FAS | Government Secrecy | Library || Index | Search | Join FAS



Security Classification of Information: Table of Contents

Chapter 5.
INFORMATION-DISCLOSURE RISKS AND BENEFITS

INTRODUCTION

If information under consideration for classification is in a classifiable area and if the government can control that information, then two of the three major requirements for determining whether information should be classified have been met (see Chapter 3). The third and final requirement is met when the following question is answered affirmatively: "Should this information be classified because its unauthorized disclosure reasonably could be expected to cause damage to the national security?" This final requirement, based on Sect. 1.1(a)(3) of EO 12356,1 is the most difficult to establish of the three major requirements.*

The answer to the question of whether unauthorized disclosure of information reasonably could be expected to cause damage to national security is at least a two-step process. The first step is to determine whether an unauthorized disclosure could damage the national security. That step presumably also requires that the specific damage to the national security be identified. The second step is to determine whether it is reasonable to expect such damage. However, even when unauthorized disclosure of information reasonably could be expected to cause damage to the national security, it does not necessarily follow that the information should be classified. There is in practice a third step to this process—a step not mentioned in EO 12356 but a step that is generally followed by those who determine whether information should be classified. That third step balances the damage to national security which reasonably could be expected to be caused by an unauthorized disclosure against the benefit to the nation of having the information unclassified. That third and final step determines whether information-disclosure damages exceed information-disclosure benefits and thereby determines whether the information should be classified.

Balancing information-disclosure risks and benefits (classification benefits and costs) is not required by EO 12356. That EO requires only a finding that disclosure of information reasonably could be expected to cause damage to national security before the information can be classified. EO 12356 requirements imply that if any damage to national security reasonably could be expected by disclosing information, then that information should be classified to prevent that damage from occurring. However, there are many adverse impacts to the nation that can result when information is classified.2 Information should not be classified if the costs of classification (the benefits of having the information unclassified) exceed the benefits of classification (avoiding the damages caused by disclosing the information). Therefore, in practice, both classification costs and classification benefits are usually considered when determinations are made to classify information.

Although EO 12356 does not require that classification benefits (preventing damage to national security) exceed classification costs (the benefits to the nation of information disclosure) before information can be classified, such a balancing procedure is a sound approach to making classification decisions. Therefore, the remainder of this document assumes that such a balancing process will be carried out when those decisions are made.

Most of the rest of this chapter is concerned with developing criteria for determining the risks and benefits to the national security of disclosing information under consideration for classification and describing factors that can be used to evaluate those criteria. Principles to use when balancing those risks and benefits will be considered in the next chapter. However, before those criteria and factors are discussed, it is important to consider the meaning of the term "national security." Focusing on what is truly meant by national security will help ensure that only damage to national security is included as an information-disclosure risk when classification decisions are made. Damages that do not directly injure the national security (e.g., damages to the general welfare of the nation) should not be considered as information- disclosure risks. Therefore, in the following section of this chapter the meaning of the terms national security and national defense are discussed with respect to classification of information.

THE MEANING OF NATIONAL SECURITY AND NATIONAL DEFENSE

The Meaning of National Security

The fundamental meaning of national security is protection from physical attack by foreign nations or from the reasonable apprehension of such attack.3 In a very broad sense, national security could encompass many other factors,4 including the following:

Current assets and national interests, as well as the sources of strength upon which our future as a nation depends . . . [These] range widely from political assets such as the Bill of Rights, our political institutions and international friendships, to many economic assets which radiate worldwide from a highly productive domestic economy supported by rich national resources. It is the urgent need to protect valuables such as these which legitimizes and makes essential the role of national security.5

However, such a definition is too broad to guide classification decisions, since classification of information should be minimized,6 and such a broad definition would result in far too much information being classified.

The Supreme Court indicated, in Cole v. Young, that the term national security related "only to those activities which are directly connected with the Nation's safety, as distinguished from the general welfare."*, 7 National security "was intended to comprehend only those activities of the Government that are directly concerned with the protection of the Nation from internal subversion or foreign aggression and not those which contribute to the strength of the Nation only through their impact on the general welfare." 8 Thus, the Supreme Court has adopted a relatively narrow definition of national security.

The EO 12356 definition of national security is also relatively narrow.* That EO defines national security as "the national defense or foreign relations of the United States."+, 9 Congress has tended to limit the meaning of national security even more than has the Executive branch. For example, the Freedom of Information Act (FOIA) exemption for information properly classified pursuant to executive order refers to matters "to be kept secret in the interest of national defense and foreign policy" 10 (emphasis added). Foreign policy is presumably a narrower field than foreign relations. Further, the FOIA exemption for confidential sources refers to records of information compiled by an agency conducting a lawful national security intelligence investigation.11 The legislative history of that confi dential source exemption specifically states that "`national security' is to be strictly construed to refer to military security, national defense, or foreign policy."12 However, since either foreign relations or foreign policy information is of relatively minor concern in this document, further examination of the definition of national security will be limited to the meaning of the term national defense.

The Meaning of National Defense

National defense was said by the Supreme Court in Gorin v. United States to refer to "the military and naval establishments and the related activities of national preparedness."++, 13 This definition was given in a case involving espionage activities directed at national defense information (a matter very pertinent to security classification of information). The Court's relatively narrow interpretation of the meaning of the term national defense is consistent with the legislative history of the Espionage Act. A broad definition of national defense proposed by the House of Representatives for that Act was not acceptable to the Senate.14

A narrow view of what is encompassed by the term national defense was also adopted in another espionage case, United States v. Heine, where a U.S. Circuit Court stated the following:

It seems plain that the section [of the Espionage Act] cannot cover information about all those activities which become tributary to "the national defense" in time of war; for in modern war there are none which do not. The amount of iron smelted, of steel forged, of parts fabricated; the number of arable acres, their average yield; engineering schools, scientific schools, medical schools, their staffs, their students, their curriculums, their laboratories; metal deposits; technical publications of all kinds; such nontechnical publications as disclose the pacific or belligerent temper of the people, or their discontent with the government; every part in short of the national economy and everything tending to disclose the national mind are important in time of war, and will then "relate to the national defense." . . . Such an assertion of national isolationism is certainly not to be imputed to Congress.

