Trade Secrets, Confidential Information, and
the Criminal Law
John T. Cross*
The author examines the extent to which prop-
erty offences in the criminal law can be used to
police the misappropriation of trade secrets
and confidential information. After assessing
the long-standing debate on whether informa-
tion can be classified as property, he argues
that answering the question one way or the
other involves circular reasoning. When
judges label information “property,” it is to
enable them to grant the desired remedies.
Courts should instead ask more directly
whether certain information should be pro-
tected under the circumstances. It follows that
precedents holding that certain information is
property in one area of the law should not be
authoritative in others. The article then
explores efforts made in Great Britain, Canada
and the United States to apply criminal prop-
erty offences to industrial espionage. Since
these provisions presume the misappropriation
or damage to tangible property, they are
unsuitable to cases involving information. The
author then suggests an alternative approach
based upon the different types of values inher-
ent in confidential information: “use value”
and “monopoly value.” Unfortunately, theft
and vandalism provisions in existing criminal
statutes protect only “use value,” although it is
the monopoly on information that is its pri-
mary source of value to the owner. The author
concludes that legislation is required to define
the mens rea and actus reus requirement in
terms designed to protect the unique value of
confidential information.
L’auteur 6value le potentiel qu’a le droit crimi-
nel de pr6venir l’appropriation malhonnete de
l’information confidentielle et des secrets
commerciaux en ]a qualifiant d’atteinte au
droit de proprit6. L’auteur expose le long
drbat sur la question a savoir si l’information
peut etre l’objet d’un droit de proprirtd; il con-
clut que dans un cas comme dans l’autre la
raponse implique un raisonnement circulaire.
L’attribution du terme << proprirt6 >> par les
juges depend du rsultat qu’ils veulent obtenir.
Les tribunaux devraient plutrt centrer leur rai-
sonnement sur l’importance de prot~ger ou
non l’information en question dans les circons-
tances, sans se sentir lids par la jurisprudence
ant~rieure qui aurait caractdris6 autrement ce
meme type d’information dans un autre con-
texte juridique. L’auteur examine les decisions
portant sur l’espionnage industriel rendues en
Grande-Bretagne, au Canada et aux ltats-
Unis, pour conclure que le droit criminel
actuel est difficilement applicable parce que
l’atteinte criminelle au droit de propri~td pr6-
suppose qu’il porte sur une chose tangible dont
on veut protrger la libre utilisation par son
propridtaire. Dans le cas de l’information, il
s’agit de protrger non pas seulement l’utilisa-
tion que peut en faire son propridtaire, mais
6galement le monopole qu’il a sur l’inforna-
tion. C’est en effet le monopole qui conftre h
l’information sa valeur et c’est en fonction du
monopole qu’il faudrait, selon l’auteur, repen-
ser le droit criminel pour accorder une meil-
leure protection 4 l’information confidentielle.
* Assistant Professor of Law, University of Louisville School of Law. B.S., Bradley University,
1981; J.D., University of Illinois, 1984.
McGill Law Journal 1991
Revue de droit de McGill
1991]
TRADE SECRETS
Synopsis
Introduction
I.
Applying Property-Based Criminal Laws to the Misappropriation of
Information
A.
B. Specific Applications of Criminal Offences
Information as “Property”
II.
A Value-Based Analysis
A. The Values Inherent in Information
B. The Values Protected by the Criminal Law of Property
1. Theft
2. Other Property Crimes
I. The Future of the Criminal Law and Information
Conclusion
Introduction
If nature has made any one thing less susceptible than others of exclusive prop-
erty, it is the action of the thinking power called an idea …..
Its peculiar char-
acter, too, is that no one possesses the less because every other possesses the
whole of it. He who receives an idea from me receives instruction himself with-
out lessening mine; as he who lights his taper at mine, receives light without
darkening me.
– Thomas Jefferson1
Notwithstanding Jefferson’s view of the nature of ideas, the law has histor-
ically recognized the need to keep certain types of information secret. In early
societies, secrecy was important primarily for reasons of state. Indeed, the Ham-
‘Writings of Thomas Jefferson, vol. 6, by H.A. Washington, ed., (1854) in El. Du Pont de
Nemours and Co. v. United States, 288 F. 2d 904 (1961) at 910.
REVUE DE DROIT DE McGILL
[Vol. 36
murabic Code of 2100 B.C. –
generally considered the world’s oldest written
code of laws –
contained a provision punishing those who were caught prying
into “forbidden” secrets.2 As manufacturing and trade increased in importance,
secrecy also proved to be a valuable asset in the world of commerce. Manufac-
turers and merchants quickly realized that information that was hidden from
competitors could provide a distinct advantage in manufacturing or commerce.3
Those without the information, on the other hand, would naturally use any
means at their disposal to obtain it. Because of the commercial advantages
afforded by secret information, most legal systems of the time developed means
to preserve secrecy.4
Today, we are in a period commonly referred to as the “Age of Informa-
tion.” Tremendous advancements in technology have made information vitally
important to the workings of many commercial enterprises. Without the neces-
sary “know-how” 5 to operate modem manufacturing and communications
equipment, a business would be hard-pressed to survive.6 Managing these vast
quantities of information has consumed an increasing amount of energy. This
trend is even reflected in the makeup of the labour force. By 1976 the number
of people working in information-related industries in the United States sur-
passed the number of workers in traditional occupations.7
2D.D. Fetterly, “Historical Perspectives on Criminal Laws Relating to the Theft of Trade
Secrets” (1970) 25 Bus. Law. 1535 at 1535 n. 3. The sanction for the offence was the loss of an
eye.
31n addition to its other functions, one purpose of the medieval system of guilds was to preserve
secret business information (Institute of Law Research and Reform, Edmonton, Alberta, Trade
Secrets (Report No. 46) (Edmonton: University of Alberta, 1986) at 39 [hereinafter Trade Secrets
Report]).
4The law of the Roman Empire afforded a cause of action to a slaveholder whose secrets had
been uncovered by bribery or other wrongful inducement of his slaves (A.A. Schiller, “Trade
Secrets and the Roman Law: The Actio Servi Corruptit” (1930) 30 Columbia L. Rev. 837). This
law addressed a serious problem of the time. Historians have documented instances of industrial
espionage during the period of the Roman Empire (Trade Secrets Report, ibid. at 39).
The modem Anglo-American law of confidential information dates from the mid-nineteenth
century. It is generally considered that the first Anglo-American case to recognize a cause of action
for misappropriation of valuable commercial information is Morison v. Moat (1851), 9 Hare 241,
68 Eng. Rep. 492 [hereinafter Morison].
5The term “know-how” has become a term of art in the jurisprudence of intellectual property.
It is generally used to refer to the skills acquired by a worker in the course of his employment.
See generally S. Lowry, “Inevitable Disclosure Trade Secret Disputes: Dissolutions of Concurrent
Property Interests” (1988) 40 Stanford L. Rev. 519.
6Indeed, information is one of the most valuable assets of many modem manufacturing and com-
mercial firms. See R.G. Hammond, “The Misappropriation of Commercial Information in the
Computer Age” (1986) 64 Can. Bar Rev. 342 at 344.
7Comment, “The Coming Jurisprudence of the Information Age: Examinations of Three Past
Socioeconomic Ages Suggest the Future” (1984) 21 San Diego L. Rev. 1077 at 1099. It has been
estimated that the production of information and information-processing systems is one of the three
1991]
TRADE SECRETS
As information has grown in importance, so have the benefits of secrecy.
Because information plays such a crucial role in a technology-driven society,
parties with exclusive knowledge of valuable information enjoy a tremendous
advantage over their competitors. Nevertheless, the development of technology
has also created a dilemma: although technology creates greater incentives to
keep information secret, it also makes secrecy more difficult to maintain. Tech-
nology itself has produced a number of new devices which make the task of
espionage much easier.8 The increased reliance on computers has been a major
factor in this phenomenon, since computers are quite accessible to the industrial
spy.9 As a result of these factors, the level of industrial espionage has risen
sharply over the past few decades, generating an estimated annual cost to busi-
nesses in the United States alone of as much as twenty billion dollars.’I
Historically, most modem legal systems have provided a civil remedy for
the wrongful taking of secret information. While societies have not hesitated to
resort to criminal penalties when government secrets are unearthed or revealed,
they have proven less willing to use the criminal law to police private industrial
espionage.”
The past 25 years have seen a reversal in this historical trend. The increase
in espionage has given rise to a widely held perception that civil remedies, by
or four most important sectors of both the Canadian and United States economies. See Hammond,
ibid. at 347.
8The substantial increase in industrial espionage is not, of course, attributable entirely to
advances in technology. Among the other significant factors is the increase in employee mobility
among competitive firms. See Trade Secrets Report, supra, note 3 at 39-41.
9See Note, “Criminal Liability for the Misappropriation of Computer Software Trade Secrets”
(1986) 63 U. Det. L. Rev. 481 at 482 [hereinafter “Criminal Liability”]. I will use the terms “indus-
trial espionage” and “industrial spy” to refer to the act and actor, respectively, involved in the
wrongful acquisition of the confidential information of another. These terms are somewhat mis-
leading because this article is not limited to purely commercial information. See, for example, the
case discussed in the text accompanying notes 75-81. These terms will nevertheless be used for
want of a more exact alternative.
Although not limited to purely commercial information, this article will not discuss confidential
information of an exclusively personal nature. The law’s treatment of personal secrets involves a
number of additional issues, not least of which is the evolving right of “privacy.” Because these
additional issues complicate the analysis, certain theories set forth in this article may not apply to
such personal secrets.
10R. Eells & P. Nehemkis, Corporate Intelligence and Espionage (New York: MacMillan Pub-
lishing House, 1984) at 118. See also B.J. George Jr., “Contemporary Legislation Governing Com-
puter Crimes” (1985) 21 Crim. L. Bull. 389 at 391, which discusses the cost to society of computer
espionage.
“See D.M. Zupanec, “Criminal Liability for Misappropriation of Trade Secret” 84 A.L.R. (3d)
967 (1978) at 971-72. Of course, the industrial spy may also commit an independent crime-such
as breaking and entering or trespass –
in the process of obtaining the confidential information.
Nothing in this article is meant to suggest that these independent offences are unavailable in cases
of industrial espionage. Because the penalties are usually quite small, however, these independent
offenses may not provide the necessary deterrent in many cases of industrial espionage.
McGILL LAW JOURNAL
[Vol. 36
themselves, do not provide a sufficient deterrent to industrial espionage. 2
Because of this perception, a number of legal systems have experimented with
the criminal law to provide a deterrent. Different systems have used different
approaches.
Attempts to apply criminal law in Anglo-American countries have met
with only limited success. Because the criminal statutes of most of these juris-
dictions do not deal specifically with the misappropriation of information, pros-
ecutors have been forced to fall back on other provisions of the criminal law.
These prosecutors generally have relied on traditional property crimes, such as
larceny and fraud. Unfortunately, information and knowledge do not coincide
neatly with the law’s traditional focus on tangible property. Because of these
conceptual problems, some courts have refused to extend the criminal law’s
property offences to the realm of information.
The purpose of this article is to explore the extent to which the criminal
law of property can be used to police the misappropriation of confidential infor-
mation. In discussing this issue, the long-standing debate concerning the extent
to which information can be classified as property will be examined. The first
part of this article will attempt to demonstrate that this debate is unhelpful in
resolving the fundamental question. The article then explores the efforts that
have been made in Great Britain, Canada and the United States to apply prop-
erty offences to industrial espionage. It will be argued that these offences have
certain features that make them unsuitable for cases involving information. The
final portion of this paper will suggest an alternative approach based upon the
different types of values inherent in confidential information. It will be argued
that existing property offences fail to protect these unique values.
It should be noted that the ensuing discussion is not confined to trade
secrets, but will include all confidential information. The reasons for doing so
are twofold. First, the analysis relies quite heavily on decisions from the Com-
monwealth countries. These countries, as a rule, do not distinguish between
trade secrets and other types of confidential information, but instead treat the
misappropriation of all types of confidential information under the single head
of “breach of confidence.”‘ 3 Second, it is analytically consistent to speak of
12Fetterly offers several reasons for the reluctance of the United States courts to resort to the
criminal law. In essence, he argues that the United States was attempting to divorce itself from the
English experience with use of the criminal law in the commercial arena prior to the wholesale con-
version to the philosophy of laissezfaire during the Industrial Revolution. See Fetterly, supra, note
2 at 1539-44. He also suggests that the reluctance was due to the severe sanctions of the criminal
law of the period. For example, as recently as the 19th century the theft of any item worth more
than forty shillings was a capital crime (supra at 1544).
13For an excellent exposition of the English law of confidential information, see The Law Com-
mission, Breach of Confidence (Report No. 110) (London: Her Majesty’s Stationary Office, 1981)
[hereinafter Law Commission]. Although acknowledging the separate development of a body of
1991)
TRADE SECRETS
trade secrets and confidential information in the same breath. As the Common-
wealth approach recognizes, trade secrets are a subset of confidential informa-
tion. Although this article will draw distinctions between trade secrets and other
types of confidential information, it will do so only after isolating those char-
acteristics of trade secrets that make them deserving of separate treatment.
I. Applying Property-Based Criminal Laws to the Misappropriation of
Information
A.
Information as “Property”
As the preceding discussion indicates, the value of confidential commer-
cial information has risen significantly in the past few decades. Attempts to cri-
minalize the misappropriation of such information are even more recent. It is
therefore not surprising that most prosecutors have been unable to point to spe-
cific criminal statutes dealing with the problem. Instead, prosecutors have gen-
erally been forced to turn to other offences of more general application to deal
with industrial espionage. At first glance, the provisions that lend themselves
most readily to the problem are the property crimes of larceny, fraud and
embezzlement.
The essential question in applying a property crime to industrial espionage
is whether the information is indeed a form of property. 14 It is far too late for
this article to do much more than summarize the debate and offer a few sugges-
tions on the issue.
The dispute over the proprietary nature of confidential information can be
traced to the middle of the eighteenth century, when the law of confidential
information was emerging in its modem form. Many of the very early trade
secret cases dealt with confidential information as a form of property right. 5
However, it was not until 1917 that the issue was clearly addressed by Justice
Holmes speaking for the majority in E.I. Du Pont de Nemours Powder Co. v.
