Security Classification of Information: Table of ContentsOn to Chapter Three.
INTRODUCTION Chapter 2.
MAJOR TYPES OF CLASSIFIED INFORMATION--
SUBJECTIVE AND OBJECTIVE SECRETSBefore starting a detailed discussion of classification principles, it is useful to describe the characteristics of the two major broad types of information that a government may want to classify—subjective and objective information. So far as the author knows, those two types of classified information were first described (as "operational" information and "scientific" information) in some detail by L. N. Ridenour 1 in 1945. Ridenour also mentioned a third type of information that governments classify—"technical" information. However, technical information and scientific information are considered, in this document, to be subsets of objective information.
The following two sections of this chapter describe the characteristics of subjective and objective secrets. The third section discusses why technical information is included, along with scientific information, in the category of objective secrets. The last section of this chapter discusses why trade secret law includes only objective secrets and does not encompass subjective secrets.
SUBJECTIVE SECRETS
One major type of information that a government sometimes classifies is unique to the government in the sense that the government has decided to conduct its affairs in a certain way. So long as the government effectively controls and protects the information about that decision, that information cannot be independently discovered by an adversary. Examples of unique (subjective) information are a government's specific military plans to invade another country (e.g., the time and place of the invasion) or a government's foreign policy about a particular international situation. When subjective information is originated, an adversary does not have that information. The adversary cannot independently generate that information. Subjective information, if properly controlled and protected, cannot be obtained by another country except through espionage or through unauthorized (either inadvertent or deliberate) disclosures of that information. This type of information has also been characterized as a "true secret." 2 L. N. Ridenour designated this type of information as "operational" information3 (an operational secret). It is better characterized as subjective information or a subjective secret. This type of information is said to have the following characteristics:1
The term "subjective information" will be used throughout this document to describe this type of information.
1. Compactness. The secret can be told in only a few words (e.g., the time and place of an invasion). This feature makes the secret easy to steal and easy to transmit.
2. Almost universally understandable (special expertise is not required to understand the secret). Therefore, anyone can steal the secret.
3. Arbitrary. An adversary needs to steal it to obtain it. The secret cannot be independently discovered.
4. Subject to change. The secret can be changed at (nearly) the last moment. If a nation learns that an adversary has discovered the time and place of an invasion, that time and place can be changed.
5. Perishable over the short term (short-lived as a secret). When an invasion begins, the adversary learns the secret. The secret has been "made public." This also means that the secret needs to be kept for only a limited time—the stringent security measures required are not permanent.
OBJECTIVE SECRETS
Another type of information that a government may want to classify is information which, even though discovered, developed, or controlled by that government, may be already known to or can be independently discoverable by another country. Scientific information (e.g., "scientific secrets"3, 4) is included in this type of information.* For example, the number of neutrons produced when a U-235 nucleus fissions, which is an important parameter that must be known when evaluating whether an atomic bomb is possible, can be discovered by competent scientists who have the proper materials and equipment. This type of information cannot be absolutely controlled. Scientific information is a fact of nature, not a secret.+
* A good illustration of the difference in a "true" secret (operational secret) and a "scientific" secret was given by L. N. Ridenour during his appearance before a special Senate committee on atomic energy in February 1946. He used the following examples to distinguish between those two types of secrets:
Scientists in other countries can independently make the same or similar scientific discoveries through their own efforts. This type of information may also be termed "objective" information or an objective secret. The characteristics of this type of information are said to be as follows:5The use of the word "secret" for the results of scientific investigation or the findings of engineering is genuinely misleading. Let me illustrate what I mean. If I say, "I am thinking of a number, but I shall keep it a secret," I have used the word "secret" in its usual sense. Apart from the possibility of my telling you the number, you have no way of knowing it or finding out what it is. On the other hand, if I say, "I know the critical mass of U-235 necessary to make a bomb, and I intend to keep it a secret," I am using the word secret in an entirely different sense. I am saying to you, not that you cannot find out what I know, but that you must find it out for yourself, without my help. This may cause you to become annoyed with me, but it cannot keep you in ignorance [L. N. Ridenour, in "Atomic Energy," Hearings Before the Special Committee on Atomic Energy, United States Senate, 79th Congress, 2nd Sess., Part 5, Feb. 15, 1946, pp. 536–537. Essentially the same information appears in the following article: L. N. Ridenour, "Secrecy in Science," Bull. At. Sci., 1(6), 3 (Mar. 1, 1946)].
+ "Secrets are made by men, not by nature" [L. W. Nordheim, "Fear and Information," Bull. At. Sci., 10, 342–346 (1954), p. 346].
1. Diffuse. Usually there are no "magic" formulas. Lengthy reports are generally required to describe scientific information. Therefore, "scientific secrets" generally are not easy to transmit.
The term "objective information" will be used throughout this document to describe this type of information.2. Usually understandable only by other scientists.
3. Not arbitrary. A "fact of nature." Others can get the answer (if they ask the right question).
4. Not subject to change. Eternal. A fact of nature has only one value.
5. Can be long-lived as a secret. Others can independently discover the information, but such discovery may require much effort. Therefore, this information may be kept secret for a relatively long period of time.