A less impossible interpretation would be to confine the clause to information about things adapted only for the national defense, and things of ambiguous use which have been already collected, or prepared, for the support of the armed services. The first would include airplanes, cannon, small arms and munitions of war, warships, forts, and the like; the second, coal, food, clothing or other supplies accumulated for the army and navy.15

In a more recent espionage case, a Circuit Court cited with approval the following jury instructions on the meaning of national defense:

And that term, the term national defense, includes all matters that directly or may reasonably be connected with the defense of the United States against any of its enemies. It refers to the military and naval establishments and the related activities of national preparedness.16

In summary, the meaning of the term national defense should be given a relatively narrow construction consistent with the judicial decisions in United States v. Heine and Cole v. Young.* Therefore, when evaluating damage to the national defense caused by informa tion disclosure, the damage under consideration should be limited to that damage reasonably and directly related 17 to those national activities concerned with protecting U.S. citizens and U.S. property from physical attack or the reasonable apprehension of such attack.

RISKS (COSTS) OF INFORMATION DISCLOSURE (CLASSIFICATION BENEFITS)

General

Before information can be classified as NSI, it must be determined that disclosure of that information reasonably could be expected to damage the national security.18 In three NSI areas (see Appendix B for a discussion of all ten NSI areas), the unauthorized disclosure of information is presumed to cause damage to the national security. Those three areas are (1) foreign government information (NSI area 3), (2) the identity of a foreign confidential source (included in NSI area 9), and (3) intelligence sources or methods (included in NSI area 4).19 Information is automatically classified if it falls within one of those three areas. Balancing the costs and benefits of classifying that information is not necessary. If information falls within one of those three NSI areas, then it is necessary only to determine the classification level and duration of that information.*

For the other NSI areas (areas 1, 2, 4 through 8, and 9), the damage to national security by disclosing the information must be determined before a classification decision is made. In those areas, it is frequently obvious that releasing certain information will help an adversary and thereby cause damage to national security. For example, if an adversary knew the details of one of our important new weapons, then that adversary could develop that weapon for its own use or could devise effective countermeasures, thereby negating the weapon's advantage. Knowledge of our battle plans or detailed military strategies would aid our adversaries in planning their military strategies. Similarly, we would obviously be at a disadvantage in our foreign relations with a nation if that nation knew our diplomatic negotiating strategies or the keys to our cryptographic codes. However, there are other circumstances where it is not obvious that releasing certain information will damage our national security and guidance is helpful in evaluating the risks to national security.

The following discussions on evaluating risks to national security caused by disclosure of the information under consideration for classification will focus on scientific or technical information (NSI area 6 and atomic energy information). Risks to the national security of disclosing information concerning NSI areas 1, 2, 4, and 5 are discussed in Appendixes C through F. Appendix B includes a brief discussion of such risks in some of the other NSI areas.

It should perhaps be mentioned that EO 12356 specifically identifies circumstances when information shall not be classified (i.e., it describes certain "benefits" of classification that shall not be considered as benefits). EO 12356 states that "in no case shall information be classified in order to conceal violations of law, inefficiency, or administrative error; to prevent embarrassment to a person, organization, or agency; to restrain competition; or to prevent or delay the release of information that does not require protection in the interest of national security."20

Six Criteria for Evaluating Risks of Information Disclosure

Six basic criteria for evaluating the risks to the national security of scientific or technical information disclosure have been established by DOE. Although those criteria were developed for evaluating declassification actions (for evaluating the risks of declassifying information), they are also applicable to classification actions (for evaluating the risks of not classifying information). The six DOE criteria,21 revised so that they are not specific to atomic energy information, are as follows:

These criteria seem to encompass all the risks that need to be evaluated when one determines whether to classify scientific or technical information.* No other major criteria have so far been identified by the author. The following subsections will discuss using those six major criteria to determine the damage to national security caused by information disclosure.

Assistance to Other Nations in Developing New Armaments

How adversaries may be assisted. Our adversaries may be helped in several ways if certain scientific or technical information important to our national security is not classified.+, ++

That information may provide the following types of assistance to those adversaries:

1. Indicating technology areas that could provide significant national security advantages. Those are areas where we are the first to recognize a national security application of a technology (e.g., when a significant technological advance—a breakthrough—occurs). By classifying the information, we buy lead time, the time to develop that application before it is developed by an adversary.**, 22 Sometimes, even the fact that we are interested in a technology area should be classified so that an adversary is not alerted to a potentially advantageous area of research and development.* Generally, basic scientific research is not classified unless there is a significant breakthrough that has a direct national security application in a new field or unless the new information represents a radical change in an existing field.+, 23

2. Indicating what can be achieved (e.g., the fact that an atomic bomb will work),++, 24 or the existence of a "feasible possibility." 25, 26

3. Indicating the principle by which something can be achieved.**, ++

4. Indicating technology areas where our knowledge is weak. The fact that certain researches are in progress must sometimes be classified to avoid identifying areas where we are not adequately prepared.++, 27 We do not wish an adversary to learn of our vulnerabilities. Therefore, information indicating that we are far behind an adversary should be classified.25

5. Indicating technology areas where our knowledge is strong. This may indicate areas that may provide significant national security advantages (see Item 1).

6. Identifying scientific or technical problems that are not readily apparent. This helps an adversary to avoid those problems and thereby saves the adversary time and money.***

7. Identifying successful methods of solving the scientific or technical problems (information concerning the methods that will work, the methods that will not work, and the method that is the best). This saves an adversary from wasting time and money on methods that will not work or are not optimum.

8. Identifying key operating parameters for a process. This tells an adversary where to spend major efforts to optimize the process, again saving the adversary time and money.

9. Providing detailed know-how concerning processes or equipment.* Classification is particularly useful for controlling design and production information (e.g., detailed drawings and special manufacturing techniques).28 A DoD task force on the Export of U.S. technology strongly emphasized the necessity for the United States to control its design and manufacturing know-how (the details of manufacturing) if it was to maintain its technological superiority over other nations.29

10. Providing data that will confirm information an adversary already has obtained by its own efforts. This may give an adversary increased confidence in the accuracy of its results, leading to faster progress by the adversary in that field.