Masland.6 Instead of using a property analysis, Holmes reasoned that the law
of trade secrets was primarily concerned with the conduct of the defendant in
acquiring the information from the plaintiff.17 The key question that courts must
law relating solely to trade secrets in the United States, the English Law Commission felt that it
was more useful to deal with all confidential information under a single legal regime (supra at 9
& 102-104).
‘4See text accompanying notes 22-50.
15See, for example, Dean v. McDowell (1878), 8 Ch.D. 345 at 354 and Peabody v. Norfolk, 98
Mass. 452 (1868).
16244 U.S. 100 (1917) [hereinafter Masland].
17 The word property as applied to trade mark [sic] and trade secrets is an unanalyzed
expression of certain secondary consequences of the primary fact that the law makes
some rudimentary requirements of good faith. Whether the plaintiffs have any valuable
REVUE DE DROIT DE McGILL
[Vol. 36
face in trade secret cases is whether the defendant violated a duty of “fair deal-
ing.” Under Holmes’ view, commercial morality, not property, formed the basis
for liability.’8
As both sides to the debate have recognized, the portion of Justice Holmes’
Masland opinion dealing with the proprietary aspects of trade secrets is dic-
tum. 9 The debate centres on the extent to which this dictum accurately depicts
the legal nature of confidential information.’ On the one hand, there is a large
contingent of judges and academics who argue that confidential information is
property. This camp focuses on the fact that the law affords protection only to
information that the owner2 has attempted to keep secret. Non-confidential
information receives no protection whatsoever.’ A number of courts have built
upon this observation to hold that trade secrets, or in some cases all types of
secret or not, the defendant knows the facts, whatever they are, through a special con-
fidence that he accepted. The property may be denied, but the confidence cannot be.
Therefore the starting point for the present matter is not property or due process of law,
but that the defendant stood in confidential relations with the plaintiffs, or one of them
(see ibid. at 102).
‘8This notion of commercial morality was to come to the fore in International News Service v.
Associated Press, 248 U.S. 215 (1918) [hereinafter INS]. Justice Pitney’s opinion in INS relied
heavily on the notion that there are certain moral limits on the scope of competition. See supra at
235-36. It is interesting to note that Justice Pitney’s opinion also contained dictum discussing a
property basis for the law of information. Pitney reasoned that, except in certain narrow circum-
stances, the news is the “common property” of all. Supra at 235.
19See Ruckelshaus v. Monsanto Co., 467 U.S. 986 (1984) [hereinafter Ruckelshaus]. See also R.
CalIman, The Law of Unfair Competition Trademarks and Monopolies, 4th ed. (Wilmette, I1.: Cal-
laghan & Co., 1983) c. 14.02 at 8-9; R. Milgrim, Business Organizations: Milgrim on Trade
Secrets, (New York: Matthew Bender & Co., 1990), c. I at 2-4; A.S. Weinrib, “Information and
Property” (1988) 38 U.T.L.J. 117 at 129-30. The Masland case actually dealt with the review of
an orderpendete lite enjoining the defendant from disclosing an alleged trade secret to third parties.
2 0A few commentators have indicated some uncertainty on the issue. See, for example, Note,
“Limiting Trade Secret Protection” (1988) 22 Val. U.L. Rev. 725 at 743.
This split is reflected in the law of several of the non-common law industrialized countries. Most
countries in Western Europe do not recognize the concept of a “trade secret,” at least as that term
is used in the United States. Nevertheless, they do provide protection for “manufacturing secrets”
(also known as “industrial secrets”) and, to a lesser extent, “commercial secrets.” See generally N.
Prasinos, “International Legal Protection of Computer Programs” (1986) 26 Idea 173 at 190-91.
France also provides protection for savoirfaire, a concept highly analogous to the common-law
concept of “know-how” discussed supra, note 5. See Prasinos, supra at 213. Among the Western
European countries, Italy treats both manufacturing secrets and commercial secrets as a form of
property right. West Germany, on the other hand, treats neither as property. France takes a middle
ground, treating manufacturing secrets (and any associated savoirfaire) as a property right, while
rejecting the property label for commercial secrets. See Prasinos, supra at 210-25.
In Japan, trade secrets are generally not considered a property right. See Prasinos, supra at 221.
21The term “owner” is of course misleading because it carries with it the baggage of property
law. This article will use the term as a shorthand way of referring to the party with the legal right
to protect confidential information. This will in most cases be the party with current control of the
information.
22Milgrim, supra, note 19, c. 1 at 63-64.
1991]
TRADE SECRETS
confidential information, are property in a wide variety of contexts.’ Certain
commentators have seized upon these cases as establishing a general theory of
property rights in confidential information.2′
In recent years, the property view has taken a stronger hold of United
States jurisprudence, as evidenced by the United States Supreme Court decision
23United States: Formulabs, Inc. v. Hartley Pen Co., 275 F.2d 52 (9th Cir. 1960), cert. denied
363 U.S. 830 (1960) (holder has right to intervene in lawsuit affecting information); Ferroline
Corp. v. General Aniline & Film Corp., 207 F.2d 912 (7th Cir. 1953) [hereinafter Ferroline Corp.],
cert. denied 347 U.S. 953 (1954) (can be protected by injunction); E.B. Muller & Co. v. FTC, 142
F.2d 511 (6th Cir. 1944) [hereinafter E.B. Muller & Co.] (a customer list, as “property,” may be
protected from a subpoena duces tecum); Herold v. Herold China & Pottery Co., 257 F. 911 (6th
Cir. 1919) [hereinafter Herold] (can be protected by injunction); Anaconda Co. v. Metric Tool &
Die Co., 485 F.Supp. 410 (E.D. Pa. 1980); Zotos International, Inc. v. Kennedy, 460 F.Supp 268
(D.D.C. 1978) (trade secret is “property” right for purposes of due process requirements); Darsyn
Laboratories, Inc. v. Lenox Laboratories, Inc., 120 F. Supp. 42 (D. N.J.), aff’d 217 F2d 648 (3d
Cir. 1954), cert. denied 349 U.S. 921 (1955) [hereinafter Darsyn] (can be protected by injunction);
Roberts v. Gulf Oil Corp., 147 Cal. App. 3d 770, 195 Cal. Rep. 393 (1983) [hereinafter Roberts]
(trade secrets are property which can be the subject of a subpoena duces tecum); Olschewski v.
Hudson, 87 Cal. App. 282, 262 P. 43 (1927) (can be protected by injunction); Murphy v. Murphy,
28 Ill. App. 3d 560, 328 N.E.2d 642 (1975) (customer list can be property right); Miller v. Ortman,
235 Ind. 641, 136 N.E.2d 17 (1956) (trade secret may be protected by contract); Mountain States
Tel. & Tel. Co. v. Dep’t of Public Service Reg., 634 P.2d 181 (Mont. 1981) (“obvious” that com-
mercial trade secrets are property for purposes of fourteenth amendment); Consolidated Boiler
Corp. v. Bogue Electric Co., 58 A.2d 759 (N.J Ch. 1948) [hereinafter Consolidated Boiler Corp.]
(because a trade secret is a “property” right, a contract can be as restrictive as necessary without
constituting an unreasonable restraint on trade); Sachs v. Cluett, Peabody & Co., 31 N.Y.S.2d 718,
177 Misc. 695 (1941) [hereinafter Sachs], order rev’d 265 A.D. 497, 39 N.Y.S.2d 853 (N.Y. A.D.
1943), aff’d 291 N.Y 772, 53 N.E.2d 241 (1944) (can be protected in equity, contains dictum sta-
ting that secret processes are “property” for other purposes); Van Products Co. v. General Welding
& Fabricating Co., 419 Pa. 248, 213 A.2d 769 (1965) (rejects Masland dictum); Brown v. Fowler,
316 S.W.2d 111 (Tex. Civ. App. 1958) (can be protected by injunction); Microbiological Research
Corp. v. Muna, 625 P.2d 696. For additional cases so holding, see the vast array of cases cited in
Milgrim, supra, note 19, c. 1 at 8 n. 15.
Commonwealth: DeBeer v. Graham (1891), 12 N.S.W.R. (Eq.) 144; Franklin v. Giddins, [1978]
Qd. R. 78 (the tangible embodiment of a secret is “property” which can be the subject of a turnover
order); Boardman v. Phipps, [1967] 2 A.C. 46; F.C.T. v. United Aircraft Corporation (1943), 68
Commonwealth L. Rep. 525; In re Keene (1922), 2 Ch. 475 (information is property for purposes
of bankruptcy); CincinnatiBell Foundry Co. v. Dodds (1887), 19 Weekly L. Bull. 84; Prince Albert
v. Strange (1849), 1 H. & Tw. 1, 2 DeG. & Sm. 652 (private drawings can be protected by injunc-
tion against dissemination).
for purposes of the criminal law are discussed at text accompanying notes 75-155.
The leading United States and Commonwealth cases on whether secret information is property
24The strongest argument in support of the property theory is advanced in Milgrim, supra, note
19, c. 1 at 60. A number of other commentators also champion this theory. See, for example, M.
Jager, Trade Secrets Law, (New York: Clark Boardman Co., Ltd., 1985) c. 4 at 12.1-20; D.F.
Libling, “The Concept of Property: Property in Intangibles” (1978) 94 L.Q. Rev. 103 at 118; S
Ricketson, “Confidential Information – A New Proprietary Interest? (Part 1)” (1977) 11 Mel-
bourne U.L. Rev. 223 at 241-422.
McGILL LAW JOURNAL
[Vol. 36
in Ruckelshaus.” The Court held that government regulations, which require
commercial entities to reveal valuable trade secrets,26 constitute a taking of
property and therefore require compensation under the fifth amendment of the
United States Constitution.27 Although the Court eventually refused to give
Monsanto the relief it sought,” the broad language used suggests that a trade
secret can indeed be property.
Justice Blackmun, writing for the majority, discussed the nature of a trade
secret and found it to be analogous to other forms of property.29 Blackmun J.
relied heavily on the fact that the laws protecting trade secrets give the holder
of the secret the right to exclude others –
“one of the most essential sticks in
the bundle of rights that are commonly characterized as property.”3 This right
of exclusivity gave Monsanto a valuable economic edge over its competitors,
which would be destroyed if the information was made available to others.3′
This “economic edge” was held sufficient to give Monsanto property rights in
the underlying information.
The majority opinion in Ruckelshaus has lent considerable support to the
“information as property” camp. Several other American courts have relied
upon Ruckelshaus when concluding that confidential information is property in
25Supra, note 19.
261n Ruckelshaus the Court faced a challenge to certain provisions of the Federal Insecticide,
Fungicide, and Rodenticide Act (FIFRA), 61 Stat. 163, as amended, 7 U.S.C. 136 et. seq. The
provision at issue required an applicant for registration of products covered by FIFRA to disclose
the chemical contents of the product and any test results relating to the product to the Environmen-
tal Protection Agency (“EPA”). In order to prevent redundant testing, FIFRA authorized EPA to
use the information submitted by the original applicant in evaluating subsequent applications sub-
mitted by others for similar products; provided that the subsequent applicant compensated the orig-
inal applicant for use of the information (3(c)(1)(D)). The Court viewed these provisions as tan-
tamount to a compulsory license of the original applicant’s data (supra, note 19 at 992).
The Monsanto Company sought io register a new pesticide under FIFRA, and filed certain sup-
porting data with EPA. Because the data concerning the pesticide in question was allegedly a val-
uable trade secret, Monsanto sued the EPA in federal court to challenge the data disclosure pro-
visions. The gist of Monsanto’s claim was that the compulsory licensing provisions of FIFRA
constituted a “taking” of Monsanto’s “property.” Monsanto sought declaratory and injunctive relief
to prevent the EPA from enforcing the provisions (supra at 997-98).
27U.S. Const. amend. V.
28The Court held that although Monsanto’s claim was valid with respect to some of the infor-
mation that it had submitted to EPA, it could not obtain equitable relief. Instead, Monsanto’s proper
course of action would have been to seek compensation for the wrongful taking under the provi-
sions of the Tucker Act, 28 U.S.C. 1491. See Ruckelshaus, supra, note 19 at 1018-19.
29 Ibid. at 1002-03. The Court cited case authority indicating that trade secrets are assignable.
They can form the res of a trust, and can pass to a trustee in bankruptcy.
30 Ibid. at 1011 (quoting Kaiser Aetna v. United States, 444 U.S. 164 (1979) at 176).
3 1Ibid. at 1012.
1991]
TRADE SECRETS
other contexts.32 The U.S. Supreme Court itself recently confined the property
rationale of Ruckelshaus in Carpenter v. United States,33 explored below.’ 4
Others reject the view that confidential information is a species of property,
arguing – much as Holmes argued in Masland –
that the conduct of the de-
fendant in either obtaining or disclosing the information is the essence of the va-
rious causes of action.” Lacking some evidence of objectionable conduct by the
defendant, information will not be protected by the courts, regardless of secrecy
or commercial value. This additional scienter requirement distinguishes informa-
tion from tangible property, which is protected against all taking, intentional or
otherwise. Because of this difference, many courts36 and commentators37 have
tert, 841 F.2d 388 (Fed. Cir. 1988) [hereinafter Formax].
32See Sheets v. Yamaha Motors Corp., U.S.A., 849 F.2d 179 (5th Cir. 1988); Formax, Inc. v. Hos-
33U.S., 108 S. Ct. 316 (1987) [hereinafter Carpenter cited to S. Ct.].
34The Carpenter decision is explored in depth at text accompanying notes 138-50.
351f the defendant is a competitor of the owner of the information, the wrongful act will typically
be the acquisition of the information by improper means. A cause of action can also exist, however,
in cases where the defendant’s acquisition of the information was perfectly proper. Take, for exam-
ple, the situation where an employer entrusts an employee with a trade secret. In such cases, the
objectionable conduct is the defendant’s disclosure of the information to others in breach of a duty
of confidence.
36United States: Northern Petrochemical Co. v. Tomlinson, 484 F.2d 1057 (7th Cir. 1973); Mon-
olith Portland Midwest Co. v. Kaiser Aluminum & Chemical Corp., 407 F.2d 288 (9th Cir. 1969);
Chemical Foundation, Inc. v. General Airline Works, Inc., 99 F.2d 276 (3d Cir. 1938), cert. denied
305 U.S. 654 (1938) (invention is not “property” subject to execution); International Industries,
Inc. v. Warren Petroleum Corp., 99 F. Supp. 907 (D. Del. 1951), aff’d 248 F.2d 696 (3d Cir. 1957),
cert. dismissed 355 U.S. 943 (1958); Jet Spray Cooler, Inc. v. Crampton, 377 Mass. 159, 385
N.E.2d 1349 (Mass. 1979); Kubik, Inc. v. Hull, 56 Mich. App. 335, 224 N.W.2d 80 (1974) [here-
inafter Kubik]; Wiebold Studios, Inc. v. Old World Restoration, Inc., 19 Ohio App. 3d 246, 484
N.E.2d 280 (1985). See also the dissent of Justice Brandeis in INS, supra, note 18 at 251-56.
