launch, of launch-on-warning. That has to be an inherent danger in the sense that no human being could really make a reliable, good decision, under the pressure of time that is allowed under this procedure....

What is the launch-on-warning policy and why does it increase the danger of a retaliation?

Launch-on-warning means firing one's own weapons between the time that we would detect an incoming missile strike and the time that those incoming warheads land on their targets. And that flight time would range between ten and thirty minutes, depending on the location of the launch. And so, launch-on-warning basically calls for firing one's own forces before they're destroyed on. And that only allows a few minutes for assessing an incoming attack, determining that it is real, and the country of origin, and its magnitude. A few minutes more for deciding whether to retaliate and how to retaliate, and a few minutes more for transmitting the orders and having them implemented by crews in the field. You add it all up, and it's only about a ten minute period of time that is allowed for this option.

Russia relies heavily on this because of the acute vulnerability of its forces. They no longer disperse their weapons into the sanctuaries of the oceans and the forests for their protection, because they lack the resources to keep these forces operating at a high tempo....So the vast bulk of the Russian nuclear arsenal is sitting in silos or in garages or at dockside, which puts them into a very vulnerable position. And if those forces don't fire from dockside, or their garages or silos, very quickly, they could be destroyed. Russian planners appreciate this fact very well, and they have geared their entire operation to fire those weapons out quickly before they're destroyed, or before the command and control network is disrupted, which in and of itself, could be a decisive failure in the strategic forces of Russia. So, for reasons of vulnerability of command and forces, today, Russia relies more on launch-on-warning than it ever has in the past, and, at the same time, Russia's early warning and command systems are deteriorating....

The end of the cold war surely means that they don't have the enemy that they once had. Is it inconceivable to the Russians that the West would ever launch an attack on them, or is there still the mindset there that an attack is possible? That a vestige of the cold war still continues?

I believe that the mindset of the Russian military planner is not that much different from its mindset during the cold war. And we certainly operate our nuclear forces as though the cold war never ended. We have thousands of strategic warheads, on each side, ready to be fired within a few minutes.... Now, that, to me, would represent a threat that has to be treated seriously by both US and Russian planners....I don't think the mindsets have changed enough, certainly, to take much comfort from the end of the cold war. And I believe that it is conceivable that, under the right brew of circumstances, we could find ourselves in a very tense and dangerous relationship with Russia....

In the comfort of our living rooms, we might imagine that the cold war is over and the threat has completely dissipated, but I think out there, within the military, the operations have not changed and the mindsets have not changed as much as you might imagine. I might also point out that many people say that, "Oh, the Russians surely can't imagine an attack against them by the United States, by the West," but the Russian military is observing, on a daily basis, all of the cold war operations that they observed during the cold war. And that includes...reconnaissance aircraft that we fly around the Russian borders, looking for holes in their air defense system where we could run bombers in during war-time. It consists of reconnaissance missions conducted by submarines, that occasionally go bump in the night to remind Russia of our presence. It includes anti-submarine warfare operations that we conduct every day against Russian forces, which we trail, and which we threaten. And there is a large set of daily activities that remind both sides' military establishments that there's still a state of virtual cold war, in terms of the operational world that they live in.

Do you see the Norwegian case as an isolated case or did it indicate something on a much wider scale?

Almost all false alarms in the history of the US nuclear warning network, and probably the Russian, as well, have been idiosyncratic. They have been unique in some respect. I recall two major false alarms on the US side, one in `79 and one in 1980. One was caused by the inadvertent insertion of a tape that simulated an all out attack against the United States, that was inserted by crews in the early warning system, who were just testing out the system. And that information was transmitted to the combat system, and looked like a real attack. In the second incident, in 1980, a 23-cent computer chip failed and generated information that looked like a large Soviet attack against the United States. These events are somewhat unique, they're not going to reoccur in exactly the same way that they did the first time around....Every false alarm in US history that I have studied indicates that it was unique in every case, and not to be replicated in the future. But I'm not sure that there's much comfort to be drawn from that, because there is a history of false alarms on both sides, and they will recur, and the only question really is how often, and what's the context, and what's the nature of the false alarm, and in every case, it will be a unique situation.