Items 1 through 3 indicate the importance of classifying certain research and develop ment work, an action that is consistent with trade secret law which allows this type of information to be a trade secret (see Appendix A). Items 6 through 10 indicate the importance of classifying design details and test results (also important in trade secret matters), information that can be discovered by adversaries but only after expenditure of much effort over a long time period.

Extent of assistance to adversaries. In determining whether to classify certain information, it is not enough to conclude that release of that information could help an adversary. The extent of such help must also be determined for each of the ways by which the adversary may be helped. Expertise in the scientific or technical areas of interest is required for this determination. Intelligence information is also needed because knowledge of the adversary's starting technology base is necessary to quantify the extent to which the new information could help that adversary.

When determining whether release of scientific or technical information could help an adversary, the relative technological position of the United States and an adversary is a major factor to consider. The general opinion is that if the United States is far ahead of an adversary in a technology, then the information should be classified to protect that lead,25 especially if the technology is critically important to national security30 and if there is a long development time for a weapons system using that technology.+, 31,32 This opinion is consistent with the conclusion of the Bucy Report, a DoD analysis of export control of U.S. technology. That report stated that controls should be placed on all key elements of a technology when the United States has a major lead in that technology.33 Such controls can help maintain this long lead time, especially when the lead is a result of a "revolutionary" advance in the underlying technology.34 Information about a revolutionary advance is less likely to be independently discovered by others than is information about an "evolutionary" advance. When the United States has only a small lead in the technology, which is usually the situation with a mature technology where the lead is a result of evolutionary advances, then controls on dissemination of the technology have only short-term effectiveness (the adversary's technology will also evolve) and it may not be worthwhile to institute such controls.33

One aspect of determining whether release of information will help an adversary is to consider the time frame during which the information would be helpful to the adversary. Would the information be helpful in the near future (short term) or only in the long term?35

In determining whether to classify scientific or technical information, it must be remembered that our country may have several adversaries who may be at different stages of scientific or technical development. This is a particularly important consideration when the classification decision concerns non-state-of-the-art information. State-of-the-art scientific or technical information related to national security is frequently classified. Sometimes there is a tendency not to classify non-state-of-the-art information in that same area because that information is not important in our national security systems (i.e., not important in our systems that have been designed with respect to "high technology" adversaries). This may not be a wise decision. Although non-state-of-the-art information may not be useful to our adversaries whose level of technological achievement may be only slightly less than our technological level, that information could be very helpful to less-advanced unfriendly nations.*,36 A pertinent example concerns the manufacture of fissile materials for nuclear weapons. A principal adversary, the U.S.S.R. [or its successor(s)], long ago mastered certain technologies for producing uranium enriched in the fissile U-235 isotope. If the United States revealed its classified uranium enrichment technologies, then that information would probably not significantly assist the U.S.S.R. However, such information would be very valuable to certain "Nth-power" countries who have not yet produced nuclear weapons but who are very interested in obtaining that capability. Therefore, for nonproliferation reasons,+ the United States continues to classify certain uranium enrichment technical information, information that is probably also possessed by some of our adversaries.++

Extent of effort to obtain data as a measure of assistance. Scientific or technical data are different from much other information that is classified for national security reasons because these data can be independently obtained by adversaries through their own (non intelligence, nonespionage) efforts (once the adversary realizes that these data may be useful). However, substantial work may be required by an adversary to obtain that data. Classifying the data that we have obtained will then require an adversary to use its own resources (sometimes substantial) to get that information, resources that otherwise could be used by the adversary (e.g., on other weapon systems)37 to the adversary's advantage and to our probable detriment.*

That rationale was given by the Atomic Energy Commission (AEC), in 1953, for not declassifying certain atomic energy information:

Relatively little has been released about those nuclear constants which are very difficult to measure and which are essential to the design of advanced reactors. . . . We have released very few details on [the distortion and corrosion of fuel elements in nuclear reactors, the fabrication of fuel elements, and their performance under irradiation] in the belief that the data are critical to our business and cannot be understood without time- consuming activities involving unique facilities. . . . [A]s long as we feel that we are retarding unfriendly nations in their efforts in this field—by making them perform the work themselves—we can afford to take a measure of slowdown in our own progress.38

This "extent of effort" factor was again mentioned by the AEC in 1974 as a factor taken into account when making declassification decisions on atomic energy information.39 The amount of time, money, or effort expended to obtain information, particularly information from research and development activities, is also a factor in determining whether that information is a trade secret (see Appendix A).

There is only one instance (known to the author) where quantitative guidance was provided on how much effort an adversary should be required to spend to get information that is relatively straightforward to obtain before that information should be classified by the United States. That instance concerned atomic energy information generated during the Manhattan Project. The Smyth Report40 was a general account of all U.S. work during World War II to develop and produce atomic bombs. It was published in August 1945, shortly after two atomic bombs were dropped on Japan. In order to decide which atomic energy information could be released in the Smyth Report, Manhattan Project leaders established several guiding principles. One principle was that information could be made public if "it could be discovered by a small group (15 of whom not over 5 would be senior men) of competent scientists working in a well-equipped college lab in a year's time or less."41 Thus, Manhattan Project management established the principle that if an adversary could obtain scientific or technical information by an effort of less than 15 "scientist-years," then it would not be worthwhile for the United States to classify that information.+

This criterion assumes that the scientific and technical principles are known so that the effort to obtain the information is relatively straightforward (e.g., measurement of neutron capture cross-section of elements). This well-established precedent of a 15 person-year measure of effort seems as appropriate today as it was over 45 years ago. Since it was accepted by the senior officials of the Manhattan Project, a DOE predecessor organization, then it is still appropriate to use that measure of effort for classification decisions within DOE (i.e., the precedent exists and has not yet been overruled).