(Engineers) Ltd. (1969), R.P.C. 41, 47-48; Morison, supra, note 4.
Commonwealth: Oxford v. Moss (1979), 68 Crim. App. Rep. 183, 185-86; Coco v. A.N. Clark
37See Law Commission, supra, note 13 at 9; A. Coleman, “Trade Secrets and the Criminal Law
in Canada” (1988-81) Eur. Intell. Prop. Rev. 15 at 18; Hammond, supra, note 6 at 349-52 and at
373 where the author writes: “Indeed, it would be helpful if the word [property] were banished
altogether from this arena.”; R.G. Hammond, “Theft of Information” (1984) 100 L.Q. Rev. 252 at
256-60 [hereinafter “Theft of Information”]; R.G. Hammond, “Quantum Physics, Econometric
Models and Property Rights to Information” (1981) 27 McGill L.J. 47 at 53-56 [hereinafter “Quan-
tum Physics”]; G. Jones, “Restitution of Benefits Obtained in Breach of Another’s Confidence”
(1970) 86 L. Q. Rev. 463 at 466 & 473; R. Klitzke, “Trade Secrets: Important Quasi-Property
Rights” (1986) 41 Bus. Law. 555 at 556 & 563-67; J. Stuckey, “The Equitable Action for Breach
of Confidence: Is Information Ever Property?” (1981) 9 Sydney L. Rev. 402 at 404-11; D. Wiesner
& A. Cava, “Stealing Trade Secrets Ethically” (1988) 47 Md. L. Rev. 1076 at 1080-81; Note,
“Theft of Trade Secrets: The Need for a Statutory Solution” (1981) 120 U. Penn. L. Rev. 378 at
383-84 [hereinafter “Statutory Solution”]; Note, “Trade Secret Protection for Mass Market Com-
puter Software: Old Solutions for New Problems” (1987) 51 Alb. L. Rev. 293 at 311.
The National Conference of Commissioners on Uniform State Laws similarly appears to have
based the Uniform Trade Secrets Act upon a policy of maintaining standards of commercial ethics,
instead of the protection of a “property” interest. See comments to Uniform Trade Secrets Act 1.
REVUE DE DROIT DE McGILL
[Vol. 36
argued against protecting trade secrets and other types of confidential informa-
tion under the legal rules pertaining to tangible property.
However, it is the author’s view that the debate cannot be resolved by sim-
ply determining whether information is property. Clearly, the per se property
view is too broad. There are several conceptual difficulties with extending all
of the rules governing tangible personal property to the realm of information.
The above quote from Jefferson illustrates the primary difficulty; namely, that
information by its very nature is incapable of exclusive possession. One can
share an idea with any number of others without diminishing the usefulness of
that information.” Therefore, unlike tangible personal property, someone can
“take” an idea from another without the original holder ever realizing that a
“taking” has occurred.39 Because most of the traditional rules of property focus
on possession and use, they often do not make sense in a world where exclusive
possession is the exception instead of the norm.4″
There are also strong policy objections to applying property rules to infor-
mation. Property rights involve some degree of exclusivity. Treating informa-
tion as a form of property may lead to it being locked up in the hands of a small
segment of society.4 This runs counter to the widely held notion that new ideas
– which can be shared with no direct loss to the holder –
should form part
of the common pool of knowledge.42 Although the law creates no obligation to
share new ideas with others, any exclusive legal rights to information should
disappear once the holder elects to disseminate the protected knowledge to soci-
“Quantum Physics,” supra, note 37 at 54.
38 0f course, even though the utility of information to the owner does not change as additional
people discover the secret, the value of that information to the owner may be affected. A later seg-
ment of this article will point out the subtle, but important, difference between utility and value.
See text accompanying notes 158-61.
39This feature also makes it difficult to determine who “owns” a given bit of information. See
40Professor Hammond also lists other conceptual problems that arise when treating information
as property. In addition to those discussed above, he notes, inter alia, that (a) information often
does not depreciate with use, (b) once used, information can quite easily enter the public domain,
(c) since creation of information in a modem technological society is often a joint activity, appor-
tionment of any rights of “ownership” is difficult, and (d) the “exclusion” of others from confiden-
tial information can be obtained only at a considerable cost to society (ibid.). In addition, it has
been suggested that treating confidential commercial information as property could detrimentally
affect the mobility of labour (Trade Secrets Report, supra, note 3 at 122).
41This problem has been recognized by other commentators (see, for example, “Quantum Phys-
42Professor Hammond, among others, has argued that a sound legal policy concerning informa-
tion should encourage dissemination of that information to society (ibid. at 55-56). See also Note,
“Electronic Crime in the Canadian Courts” (1986) 6 Oxford J. Leg. Stud. 145 at 150.
ics,” supra, note 37 at 69-70).
1991]
TRADE SECRETS
ety at large.43 Societal progress, especially in today’s “Age of Information,” is
largely dependent upon the free flow of information.”
Even if not all confidential information can be labelled as property, several
commentators have argued that a subclass of confidential information –
trade
secrets –
does deserve a per se property treatment.’ As explored in greater
depth below,46 trade secrets exhibit certain special characteristics that distin-
guish them from confidential information in general. The U.S. Supreme Court
in Ruckelshaus relied heavily upon these special characteristics in defining a
trade secret as a form of property. Nevertheless, important differences exist
between trade secrets and tangible property. Like other forms of confidential
information, the conceptual and policy arguments against property-based treat-
ment still exist. Trade secrets present a further conceptual difficulty. Although
the holder of a trade secret may have a right to prevent disclosure in certain cir-
cumstances, this right disappears as soon as the “secret” becomes public knowl-
edge –
for example, if another person independently generates and discloses
the underlying idea. As others have recognized, it is difficult to treat as property
something that disappears when it is revealed to the public.47 Therefore, trade
secrets are no more deserving of a per se property treatment than confidential
information in general.
Members of the anti-property camp, however, have also oversimplified the
issue. In certain situations, the existing rules of property can be used to protect
confidential information. To draw any meaningful distinctions, one must recog-
nize that the term property is a conclusion not an analysis. To argue that “X is
property, ergo, it should be protected” is simply a circular argument.4″ One
43″The general rule of law is, that the noblest of human productions – knowledge, truths ascer-
tained, conceptions, and ideas – become, after voluntary communication to others, free as the air
to common use” (INS, supra, note 18 at 250, Brandeis J. (dissenting)).
The patent and copyright laws are, of course, an exception to this desideratum. Under the patent
and copyright laws, individuals are given exclusive rights in inventions and works of art. Never-
theless, the patent and copyright laws are not inconsistent with the goal of dissemination of infor-
mation. They instead represent a decision to allow a limited monopoly to certain individuals in
return for dedication of the invention or work of art to the public at the end of the period of exclu-
sivity. In the end, society will benefit from the individual’s creative activity.
‘See B. Kaplan, An Unhurried View of Copyright (New York: Columbia University Press, 1967)
at 2 and D.G. Baird, “Common Law Intellectual Property and the Legacy of International News
Service v. Associated Press” (1983) 50 U. Chi. L. Rev. 411 at 415.
45The American commentators, in accordance with the American custom of treating trade secrets
under different rules than other confidential information, typically focus their attention exclusively
on trade secrets. See sources cited in notes 24 and 37. The Alberta Institute also limited its study
to trade secrets, even though Canadian law has tended to follow the more inclusive English
approach. See Trade Secrets Report, supra, note 3.
46See text accompanying notes 159-62.
47J. Stedman, “Trade Secrets” (1962) 23 Ohio St. L.J. 4 at 21.
4″The tautological nature of the term property is also discussed in Callman, supra, note 19, c.
14.02 at 9-10 (arguing that “property” is merely a “magic word” for recognizing an exclusive right
McGILL LAW JOURNAL
[Vol. 36
should instead ask whether X should be protected under the particular facts and
circumstances, and second, what form such protection should take. It is only
after deciding that X is to be protected in a certain fashion in a given fact sit-
uation that one can meaningfully employ the term “property.” In other words,
X is “property” only if some legal rule will afford the holder of X exclusive
rights with respect to a certain class of people.49 Viewed in this light, the law
of “property” focuses on the relationships between people, not the protection of
things.5
What distinguishes property rights from other types of rights is the class of
people excluded from X. Although a precise definition of what constitutes a
property right is far beyond the scope of this article, such a right must at the
very minimum exclude some people who are not in privity with the holder of
X.5 Without this minimum requirement, there is no discemable difference
between property rights and contractual rights. 2 In brief, people do not have a
property right unless the law protects them in their use of X against third parties,
even when those third parties have not agreed to respect the owner’s rights.
Under the foregoing analysis, then, the answer to the question of whether
X is property depends upon the context in which the question is asked. Some-
thing may be property in some situations, but not in others. It is therefore mean-
ingless to conclude that all of the law’s rules governing property can be applied
in a pecuniary value); “Quantum Physics,” supra, note 37 at 57-58; “Theft of Information,” supra,
note 37 at 257; D. Vaver, “Civil Liability for Taking or Using Trade Secrets in Canada” (1981) 5
Can. Bus. L.J. 253 at 262-63. See also Stuckey, supra, note 37 at 405-406, who argues that infor-
mation can be property for limited purposes.
49A number of commentators have argued that a right is a “property” right only if it affords the
holder of X rights of exclusivity as against the entire world, not merely a limited class of people.
See Trade Secrets Report, supra, note 3 at 110; Stuckey, ibid. at 405. Under this view, the laws
protecting confidential information could rarely, if ever, give rise to a property right. Since under
Anglo-American law the holder of confidential information is not protected against one who inde-
pendently comes up with the same information, the rights of that holder would not satisfy this def-
inition of a property right. See Stuckey, supra. In the long run, this may be a more workable def-
inition. Nevertheless, the modem trend seems to label even non-universal rights of exclusivity as
“property” rights. See S. Ricketson, “Confidential Information – A New Proprietary Interest?
(Part I)” (1978) 11 Melbourne U. L. Rev. 289 at 305-306. In accordance with the modem trend,
this article will use the less stringent definition of a property right set out in the text.
5See, for example, Curry v. McCanless, 307 U.S. 357 (1939) at 366 (“[Intangible property]
rights are but relationships between persons, natural or corporate, which the law recognizes by
attaching to them certain sanctions enforceable in courts.”); B. Ackerman, Private Property and
the Constitution (New Haven: Yale University Press, 1975) at 26; Weinrib, supra, note 19 at 129
(the label “property” is simply a recognition that the law imposes a requirement of good
behaviour).
51Ackerman, ibid. at 27; Ricketson, supra, note 49 at 306.
521t is interesting to note that the “contract” analogy is often used –
especially in the Common-
as the basis for protecting confidential information. See Jones, supra, note 37
wealth countries –
at 463.
1991]
TRADE SECRETS
to confidential information without first determining why it is important to
attach the label property to the right under the given facts. The Ruckelshaus case
provides a good example. The fifth amendment prohibits the federal govern-
ment from depriving someone of property without just compensation. As recog-
nized by both sides in the Ruckelshaus litigation, this provision is potentially
applicable to governmental actions that destroy the confidentiality of someone’s
information. In order to resolve the issue, the analysis should focus on why the
fifth amendment treats property differently than other types of interests that a
person may hold. Generally speaking, people have certain unique rights in their
property that do not exist in their non-property interests.53 Because the fifth
amendment limits its protection to property, it only covers situations in which
otherwise applicable law’ grants rights to the holder which are of sufficient
importance to merit protection against government actions. If interests in con-
fidential information are sufficiently similar to the interests that the fifth amend-
ment was enacted to protect, one may conclude that the information is property
for the purposes of the fifth amendment.
At this point, however, the analysis should stop. It is important to realize
that the decision in Ruckelshaus to apply the label property to confidential infor-
mation is binding only for purposes of the fifth amendment. Therefore, even
assuming that Ruckelshaus reached the correct conclusion in that case, the deci-
sion is of little precedential value in determining whether the same information
is property under tax laws.55 The reasons for affording special treatment to prop-
531t may be sufficient, for example, if the owner has the right to exclude others from using the
item.
54Although both the fifth and fourteenth amendments to the United States Constitution protect
“property,” the courts will look beyond the Constitution – normally, to state law –
to determine
whether a particular interest is property under those provisions. See Ruckelshaus, supra, note 19
at 1001; Board of Regents v. Roth, 408 U.S. 564 (1972).
55The proprietary nature of trade secrets for purposes of the law of income taxation was a hotly-
debated issue while the United States federal income tax laws provided preferential treatment for
gains realized upon certain sales of “capital assets.” See generally B.M. Harding Jr., “Obtaining
Capital Gains Treatment on Transfers of Know-how” (1984) 37 Tax Law. 307. When Congress
abolished the preferential treatment in 1986, the issue of whether information could be property
became less significant for purposes of income tax law. Nevertheless, the issue may still arise in
some cases. The issue also remains relevant outside the realm of the federal income tax. There is
still some debate over the extent to which a trade secret is “property” for the purposes of ad valo-
rem taxation. See Roberts, supra, note 23 (treats trade secret as property); Registrar & Transfer
Co. v. Director of Division of Taxation, 157 N.J. Super. 532, 385 A.2d 268 (Ch. 1978), rev’d 166
N.J. Super. 75, 398 A.2d 1335 (trade secret is not property); Milgrim, supra, note 19, c. 6 at 50
n. 3, (sales and use taxes); Commerce Union Bank v. Tidwell, 538 S.W.2d 405 (Tenn. 1977) (sales
tax); State v. Central Computer Services, 349 S.2d 1160 (Ala. 1977) (estate and use taxes); Mil-
grim, supra, note 19, c. I at 31.
The Commonwealth decisions have also exhibited some disagreement. See Musker v. English
Electric Co. (1964), 41 Tax. Cas. 556; Jeffrey v. Rolls-Royce Ltd. (1962), 40 Tax. Cas. 443 [here-
inafter Jeffi-ey]; EC.T. v. United Aircraft Corp. (1943), 68 Commonwealth L. Rep. 525.