What is not unique, what is generally true about the Russian situation, is that their early warning and command systems have fallen on hard times, and they are deteriorating in physical respects. There are holes in the radar and satellite constellations. Russia has almost no ability to monitor the oceans from space, so they rely heavily on ground radars, and the ground radars are not being maintained properly. The crews that operate them are not as proficient as they once were, their morale is not as high. In every sense of the word, the Russian early warning and command system is suffering. And that's a trend that is almost certain to produce more false alarms in the future....

When the Russian command and control system was designed and implemented originally, was the level of sophistication of the safety procedures to prevent a mistaken or unauthorized launch satisfactory at that point in history?

Well, tight central control over nuclear weapons is a core value of the Russian military and political culture, we all know that, going back to Stalin. And so, as you can imagine, the Russian designers went to extraordinary lengths to ensure tight control over nuclear weapons. And the architecture of their system of control and safeguards is much more impressive than that of the United States. For example, they can monitor the status of missiles in their holes in Siberia from Moscow, and can actually both fire them from Moscow or override unauthorized launch commands from Moscow. There are electronic feedback loops all the way through the system, from bottom to top. This is unique. The United States has nothing close to that degree of ability to monitor and control its nuclear weapons. So, Russia's system of control, on paper, is extraordinarily impressive. The problem is that it wasn't designed for the kinds of upheavals that we have observed and they have experienced since the end of the cold war, which include coups, secession of parts of the country on which nuclear weapons are stationed, deepening civil/military tensions..., a complete breakdown in some areas of the military, in training and discipline and morale, a lack of adequate resources to troubleshoot these systems and make sure that they're overhauled and repaired properly. No command system, however impressive on paper, can function well under the circumstances of acute political instability and lack of resources. And that's the situation in Russia today, a set of safeguards that are impressive on paper but which are not being maintained and which were not designed for some of the threats that Russia now has to deal with.

...Imagine, a year or two ago, the head of the prestigious institute that designed all of the nuclear control and communications networks for the Soviet and then Russian Strategic Rocket Forces, that institute went out on strike, including the director of the institute. [They] went out into the streets to protest pay arrears, to protest the lack of resources to overhaul this vital part of the nuclear arsenal..., to protest the lack of resources to troubleshoot this system which is aging and needs to be repaired, and is generating more problems all the time. So, when you have a situation in which the designers believe that there's a serious concern, and they're actually going on strike to protest it, I take that to be a serious sign that something is awry.

Why were nuclear briefcases introduced in the `80s?

The nuclear briefcases were deployed because of the threat of Pershing II missiles that were going to be deployed in Europe in the early 1980s, which in the Russian view could have been launched and reach Moscow in under ten minutes, which meant that they could catch the leadership and decapitate the top leadership before they would have a chance to decide on retaliation, unless they had, in their possession, a handy-dandy device that would allow them to give authority to launch Soviet weapons immediately on receiving warning of an incoming Pershing attack. The Pershings were not deployed as part of the arms control deal cut in the early 1980s, but the submarine threat remains. And, in fact, the submarine threat grew in the eyes of the Russian planner, so that, today, they still require use of the nuclear briefcase so that the President, or his successor, can give permission to launch nuclear weapons immediately upon determining that Moscow is under attack by Trident submarine missiles. This was the underlying consideration during the Norwegian incident that activated the famous nuclear briefcase....

The nuclear briefcase is designed to receive early warning information and display it, and to allow the political leadership to give permission to the military to launch strategic forces. And that permission code that is contained in the nuclear briefcase carried by the President, is, by itself, neither necessary nor sufficient to initiate a missile strike. It simply is a mechanism, a device, that allows the President to, in a very efficient way, tell the military, "Go ahead and fire the missiles." ...[The] briefcase does not transmit a signal that directly initiates a missile strike, it transmits a signal that is received by people, mainly at the General Staff headquarters, who then decide whether it's truly authentic and what to do about it. And then they format a message that contains the real launch authorization and unlock codes, that they then would transmit to the submarine crews and the land rocket crews and others to carry out the orders. The General Staff possesses the wherewithal to initiate a strategic launch on its own, with or without the permission of the President, Defense Minister, or Chief of the General Staff. It's an arrangement identical to the one that exists in the US system, where our President carries codes around that he would use to give permission to use nuclear weapons. But our military carry the real codes that would be put into a message that would be sent out, that would actually trigger the crews in the field to fire their weapons....

