207. Report of the Inspection Study Group0

[Here follow a title page, preface, contents, Part A. History of the Inspection Study Group Project, Part B. Task of the Inspection Study Group, and Part C. General Remarks Concerning Inspection.]



Need for Experiment and Testing. It appears to us that the resources of existing technology will allow an inspection system that will support an arms control agreement. We are impressed, however, with the gap in our knowledge of such capabilities and therefore believe that a program of experiment and test is much needed to determine and perfect inspection equipment, techniques and methodology, and to permit the determination of a minimum level of inspection needed to support various objectives of arms control. Knowing then what type and degree of inspection is necessary to achieve specific inspection goals, an evaluation can and should be made of the minimum rights of access essential to meaningful inspection, and the policy implications of these necessary rights should be developed. On the basis of field test results and access requirement studies, longer range research and development programs can be formulated.

[Page 525]

There is a very great need for determining what is an acceptable balance of cost and effectiveness in an inspection system: that optimum employment of people, techniques and equipment which correctly balances the (unacceptable) cost of additional information with the cost of less—and insufficient—information. Even before this there is much testing to be done to determine whether a certain device will indeed do a specific inspection job, and how well. There is little question but that the inspection concepts we have proposed in Appendix III,1 for example, are redundant in combination and in many individual cases more thorough than will be required in the actual inspection situation. The technical “minimums”, however, cannot be satisfactorily determined until actual field tests of inspection equipment and techniques have been carried out. Considering the abstract physical properties of equipment and techniques is not sufficient. It is necessary to take them off the shelf, into the field and try them, individually and in combination, as nearly as possible in the context of an actual arms control inspection situation.

“Critical Components” Philosophy. Zonal inspection is itself a form of sampling.2 Within the inspection process, another sampling method which may cut down the number of inspectors and kinds of inspection is to locate the critical points in the production, handling, or use of the item to be inspected. We need to look at the things that give the greatest gain in information. Much productive research could be conducted to establish for each inspection target the critical components—those items, stages or people without which the inspected thing cannot be produced or distributed. (Such research will have to keep pace with new technology, especially when an inspection system is actually in operation.) It must be kept in mind, however, that this is and must remain a sampling technique, and that there should be no corresponding diminution of inspection rights; that is, any limitation to inspection of critical components should be one of choice for efficiency within a secure inspection framework, not a restriction written into an agreement. As a matter of principle, no agreement should have as one of its requirements limitation to critical components, for it would always be technically possible to circumvent such an agreement.
Transport Inspection Problems. In our preliminary description of the general characteristics of an inspection system for zonal disarmament, by far the greatest number of inspectors and consequently the major portion of the personnel costs are those connected with transport inspection, where there must be provided a capability to detect a sufficient [Page 526] fraction of any illegal movement out of zones to deter efforts at evasion. The Group looked into countrywide inspection at all zonal boundaries and found it impractical because of the very large number of inspectors required; we therefore suggest restriction to the boundaries of the single zone selected for inspection. Careful advance planning can reduce the interval between announcement of the zone and arrival of inspectors; a moratorium on traffic out of the zone for even a few hours, checked by air surveillance, would help; and the zonal inspection forces can first be assigned entirely to seal off the zone, then move to their regular assignments. Even then, transport inspection will require substantial numbers of inspectors. A priority research task is to develop equipment and techniques which will reduce the numbers necessary for this critical portion of the total inspection process.
Need for Precise Definition and Designations. In the past the USSR has been able to evade the general purposes of declarations to international organizations by using special designators for equipment which cannot be compared with those in general use either internally or internationally. It is essential that arms declarations agreements include explicit provisions for giving all designators for the item under consideration, definition of locations by name and geographic coordinates, as well as meaningful and internationally acceptable specifications for items to be declared. Similarly, all necessary inspection rights and procedures must be defined explicitly and precisely, with their practical application spelled out in detail. Zonal design criteria must be decided upon which will balance potential gerrymandering against the requirements of efficient and secure inspection. (See Appendix VI for further discussion.)3
Intelligence Capabilities. Unilateral intelligence probably has the capability to assess the accuracy of armaments declarations relating to strategic aircraft, fixed strategic missile launch facilities and AICBM installations, location and class of airfields and ports, the number and location of major surface ships, and the number of missile-carrying submarines and the number of launchers they contain. In other areas of armaments declarations, intelligence can aid in evaluating their general reasonableness. There is particularly a weakness in capability to evaluate operational inventories of missiles, including those on submarines, and declarations concerning mobile land-based missile launchers. (See Appendices II and VII.)4
Intelligence can aid in the identification of evasions by discovering indications of suspicious activity and guiding inspection efforts to [Page 527] the extent consistent with the protection of intelligence sources. Intelligence may not, for various reasons, justify a smaller inspection effort, but it can increase the deterrent influence and effectiveness of one in being.
Addition of Fixed Launch Facilities as Inspection Target. The relative confidence in intelligence and inspection capabilities in connection with fixed launching sites for strategic missiles, compared with the lack of confidence in ability to verify declarations on the missiles themselves, leads us to suggest that any control agreement, in addition to numbers of missiles and their deployment, include as well inspection of fixed launch facilities.
Use of Indications Intelligence. The method of indications intelligence, i.e., the comparison of intelligence information with a list of “indicators” which taken together suggest the occurrence of some significant event, should be developed in application to arms control. It may help to disclose evasions of the agreement which are not individually susceptible to detection or measurement, or which in isolation are seemingly not significant. (See Appendix VII.)

[Here follow Part E. Intelligence and Arms Control Inspection, Part F. Some General Characteristics of an Inspection System, Part G. Research and Development, Test and Evaluation Programs on Inspection Techniques, and seven appendices.]

  1. Source: Kennedy Library, National Security Files, Kaysen Series, Inspection Study Group. Secret. The typescript report is 12 pages, plus 4 pages of front matter and 7 appendices, totaling 98 pages. Regarding the interagency Inspection Study Group (also known as the Foster Panel), see footnote 3, Document 140. In an August 31 letter to McGeorge Bundy, Foster enclosed copies of the report “for yourself and for the President if you find, as I hope you will, that it is worthy of his attention.” He called the report “the most thorough investigation of inspection and verification, and of the relations of intelligence to arms control inspection, that has so far been made,” though he emphasized that it was “but a beginning, and that its preliminary estimate of the scope of inspection that might be required for comprehensive arms control on a zonal basis, for example, is only illustrative.” (Kennedy Library, National Security Files, Departments and Agencies Series, ACDA, Disarmament, General, 7/62-6/63)
  2. Appendix III is entitled “Some General Characteristics of an Inspection System To Verify Comprehensive Arms Control Programs.”
  3. The Inspection Study Group did not concern itself with the technical problem of statistical sampling in inspection. [Footnote in the source text.]
  4. The title of Appendix VI is “Some Aspects of Zonal Schemes.”
  5. Appendices II and VII are entitled respectively “Intelligence and ACDA Plan 1,” and “Findings and Problems.”