REVUE DE DROIT DE McGILL
[Vol. 36
erty may be altogether different under tax laws than under the fifth amendment.
Because of these differences, Ruckelshaus should not be deemed binding prec-
edent in the realm of taxation. 6
A number of courts and commentators have failed to grasp that the label
property is merely a conclusion. A good deal of the literature advocating a per
se view of confidential information as property has supported that conclusion
merely by citing a string of cases from the law of bankruptcy, wills, and other
unrelated areas of the law. 7 A classic example can be seen in the cases dealing
with the availability of an injunction to prohibit the dissemination of informa-
tion. The equitable remedy of injunction has traditionally been available only to
protect property interests.” Because an injunction is also typically the remedy
of choice in confidential information litigation,59 many courts have had to deter-
mine whether such information is property. Most courts of equity have con-
cluded that the information is property, and can therefore be protected by
injunction.’
Modem commentators often cite these injunction cases as support for the
property theory.6′ The problem, of course, is that these commentators have
failed to treat these opinions for what they were; namely, attempts to fit infor-
mation into an existing pigeonhole of the law. The property analysis in the
injunction cases is simply a means to an end. Because an injunction appears to
be a particularly appropriate remedy for the misappropriation of information,
judges are willing to label the information as property in order to grant the
desired remedy. What these cases really decide, then, is that information may
in certain instances be protected by injunction.’ This in no way implies that the
56See Lord Radcliffe’s opinion in Jeffrey, ibid. After concluding that information was “property”
for purposes of the tax laws, Lord Radcliffe made it clear that his conclusion did not necessarily
support an overall view of information as property. See supra at 492.
57Carpenter, supra, note 33 at 320; E.B. Muller & Co., supra, note 23; Consolidated Boiler
Corp., supra, note 23; Sachs, supra, note 23; Jager, supra, note 24, c. 4 at 16; Limiting Trade Secret
Protection, supra, note 20 at 742.
58See Callman, supra, note 19, c. 14.02 at 10; Milgrim, supra, note 19, c. 1 at 60-62; Stuckey,
supra, note 37 at 405.
59From a practical standpoint, an injunction offers a number of advantages over the retroactive
remedy of damages. Most importantly, the holder of the information can prevent any further loss
of confidentiality, thereby preserving the value of the secret. In addition, because the various
causes of action for the loss of confidential information focus on whether there has been some
objectionable conduct on the part of the defendant, an equitable remedy seems particularly appro-
priate in such cases.
6See Herold, supra, note 23; Ferroline Corp., supra, note 23; Darsyn, supra, note 23; Sachs,
supra, note 23; see Kubik, supra, note 36 (equity intervenes in trade secret cases to “foster the
integrity of confidential relationships.”)
61See, for example, Milgrim, supra, note 19, c. I at 60-62.
62Stuckey, supra, note 37 at 418, supports the view that an injunction should be available, and
would allow the court to make an analogy to the law of property to reach that end.
1991]
TRADE SECRETS
same information should be treated like tangible personal property in other
areas of the law.63
Because of these difficulties, commentators have suggested that the label
property is of little use, and should be abandoned when dealing with confiden-
tial information.’ As a practical matter, however, this would be impossible. To
date, neither the courts nor the legislatures have created a separate legal hierar-
chy to deal with all aspects of the protection of information. Although there are
doctrines and statutes that govern certain aspects of the problem,” the courts
must usually look to the existing body of law for ways to protect information.
This general law, however, constantly draws distinctions based upon whether
something is or is not property.66 Therefore, unless society is willing to create
a parallel legal hierarchy to protect information, the courts must continue to face
the issue of whether information can be classified as property under existing
statutes and common law rules. The alternative would be to deny any sort of
protection to information.67
Regardless of the outcome of the debate, in certain areas such as the
injunction cases the analogy has worked quite well. The problems arise when
courts blindly rely upon precedent in cases which present different factual,
legal, and policy circumstances.6″ The extent to which the property offences of
63Vaver, supra, note 48 at 274-75 (“However convenient the terminology [property], it must not
be used as a substitute for analysis or to attribute characteristics pertaining to one sort of ‘property’
to the special nature of confidential information.”); Trade Secrets Report, supra, note 3 at 137-38;
“Electronic Crime in the Canadian Courts,” supra, note 42 at 149.
supra, note 6 at 373.
64See Coleman, supra, note 37 at 18; “Statutory Solution,” supra, note 37 at 384; Hammond,
65The statute books are replete with provisions designed to deal with bits and pieces of the over-
all problem of confidential information. The provisions of FIFRA covering trade secrets at issue
in the Ruckelshaus decision are but one example. See supra, note 26. In addition, the turnover pro-
visions of the Freedom of Information Act make an exception for “trade secrets and commercial
or financial information obtained from a person and privileged or confidential” (5 U.S.C.
552(b)(4)). Certain state criminal provisions specifically covering trade secrets are discussed at
text accompanying notes 199-212.
Although numerous specific provisions exist, there is no sui generis body of law which recog-
nizes the unique characteristics of information. Professor Hammond has argued that the law must
create new legal rules and principles if it is to protect information adequately. See “Quantum Phys-
ics,” supra, note 37; Coleman, supra, note 37 at 18.
66In Ruckelshaus, for example, the government’s actions could be challenged under the fifth
amendment only if the information was capable of classification as “property,” as that term is used
in the amendment.
67Some commentators have argued that legal regimes other than property law could be applied
to information. The Alberta Institute treats trade secrets as a type of “entitlement,” borrowing from
Professor Calabresi. See Trade Secrets Report, supra, note 3 at 141. See also “Quantum Physics,”
supra, note 37 at 52, which suggests that information is a societal resource, and that the law per-
taining to natural resources could therefore be closely analogous.
68These differences are discussed above at text accompanying notes 38-44.
McGILL LAW JOURNAL
[Vol. 36
criminal law can be used to police industrial espionage is one of these problem
areas. Although Anglo-American law draws sharp distinctions between the bail-
iwicks of the criminal and civil law, there are a number of conceptual similar-
ities. Therefore, the reasoning of the civil cases may prove useful, even though
the conclusions reached in those cases are not binding.
B. Specific Applications of Criminal Offences
The increased frequency of industrial espionage has troubled courts and
legislatures alike, because such activity is inimical to basic notions of “fair
play” that society imposes even upon competing enterprises. There is a wide-
spread sentiment that even in a capitalist society minimum standards of conduct
exist, which define the permissible scope of competition.69 Because the actions
of the industrial spy fall below this minimum, the law should deter those who
contemplate appropriating the creative activity of others.”
As discussed above, Anglo-American law has relied primarily upon civil
remedies to police the area of industrial espionage.7′ Recent cases, however,
reveal a marked increase in the use of criminal law. In the United States, the
impetus for this change came in the 1960s with a highly-publicized rash of
industrial espionage in the pharmaceutical industry. Although the origin of the
change is not as clear, the Commonwealth countries have experienced a similar
increase in the use of criminal law. This trend reflects both the enhanced value
that modem society places in confidential information and a growing frustration
with the limits of traditional civil remedies.72
691t is often claimed that the fundamental purpose of the law’s protection of confidential com-
mercial information is the preservation of commercial morality. See Wiebold Studios, Inc. v. Old
World Restoration, Inc., 19 Ohio App. 3d 246, 484 N.E.2d 280, 284 (1985); Boardman v. Phipps
(1967), 2 A.C. 46, Upjohn J. (dissenting); Jones, supra, note 37 at 465-66; Klitzke, supra, note 37
at 556; Wiesner & Cava, supra, note 37 at 1080-81; Restatement of Torts, 756, comment a. This
is perhaps most eloquently stated in the majority opinion in the United States Supreme Court’s INS
decision. Indeed, the INS majority would take the concept of commercial morality one step further.
Justice Pitney’s opinion implies that the existence of competition between two people creates a
higher duty of “fair dealing” than would exist between non-competitors. See INS, supra, note 18
at 235-36 & 239-40.
70Failure to provide this adequate deterrent, the argument continues, will also significantly
decrease the incentive for commercial enterprises to generate useful information.
In recent years, some economists have expressed doubt as to whether the legal protection of pat-
ents, copyrights, and trade secrets actually provides a significant incentive to engage in creative
activity (see discussion in “Theft of Information,” supra, note 37 at 262). The only empirical evi-
dence relates to the Netherlands’ abolition of its patent system from 1869 to 1912. Although there
was a small decline in the level of innovation during that period, the evidence was not conclusive
(Trade Secrets Report, supra, note 3 at 114).
71Zupanec, supra, note 11 at 971-72; Trade Secrets Report, supra, note 3 at 126. The possible
721t is admittedly open to debate whether the misappropriation of information is a proper subject
for the criminal law, or whether society should rely solely upon the civil law to provide the needed
reasons for this reluctance to use the criminal law are discussed supra, note 12.
1991]
TRADE SECRETS
Some Western European countries have enacted specific criminal statutes
dealing with industrial espionage. 3 Most Anglo-American jurisdictions, how-
ever, have not followed suit. Accordingly, many of the prosecutions in those lat-
ter countries have proceeded under more general provisions of criminal law.
Prosecutors typically turn first to the property offences, most commonly, some
form of larceny, fraud, and vandalism/mischief. From a purely conceptual point
of view, the acti rei of these property offences provide the closest analogy to the
acts that society is trying to punish. For example, both the crime of larceny and
the act of industrial espionage involve the appropriation of something of value
that was previously under the exclusive control of another. Therefore, in the
absence of a separate body of criminal law relating to information, it is logical
to attempt to prosecute industrial spies under a property offence.
Certain formidable obstacles exist to any attempt to apply general property
offences to the misappropriation of information. Property crimes were originally
created to protect tangible personal property. The resulting legal rules concen-
trate upon those characteristics of tangible personalty that are most valuable to
society. Because information does not necessarily share all of these important
characteristics, certain key elements of property offences must be stretched to
the breaking point if they are to be extended to industrial espionage.
A great amount of judicial time and effort has been expended in attempting
to mold property offences to the misappropriation of information. Therefore,
this article will rely upon case law to illustrate the problems. In a series of deci-
sions from Great Britain and Canada, the courts have wrestled with the logical
difficulties and reached divergent conclusions. A review of these cases will be
useful since there are marked similarities between the relevant criminal statutes
of Great Britain and Canada. 4
deterrence. This somewhat more fundamental issue is well beyond the scope of this article. For a
taste of the debate, compare the views of the law reform committees in New Zealand (which argues
in favour of criminal sanctions) and Great Britain (which argues against use of the criminal law),
summarized in the Trade Secrets Report, supra, note 3 at 189-91. The report went on to conclude,
after a detailed discussion, that criminal sanctions were both proper and necessary (supra at 128-33
& 218-20). See also E.W. Kitch, “The Law and Economics of Rights in Valuable Information”
(1980) 9 J. Leg. Stud. 683 at 692-93; C. Webber, “Computer Crime or Jay-Walking on the Elec-
tronic Highway” (1983-84) 26 Crim. L.Q. 217 at 235-36; “Criminal Liability,” supra, note 9 at
490.73France, West Germany, and Switzerland have also experimented with criminal sanctions
against industrial espionage (see generally Fetterly, supra, note 2 at 1539). The Swedish Commis-
sion on the Protection of Trade Secrets also recommended the adoption of strong criminal penalties
(Trade Secrets Report, supra, note 3 at 131). See also Remington-Rand Corp. v. Business Systems,
Inc., 830 F.2d 1260 (3d Cir. 1987), which discusses a provision of the Netherlands Penal Code pun-
ishing employees who breach the confidence of their employers’ secrets.
“In both Great Britain and Canada, the criminal law is a matter of predominantly national
instead of local concern. Although Canada, like the United States, is a federation of semi-
independent provinces, the Canadian Constitution, unlike that of the United States, specifically
REVUE DE DROIT DE McGILL
[Vol. 36
The first case of note is the British decision in Oxford v. Moss.75 In Oxford,
an engineering student at Liverpool University surreptitiously obtained a copy
of an upcoming exam paper, read it, and returned it to the file. Once the stu-
dent’s act was uncovered, he was charged with theft of the information set forth
on the examination paper.76 Although it was clear that the student had performed
the acts alleged, the Divisional Court upheld the magistrate’s dismissal of the
case. Justice Smith’s majority opinion focused on the two key elements of the
offence of theft, namely, whether the information was property, and if so,
whether the defendant intended to deprive Liverpool University of that property.
In his view, the facts of the case failed to satisfy either element. Smith first
found that the exam paper did not satisfy the definition of property under the
Theft Act.77 Unfortunately, his opinion did not discuss the issue in depth, but
simply stated that information fell outside the statutory definition.”
Most of Smith’s opinion focused on the intent element of the offence.
Smith reasoned that because the examination paper had been returned to the file,
the student at worst intended to borrow the examination paper in order to gain
an unfair advantage over his classmates.79 The student’s actions accordingly
were not intended to deprive the university of possession of the test questions,
nor resulted in any actual deprivation. Even though the student had almost cer-
tainly destroyed the utility of the examination questions, the university retained
full ability to possess and use the information. Justice Smith therefore con-
cluded that even though the student’s conduct was “to be condemned, and to the
layman … would readily be described as cheating,”8 it was not conduct that
would support a prosecution for the offence of theft.”s
grants the exclusive power to enact general criminal statutes to the federal government (Constitut-
tion Act, 1867 (U.K.), 30 & 31 Vict., c. 3, s. 91(27)). Further, Canadian criminal law has borrowed
quite heavily from that of Great Britain.
75(1979), 68 Crim. App. Rep. 183 [hereinafter Oxford].
76The student was charged under s. 1 of the Theft Act (U.K.), 1968, c. 60. Except as discussed
infra, note 77, s. 1 codifies the main features of the common-law offence of larceny.
71Section 4(1) of the Theft Act, ibid. defines “property” to include “money and all other property,
real or personal, including things in action and other intangible property.”
At common law, larceny was confined to the taking of tangible personal property. Most juris-
dictions today define the “property” that can be the subject of larceny in a much more inclusive
fashion. For example, the United States Model Penal Code defines property as “anything of value,”
a definition that encompasses both tangible and intangible property (Model Penal Code 223.0).
Therefore, the fact that information is intangible does not in and of itself remove industrial espi-
onage from the reach of modem theft statutes. Nevertheless, as will be discussed below, the intan-
gibility of information does make it difficult to apply certain elements of the offence.