The deterioration of the command and control system, if you were summing up generally about it, how would you say it compared now to ten or twenty years ago, just the level of functioning of it?

Ten or twenty years ago, the Russian nuclear command system and early warning network was operating at a very high tempo, I think quite efficiently and performing quite well, although they have always had gaps in radar coverage and early warning coverage, they've always had problems in their control of nuclear weapons. But the situation today, I would say, is many times worse than it was then, primarily because of the lack of funds to maintain the system, and not only the hardware but also to maintain the people and to ensure their adherence to nuclear weapon safety rules, their motivation to perform well, and to ensure proper training and proficiency of the people who operate these networks.

Is morale is lower in the Strategic Forces than ever?

The people I talk to who are close to the Strategic Rocket Forces and other elite nuclear units, including naval units, tell me that morale is lower than it has been historically. And this is manifest in increasing rates of suicide. Within the Strategic Rocket Forces, for example, where suicide officers have been established newly established at bases around the strategic installations to try to contain this problem. There clearly are signs of frustration within the nuclear navy. There are reports of desertion by crews who refuse to go out on a combat mission. Of families, of crew members, who are staging strikes or protest actions because of lack of pay or food or housing or something....

Is it possible, given that low morale, that the normal checks and balances in the command system could be short-circuited, that there might be a way a relatively low level commander could trigger a nuclear launch, either from a silo or an SS-20 or a submarine?

The likelihood of a successful unauthorized launch of strategic forces by the lowest echelon of command and control, I think, is fairly remote, but is increasing. Those crews need special unlocking codes in order to be able to physically fire their weapons. But in some cases, those coding systems are breaking down and are not being fixed. Some of the alarms that would notify others that an unauthorized action was taking place may not be functional any longer, they may be broken and not fixed. So that there's a general deterioration that can only increase the potential danger of an unauthorized launch, but I think that's among the least plausible of all of the scenarios of a breakdown of Russian control.

Now, the submarines are in kind of a category by themselves....A submarine is a vehicle that is not monitored continuously by a higher authority, nor can a higher authority take immediate action to override the conduct of a submarine crew. So there's autonomy on the submarine force that's lacking, say, in a land-based rocket force. So it's hard to know, because we haven't studied the blueprints, but it's certainly conceivable that a submarine crew could figure out how to circumvent safeguards, and at least pose a plausible blackmail threat of use even if they actually hadn't succeeded in bypassing the safeguards....

I believe that a department has been formed in the Strategic Rocket Forces to investigate any alleged cases of unauthorized or attempted unauthorized launches. If so, what does that tell us about their potential seriousness?

Well, a new office or a unit was established some years ago to address questions of the unauthorized use of nuclear weapons, and that tells me that during the period of the dissolution of the Soviet empire, and the period when nuclear weapons were being relocated in large numbers, and when the country, basically, was unstable, that there was a perceived need to address the potential risk of the unauthorized use of nuclear weapons....

We talked about the possibility of an unauthorized launch from a low level, is it possible that someone at a higher level who decided that he wanted to hold the Russian Army for ransom, or the world, would he ever have the capacity to launch a missile in an unauthorized manner?

The only echelon at which we're sure the ability to launch exists is the level of the General Staff, the highest echelon, their command posts and alternates. There is some belief within the American intelligence community that the ability to launch resides at somewhat lower levels, in the Strategic Rocket Forces, for example, and the Navy and Air Force, but there is a lively debate over the level at which the ability to launch resides in the Russian nuclear control system. Personally, I think it resides, in peacetime, at the level of the General Staff, its main headquarters and alternates, and, in a crisis, should one ever occur, that ability is distributed to lower levels. And this is where the danger would be greatest for an intermediate level commander to be in a position to fire without authority. Now, I think that...even in peace time, there are ways to circumvent safeguards. The question, again, is how much time and impunity is offered to an aberrant unit to do these sorts of things. I'm reminded of the situation in Ukraine, when Ukraine essentially could have taken control over strategic missiles based on its territory, and by the Russian General Staff's own estimation, bypass the existing locking devices, the safeguards on those missiles, within a matter of days to weeks. So, clearly, all of these safeguards only work for a period of time....

