99. Memorandum From Victor Utgoff of the National Security Council Staff to the President’s Assistant for National Security Affairs (Brzezinski) and the President’s Deputy Assistant for National Security Affairs (Aaron)1
SUBJECT
- Observations on the Continued Need for a Triad of Strategic Forces
During the recent Mid-Term Review of Defense Programs, the President questioned Harold’s assumption that we must maintain our Triad of strategic forces. This memo reviews the arguments for maintaining a full Triad.
The first thing to note in addressing the overall issue is that, given the continued viability we project for the bomber and ICBM forces, and the modernization efforts already underway, the question is not the general one of whether we want a Triad. It is instead the more specific one of whether the modernization of the ICBM force to maintain a Triad (1) adds wholly unique and necessary capabilities to the Dyad of bombers and SLBMs we are already modernizing, or (2) hedges in some unique way against the possibility of a catastrophic loss of one leg of our strategic forces posture, or (3) provides capabilities present in the underlying Dyad, but at less cost.
Unique and Necessary Contributions of the ICBM Force
It is commonly argued that the unique combination of good C3, high readiness and short flight times make the ICBM force uniquely capable of rapid destruction of hard targets. There are several problems with this claim, however.
First, the requirement for rapid destruction of targets may not be very great. The ICBM modernization study now being completed by OSD argues that there are [less than 1 line not declassified] targets (including [less than 1 line not declassified] hard targets) other than missile silos and launch control centers that change value so rapidly that striking them with missiles might do significantly more damage than striking them later with bombers or cruise missiles.
[Page 440]However, the need to kill these targets is realistic only in scenarios in which the crisis deepens from modest levels (where actions to reduce the values of potential targets don’t seem necessary to the Soviets) to the missile shooting level in a matter of less than a day or two, and such scenarios don’t seem likely.
With respect to the missile silos and launch control centers, while we might like to be able to destroy such targets before they launch their missiles, it doesn’t seem likely that we would be able to. Soviet missile readiness levels, warning systems, doctrine, and exercises indicate that they are capable of and apparently plan on launching their missiles under attack. We might want to attack such targets anyway—to destroy missiles that failed to launch, or to prevent reload of the silos . . . However, if this is our purpose, a half a day or so delay in hitting these targets probably wouldn’t make much difference.
To the extent that there is a real requirement to be able to destroy the [less than 1 line not declassified] time-sensitive targets, SLBMs can be made capable of the task. Bill Perry noted in the Defense Review meeting that while it would be a tough technical problem, SLBMs can be made to achieve ICBM accuracies by the late 80’s. Further, contrary to popular opinion, submarine C3 is good enough to rapidly generate large nuclear strikes. Against worst case jamming, TACAMO aircraft operating off our coasts can complete transmission of an emergency action message (EAM) within 45 seconds with virtually perfect confidence that it will be received by any submarine at normal operating depth that is within 800 nautical miles. This implies that two TACAMO aircraft can guarantee coverage of about 4 million square miles of ocean—enough operating area for at least five SSBNs carrying a total of more than 600 warheads. These aircraft can also guarantee delivery of an EAM over an additional 10 million square nautical miles within 15 minutes—which would allow launch of more than 1500 additional warheads.
One might argue that other elements of the SSBN C3 networks may fail to get the message to the TACAMO aircraft, but in this regard, the SLBM C3 network seems no worse off than the C3 network supporting the ICBM force. Given the vulnerability of ground based communications stations and transmission lines, the ICBM force seems likely to be controllable only from our airborne command posts, and getting the word to TACAMO isn’t much tougher in principle than talking to our other airborne command posts.
The Increment of Hedging Provided by a Modernized ICBM Force
One of the strongest arguments made in favor of modernizing the ICBM force is that it substantially improves the degree to which our overall strategic posture is hedged against catastrophic failures in [Page 441] any one force component. This argument is based on two ideas—first, that failure of any one force component in a strategic force divided into three essentially equal components leaves you with ⅔ of your original force, whereas with a strategic force divided into 2 essentially equal components you would only have ½ your original force left. Second, since there is no reasonable way to establish the relative probabilities of unforeseen catastrophic failures in each strategic force component, we should adopt a conservative approach and construct our strategic force so as to make the loss of our most capable force component at least marginally acceptable. While these ideas are not unreasonable, a deeper look yields some interesting insights.