The effort necessary for an adversary to get certain information can reasonably be estimated from knowing our efforts to get that information. Generally, most advanced nations are at about our level of technology. One of those nations would be expected to expend about the same effort to get the information as we spent. A nation not at our advanced level of technology would be expected to require relatively more effort than we spent to get that information. However, that low-technology nation may be able to buy that expertise from a technologically advanced nation at a cost approximately equal to the development effort. Therefore, estimating the cost to an adversary to get the information under consideration for classification should not be difficult.

The preceding paragraphs considered the effort required to get information as a measure to help decide whether to classify information. The time required to get information is sometimes a more important factor than the effort.* Delaying the date by which an adversary acquires a weapon system or countermeasures to a weapon system may determine the outcome of a battle or a war.+ Delays in Brazil's and South Africa's uranium enrichment programs (caused by classification and other nonproliferation efforts of the United States and other nations) are said to have given more time for the leaders in those countries who opposed building nuclear weapons a chance to gain political control and terminate those programs.42 The time factor will be further considered in a subsequent chapter, where duration of classification is discussed.

Assistance to Other Nations in Improving Their Armaments

The factors that were discussed in the preceding section on risks of assisting other nations in developing new armaments also apply to the risks of assisting other nations in improving their armaments. However, the risks generally may be less with respect to assisting other nations in improving their armaments (where the improvements are generally by evolutionary advances) than assisting other nations in developing new armaments (which generally result from revolutionary advances).

Assistance to Other Nations in Manufacturing Materials for Armaments

The factors that were discussed in a preceding section on risks of assisting other nations in developing new armaments also apply to the risks of assisting other nations in manufacturing materials for armaments. As mentioned in the preceding section, the degree of risk may be less for materials that are used to improve armaments (evolutionary advances) than for materials that are used to make new armaments (revolutionary advances). Within DOE, this criterion is of particular importance because it encompasses the production of special nuclear materials, the materials needed to make nuclear weapons. Production of nuclear weapons was a revolutionary advance in armaments. Therefore, information important to the production of special nuclear materials (e.g., how to produce militarily significant quantities of uranium enriched in the U-235 isotope) should be stringently classified.

Detrimental Effects on U.S. Foreign Relations

U.S. foreign relations, including arms control negotiations and treaty obligations, could be adversely affected by disclosing (i.e., not classifying) certain information. For example, the U.S. is very interested in preventing other nations from developing nuclear weapons. Major diplomatic initiatives by the United States and other countries (e.g., the Nuclear Non- Proliferation Treaty) have helped limit the spread of these weapons. Most nuclear weapon countries cooperate with the United States and closely control the materials and technologies of importance in developing and producing these weapons. If the United States released (e.g, did not classify or declassified), for apparent economic reasons, technical information that is perceived to be helpful in producing special nuclear material or otherwise in developing nuclear weapons, then some of these other countries might decide to release similar information, to the major detriment of these international control arrangements and to world political stability.

Any Other National Security Impact or Significance

The criterion of "any other national security impact or significance" encompasses those national security impacts not included in the other risk criteria. It includes, for example, the extent to which the information would assist other nations (e.g., adversary nations) in assessing or countering U.S. capabilities or determining U.S. limitations or vulnerabilities.

Detrimental Effects on Classification Program Credibility

For classified projects, if the information disclosed (not classified) is very important to the success of the project and if project personnel believe that reasons for not classifying that information did not adequately consider national security matters (e.g., if short-term cost savings or "political" purposes appear to override important national security factors), then the credibility of the classification program will be diminished by not classifying (or declassifying) that information. Therefore, not classifying (or declassifying) information for unsound reasons (unsound as viewed by rank and file project members) may lead to poor attitudes by some project personnel towards protecting the project's remaining classified information.*, 43

The credibility of the classification program may also be affected by the expected breadth of dissemination of the classified information and the planned end use of that information (or a product embodying that information). The extent of dissemination or the planned end use may be such that classification of the information will soon be impractical. "Classification under such circumstances may degrade the classification system by attempting to impose security controls in impractical situations."44 However, information could be classified until the information was widely distributed or the product that revealed this information was put into use.

BENEFITS OF INFORMATION DISCLOSURE
(COSTS OF CLASSIFICATION)

General

The benefits of information disclosure (having the information unclassified) are sometimes better understood and evaluated if they are considered as the costs that would accrue if the information was classified. In those instances, determining the costs of classification will establish the benefits of not classifying the information (the benefits of information disclosure).*

The benefits of information disclosure are broader in scope than are the risks of information disclosure. Risks of information disclosure pertain only to damages to the national security (see earlier in this chapter). Benefits of information disclosure include benefits to the general welfare of the nation and are therefore much broader in scope.

Eight Criteria for Evaluating Benefits of Information Disclosure

DOE's current procedure for evaluating a proposed declassification action+ includes the consideration of five benefits from having scientific or technical information be unclassified.45

Those criteria, revised so that they are not specific to atomic energy information, are as follows:

The following additional benefit was listed in a 1987 DOE summary of classification policy:46

With respect to scientific and technical information, another benefit should be included:

This criterion was one of the criteria used by the Tolman Committee in evaluating Manhattan Project atomic energy information for declassification.*

A "catch-all" benefit criterion should be included, similar to inclusion of the "any other national security impact or significance" criterion that was included as a risk criterion. That final classification benefit criterion is

An evaluation of those eight categories of benefits + is presented in the following subsections, usually as costs of classification rather than as direct benefits of having the information unclassified.

Benefits to the Progress of the U.S. Program

The benefits to the progress of a U.S. program for which scientific or technical information was developed include avoiding the following delays or inefficiencies (which may be translated to costs) to the program if that information is not classified:

As mentioned previously in the section on risks of information disclosure, the relative technological position of the United States and an adversary is a factor to consider when determining whether release of scientific or technical information would help an adversary. When the United States is far ahead in the technology, then the information should probably be classified to protect that lead. In this section on benefits of information disclosure, the question is whether the United States can be helped by an increased rate of progress in the scientific or technical field, if the information is not classified. The general opinion is that if the United States is slightly behind an adversary, then the information that the United States is developing should not be classified. In that situation, because the United States needs to catch up with the adversary as fast as possible, all the work in that area should be unclassified to take advantage of the value of unrestricted communications.25 For similar reasons, the technology should be unclassified when the United States is on a par with or only slightly ahead of the adversary and possibly when the United States is far behind an adversary.*, 30

Benefits to the U.S. Program by Eliminating Classification Costs

The benefits to a U.S. program of not classifying scientific or technical information include avoiding the following direct costs to the program:

Although some of those classification costs were discussed in an earlier report,2 little information is available to aid in estimating those costs.* It would be useful if federal agencies that frequently sponsor major classified technical projects would also sponsor studies on the costs of classification.