7 8Oxford, supra, note 75 at 185-86.
79Ibid. at 185.
“Olbid.
81 Wien J. also wrote a short opinion in the case. His opinion focused solely on the issue of
whether the information was property, and did not reach the additional element of “deprivation.”
1991]
TRADE SECRETS
The Canadian courts have built upon Oxford in several criminal cases. The
first case to be decided was R. v. Kirkwood,”2 a decision of the Ontario Court
of Appeal. In Kirkwood, the defendant copied videotapes which he had obtained
from various sources, and rented or sold the copies to others. These acts led to
an indictment under the fraud provisions of the Criminal Code.3 Although the
labels of the two offences are different, the provisions of the Theft Act at issue
in Oxford and the fraud provisions that were applied in Kirkwood deal with the
a wrongful taking of property possessed by another.84
same fundamental evil –
The primary difference between the offences of theft and fraud relates to the
means by which the appropriation is accomplished.”
Notwithstanding the similarity between the governing provisions, the court
in Kirkwood found that the defendant could be prosecuted under the fraud pro-
visions. Justice LaCourci~re delivered the opinion of the court, most of which
dealt with defining the circumstances constituting fraud. s6 In the latter part of
the opinion, however, LaCourcire focused upon the element of deprivation.
Because Kirkwood had merely copied the videotapes, LaCourci~re found that
the original owner had not been deprived of his tapes or the material on those
tapes. Nevertheless, Kirkwood had deprived the owner of something of value;
namely, the right to earn a profit from commercial sale or rental of the original
videotapes. LaCourci~re therefore ruled that the deprivation element could be
satisfied if the defendant either deprived the victim of something in the victim’s
On the former issue, Wien J. agreed that the confidential information did not satisfy the s. 4 def-
inition (ibid. at 186).
82(1983), 5 C.C.C. (3d) 393,42 O.R. (2d) 65, (Ont. C.A.) [hereinafter Kirkwood cited to C.C.C.].
83R.S.C. 1970, c. C-34, s. 338(1) [hereinafter Criminal Code, 1970] as am. S.C. 1974-75-76, c.
93, s. 32. See now R.S.C. 1985, c. C-46, s. 380(1) [hereinafter Criminal Code, 1985].
84Compare Theft Act, supra, note 76, s. 1(1): “A person is guilty of theft if he dishonestly appro-
priates property belonging to another with the intention of permanently depriving that other of it;
…” to the Criminal Code, 1985, s. 380(1): “Every one who, by deceit, falsehood or other fraudulent
means, whether or not it is a false pretence within the meaning of this Act, defrauds the public or
any person, whether ascertained or not, of any property, money, or valuable security ….
The defendant’s actions in Kirkwood almost certainly constituted an offence under the criminal
provisions of Canadian copyright laws. The Trade Secrets Report, supra, note 3 at 95, offers a plau-
sible explanation for the Crown’s decision to proceed under the general fraud provisions instead
of the more specific copyright provisions. Under the copyright law then in force, the maximum fine
for defendant’s actions would have been a mere $10 per copy, and $200 for the entire transaction.
5At common law, a defendant committed larceny when he deprived the victim of property with-
out the victim’s consent. Fraud, on the other hand, was present in the situation where the victim
consented to the deprivation, but the consent was wrongfully obtained.
6An important issue in Kirkwood was whether fraud could exist in the absence of a prior rela-
tionship between the parties. The court held that it could, as long as the defendant acted “dishon-
estly” (supra, note 82 at 397-98). Another interesting feature of the Kirkwood decision is that it
erodes the distinction between the theft and fraud provisions of the Criminal Code, 1970. Typically,
the offence of fraud required some proof of a relationship between the parties. In finding that Kirk-
wood had satisfied the “fraud” element by acting “dishonestly,” the court has created a potential
for overlap between the theft and fraud provisions (Trade Secrets Report, supra, note 3 at 97-98).
8
8
McGILL LAW JOURNAL
[Vol. 36
possession, or diverted to the defendant something that the victim was entitled
to receive.87 In essence, LaCourci~re found the requisite property interest not in
the information itself, but in the victim’s right to receive income from the infor-
mation.”
Three days following its Kirkwood decision, the Ontario Court of Appeal
ruled upon a fact situation even closer to Oxford in R. v. Stewart.89 The court
in Stewart (C.A.) was called upon to construe not only the fraud provisions, but
also the theft provisions of the Criminal Code, 1970.90 Stewart, a consultant,
was approached by an official of a union which was attempting to organize the
work force of a large hotel complex. The union official hired Stewart to obtain
the names, addresses, and telephone numbers of all of the hotel’s employees.
The defendant contacted a security worker at the hotel, and offered to pay him
two dollars for each name. 9′ Stewart was subsequently arrested and charged
with the offences of counselling mischief, fraud, and theft of information.
Although the trial court acquitted Stewart on all three counts, the prosecution
appealed only the fraud and theft issues.92
Unlike the unanimous decision in Kirkwood, the Ontario Court of Appeal
split on the applicability of both the fraud and theft provisions. Justices Cory
and Houlden ruled in favour of the prosecution, while LaCourci~re – who had
written the opinion of the court in Kirkwood sustaining prosecution –
dis-
87Kirkwood, supra, note 82 at 399, relying upon R. v. Clan, Hudson, and Hartnett, [1978] 2
S.C.R. 1175, 41 C.C.C. (2d) 145 [hereinafter Olan] and R. v. Renard (1974), 17 C.C.C. (2d) 355
(Ont. C.A.).
8 professor Coleman argues that this aspect of the Kirkwood decision could have a profound
impact on intellectual property law. Because the court’s opinion relies upon the loss of potential
profit that the owner can derive from information, Coleman argues that it could lead to a significant
extension of the penalties for copyright infringement (Coleman, supra, note 37 at 16).
‘9(1983), 5 C.C.C. (3d) 481, 42 O.R. (2d) 225 (Ont. C.A.) [hereinafter Stewart (C.A.) cited to
C.C.C.] rev’d Stewart v. R. (1988), 41 C.C.C. (3d) 481, 50 D.L.R. (4th) I (S.C.C.) [hereinafter
Stewart (S.C.C.) cited to C.C.C.]. The Supreme Court’s reversal is discussed below at text accom-
panying notes 116-24.
9cThe primary theft provision at issue was Criminal Code, 1970, s. 283(1), now Criminal Code,
1985, s. 322(1). The relevant portions provided as follows:
283(1) Every one commits theft who fraudulently and without colour of right takes, or
fraudulently and without colour of right converts to his use or to the use of another per-
son, anything whether animate or inanimate, with intent,
(a) to deprive, temporarily or absolutely, the owner of it or a person who has a special
property or interest in it, of the thing or of his property or interest in it,
(d) to deal with it in such a manner that it cannot be restored in the condition in which
it was at the time it was taken or converted.
91Stewart (C.A.), supra, note 89 at 484.
92The offence of mischief was later to be used with success in a case of industrial espionage pre-
senting somewhat different circumstances; see below, text accompanying notes 125-30.
1991]
TRADE SECRETS
sented.93 The decision in Stewart (C.A.) analyzes the property question in more
depth than Kirkwood. First, although the theft statute differs from the fraud pro-
visions insofar as it defines the subject matter as “anything, whether animate or
inanimate …. ” instead of property, all three Justices acknowledged that prior
decisions had incorporated concepts of property into the definition. Therefore,
in order to apply either the theft or the fraud provisions to Stewart’s actions, the
Justices had to find some form of property right which Stewart had taken from
the hotel. Cory and Houlden found that the property element was satisfied for
purposes of both the theft and fraud provisions. Although the opinions focus on
theft, they incorporate the same basic arguments when discussing fraud. It is
notable that neither Justice found the property right in the potential profits that
could be derived from the information, as the same court had done only three
days earlier in Kirkwood. Indeed, since the list of employees had no real com-
except possibly to the union which sought the information –
mercial value –
such a theory would have proven problematical.94 Both Justices instead argued
that the information was itself property for purposes of the fraud and theft pro-
visions of the Criminal Code, 1970. Houlden’s opinion cited the British
Exchange Cases95 and the American Chicago Board of Trade case96 –
all civil
cases –
as establishing a general principle that confidential commercial infor-
mation was a species of property.97 Although Houlden recognized that his con-
clusion was inconsistent with Oxford, he nevertheless argued that the cited prec-
edent was binding. 9s
Cory’s opinion for the most part agreed with Houlden. In two respects, it
is arguable that it went further. First, Cory’s opinion suggested an alternative
property argument. In Cory’s view, the defendant’s acts destroyed the hotel’s
ability to protect the list under Canadian copyright laws. Cory therefore rea-
93Neither Justice Cory nor Justice Houlden had sat on the appellate panel in Kirkwood. It is inter-
esting to note that Justice LaCourci~re had indicated in dictum in his opinion in Kirkwood that the
offence of theft might not apply to the facts in that case. See Kirkwood, supra, note 82 at 399-400.
Professor Coleman finds it difficult to reconcile LaCourci~re’s opinions in the two cases. See Cole-
man, supra, note 37 at 17. The approach to the problem set forth in Parts II and III of this article
suggests that Kirkwood and Stewart are indeed factually distinguishable.
94Both parties had conceded that the hotel had not lost any commercial advantage by the loss
of secrecy. See Stewart (C.A.), supra, note 89 at 492. The value of the list was also relevant in
the case insofar as the offence of counselling theft required that the property have a value in excess
of $200. Houlden’s opinion finessed this issue by noting that the defendant had offered the security
guard an amount well in excess of $200 to obtain the list. See supra at 495-96.
95Exchange Telegraph Co. Ltd. v. Gregory & Co., [1896] 1 Q.B. 147; Exchange Telegraph Co.,
Ltd. v. Central News, Ltd., [1897] 2 Ch. 48; and Exchange Telegraph Co. (Ltd.) v. Howard and
London & Manchester Press Agency (Ltd.), [1906] 22 T.L.R. 375.
96Board of Trade of City of Chicago v. Christie Grain & Stock Co., 198 U.S. 236 (1905) [here-
inafter Chicago Board of Trade].
97Stewart (C.A.), supra, note 89 at 493-94.
9S1bid.
REVUE DE DROIT DE McGILL
[Vol. 36
soned that the hotel had been deprived of a property right not only in the infor-
mation itself, but also in the rights afforded authors under copyright laws.99
Even more significantly, Cory intimated that all information, regardless of
its confidentiality, might be treated as a form of property, provided that the orig-
inal holder had expended time and effort in compiling it.”
In addition to finding that the information was property, the court also had
to face the additional element of intent for the theft charge. The Criminal Code,
1970 requires proof that the defendant had acted with one of an enumerated list
of intents.” Houlden’s opinion was the only one to deal with this issue. He rea-
soned that Stewart’s intent satisfied s. 283(l)(d), which states:
(d) to deal with it [the property] in such a manner that it cannot be restored in the
condition in which it was at the time it was taken or converted.
In Houlden’s view, Stewart had destroyed the confidentiality of the infor-
mation. Although the information itself remained in the hands of the hotel, its
character had been forever changed.” Accordingly, the prosecution had satis-
fied all of the elements of the offence of theft under s. 283(1).
As indicated above, LaCourci~re wrote a dissenting opinion. Unlike Cory
and Houlden, LaCourci~re undertook separate analyses of the theft and fraud
charges, eventually concluding that neither offence was satisfied. For the theft
charge, LaCourci~re relied upon Oxford and other English cases for the propo-
sition that confidential information was not property under the statute. The civil
cases discussing confidential information as a form of property were, in
LaCourci~re’s view, inapposite.”0 3 Although theft was not limited to tangible
property, LaCourci~re felt that it would clearly exceed Parliament’s intent if the
theft statute were extended to confidential information.’ 4
The fraud count presented LaCourci~re with an issue of greater complex-
ity. A prior decision of the Supreme Court had construed the fraud provision to
extend not only to situations where the defendant deprived the victim of prop-
erty, but also to cases in which the defendant caused “prejudice to the economic
interest of the victim.”‘0 5 Although the portion of the opinion dealing with theft
resolved the property issue, the broader prejudice rationale could potentially
apply to the facts of Stewart (C.A.). On this issue, LaCourci~re relied upon the
fact that the hotel had no intention to use the list of employees in “a commercial
991bid. at 498-501.
‘Ibid at 497-98.
’01The intents are enumerated in subsections (a) through (d) of Criminal Code, 1970, s. 283(1)
(now Criminal Code, 1985, s. 322).
“2Stewart (C.A.), supra, note 89 at 495.
’03Ibid, at 489.
’41bid. at 491.
10501an, supra, note 87.
1991]
TRADE SECRETS
way.”‘ Stewart accordingly had in no way caused any prejudice to the eco-
nomic interest of the hotel. 7 This argument allowed LaCourci~re to distinguish
Stewart (C.A.) from Kirkwood, because in Kirkwood the information contained
on the videotapes was clearly intended to be used for commercial gain.
The sharp differences between the majority and dissenting opinions in
Stewart (C.A.) reflect the basic conflict between policy and statutory interpre-
tation. On the one hand, policy dictates that society sanction the defendant’s
appropriation of the information. As Cory recognized, considerable effort often
goes into the accumulation of such information. There is something troubling in
allowing a third party to use the compiled information without the permission
of –
the owner. On the other hand, the current
criminal law cannot be easily molded to fit industrial espionage. Although the
original holders have been injured, they still retain full possession and use of the
information. All that they have lost is the confidentiality.
or at least compensation to –
These difficulties led the Alberta Court of Appeal to reject Stewart (C.A.)
in R. v. Offley 8 Offley involved a confidential police computer data base. The
defendant in Offley operated a business which performed security checks on job
applicants for employers. Defendant offered to pay a member of the Edmonton
City Police Department two dollars per name for running the name through the
police data bank. This offer was reported and, following a sting operation, the
defendant was arrested, charged, and convicted of the offence of counselling
theft of information contained in the police database.”
The Court of Appeal reversed the conviction. Justice Belzil’s opinion
rejected the view that the defendant had in any way deprived the Edmonton
Police of property because they had at all times retained full use of the infor-
mation.”‘ In addition, Belzil rejected Stewart (C.A.)’s interpretation of the
intent element of the offence of theft.”‘ Justice Belzil argued that it made no
sense to talk of returning something to the owner in a condition different from
that when it was “taken””‘ because there had never been a taking in the first
place.” 3 Therefore, although Offley had affected the confidentiality of the data
base, that did not constitute a deprivation of property.”‘
16Supra, note 89 at 492.