To begin our analysis, we might first examine the significance of the 17% difference between having ⅔ rather than ½ the total strategic force remain effective, given a catastrophic failure? At Tab A I have constructed illustrative 1990 strategic force postures for a full Triad, and a slightly augmented Dyad in which our ICBM forces are not modernized. For these national forces, you have [less than 1 line not declassified] remaining after the strongest leg of the Triad fails, versus [less than 1 line not declassified] after the stronger leg of the Dyad fails. The significance of such a difference is clearly a judgment call. While moving to sharply lower strategic force levels than say [less than 1 line not declassified] will make a 15–20% difference more significant, I don’t see this difference much of a motivation for modernizing the ICBM force in the foreseeable future.
We should also ask how likely the total failure of one leg really is? The only failure that our strategic forces have had that approaches the level one might call catastrophic occurred in the late 1960’s. [4 lines not declassified] While this was a very serious problem and took a number of years to fix, even if we had discovered it in the course of a nuclear exchange, there is good reason to believe that we could have strung out our missile launches and realized at least ⅓ of the original expected effectiveness in each missile leg (I grant you the failure was in 2 legs, but I’ll get to that point below).
However, if we assume that hedging against a ⅔ loss of effectiveness in any one leg of our posture is sufficient, a Dyad of forces should be acceptable. The reasoning behind this is that there seems to be a reasonable consensus that a Triad of forces provides sufficient hedging against a complete catastrophic failure of any one leg. However, a Dyad hedges as well against ⅔ type losses as a Triad does against complete losses (½ your force plus ⅓ of the other ½ is equal to ⅔).
Another question we might ask is—exactly how does modernizing the ICBM force to maintain a second missile force allow any greater degree of hedging than is possible by increasing the sizes of each leg of the underlying Dyad? Given the [10 lines not declassified]
[Page 442]Basing mode failures can be divided into two categories, failures by natural causes, and failures due to an unforeseen ASW vulnerability that gets exploited by our opponents. As far as natural causes are concerned, our oldest SSBNs have been operating for nearly 20 years and the complete gamut of required submarine evolutions through launch of the missile has been run through hundreds of times. Given this testing, and the fact that the Trident SSBN is already on the way and will have been in operation for many years by the time the current SSBNs will be retired in large numbers, the risk of a catastrophic failure of our SSBN force by natural causes seems extremely small.
The relatively significant possibility is that of a catastrophic ASW breakthrough—and even this doesn’t seem very large. Thus, the ICBM force’s significant marginal contribution to hedging against catastrophic failures—is in its ability to hedge against the loss due to a catastrophic ASW breakthrough, of whatever capabilities are provided uniquely by missiles.
Note, however, that the bomber forces can do anything except destroy their targets in less than [less than 1 line not declassified]—which takes us back to the issue raised on page 2—given the likelihood that any homeland-to-homeland nuclear exchange would take place only after a deep crisis lasting a period of days or weeks—how important is it to hedge against the loss of our ability to deliver nuclear weapons on target within less than [less than 1 line not declassified] of the decision to do so? Wouldn’t the value of the Soviets’ time-sensitive targets have been reduced to a minimum long before an exchange of nuclear weapons begins?
Can Modernizing the ICBM Force Give US Capabilities Provided by the Two Legs of the Triad but at Less Cost?
In my opinion, the arguments we have been through so far seem to make less than a compelling case for modernizing the ICBM force to retain a full Triad. Thus, if we want to make a reasonably strong objective case for modernizing the force, we need a positive answer to this last question.
The most up-to-date source we have for judging the relative costs of modernizing our strategic forces is the PD–18 follow-on study on modernizing the ICBM force.2 That study estimates the costs of deploying and operating additional strategic forces through the year 2000 at $5–7 million per arriving warhead for an ICBM system, $8–11 million per arriving air launched cruise missiles, and $15–17 million per arriving Trident warhead. While these figures indicate that arriving ICBM RVs would be somewhat cheaper than arriving ALCM and Trident [Page 443] weapons, they assume that we deploy the MAP system that DOD currently favors and that the Soviets do not fractionate their missile payloads to the maximum allowed by SALT in order to maximize their ability to attack the MAP system.