Costs of declassifying documents were discussed at a 1971 congressional hearing on U.S. Government information policies and practices. A DoD representative stated that much declassification required document-by-document review. Such review was difficult to accomplish because the experts who must be used to make declassification judgments were fully occupied in other duties.56 He was not able to estimate the costs of declassification. Congressman Horton suggested that it would be a good idea if DoD conducted a study to find out how much it would cost to declassify documents,57 but such a study does not appear to have been carried out.

Sometimes the most important benefits to the progress of a program are not due to the reduction of direct costs, but because of the indirect benefits. For example, there will usually be more rapid progress in a scientific or technical field when restrictions on communications between workers in such a field are minimized. This is particularly true when workers on a classified project are able to discuss progress with their peers in academia or those working on unclassified projects. A hindrance to rapid progress in some scientific and technical atomic energy areas is caused by the "born classified" concept. Scientists in unclassified university and other research facilities cannot work in some areas of atomic energy because of classification restrictions. Such restrictions will, of course, hinder the rate of scientific progress in that field. However, such restrictions are sometimes essential for national security reasons. See Security Classification of Information, Volume 1. Introduction, History, and Adverse Impacts,2 for further information on indirect benefits of not classifying information.

Benefits to General Scientific and Technological Progress

Overall U.S. scientific and technical progress is benefitted by not classifying scientific or technical information because the following costs are avoided:

An important factor is the short-term and long-term effects of classification on the nation's scientific and technical progress. Generally, short-term (e.g., a year or two) classification of scientific and technical information will not have major impacts on the nation. However, there may be adverse long-range consequences when those constraints are extended.58

One of the earliest groups to consider the short-term and long-term effects of classifying scientific or technical information was the Tolman Committee, which was concerned with declassifying Manhattan Project scientific and technical information. That committee stated the following about classification of scientific information:

It is not the conviction of the Committee that the concealment of scientific information can in any long term contribute to the national security of the United States. . . . If we are looking to the national welfare or national security as they may be two decades from now the Committee has no doubt that the greatest strength in both fields would come from a completely free and open development of science.

Thus the Committee is inclined to the view that there are probably good reasons for keeping close control of much scientific information [related to atomic energy] if it is believed that there is a likelihood of war within the next five or ten years. It is also their view, however, that this would weaken us disastrously for the future—perhaps twenty years hence.59

Benefits of Technology Transfer

Information about certain processes, materials, or devices, which has been developed as part of a classified program, can sometimes be used to develop a new commercial process or to significantly improve an existing industrial process. Not classifying that technology and transferring it to private U.S. industry could help U.S. industrial competitiveness in the world marketplace and thereby help the U.S. economy. The expertise necessary to help evaluate the potential benefit to the U.S. economy by transferring technology to private industry can usually be obtained from consultants familiar with that industry.

Benefits to U.S. Foreign Relations

This category includes benefits to U.S. foreign relations, arms control negotiations, and treaty obligations. Those benefits should be considered in some classification decisions. For example, certain technology might be important to the success of worldwide efforts to detect and deter terrorism involving conventional or chemical weapons or to detect and deter the spread of nuclear, chemical, or biological weapons. Classifying that technology would not benefit those international efforts.

Importance of the Information to Public Discussion and Education

The effective governing of our nation is based on input from the governed to the governors (including voting the governors out of office). If citizens do not know what their government is doing, they can hardly have effective input into this governing process. Therefore, when information concerning our government's activities is classified, then there are costs to free and open discussion of our government's activities and consequently there are costs to the processes of "representative" government.

Generally, it is important to have public input into the government's long-term, strategic decisions. Public discussion of short-term, operational information or of detailed technical matters is of lesser importance.* Therefore, as a general rule, information pertinent to strategic decisions will have greater public-discussion benefits than will operational information or technical details.

Some factors to be considered when evaluating the importance of information to public disclosure and education include the following:

Any Other Significant Benefit to the United States

This benefit criterion is meant to encompass any significant benefit to the United States which was not included in the previous seven criteria for determining benefits of not classifying certain information. Note that this is stated as "any other significant benefit." The comparable risk criterion was stated as "any other national security impact or significance."

One benefit of disclosing information about military capabilities, which might otherwise be classified, is to help deter an adversary. For a deterrent effect, one wants an adversary to know of the existence of a force (weapon) and to give credence to that force level (weapon capability).30 "A truly secret weapon is valueless for the purpose of deterrence."63 It has been reported that the United States deliberately let the U.S.S.R. know some of the details of our nuclear deterrent force so that this force could be an effective deterrent.64

Sometimes, classified information may be deliberately released by the government to "send a message" to an adversary. During the Cuban missile crisis in 1962, President Kennedy released much classified intelligence information about U.S.S.R. missile bases in Cuba. The purpose of this release was to inform the U.S.S.R. that the United States had precise information on the extent of the Soviet threat so that the U.S.S.R. would understand why the United States was taking such strong actions in response to that threat.65

Benefits to the Credibility of the Classification Program

Classification of information must have the enthusiastic support of the persons working in classified programs if that classification is to be effective. For example, if personnel in a classified program believe that certain classified technical information in that program is widely known and used in U.S. or foreign industry (as unclassified information), or that this classified information would be obvious to an expert in the field, then those persons may not enthusiastically support the classification of this information.* That attitude may also diminish those persons' respect for information that truly should be classified, information whose unauthorized disclosure would significantly damage the national security.66, 67 Thus, if information that is common knowledge or of little importance is not classified (or is declassified), then the remaining classified information will probably be better protected by project personnel.+

It should also perhaps be noted that consistency in classification matters is important to maintaining credibility in classification. When classification guidance is inconsistent (or illogical), then respect for classification is lost. Also, it is more difficult for persons working on a classified program to remember what is classified if much of the program is classified and if the classification guidance is illogical or inconsistent.68

DOE's policy on the credibility of its classification program has been stated as follows:

The effectiveness of Department of Energy (DOE) classification policies depends significantly on the willing cooperation of those who must implement the resulting classification guidance. It is important that they have confidence that the information we are continuing to classify is indeed truly sensitive for national security reasons. This confidence is the manifestation of the DOE classification program's credibility.