TECHNICAL INFORMATION AS AN OBJECTIVE SECRET
A third type of information does not exactly fit into either the subjective (operational) or objective (scientific) categories of secrets. That third type is technical information.5 It concerns, for example, the technical design and performance of new weapons (as contrasted to the scientific basis for their design).1 Technical information has characteristics similar to scientific information in that it is diffuse and usually completely understandable only by other technical people. However, unlike scientific information, technical information is not a fact of nature. It is a method, process, technique, or device that is employed to create a product from a fact of nature. A fact of nature (scientific information) may be exploited by several methods (the details of which are technical information) to accomplish essentially the same objective. Thus, there are usually several technical solutions to the problem of how to use a fact of nature to achieve a certain result. Another difference between scientific information and technical information is that technical information is subject to change (improvements can always be made to a technical process). However, like scientific information, technical information is eternal—if the same technical procedure is always followed, then future results will be the same as those achieved in the past.*
* Ridenour's 1945 article [L. N. Ridenour, "Military Security & the Atomic Bomb," Fortune, 32, 170–223 (November 1945)] also discussed "security" of operational, scientific, and technical information, some aspects of which are of interest to classification of these types of information. Ridenour was of the opinion that security of operational information was accomplished mainly by concealing this information (i.e., classifying and stringently protecting this information—security by concealment). However, since "scientific secrets" are facts of nature, they cannot be concealed forever from an adversary. Security by concealment was not the long-term solution. For a nation to maintain security in scientific matters (i.e., stay ahead of an adversary), Ridenour believed that "security by achievement" was the answer. That is, he believed that scientific information should not be classified and that the United States should rely on the ability of its scientists and engineers, if they had access to all pertinent information and could freely discuss this information with others, to keep the United States significantly ahead of its adversaries. For technical information (e.g., design of weapons), both security by concealment and security by achievement were applicable ("achievement" to get a new weapon quickly; "concealment" to ensure surprise in its first use and to reduce the application of countermeasures by an adversary).
It was mentioned earlier in the chapter that technical information is used to exploit scientific information—that technology implements facts of nature. There are two parts to the use of technology to create a product (e.g., a weapon system). First, one has to know the technical information that can implement the scientific information. Second, one has to be capable of using that technical information; one has to have the industrial skills (know-how and hardware) and the specific materials to implement the technical information.*, 6 Secrecy is the only way to protect scientific information. Secrecy can also protect technical information. However, other protective measures such as export controls must be used to keep an adversary from acquiring industrial skills and strategic materials.7
With respect to the ability to keep secrets, Ridenour mentioned that it was to some extent easier with operational information than with technical information. Only a few people need to know operational secrets (e.g., staff planners), at least initially. However, hundreds or thousands of persons need to know technical secrets about weapons systems (e.g., the designers, manufacturers, and users of those weapons).
The "security by achievement" policy seems to have been accepted, or strongly considered for acceptance, by the Department of Defense (DoD) in 1971. A new DoD policy was to not classify scientific or technical information for technological areas in which lead time advantage was indispensable and in which the United States had a world wide superiority. The absence of classification was expected to lead to greater interest and faster achievement of U.S. objectives, thereby maintaining U.S. superiority in that area [G. MacLain, "Updating Security Classification Management and its Support by DCAS Organization," J. Natl. Class. Mgmt. Soc., 7, 9–14, (1971) pp. 10–11].* The protection of scientific and technical information can be thought of as a three-tiered process: (1) protect the scientific secret, (2) protect the technical information that exploits the scientific secret, and (3) protect the industrial skills and strategic materials that are necessary to exploit the technical information. This three-tiered process is consistent with the three classes of secrets concerning atomic energy information and the Manhattan Project that were described by General L. R. Groves in 1945 [U.S. Congress, House of Representatives, Committee on Military Affairs, "Atomic Energy," Hearings on HR 4280, An Act for the Development and Control of Atomic Energy, Oct. 9 and 18, 1945, 79th Congress, 1st Sess., 1945, Testimony of Gen. L. R. Groves at pp. 12–13]:
Although three specific types of information that a government may want to classify have been mentioned—subjective, objective, and technical—it seems that an optimum categorization would be to specify only two major types, subjective and objective, with scientific and technical information being subsets of objective information. Scientific and technical information share many characteristics with respect to classification matters, and both differ significantly from subjective (operational) information. Therefore, in this document, scientific and technical information will generally be considered, for classification purposes, as a single type of information.The secrets, as they are loosely termed in the public discussion, are divided properly, I think, into about three classes. One class of these secrets consists of established scientific facts which were not secret at all. . . .
The second classification of secrets is the scientific developments which went beyond this, and most of those developments were not basic. They were made up of hundreds and hundreds of problems that had to be solved before this work could be done.