The costs of the alternative ICBM modernization schemes such as air mobile ICBMs, and the offshore submersible idea would probably be significantly higher than the MAP system, which could make ALCMs the least expensive option though we should realize its costs are quite uncertain, too. An ICBM system deployed on trucks that would disperse and hide whenever the Soviets order their forces to high readiness might cost significantly less than the MAP system, but DOD has not yet examined this system in any detail.
Summary and Conclusions
Aside from the institutional considerations, in order to believe that modernization of the ICBM force to maintain a Triad is necessary on other than perceptual grounds, it would seem necessary to subscribe to some combination of the following statements:
- —
- There is a significant probability that given a failure in one leg of our strategic force, this failure will reduce the effectiveness of the affected force by more than 67% and that the advantage (maximum 15–20%) in total force effectiveness of a Triad relative to a Dyad under such circumstances will make a significant difference.
- —
- There is a significant probability that we might find ourselves in a situation where the Soviets had achieved an ASW breakthrough negating the effectiveness of the SSBN force and a crisis had reached the shooting level so fast that the Soviets did not have enough time to reduce their time-sensitive targets down near their minimum values.
- —
- There is sufficient value and a high enough chance of neutralizing Soviet ICBMs before they can be launched to justify hedging against an ASW breakthrough that made the SSBN force unable to perform this task.
- —
- Modernizing the ICBM force allows us to buy at less cost capabilities that could be provided by further modernizing the other strategic forces.
While the first three statements have some validity, I believe that unless DOD can identify an ICBM modernization scheme that we can confidently expect to be more cost-effective in providing arriving weapons than CMC/ALCM or unless the political and institutional costs are too great, we should start moving toward a Dyad of SLBMs and strategic aircraft.
The above arguments also suggest the following:
- —
- The importance of ICBM modernization schemes that could provide arriving warheads at the lowest possible cost (such as the [Page 444] truck-mobile scheme mentioned by Harold Brown) should receive more emphasis.
- —
- A careful study should be made to determine whether DOD’s concerns about the security problems associated with moving strategic systems onto private land are reasonable—particularly if such movements occur only in deep crisis situations.
- —
- As part of the assessment of the feasibility of ICBM systems that are normally visible and countable but disperse in times of crisis, the intelligence community should be asked to assess the level of confidence we can have that we would detect Soviet moves to the levels of military readiness at which they would be willing to execute a counterforce attack. This assessment should identify the numbers and types of strategic forces and other important Soviet assets that might have to remain vulnerable to a US retaliatory strike in order to achieve strategic surprise.
- —
- DOD should be asked to bring its now overdue targeting study to a rapid close so that we can more fully appreciate how the issues it raises bear on the question of whether to modernize the ICBM force.
- —
- DOD should be asked for descriptions of the various classes of time-sensitive targets, the numbers of targets in each class, the time required to reduce them to their minimum values, and a comparison of their relative day-to-day and minimum values.
- —
- The intelligence community should be asked to assess the likelihood that we could prevent the launch of significant numbers of Soviet ICBMs.
- —
- Improvements in C3 systems that support the SLBM force (more TACAMO and improvements in the systems that can talk to it, such as E4B aircraft) should be supported in the upcoming FY 80 budget review.
- —
- On a very close-hold basis, a set of talking points making the case for a Dyad should be worked out to see if such a move can be explained in convincing terms to the public.
Even if our continuing investigations show that moving to a Dyad is the right thing to do, such a move would have to be made relatively slowly, given the upcoming SALT ratification debates and the enormous amount of change implied by the strategic forces decisions that have already been made during this Administration.
Fritz Ermarth has discussed this memo with me and has no significant difficulties with what it says.3
[Page 445]- Source: Carter Library, National Security Affairs, Brzezinski Material, Brzezinski Office File, Subject Chron File, Box 89, Doctrine/C3I: 1978. Top Secret. Sent for information. At the top of the memorandum, Brzezinski wrote: “VU this is a good + and provocative memo. Set up a discussion of it among our people—and at some point we ought to draw in the DOD. In the meantime develop the [illegible] memos mentioned on pp 6–7. ZB.”↩
- See Document 93.↩
- Ermarth initialed above his name and wrote at the bottom of the page: “I’ll try for some contribution of my own shortly. F.E.”↩
- No classification marking.↩