Erosion of credibility occurs whenever national security sensitivity is diminished through compromise, technological progress in the public sector, or other valid reasons. If declassification is chosen as an appropriate action to pursue under a given set of circumstances, then, as a policy, credibility of the DOE classification program will be one of the factors considered in the risk-benefit analysis that is performed to support each recommendation for declassification of Restricted Data or Formerly Restricted Data.68

A classification program's effectiveness is improved when classification efforts are focused on information that is truly important to national security.

SUMMARY

The criteria for determining the risks and benefits of disclosing scientific or technical information are summarized in Table 5.1. Factors for evaluating those risks and benefits of information disclosure are summarized in Tables 5.2 and 5.3, respectively.

Table 5.1. Summary of criteria for determining risks and benefits of disclosure
of scientific or technical information for use when considering
whether to classify that information
Criterion numbera Description
Criteria for determining risks
RC1 The extent to which the information could assist other nations in developing new armaments
RC2 The extent to which the information could assist other nations in improving their armaments
RC3 The extent to which the information could assist other nations in producing materials for armaments
RC4 The detrimental effects that release of the information could have on U.S. foreign relations, arms control negotiations, or treaty obligations
RC5 Any other national security impact or significance (e.g., the extent to which the information would assist other nations in assessing or countering U.S. capabilities and limitations)
RC6 The detrimental effects that release of the information could have on the credibility of the classification program
Criteria for Determining Benefits
BC1 The benefits to the progress of the U.S. program that could be achieved by release of the information
BC2 The benefits to the U.S. program that could be achieved by eliminating the costs of classifying the information
BC3 The benefits that could be realized by U.S. progress in science and technology, in general, if the information was not classified
BC4 The benefits to the U.S. economy that could be realized by transferring the technology to U.S. industry for commercialization
BC5 The benefits that release of the information could provide to U.S. foreign relations, arms control negotiations, or treaty obligations
BC6 The importance of the information to public discussion or education.
BC7 Any other significant benefit to the U.S.
BC8 The benefits that not classifying the information could have on the credibility of the classification program
aRC = risk criterion; BC = benefit criterion.


Table 5.2. Summary of some evaluation factors to use when determining

the risks of disclosing scientific or technical information

Factor numbera Description of evaluation factors
Factors relevant to assisting other nations with respect to armaments
REF1 Disclosure indicates technology areas that could provide significant national security advantages
REF2 Disclosure indicates what can be achieved in a technology area
REF3 Disclosure indicates the principle by which a technical result can be achieved
REF4 Disclosure indicates technology areas where our knowledge is weak
REF5 Disclosure indicates technology areas where our knowledge is strong
REF6 Disclosure identifies scientific or technical problems that are not readily apparent
REF7 Disclosure identifies successful methods of solving scientific or technical problems
REF8 Disclosure identifies key operating parameters for a process
REF9 Disclosure provides detailed know-how about processes or equipment
REF10 Disclosure could save an adversary a significant effort if otherwise the adversary would have to get the information by its own technological efforts
REF11 Disclosure could significantly reduce the lead time that the United States has over an adversary in that technology
REF12 Disclosure could provide independent confirmation of data already obtained by adversary
Factors relevant to foreign relations
REF13 Disclosure could lead to a perception that the United States is not totally supporting international arms control efforts
Factors relevant to other national security impacts
REF14 Disclosure could help an adversary in assessing or countering U.S. capabilities or limitations
Factors relevant to classification program credibility
REF15 Disclosure could lead to a perception that program management is releasing, for unsound reasons, information important to the nation's security
REF16 Protection of the information may be impractical and its classification may adversely affect the classification program's credibility by such an attempt to impose security controls in impractical situations
aREF = risk evaluation factor.


Table 5.3. Summary of some evaluation factors to use when determining

the benefits of not classifying scientific or technical information
Factor numbera Description of evaluation factors
Factors relevant to the specific program
BEF1 Reduced costs of inefficient communications between program personnel
BEF2 Elimination of costs of delays while program personnel await security clearances
BEF3 Elimination of costs of inadequate program management because of inadequate peer review of the scientific or technical efforts
BEF4 Increased productivity because ability to publish results and discuss results with peers will lead to improved motivation of scientists and engineers
BEF5 Improved ability to recruit top-notch scientists and engineers because of fewer restrictions on discussions with peers
BEF6 Reduced delays caused by lengthy times required for classified procurements
BEF7 Reduced costs of hardware because it can be fabricated as unclassified
BEF8 Reduced costs of hardware because it can be procured as unclassified
BEF9 Reduced costs of hardware because of more vendor competition in procurements
BEF10 Reduced costs of document preparation
BEF11 Reduced transportation costs for documents or materials
BEF12 Reduced document storage costs
BEF13 Elimination of costs of reviewing documents for declassification or downgrading.
BEF14 Elimination of costs of destroying or otherwise disposing of classified documents, materials, or hardware
BEF15 Elimination of pro rata share of classification office and security department staff costs
BEF16 Elimination of pro rata share of costs of protecting facilities
BEF17 Elimination of personnel security costs
BEF18 Elimination of employee time spent in classification and security briefings not directly related to production
Factors relevant to general progress in science and technology
BEF19 Improved progress in the scientific or technical field because information is available to all workers in that field
BEF20 Improved progress in related fields because information can contribute to cross-flow of ideas and information to those related fields
Factors relevant to technology transfer to industry for commercialization
BEF21 Improved U.S. economy because new and better products are available
BEF22 Improved U.S. economy because improved products are available
Factors relevant to foreign relations
BEF23 Information could benefit international efforts in control of terrorism or the spread of nuclear, chemical, or biological weapons of war
Factors relevant to public discussion and education
BEF24 Information is relevant to policy, not operational or scientific and technical details
BEF25 Information is relevant to performance or integrity of governmental officials
BEF26 Information is relevant to significant governmental expenditures
BEF27 Information is relevant to a current issue of national importance
Factors relevant to any other significant benefit to the United States
BEF28 Information provides other significant benefits to the United States which are not included in the previously stated factors for determining benefits of not classifying information
Factors relevant to classification program credibility
BEF29 Improved attitudes of program personnel in protecting classified information
aBEF = benefit evaluation factor.