’01lbid.
10(1986), 70 A.R. 365 [hereinafter Offley].
10Ibid. at 365-66. As in Stewart (C.A.), the defendant was indicted under Criminal Code, 1970,
s. 283(1).
“Offley, ibid. at 368.
“‘See above text accompanying notes 101-102.
112Criminal Code, 1970, s. 283(1)(d) (now Criminal Code, 1985, s. 322(I)(d)).
lI3Supra, note 108 at 368.
1″Justice Belzil’s opinion at times fails to distinguish clearly the issues of (a) whether the infor-
mation in the police data bank was property, (b) whether defendant had “deprived” the police
department of that property, and (c) whether the defendant’s intent fell into one of the enumerated
categories. It appears, however, that he would answer each question in the negative.
McGILL LAW JOURNAL
[Vol. 36
The court in Offley recognized that its decision was essentially irreconcil-
able with its sister court’s opinion in Stewart (C.A.). 1″ 5 Both cases involved
information that the owner had kept in confidence. In addition, the defendants
in both cases bribed someone with access to the information with the hope of
obtaining the information for commercial use. Further, each case involved only
a limited breach of confidentiality because the information was revealed only to
the defendant –
not the public at large. Because the two cases involved essen-
tially the same facts, the results reflect a basic difference of opinion as to the
applicability of criminal law property offences to industrial espionage.
While Offley was pending, the decision of Stewart (C.A.) had been
appealed to the Supreme Court of Canada. In 1988, the Court reversed the
Ontario Court of Appeal.”6
The decision in Stewart (S.C.C.) followed the format of LaCourci~re’s
appellate dissent by distinguishing between the fraud and theft charges. On the
theft charge, the Court relied heavily upon the fact that the defendant had
merely copied the information, as opposed to taking the physical embodiment
of that information.” 7 The information itself, the Court reasoned, was not the
sort of property right that could be the subject of a prosecution for theft. The
Court held that in order to be property for purposes of s. 283, the item in ques-
tion must be capable of being taken.”8 Since the information itself could not be
taken, industrial espionage could not form the basis of a prosecution for theft.
The Court also offered a public policy justification for this conclusion. If
information was considered property for purposes of theft, a whole host of
unforeseen consequences could arise. For example, the Court hypothesized that
people who steal information and who cannot erase it from their memory might
be guilty of a separate offence for each day that they retained the information.” 9
Because these potential consequences had not been fully explored, the Court
indicated that the decision of whether to treat information as full-fledged prop-
erty under s. 283 should be left to Parliament. 2 This public policy rationale also
allowed the Court to sidestep various civil cases which had treated information
in a fashion similar to tangible property. Because the policy implications of
“5 Offley, supra, note 108 at 367-68.
“6 Stewart (S.C.C.), supra, note 89.
‘”lbid. at 484.
“5 It is unclear why the Canadian courts continue to cling to this vestige of the common law.
As noted by Justice LaCourci~re in his decision in Stewart (C.A.), supra, note 89 at 487, s. 283(1)
was amended in 1954 to delete the words “capable of being stolen.” The Supreme Court neverthe-
less indicated that the current language of Criminal Code, 1970, s. 283, reaches only property that
is capable of being stolen (ibid. at 489).
“glbid. at 491.
120Ibid. at 492.
1991]
TRADE SECRETS
treating something as property in the civil and criminal realms were entirely dif-
ferent, the Court refused to be bound by these civil precedents.’
The Court also addressed the alternative argument used by Justice Houlden
in Stewart (C.A.), namely, that even though the owner retained full use of the
information, it had been deprived of the confidentiality of that information.”
The Court disposed of this argument in short order. Once confidentiality is
divorced from the information itself, bare confidentiality cannot be considered
a form of property under even the broad ambit of s. 283.123 Confidentiality is
merely a condition which does not itself generate any legal implications. 4
The portion of the opinion dealing with the fraud charge was much more
cursory. In essence, the Court quoted LaCourci~re’s dissenting opinion with
approval. This implies that the Supreme Court, like LaCourci~re, would have
reached a different result in a case involving a trade secret or other information
with commercial value.
In Stewart (S.C.C.), the Supreme Court rejected the application of the gen-
eral property offences of theft and fraud in a case of industrial espionage. The
effect of Stewart (S.C.C.) on future cases, however, is unclear. In all likelihood,
the offence of theft will rarely be available to prosecute industrial spies in Can-
ada. By holding that “pure” information could not be the subject of theft, the
Court confined s. 283 to situations in which the defendant took some physical
embodiment of the confidential information, such as papers or photographs.
The offence of fraud, however, remains as a potential tool in these cases.
The court in Stewart (S.C.C.) was careful to distinguish between the type of
information at issue in that case and other types of information, such as trade
secrets, that could be used commercially. In cases involving the latter, the ele-
ments of fraud might be satisfied. Therefore, although the Supreme Court’s
opinion is silent on the validity of Kirkwood, the rationale used by the Court
lends support to that decision. Unlike the hotel’s list of employees, the video-
121Ibid. at 490. The Court also questioned whether the civil cases actually established a general
view of confidential information as property, or whether, as discussed above at notes 53-63, they
used the property label in order to grant an appropriate remedy given the strictures of legal rules
designed for tangible personalty.
12’It is unclear why the Court discussed this alternate theory. Both the prosecution and defense
had agreed that the hotel did not really benefit from maintaining the confidentiality of the list of
hotel employees. See Stewart (S.C.C.), supra, note 89 at 492. Defendant’s interference with this
confidentiality, therefore, caused no harm to the hotel.
231bid. at 493.
124The Court also addressed Justice Cory’s alternate contention that Stewart had deprived the
hotel of any copyright rights it may have held in the information. See above text accompanying
note 99. On this point, the Court found that no deprivation had occurred. Insofar as the hotel still
retained the “original” and the right to reproduce that original, its copyright rights remained intact
(ibid. at 495).
REVUE DE DROIT DE McGILL
[Vol. 36
tapes at issue in Kirkwood clearly displayed the type of commercial value that
could be affected by the industrial spy. Although the fraud statute presents other
unique barriers to the prosecutor, the statute has at least some potential for use
in the proper case.
A final Canadian case of note is the 1984 decision of the Ontario High
Court of Justice in Re Turner and the Queen.” Unlike the decisions discussed
to this point, Turner involved the application of the mischief provisions of the
Criminal Code, 197026 The facts in Turner are also somewhat different. The
defendants, operating from their office in Toronto, had on several occasions
gained telephone access to the computer records of a data storage firm in Mil-
waukee, Wisconsin. After infiltrating the records, the defendants encoded them
so that they could not be retrieved in the future. 27 There was no evidence that
the defendants made any copies of the information. Further, defendants did not
erase any of the relevant data. The defendants simply made it impossible for the
owners of the information to recall it.
Probably because the defendants had not appropriated the information, the
prosecutor did not rely on the theft and fraud provisions, but instead opted to
use the mischief provisions. Like theft and fraud, mischief requires an act which
affects property. I” However, the mischief provisions place certain restrictions
on the term property that do not exist in the theft and fraud provisions. In par-
ticular, the mischief provisions define property to include only corporeal prop-
erty.29 The defendants relied upon this limitation to argue that because their acts
did not change the physical nature of the computer tapes themselves –
the
125(1984), 13 C.C.C. (3d) 430 (Ont. H.C.) [hereinafter Turner].
126 Ss 385 & 387 (now Criminal Code, 1985, ss 428 & 430). The Commonwealth offence of
mischief typically bears the label “vandalism” in United States jurisdictions. Although the defend-
ant had also been indicted for fraud, only the mischief charge was before the Ontario Court of High
Justice (Turner, ibid. at 431).
It is interesting to note that the defendant in Stewart had also been charged with mischief under
the above provisions. The Ontario Court of Appeal never faced this issue because the prosecution
did not appeal the trial court’s verdict of acquittal on the mischief charge (supra, note 89 at 483).
127Turner, supra, note 125 at 431.
12SThe Criminal Code, 1970 reads:
387(1) Every one commits mischief who wilfully
(a) destroys or damages property,
(b) renders property dangerous, useless, inoperative or ineffective,
(c) obstructs, interrupts or interferes with the lawful use, enjoyment or operation of
property, or
(d) obstructs, interrupts or interferes with any person in the lawful use, enjoyment or
operation of property.
See now Criminal Code, 1985, s. 430.
129Criminal Code, 1970, s. 385 (now Criminal Code, 1985, s. 428). The Turner opinion dis-
cussed two proposed amendments to ss 385 & 387(1) that would have extended the mischief pro-
visions to computer data (supra, note 125 at 433).
1991]
TRADE SECRETS
tapes could still be erased and re-used –
the element of damage to corporeal property.
the prosecution had failed to satisfy
The Turner court rejected the defendant’s argument. Justice Gray argued
that mischief was not a crime against possession, but rather against enjoyment
of property. Because the defendants had physically altered the computer tapes
in a manner which reduced the owners’ ability to use those tapes, they had sig-
nificantly impaired the owners’ enjoyment of the accumulated information.
Under this rationale, the limitation of the definition of property to corporeal
items did not present an obstacle, because the tapes themselves were clearly tan-
gible. 3
The unique facts of Turner make the case of little use in resolving whether
the crime of mischief can reach other cases of industrial espionage. Part of the
difficulty is that the Canadian mischief provisions apply only to corporeal prop-
erty. That distinction, however, is relatively minor. The more significant facet
of Turner is that acts of the defendant did, in effect, result in a loss of the infor-
mation to the owner. Because the victim’s use of the tapes was deleteriously
affected, the court did not face the issue of how the law would apply if the tapes
themselves had not been altered.
The above decisions have evoked a great deal of scholarly interest in Can-
ada, Great Britain, and other Commonwealth countries. Most of the discussion
has focused on the decision in Stewart (C.A.), and predates the reversal of that
decision by the Supreme Court. 3′ The weight of academic opinion is decidedly
against its interpretation of the theft and fraud provisions. 132Most commentators
argue that the court’s interpretation of the Criminal Code, 1970 went beyond the
historical reach of Commonwealth property offences.133 It remains to be seen
whether Commonwealth jurisprudence will follow the lead of the Supreme
Court of Canada’s holding in Stewart (S.C.C.).
The British and Canadian courts are not alone in their efforts to apply prop-
erty offences to acts of industrial espionage. Courts in the United States have
also attempted to fit espionage into existing provisions of the criminal law. The
United States cases can be divided into two categories. Unlike the Common-
130Ibid. at 434. Even though Justice Gray found the requisite property interest in the computer
tapes instead of the intangible information stored thereon, he cited Stewart (C.A.) as authority.
13’In addition, there is some published commentary comparing the decisions of the Ontario
Court of Appeal in Kirkwood and Stewart and the Turner decision (see Coleman, supra, note 37
at 16-18; “Electronic Crime in the Canadian Courts,” supra, note 42 at 146-48).
132The most in-depth analysis is that contained in the Trade Secrets Report, supra, note 3 at
84-93, which lists five major objections to the decision. See also F.R. Moskoff, “The Theft of
Thoughts: The Realities of 1984” (1984-85) 27 Crim. L.Q. 226 at 229-33, which supports the
Ontario Court of Appeal; “Theft of Information,” supra, note 37 at 259 which intimates that the
result might be justified had the information constituted a trade secret.
133See, for example, Moskoff, ibid. at 232-33.
McGILL LAW JOURNAL
[Vol. 36
wealth countries, some state legislatures have enacted specific criminal statutes
dealing with industrial espionage. In jurisdictions without these specific stat-
utes, however, prosecutors have been forced to turn to general offences. This
article will begin with general offences and move to a discussion of specific
offences. 3″
In the state courts, there have been relatively few prosecutions for indus-
trial espionage under the general property offences. This is clearly not due to a
lack of relevant statutory authority, for every state has enacted some form of
general theft provision. For some reason, state prosecutors have not taken an
active role in prosecuting industrial spies.’35 In addition, in the few cases which
rely upon the general criminal provisions, prosecutors have had mixed suc-
cess. 136
The situation in the United States federal courts is altogether different.’37
Federal prosecutors have been quite active in the area of industrial espionage,
and have achieved a considerable rate of success. A large part of the success of
the federal prosecutors can be attributed to the expansive view of the scope of
the federal property offences adopted by the federal courts.
The most significant federal decision is the recent opinion of the United
States Supreme Court in Carpenter.”t s The facts of Carpenter are conceptually
similar to those of several of the Canadian cases discussed above. The defen-
dants in Carpenter were a reporter for the Wall Street Journal and a stockbroker.
The reporter gathered information from corporate executives for later publica-
tion in the paper. Although none of the information was inside or “hold for
release” information,’39 a general publication of the information could have had
a substantial impact upon stock prices. The Journal’s policy was therefore to
utes dealing specifically with industrial espionage. See Kitch, supra, note 72 at 692.
134See text accompanying notes 196-212.
135Trade Secrets Report, supra, note 3 at 216. This is true even in states which have enacted stat-
13 6See People v. Home Ins. Co., 197 Colo. 260, 591 P.2d 1036 (1979) (non-trade secret infor-
mation not “property”); State v. McGraw, 480 N.E.2d 552 (Ind. 1985) (theft conviction overturned
based upon finding that trade secrets are not “property”); Commonwealth v. Engleman, 336 Mass.
66, 142 N.E.2d 406 (1957) (trade secrets are not “property” for purposes of larceny statute); Han-
cock v. State, 402 S.W.2d 906 (Tex. Crim. App. 1966) (conviction for theft sustained). See also
the discussion in J.R. Vandevoort, “Trade Secrets: Protecting a Very Special ‘Property”‘ (1971) 26
Bus. & Law. 681 at 684.
137Unlike Canada, the general criminal law in the United States is usually a matter of state, not
federal, concern. Nevertheless, the United States Congress has enacted a number of federal crim-
inal statutes which incorporate in a wholesale fashion certain basic elements of the common law
property offences. These criminal statutes are typically enacted in connection with some federal
interest, such as the mails (18 U.S.C. 1341) or the transport of property across state lines (18
U.S.C. 2314).
13 t Supra, note 33.
1391bid. at 319. If the information had fallen into one of these categories, defendants’ acts would
have generated securities law consequences.