When he was announcing de-targeting, Clinton said, "American children can sleep safely in their beds at night, there's now nothing to worry about, Russian nuclear weapons aren't pointing at us any more." Was that really the case? Did the de-targeting make that much of a difference?

The 1994 de-targeting agreement was entirely cosmetic and symbolic and had absolutely no effect on the combat readiness of US and Russian nuclear forces, or on the danger or risk of unauthorized or accidental or inadvertent use of those weapons. Neither side stripped out their wartime targets, and, as a result, it's as trivial as changing channels on a television set with a remote control, to insert the war time target coordinates into the missile. In fact, they're already sitting in the missile, they're just changing files internally, on command. Therefore, if there's an intentional launch the crews go through the procedure of targeting missiles just before launch, and that's a step that, in fact, has been a standard launch procedure forever. So, this de-targeting agreement that requires targets to be reloaded into the missiles, hasn't added a single second to the time needed to fire the missiles intentionally....

We've talked of problems, what would you say are the chief solutions here?

Well, I believe that we need to stand down the nuclear arsenals, take them out of play, so they're not poised for immediate launch, so that they can no longer be susceptible to mistaken launch-on-warning, or to unauthorized or accidental firing....The problem that we face today is not deterrence. We don't need to keep thousands of warheads on high alert, poised for immediate launch, in order to deter one another. I'm not even sure we needed that many weapons during the cold war, but we certainly don't now, at the end of the cold war. Therefore we need to recognize that the primary challenge that we face today is not deterrence but a failure of control, particularly in Russia, because Russia depends more on nuclear weapons, depends more, currently, on quick launch of those weapons, at a time when its command and early warning networks are deteriorating. These hair trigger nuclear arsenals are inherently dangerous, and, on the Russian side at least, becoming more dangerous, because of the decline in early warning and control. So, the obvious solution to the danger of the hair trigger on Russian nuclear arsenals is to take them off high alert, so that those forces in Russia need hours, days, weeks or months of preparations, before they can be fired. That's called de-alerting. And it's a new agenda for arms control, and for US-Russian bilateral changes in their nuclear arsenals. But this is an agenda that should be pursued energetically now. Because we can't wait for decades for the nuclear arsenals to disappear, which they probably never will, from Russia and the United States. The dangers that I've tried to outline, of accident or unauthorized use of these forces, is a present danger and will only respond to changes in the operational safety of these arsenals. So we need to take steps that take Russian weapons out of play completely, and that means of course, that American weapons need to be de-alerted as well, and eventually, British, French and Chinese weapons.

What's the worst case scenario, if we don't adopt any of these measures and the Russian command and control system continues its current trend of deterioration, say, ten or twenty years hence?

...The Russian command system cannot endure the stress and strain indefinitely, there will be an incident. And I believe that we will look back at it, in hindsight, whether it's a year from now or ten years from now, I can't predict the timing of this breakdown, but we will look back at it in hindsight and decide that this was a train wreck in progress that we should have seen coming and which was preventable. And the obvious solution is not to count on nuclear arms control of the standard variety, which reduces the arsenals over the course of decades, and leaves hundreds or thousands of weapons still on high alert, but, rather, to move quickly in the months ahead, or certainly a short period of years, to stand down these arsenals so that they simply are not in any position to be used. This is not abolition. These weapons could be, if necessary in a national emergency, re-deployed, re-alerted. But, on a normal peacetime basis we have everything to gain by standing down these arsenals so that they simply cannot be fired, period. And if we don't take these measures, if we don't eliminate the hair trigger that exists on Russian nuclear forces -- and, I might say, on American, too -- because there's an inherent danger in this posture, then we are simply inviting an accident. This system is an accident waiting to happen. And, given the adverse trends in Russian early warning and control, physical, organizational and human I'm afraid that something will happen and sooner rather than later.

home . atomic suitcase bombs . a close call? . a nuclear smuggling scenario . russia's nuclear complex . debating the future . interviews . readings
join the discussion . synopsis . links . press . tapes & transcripts
frontline online . pbs online

web site copyright 1995-2014 WGBH educational foundation