REFERENCES

1. Executive Order 12356, Fed. Reg., 47 14874 (Apr. 6, 1982), Preamble; §1.1(a)(3). Hereafter cited as "EO 12356."

2. A. S. Quist, "Some Adverse Impacts of Classification," Chapter 6 in Security Classification of Information. Volume 1. Introduction, History, and Adverse Impacts, K/CG-1077/V1, Martin Marietta Energy Systems, Inc., Oak Ridge, Tenn., 1989. Hereafter cited as "Quist."

3. "The National Security Interest and Civil Liberties," Harvard L. Rev., 85, 1130–1326 (1972), p. 1133.

4. T. I. Emerson, "National Security and Civil Liberties," Yale J. World Pub. Order, 9, 78–112 (1982), pp. 78–79. Hereafter cited as "Emerson."

5. Gen. Maxwell Taylor, "The Legitimate Claims of National Security," Foreign Affairs, April 1974, p. 57, as quoted by M. M. Cheh, "Government Control of Private Ideas—Striking a Balance Between Scientific Freedom and National Security," Jurimetrics Journal, 23, 1–32 (1982), p. 16, note 106.

6. New York Times v. United States, 403 U.S. 713, 729 (1971) (J. Stewart, concurring).

7. Cole v. Young, 351 U.S. 536, 543 (1956).

8. Cole v. Young, 351 U.S. 536, 544 (1956).

9. EO 12356, §6.1(e).

10. 5 U.S.C. §552(b)(1)(A).

11. 5 U.S.C. §552(b)(7)(D).

12. Conference Report 93-1200, "Joint Explanatory Statement of the Committee of Conference," reprinted in U.S. Code Cong. & Adm. News, 6285–6293 (1974), p. 6291.

13. Gorin v. United States, 312 U.S. 19, 28 (1940).

14. H. Edgar and B. C. Schmidt, Jr., "The Espionage Statutes and Publication of Defense Information," Columbia Law Rev., 73, 929–1087 (1973), pp. 969–974.

15. United States v. Heine, 151 F. 2d 813, 815–816 (2nd Cir., 1945).

16. United States v. Morrison, 844 F.2d 1057, 1071 (4th Cir., 1988); cert. denied., 108 S. Ct. 1837 (1988).

17. Gorin v. United States, 312 U.S. 19, 31 (1940), citing, with approval, the following jury instructions: "The connection must not be a strained one nor an arbitrary one. The relationship must be reasonable and direct."

18. EO 12356, Preamble; §1.1(c), §6.1(c).

19. EO 12356, §1.3(c).

20. EO 12356, §1.6(a).

21. J. C. Tuck, Assistant Secretary for Defense Programs, U.S. Department of Energy, memorandum to Operations Office Managers, "Department of Energy Classification Procedures," Tab 1, May 1, 1990. Hereafter cited as "Tuck Memorandum, Tab 1." A. B. Siebert, Director, Office of Classification, U.S. Department of Energy, Attachment to memorandum to Distribution, "Revised Department of Energy Declassification Procedures," March 12, 1992.

22. W. P. Raney, "The Sea Lanes & Their Challenges," J. Natl. Class. Mgmt. Soc., 13, 17–22 (1977), p. 21.

23. U.S. Department of Defense, Department of Defense Handbook for Writing Security Classification Guidance, DOD 5200.1-H, March 1986, §2.2(a)(6).

24. Quist, p. 63, footnote.

25. S. J. Lukasik, "Remarks, Workshop A—Lifetime Cycles for Security Classification," J. Natl. Class. Mgmt. Soc., 7, 56–58 (1971), p. 58. Hereafter cited as "Lukasik."

26. See also T. S. Church, "Panel—Science and Technology, and Classification Management," J. Natl. Class. Mgmt. Soc., 2, 39–43 (1966), p. 43.

27. W. Gelhorn, Security, Loyalty, and Science, Cornell University Press, Ithaca, N.Y., 1950, p. 63. Hereafter cited as "Gelhorn."

28. U.S. Department of Defense, Report of the Defense Science Board Task Force on Secrecy, F. Seitz, Chairman, Office of the Director of Defense Research and Engineering, Washington, D.C., July 1, 1970, pp. iv, v, 1, 2, 10. Hereafter cited as the "Seitz Report."

29. U.S. Department of Defense, An Analysis of Export Control of U.S. Technology—A DoD Perspective, A Report of the Defense Science Board Task Force on Export of U.S. Technology, J. F. Bucy, Chairman, Office of the Director of Defense Research and Engineering, Feb. 4, 1976, pp. iii, v, xiii, 1, 3, 34. Hereafter cited as the "Bucy Report."

30. F. J. Thomas, "Remarks, Workshop A—Lifetime Cycles for Security Classification," J. Natl. Class. Mgmt. Soc., 7, 58–62 (1971), p. 61. Hereafter cited as "Thomas."

31. Seitz Report, p. 2.

32. Lukasik, p. 57.

33. Bucy Report, pp. 9–14.

34. Bucy Report, p. 12.

35. J. H. Kahan, "Remarks, Workshop A—Lifetime Cycles for Security Classification," J. Natl. Class. Mgmt. Soc., 7, 52–56 (1971), p. 55.

36. T. S. Church, "Panel—Science and Technology, and Classification Management," J. Natl. Class. Mgmt. Soc., 2, 39–43 (1966), pp. 42–43.

37. H. G. Maines, "Panel—Government Classification Management Policies and Programs," J. Natl. Class. Mgmt. Soc., 2, 82–87 (1966), p. 84.