1991]
TRADE SECRETS
keep this information confidential prior to publication. The defendants entered
into a scheme under which the reporter would relay the information to the stock-
broker prior to publication in the paper. In return, the stockbroker agreed to
share with the reporter any profits that he derived from investments using the
information. The scheme was uncovered, and the defendants were indicted and
convicted under the federal mail fraud and wire fraud statutes.”4
On appeal, the defendants attempted to rely upon a prior U.S. Supreme
Court decision which had limited the reach of the mail fraud and wire fraud stat-
utes to situations involving a deprivation of “money or property.’ 4 Defendants
argued that the corporate information that the reporter passed on to the stock-
broker did not meet this definition of property. Carpenter presented the Court
with an interesting twist on the property issue. The defendants’ acts would not
have resulted in their receiving any additional financial benefit from the owner
of the information, that is, the Wall Street Journal. The financial reward would
instead come from the defendants’ dealings on the stock market. Further, the
Journal did not itself use the information for the purpose of trading in the stock
market, but rather held it for eventual publication in the paper. Therefore, unlike
earlier cases applying the mail fraud and wire fraud statutes to the misappropri-
ation of information, the Court was not faced with the relatively straightforward
situation in which a defendant’s acts deprived the victim of money or some
other easily-identifiable property interest. 42 The only conceivable property right
was the right to the information itself.
Facing the issue head-on, the Court held that the confidential information
did give rise to a property right. The Journal’s reasonable efforts to preserve
confidentiality gave it the right to make exclusive use of the information prior
to publication.’43 This right of exclusive use, the Court felt, was a property right
for purposes of the mail and wire fraud provisions. As authority for this conclu-
sion, however, the Court cited its Ruckelshaus,14 Chicago Board of Trade,45
and INS’46 decisions –
cases decided under the civil, not the criminal law.
14018 U.S.C. 1341 & 1343, respectively. Defendants were also indicted for conspiracy under
18 U.S.C. 371. See Carpenter, ibid. at 318.
141McNally v. United States, 483 U.S. 350 (1987), where the Court construed the mail fraud stat-
ute as applying only when the defendant had deprived the victim of a “property” right (supra at
the “intangible rights to honest and impartial government” – were held
356). The rights at issue –
not to rise to the level of a property right.
14 2See, for example, United States v. Doherty, 675 F Supp. 726 (D. Mass. 1987), rev’d in part
867 F2d 47 (1st Cir. 1989), cert. denied U.S., 109 S.Ct. 3243 (1989) [hereinafter Doherty], a case
in which the defendant copied police department exams. The court found that the police depart-
ment had been deprived of “property,” namely, the additional salary that it would be required to
pay to the defendant following his ill-gotten promotion.
143Supra, note 33 at 320-21.
‘”4The Ruckelshaus decision is discussed above at text accompanying notes 25-31.
’45Supra, note 96.
146Supra, note 18.
REVUE DE DROIT DE McGILL
[Vol. 36
Once the Court found the requisite property interest in exclusivity, it could
easily dispose of the other elements of the offences. The defendants’ actions had
clearly deprived the Journal of its right to make exclusive use of the informa-
tion. 47 The Court further held that the prosecution was not required to demon-
strate that the loss of confidentiality would result in any monetary loss to the
Journal.’48 It was enough to show that the defendants had deprived the Journal
of the right to use the information prior to publication, and, if it so chose, to
select the time and place for publication.’49 Defendants’ activities also satisfied
the requirement that the actus reus be fraudulent, insofar as both defendants
knew that the scheme was in violation of the Wall Street Journal’s internal
rules. 50 Based upon these findings, the Court upheld the conviction of both
defendants.
The U.S. Supreme Court’s decision in Carpenter was not without some
precedent. A pair of influential cases from the circuit courts had applied the mail
fraud and wire fraud provisions to industrial espionage a number of years ear-
lier.’ Nevertheless, the U.S. Supreme Court’s express sanction of the use of
federal law has given federal prosecutors a powerful new tool against the indus-
trial spy. Since Carpenter, a number of decisions have relied upon the mail
fraud and wire fraud provisions in industrial espionage cases. 2 In addition, fed-
eral courts have used Carpenter as precedent even when construing other
property-based federal criminal provisions.’53
One of the recent mail fraud cases is especially interesting when compared
to Oxford. In U.S. v. Thomas,” defendant was convicted of mail fraud for
147Supra, note 33 at 321.
141bid.
149This argument borrows freely from Justice Pitney’s majority opinion in the INS case (supra,
150Supra, note 33 at 321. Rejecting the defendants’ argument, the court held that the acts could
note 18 at 239.40).
be “fraudulent” even though they violated no statute.
15’United States v. Bottone, 365 F.2d 389 (2d Cir. 1966), cert. denied 385 U.S. 974 (1966)
(defendant photocopied and took notes of information); Abbott v. United States, 239 F.2d 310 (5th
Cir. 1956) [hereinafter Abbott] (defendant had a draftsman make an extra copy of original geophys-
ical maps). See also U.S. v. Seidlitz, 589 F2d 152 (4th Cir. 1978), cert. denied 441 U.S. 922 (1979)
(obtaining computer software by way of a telephone tap).
’52Belt v. U.S., 868 F.2d 1208 (11th Cir. 1989) (employee leaked confidential business informa-
tion concerning subcontractor bids); United States v. Grossman, 843 F.2d 78 (2d Cir. 1988), cert.
denied U.S., 109 S.Ct. 864 (associate at a law firm disclosed confidential information concerning
the recapitalization of a client); United States v. Elliott, 711 F. Supp. 425 (N.D. Ill. 1989) (partner
in law firm used confidential client information to his benefit by purchasing stock in company;
information held to be property of the law firm); see generally Milgrim, supra, note 19, c. 1 at
81-90.
153Formax, supra, note 32 (prosecution under Racketeer Influenced and Corrupt Organizations
Act, 18 U.S.C. 1962 et seq.).
1958 (1989) [hereinafter Thomas].
‘5686 F. Supp. 1078 (M.D. Pa.), aff’d 866 F.2d 1413 (3d Cir. 1988), cert. denied U.S., 109 S.Ct.
1991]
TRADE SECRETS
obtaining and distributing copies of the answer key for a police department
entrance examination. The court found that the answer key was the property of
the police department for the purposes of the mail fraud statute, primarily based
upon the fact that the police department had tried to keep it confidential.
Accordingly, even though the police department (unlike the Wall Street Journal)
could not use the information commercially, it nevertheless could protect its
property from acts which could destroy its confidentiality. The Thomas decision
is clearly at odds with the attempted use of the criminal law of theft against the
student in Oxford.
The long-term impact of Carpenter is at the present time unclear. As in its
earlier Ruckelshaus opinion,’56 the Court advances a very broad view of the
extent to which rules of property can be applied to confidential information.
Read literally, the opinion suggests that all confidential information –
not
merely the subcategory of trade secrets –
is eligible for treatment as property.
The Court found the requisite property interest in the Journal’s right of exclu-
sive prepublication use. Yet, given the facts of the case, it is difficult to see how
this right of exclusive prepublication use was of any value to the newspaper.
The Journal did not itself use the information to trade in the stock market. The
real value of the information to the paper was not the ability to use it prior to
publication, but instead the ability to be the first to disseminate the information
to the public. In other words, the information had a value to the paper only inso-
far as it could generate increased newspaper sales. Because the Carpenter
Court’s analysis does not search for any decrease in the value of the informa-
tion, the Court’s interpretation would appear to apply to all types of confidential
information, not merely trade secrets.
It should be obvious that the cases discussed above are somewhat at odds.
Admittedly, it is always difficult to compare cases which apply different crim-
inal statutes. This difficulty is aggravated when the courts are in different legal
systems. Nevertheless, the above cases represent attempts to deal with the same
basic problem: how to punish the industrial spy under a statutory scheme which
was not drafted to protect information. Regardless of the differences in style, the
statutes applied in the above cases are descendants of the same common law
155Further, the court in Thomas did not need to look to the information itself for the required
property interest. Because a successful copying of the exam questions would have caused the
police department to pay more to the test takers, the court could have found a “deprivation” of the
police department’s money. Compare Doherty, supra, note 142, a pre-Carpenter case.
The court in Oxford intimated that it might have allowed prosecution had it been shown that the
student actually took the paper upon which the exam questions were printed (supra, note 75 at 185,
Smith J.). The decision in Thomas does not indicate whether the defendants had taken the exam-
ination paper. Nevertheless, because the court found the requisite “property” interest in the infor-
even if it had existed – would not in itself be enough
mation itself, such a factual difference –
to distinguish the cases.
156Supra, note 19 at 1001-1003.
McGILL LAW JOURNAL
[Vol. 36
ancestors. Accordingly, the differing results in the cases are attributable to a fun-
damentally different perception of the proper reach of criminal property
offences.
57
In the view of this author, none of the above cases deals adequately with
the unique problems that arise in the area of confidential information. This does
not necessarily mean that the courts reached an incorrect result in each case.
Imagine, for example, a court faced with construing a criminal statute, but the
only available precedent is a civil case which finds that confidential information
is property for purposes of bankruptcy or tax law. The above discussion dem-
onstrates that this civil precedent should not be particularly relevant in interpret-
ing the criminal statute. Nevertheless, the fact that the civil precedent is not
binding does not mean that the same information is not property under the crim-
inal law.
To answer the question of how information should be treated under crim-
inal property offences, one must focus on the nature of confidential information.
Although such information does not share many of the characteristics of tangi-
ble property, it does resemble tangible property in certain respects. Therefore,
the proper analysis should isolate the unique characteristics of confidential
information, identify how such characteristics differ from those exhibited by
tangible property, and then –
determine which of the criminal
law property offences can be applied in cases of industrial espionage.
and only then –
II. A Value-Based Analysis
One characteristic confidential information shares with other types of prop-
erty is that it has value to the owner. Indeed, the law governing confidential
information is largely an attempt to enable the rightful holder of information to
realize the full potential value of this information.’58 However, the values inher-
ent in information are not identical to those inherent in tangible personal prop-
erty. This suggests that there might be another approach to analyzing the extent
to which criminal property offences can be applied to cases of industrial espi-
onage. Although the criminal law of property is intended to protect value, the
current offences are crafted so as to reach only certain types of value. The val-
157Justice LaCourci~re’s opinion in Stewart (C.A.), supra, note 89 at 490, suggests that the fed-
eral criminal laws of the United States are intended to protect substantially broader values than the
common law “property” rights protected by the Criminal Code, 1970. The basis for this conclusion
is unclear. Nevertheless, this article will demonstrate that the scope of the criminal law of property
is defined not so much by the breadth of the protected values as by the types of values involved
in a particular case. See above Parts II and III.
‘5 tThis, in turn, is intended to afford a greater incentive for engaging in creative activity. In addi-
tion to the doubts expressed as to the validity of this theory (see supra, note 70), some have argued
that Anglo-American law devotes too much attention to the creation of information, and not
enough to its usage and access. See “Quantum Physics,” supra, note 37 at 55.
1991]
TRADE SECRETS
ues protected by existing offences may or may not be the same as the values
exhibited by confidential information. A value analysis would attempt to match
the values recognized as important under the criminal law with those inherent
in information. To the extent that there is significant overlap, it may be possible
to apply a given criminal offence.
A. The Values Inherent in Information
Not all information has value. This is true even with respect to confidential
sometimes in great detail –
information. For example, everyone remembers –
certain personal events that have occurred throughout his or her lifetime. This
storehouse of knowledge has no real ascertainable commercial value. It is
unlikely that the holder of personal confidential information can make any use
of such information for commercial gain. Similarly, because no one else will be
able to use the information, the holder cannot sell the information for profit.
Certain types of confidential information, however, do have value. The
ability of humans to warehouse information – whether in their minds, a file
cabinet, or a computer memory bank –
can prove to be immensely valuable.
To the extent that one stores information, one can save the cost of reproducing
that information at a later time. One may also be able to profit from the stored
information by selling it to others who likewise want to avoid the cost of repro-
ducing that information. Depending in part upon the costs of recreating a given
set of data, information can have substantial value to the holder.
Confidential information may exhibit one or two types of value.’59 The first
can be dubbed “use value.” A good deal of the information that is collected by
commercial entities will have use value. Broadly speaking, information has use
value if it can be utilized to lower the marginal costs of producing that firm’s
output. Information concerning methods of manufacture will clearly exhibit this
type of value. Information like the list of hotel employees in Stewart (C.A.) will
also have use value. Because the hotel could almost certainly have been oper-
ated more efficiently with the list of employees than it could have been without
that list, the information had use value to the hotel.
Measuring the use value of information is more difficult. The decrease in
marginal costs associated with the information is not in and of itself a true indi-
cation of use value. One must also consider the costs of obtaining, warehousing,
and retrieving the information. To use a simple example, one could conceivably
reduce the marginal cost of the business of teaching law if one were to create
a filing system for the books and papers scattered about an office. By so doing,
159As indicated above, this article does not purport to analyze purely personal confidential infor-
mation. The privacy interests inherent in personal information may significantly skew the value
analysis set forth in this article.
REVUE DE DROIT DE McGILL
[Vol. 36
one could reduce somewhat the time that it takes to find a given book. One has
no real incentive, however, to create this storehouse of knowledge. First, creat-
ing the system would involve some cost. Second, in order for the system to be
effective, one would have to spend a great deal of time reducing it to memory.
These costs, even when spread out over time, offset many times over the min-
imal reduction in the costs of locating a given book. Accordingly, the informa-
tion probably has a negative use value –
it may actually increase the cost of
operating the business.
Some types of information also exhibit a second type of value, which can
be dubbed “monopoly value.” Monopoly value is the value attributable to being
the only person who has access to the item of information. When knowledge is
exclusive, the holders of the information are placed in an enviable position vis-
k-vis their competitors. The holders may use the information to reduce their
costs of producing the same goods as their competitors, or to produce a good
of superior quality. In either case, they will be able to derive profits not available
to their competitors.”6 They will continue to enjoy these additional profits until
someone else obtains the same information. Alternatively, they can sell or
license the information to others and capture the “value” of the information
more directly. Regardless of the method used to recapture the value of the infor-
mation, the fact that the holders alone have access to the information makes the
information more valuable. In essence, then, the owners are in a position akin
to that of a monopolist. 6’
Monopoly value can only exist when the information also has use value.
Unless the information has some utility, the holders will not gain any further
advantage from their position of exclusive access. Therefore, monopoly value
is a supplemental form of value that may exist when information with use value
is not known to others in the field.