They involved things such as the purification of metals and the purification of other products and the handling of certain products that are not easy to handle safely. They involved all the research work, principally that which was done in many laboratories by thousands of scientists. There are other ways to do this. We hit one way. Probably there are several ways in which you can do those various items, but they require a tremendous amount of work.
This could undoubtedly be achieved by other nations if they spend the money, the labor, and the time and have the scientific organization with which to do it. In time they can, of course, do it. It is merely a case of relative speed. . . .
The other class of secret, which is the biggest field, is the ingenuity and the skill of the American worker and the American management, both the top management and the junior management, and that is a secret that I do not think any other nation has, and I do not think anyone is going to have it in a hurry.In subsequent chapters, much more attention will be paid to classification of objective information (e.g., scientific or technical information) than to classification of subjective information (e.g., military operational information, foreign relations information, etc.). Scientific or technical information generally is information of a more permanent nature, is not subject to rapid change, and has lasting value, as compared with operational information which usually is fluid, is subject to rapid change, and has transient value.
SUBJECTIVE SECRETS, OBJECTIVE SECRETS, AND TRADE SECRETS
Although there are two major types of classified information, subjective and objective secrets, trade secret law recognizes only one type of trade secrets—objective secrets (see Appendix A). Trade secrets include information about manufacturing processes, "recipes" for products (e.g., the formula for Coca-Cola®), and other objective information which could be discovered independently by another business. Many trade secrets are analogous to the scientific or technical information (especially technical information) that governments sometimes classify.
Subjective secrets are not generally protected by trade secret laws. Although a government will classify and protect the date on which it plans to invade another country (subjective information), a business cannot receive trade secret protection for the date on which it plans to introduce a new product. The Restatement of the Law of Torts definition of a trade secret specifically excludes this type of information:
[A trade secret] differs from other secret information in a business (see §759) in that it is not simply information as to single or ephemeral events in the conduct of business, as, for example, the amount or other terms of a secret bid for a contract or the salary of certain employees, or the security investments made or contemplated, or the date fixed for the announcement of a new policy or for bringing out a new model or the like.8One reason for the difference between trade secret law and classification rules with respect to the treatment of "sensitive" subjective information may be that businesses can effectively protect subjective secrets. Those subjective secrets usually need to be kept secret for only a relatively short time period—they relate to "single or ephemeral events"—and need to be known only by a few people. Businesses can usually keep this subjective information a secret by their own intensive information protection efforts. In contrast, the trade secrets of a business usually must be kept secret for a relatively long time period—the production processes to which they relate are used for many years—and those trade secrets eventually become known to many persons. Because trade secrets usually must be kept secret for many years and those secrets become known to many persons, those trade secrets are more vulnerable to theft. Also, there is generally a greater economic incentive for thievery of those objective secrets, as compared with that of subjective secrets.* Therefore, objective secrets need the extra protection afforded by trade secret law so that businesses depending on those trade secrets are not harmed by unscrupulous competitors who steal those secrets. Trade secret law remedies this thievery by issuing injunctions that prohibit the thief from using the trade secret or by awarding monetary damages to the injured business.
* One article states that "the only rational basis for distinguishing between trade secret information and other confidential business information is that the latter does not involve effort, expense, or innovation" [S. J. Davidson and P. L. DeMay, "Application of Trade Secret Law to New Technology—Unwinding the Tangled Web," William Mitchell L. Rev., 12, 579–621 (1986), p. 583].
The matter of monetary damages suggests another reason why trade secret law protects objective secrets but not subjective secrets. The time and effort required to develop a trade secret can be reasonably easily quantified in monetary terms, as can its current economic (market) value. Therefore, it is relatively easy to quantify monetary damages in trade secret litigation. In contrast, it is generally relatively hard to quantify the damages of unauthorized disclosures of subjective secrets such as the date of release of a new product. Legal systems generally prefer not to get involved in such ill-defined business matters, which may help explain why trade secret law does not protect subjective secrets.For governments, drastic adverse consequences can arise from the loss of either subjective secrets (invasion dates) or objective secrets (information on new weapons systems). Therefore, governments are very interested in fully protecting both types of secrets. This protection is aided by espionage statutes and other laws.
REFERENCES
1. L. N. Ridenour, "Military Security & the Atomic Bomb," Fortune, 32, 170–171, 216, 218, 221, 223 (November 1945), p. 171. Hereafter cited as "Ridenour, 1945."
2. W. Gellhorn, Security, Loyalty, and Science, Cornell University Press, Ithaca, New York, 1950, p. 9. Hereafter cited as "Gellhorn."
3. Ridenour, 1945.
4. Gellhorn, p. 11.
5. Ridenour, 1945, pp. 171, 216.
6. A. DeVolpi, G. E. Marsh, T. A. Postal, and G. S. Stanford, Born Secret: The H-Bomb, The Progressive Case, and National Security, Pergamon Press Inc., Elmsford, New York, 1981, p. 144. Hereafter cited as Born Secret.
7. Born Secret, p. 145.
8. Restatement of the Law of Torts, 4, American Law Institute, St. Paul, Minn., §757, Comment (b) (1939).