38. J. G. Beckerley, Director of Classification, Atomic Energy Commission, in Atomic Power Development and Private Enterprise, Hearings before the Joint Committee on Atomic Energy, Congress of the United States, 83rd Cong., 1st Sess., on Atomic Power Development and Private Enterprise, June and July 1953, U.S. Govt. Printing Office, June 24, 25, and 29; July 1, 6, 9, 13, 15, 16, 20, 22, 23, 27, and 31, 1953, pp. 38–39.

39. U.S. Atomic Energy Commission, Wash., D.C., internal document, 1974.

40. H. D. Smyth, A General Account of the Development of Methods of Using Atomic Energy for Military Purposes Under the Auspices of the United States Government, 1940–1945, U.S. Govt. Printing Office, August 1945.

41. V. C. Jones, Manhattan: The Army and the Atomic Bomb, United States Army in World War II, Special Studies, Center of Military History, U.S. Army, Washington, D.C., 1985, p. 559, referring to a May 21, 1945, letter from Gen. L. R. Groves to H. D. Smyth.

42. D. Charles, "In the Beginning Was Uranium . . . ," New Scientist, 136(1844), 30–35 (Oct. 24, 1992), p. 35.

43. S. Fernbach, "Panel—Science and Technology, and Classification Management," J. Natl. Class. Mgmt. Soc., 2, 48–53 (1966), p. 50.

44. U.S. Department of Defense, Security Classification of Official Information, Instruction 5210.47, Chap. VI, Sect. D (Dec. 31, 1964).

45. Tuck Memorandum, Tab 1.

46. U.S. Department of Energy, Washington, D.C., internal document, 1987.

47. J. A. Buckland, "Remarks, Workshop B—Security Costing in Relation to Classification," J. Natl. Class. Mgmt. Soc., 7, 77–98 (1971).

48. A. A. Correia, "Panel—Industrial Aspects of Classification Management in West Coast Defense/ Aerospace Industry," J. Natl. Class. Mgmt. Soc., 3(2), 134–138 (1967), pp. 137–138.

49. See, for example, E. J. Castro, "Generation of a Classified Document Cost Study," J. Natl. Class. Mgmt. Soc., 8, 111–114 (1972), which describes a methodology for estimating the direct and indirect costs for generating and handling classified documents.

50. E. H. Calvert, "Panel—Realizing Savings From Classification Management," J. Natl. Class. Mgmt. Soc., 1(2, 3, 4), 108–112 (1965).

51. J. F. Pellant, "Remarks, Workshop B—Security Costing in Relation to Classification," J. Natl. Class. Mgmt. Soc., 7, 73–77 (1971).

52. Comments by a U.S. Navy Commander during a panel discussion on costs of classification, as reported in J. Natl. Class. Mgmt. Soc., 1(2, 3, 4), 115 (1965).

53. R. D. Donovan, "Classification Management Cost Study," J. Natl. Class. Mgmt. Soc., 4(2), 72–76 (1968).

54. See also, for example, A. F. van Cook, "Remarks, Workshop B—Security Costing in Relation to Classification," J. Natl. Class. Mgmt. Soc., 7, 98–102 (1971), p. 99.

55. M. D. Aitken, "Panel—Classification Management in the Department of Defense Today," J. Natl. Class. Mgmt. Soc., 3(2), 75–80 (1967), p. 78.

56. D. O. Cooke, testifying for the Department of Defense, as reported in U.S. Government Information Policies and Practices—The Pentagon Papers (Part 2), Hearings Before a Subcommittee of the Committee on Government Operations, House of Representatives, 92nd Congress, 1st Session, June 28 and 29, 1971, U.S. Govt. Printing Office, 1971, pp. 601, 659.

57. F. Horton, Congressman from New York, in U.S. Government Information Policies and Practices—The Pentagon Papers (Part 2), Hearings Before a Subcommittee of the Committee on Government Operations, House of Representatives, 92nd Congress, 1st Session, June 28 and 29, 1971, U.S. Govt. Printing Office, 1971, p. 659.

58. Gellhorn, p. 35.

59. R. C. Tolman, R. F. Bacher, A. H. Compton, E. O. Lawrence, J. R. Oppenheimer, F. H. Spedding, and H. C. Urey, Report of Committee on Declassification, Memorandum to Maj. Gen. L. R. Groves, Nov. 17, 1945, p. 3.

60. Adapted from Emerson, pp. 92–93, 97.

61. J. A. Goldston, J. M. Granholm, and R. J. Robinson, "A Nation Less Secure: Diminished Public Access to Information," Harvard Civil Rights–Civil Liberties Law Review, 21(2), 409–491 (1986), p. 458.

62. Emerson, pp. 93, 97.

63. A. H. Katz, "Classification: System or Security Blanket?," J. Natl. Class. Mgmt. Soc., 8, 76–82 (1972), p. 80.

64. W. J. Coughlin, statement to the Foreign Operations and Government Information Subcommittee, as reported in Government Information Plans and Policies, Part 1, Hearings Before a Subcommittee of the Committee on Government Operations, House of Representatives, 88th Congress, 1st Session, March 19, 1963, U.S. Govt. Printing Office, Washington, D.C., 1963, p. 35.

65. U.S. v. Sims, 471 U.S. 159, 180 (note 24) (1985).

66. See, for example, W. J. Howard, "Panel—The Executive Views Classification Management," J. Natl. Class. Mgmt. Soc., 1(2, 3, 4), 73–76 (1965), p. 73; R. C. Arnold, "Panel—Classification in the Department of Defense Today," J. Natl. Class. Mgmt. Soc., 3(2), 83–87 (1967), p. 85.

67. F. M. Kaiser, "The Impact of Overclassification on Personnel and Information Security," Government Information Quarterly, 3(3), 251–269 (1986), p. 253.

68. R. G. Jordan, Union Carbide Corporation, Nuclear Division, as reported in an internal document of the U.S. Atomic Energy Commission, Washington, D.C., 1965.

69. U.S. Department of Energy, Washington, D.C., internal document, 1985.

On to Chapter Six.




FAS | Government Secrecy | Library || Index | Search | Join FAS