In many ways, the distinction drawn above between use value and monop-
oly value is an alternative way of stating the traditional distinction between a
160In the situation where the secret results in a reduction in marginal cost, the owners can either
undercut the price of their competitors and capture a larger share of the market, or match the market
price and thereby realize additional profits from the sale of each unit of product. The optimal strat-
egy will depend, inter alia, on the price elasticity of demand for the product.
In the alternative situation where the secret allows the owners to produce a good of higher qual-
ity, they can, assuming a normal good, command a higher price than their competitors for the
product.
16 10f course, the rights of the owners of the information are not a legal “monopoly” in the tech-
nical sense of that word. Unlike the case of a true legal monopoly, such as a patent, any third party
can use confidential information obtained in an acceptable manner. Nevertheless, because the sec-
ond type of value that information can exhibit can be fully realized only while tie owners have
exclusive knowledge of the information, the term “monopoly” will be used as a matter of
convenience.
1991]
TRADE SECRETS
trade secret and other forms of confidential information. 62 A trade secret is
information that has both use value and monopoly value. It is monopoly value
that makes a trade secret such a valuable commodity in the modem world of
commerce, and that justifies society’s efforts to protect the holders of trade
secrets from the acts of the industrial spy. Therefore, this article will use the
term trade secret as a shorthand for information with both use and monopoly
value.
Nevertheless, thinking in terms of the above-described value analysis can
prove beneficial. 63 Most modem jurisprudence on confidential information has
failed to recognize that the value of a given bit of information may be quite dif-
ferent for different potential users. The holders of information, for example,
may derive both use value and monopoly value from the information. A com-
petitor who is allowed access to the information would derive both use value
and, to a lesser extent, monopoly value from the same information.” A user
who was not in competition, however, would derive only use value. An analysis
that focuses simply on whether the information is a trade secret overlooks the
fact that the answer depends upon the perspective from which the question is
asked.
The examination questions appropriated by the student in Oxford can be
used to illustrate this proposition. Under the traditional analysis, it is unclear
whether the examination questions would constitute a trade secret. Viewed from
the perspective of the university, the questions had only use value. Even before
the breach of confidence took place, the university did not enjoy any apprecia-
ble commercial advantage over its competitors.’65 To the student, however, the
information had both use value and monopoly value. Had the student’s plan suc-
162For example, the Uniform Trade Secrets Act, 1(4) defines a trade secret as “information …
that: (i) derives independent economic value, actual or potential, from not being generally known
to, and not being readily ascertainable by proper means by, other persons who can obtain economic
value from its disclosure or use ….
1631t is interesting to note that one of the early federal criminal decisions made reference to these
two distinct components of value inherent in information, but failed to recognize how they affected
the applicability of the mail fraud provisions. See Abbott, supra, note 151 at 314.
164Because the original owner will also retain the information, the amount of monopoly value
realized by the competitor will be somewhat less than that realized by the owner before secrecy
was lost. Of course, if there are only two firms in the industry, a loss of secrecy to the sole com-
petitor will destroy all monopoly value associated with the information. Monopoly value can exist
only if there are others in the industry who do not have access to the information.
165That other universities might be willing to pay Liverpool for a copy of the questions does not
itself necessarily mean that the information had monopoly value. Instead, it could indicate merely
that the costs of purchasing the information were less than the expense of independently creating
similar questions. This situation could exist regardless of whether the information had monopoly
value.
It would be an indication of monopoly value, however, if another university was willing to pay
more for an exclusive right to use the information than it would pay for a nonexclusive right.
McGILL LAW JOURNAL
[Vol. 36
ceeded, he would have been placed in a beneficial position with respect to his
competitors –
the other students in the class. This advantage could have
resulted either in higher test scores or the possibility of reaping a profit from the
sale of the questions.166 Therefore, the respective values of the information to
the university and the student were different in kind. To the university, the test
questions were merely confidential information. To the student, they were a
form of trade secret.
B. The Values Protected by the Criminal Law of Property
As demonstrated, confidential information may possess two distinct, but
related, types of value. One cannot ascertain whether one or both types of value
are present without identifying the current holder of that information. As the
information changes hands, its value –
especially its monopoly value – may
change. These features of confidential information make it difficult to apply the
criminal law of property to cases of industrial espionage.
Industrial espionage may in theory affect one or both of these components
of the total value of information. Normally, however, industrial espionage will
affect only monopoly value. In most instances of espionage, the spy does not
actually take the information from the original holder, but instead merely copies
the information. 67 Copying, of course, leaves the information itself in the hands
of the owner. Because the owner still retains possession and use of the informa-
tion, its use value remains unaffected. The owner will be able to produce the
product as cheaply and efficiently as before. Only monopoly value will be
affected, and even then only if the information is made available to one or more
competitors of the original owner.
Therefore, the world of industrial espionage presents essentially four sce-
narios. First, one may distinguish situations like Turner, in which the owner is
deprived of the information, from cases like Stewart (C.A.), Kirkwood, and Car-
penter, in which the owner retains the information. Each of those categories can
be further subdivided on the basis of whether the information had monopoly
value –
or whether it had only use value. 161 These
distinctions result in the following scenarios:
that is, was a trade secret –
’66Unlike the hypothetical university in the prior footnote, the other students in the class did not
have the alternative of independently creating similar information. Their concern would have been
with the exact questions which were to appear on the test, not merely with a set of questions rel-
evant to the subject matter.
‘ 67The reasons, of course, are not altruistic. If spies simply copy the information, it is much less
likely that their efforts will ever be detected by the owners.
16sSituations in which confidential information has no use value are not considered in this anal-
ysis. Indeed, it is unlikely that the owner of commercial information that has no use value will be
greatly concerned with the loss of secrecy.
1991]
TRADE SECRETS
A.
B.
C.
D.
A defendant deprives a victim of a trade secret;
A defendant deprives a victim of confidential information that is not a
trade secret; 69
A defendant copies 7′ a trade secret of a victim; and
A defendant copies a victim’s confidential information that is not a trade
secret.
Analyzing the issue in terms of these scenarios enables one to focus on what
elements of a property right are covered by each of the general property
offences.
1.
Theft
Most prosecutions for industrial espionage under the criminal law of prop-
erty have relied upon some form of the crime of theft. Reduced to its fundamen-
tals, theft is a crime against the right of possession.’ The defendant must inten-
tionally act in a manner which causes the victim to lose either his present or
pending possession of property. Further, the means by which the deprivation is
accomplished must somehow be wrongful. Although both the common law and
modem statutes draw distinctions based upon whether the deprivation was
a loss of the right
accomplished by force or by fraud,’
remains unchanged. Therefore, the discussion of theft in this
of possession –
article will include not only the offences actually bearing the name theft, but
also the various fraud and embezzlement provisions scattered among the crim-
the common thread –
’69More precisely, the above presupposes that the information is not a “trade secret” because of
the absence of monopoly value, not because it fails to meet one of the other elements of the def-
inition of a trade secret.
17For purposes of this discussion, “copying” is any sort of wrongful duplication.
171At common law, the crime of larceny applied only when the victim had either actual posses-
sion of property or fell within one of the narrow categories of constructive possession. Therefore,
agents who collect accounts for their principals, and immediately convert the accounts to their own
use, could not be prosecuted for larceny, because the victims (the principals) had never received
actual possession of the funds. See W.R. LaFave & A.W. Scott, Jr., Criminal Law, 2d. ed. (St. Paul,
Minn.: West Publishing Co., 1986) at 703-704. The agents could, however, be guilty of the related
crime of embezzlement (supra at 704-706). This article will use the broader term “theft” to avoid
the necessity of drawing these technical distinctions, which are of little relevance to the more basic
issue at hand.
172The common law further distinguished between larceny and larceny by trick, again focusing
on the means used to accomplish the deprivation.
REVUE DE DROIT DE McGILL
[Vol. 36
inal codes of various jurisdictions.’ All of the cases discussed above except
Turner essentially applied some variant of theft. Regardless of the label, the
objectionable conduct by the defendant in each case allegedly involved a
wrongful intentional deprivation of possession.
The foregoing value analysis is useful when applied to the crime of theft.
Theft’s focus on the right of possession is essentially a recognition of the impor-
tance of use value. There is a direct correlation between possession and use
value. If one loses the possession of property, one also naturally has lost the
ability to make economic use of that property.” Therefore, the offence of theft
is intended as a deterrent to those who would wrongfully interfere with a lawful
possessor’s right to use property to economic advantage.
On the other hand, theft is not very well suited to protect monopoly value.
There will be no direct logical connection in many cases between the fact of
possession and the existence of monopoly value. First, of course, one can pos-
sess confidential information that has no monopoly value. The examination
questions in Oxford again provide a good example. As discussed above, the
exam questions in that case had only use value to the university. The student’s
copying of that information accordingly could not have resulted in the loss of
any monopoly value to the university.
There is a second reason why a direct logical connection between posses-
sion and monopoly value can not be drawn. Take for example a situation in
which X owns a piece of machinery that is used in his business. If X is the only
person in the industry who owns such a machine, he will realize both use value
and monopoly value from his exclusive position. With the obvious exception of
land, however, most types of property –
can
be duplicated.’75 If someone duplicates X’s machine there will be no effect on
X’s possession of the original machine and the use value that he derives from
the fact of possession. Such duplication, however, may have a substantial effect
on the monopoly value that X derives from continued possession of the
tangible as well as intangible –
173This comports with the law of the Commonwealth countries and a number of jurisdictions in
the United States. The Model Penal Code would define “theft” even more broadly to incorporate
the historically distinct crimes of receiving stolen property and extortion (Model Penal Code
223.1(1)).
1740f course, the act of realizing the use value of property may involve a loss of possession. For
example, instead of using a piece of machinery which incorporates a secret process in a business,
one may elect to lease or sell it to a competitor. In so doing, one loses the right to possess the
machinery. This situation involves a voluntary transfer of possession in return for consideration,
and is therefore instantly distinguishable from the case of larceny, which involves a wrongful
taking.
175Certain types of chattels may also be sufficiently unique to be incapable of duplication, for
example, an original painting by a famous deceased artist. Unlike land and these unique chattels,
information is unique in that it can often be duplicated at a relatively low cost once someone
obtains access to it.
1991]
TRADE SECRETS
machine. As copies of the machine are distributed to X’s competitors, X will
lose the edge that he previously enjoyed over those competitors. A business
entity can therefore experience a loss in the monopoly value of property even
in situations where its possession of that property remains unimpaired.’76
In order for the offence of theft to apply to industrial espionage, the
defendant must affect the use value of the information. Scenarios A and B sat-
isfy this criterion. In both of these scenarios, the defendant has taken the con-
fidential information (and its associated use value) from the victim. The victim
has lost all of the benefits which previously flowed from the right to “possess”
the information. Because theft is designed to protect those benefits, it should be
available in these scenarios.
This conclusion may appear surprising. Using the value analysis, it
becomes apparent that the offence of theft may be available even in cases where
the information is not a trade secret. The key issue in determining the availabil-
ity of the offence is whether the owner lost use of the information. All types of
confidential information, not merely trade secrets, may exhibit use value.
Although this conclusion may appear counterintuitive, 77it is consistent with the
interests protected by the criminal law of theft.
Therefore, information can be property for purposes of the criminal law. 7 ‘
The value analysis helps to identify those rights associated with confidential
information that constitute property, and the extent to which criminal property
offences protect those rights. The offence of theft does not reach all components
of a property right; rather, it focuses more narrowly upon the benefits the owner
accrues from the right of possession. Only in scenarios A and B does this type
of property right exist. In scenarios C and D, on the other hand, theft will not
be available. The defendant in those scenarios has not affected the use value of
the information because possession has in no way been affected. At most, the
defendant has decreased the monopoly value of the information.
176The converse of this proposition, however, is not true. Losing possession of a given item of
property will also automatically result in a loss of all monopoly value associated with that property.
177See the opinions of Justice LaCourci~re in Stewart (C.A.), supra, note 89 at 492 and of Justice
Lamer in Stewart (S.C.C.), supra, note 89 at 496-97. Both of these opinions suggest that the
defendant could have been found guilty under the fraud charge had the list of hotel employees con-
stituted a trade secret.
1781ndeed, the information need not even be confidential to qualify as “property.” Even non-
confidential information can exhibit substantial use value. Since the crux of the offence of theft
is a loss of use value, a defendant who wrongfully deprives the owner of any useful information
can conceivably be charged with theft.
This further assumes, of course, that the governing statute does not limit the definition of prop-
erty to tangible property. Because industrial espionage generally does not physically affect the tan-
gible embodiment of the confidential information, it would fall outside a statute which protects
only tangible property.
McGILL LAW JOURNAL
[Vol. 36
Although the offence can be used in the proper case, theft as a practical
matter will rarely be available against the industrial spy. The vast majority of
cases of industrial espionage will fall into scenarios C or D. Indeed, of the cases
discussed above, only Turner would fall within Scenarios A or B.’79 Most indus-
trial spies do not actually take the confidential information from the original
owner; preferring instead to copy the information.8 ‘ The industrial spy who
merely copies confidential information has affected neither the owner’s posses-
sion of that information nor the use value that may be derived from possession.
The owner’s marginal cost of production has not been increased by the dissem-
ination; rather, the marginal cost of at least one of his competitors has been
decreased. Because theft deals with the loss of possession or use of a piece of
property, not the loss of all value that may flow from that property, it does not
deal adequately with a situation where the owner loses monopoly value, but
retains possession.
The same conclusion holds even if one relies upon the “control” argument
used in Carpenter. The Court in Carpenter, recognizing that the newspaper
retained full use of the information, argued that it had nevertheless lost the right
to control that information. That right of control, the Court reasoned, was prop-
erty that the defendants had taken from the newspaper. Even assuming that the
newspaper had a right of exclusive control,’8’ which was misappropriated by the
actions of the defendants, the offence of theft would still be inapplicable. In the
case of confidential information, exclusive control is not synonymous with pos-
session. Possession, regardless of whether it is exclusive, generates use value.
The use value of a given bit of information to a commercial entity will be the
same regardless of whether that information is known to that entity or to every-
one in the industry. The right of exclusive control, on the other hand, gives rise
‘”The prosecution in Turner could not rely upon the theft provisions of the Criminal Code, 1970
because of the statutory requirement of an “appropriation.” See 283(1), now Criminal Code, 1985,
s. 322(1). Although Turner had deprived the owner of the information, he had not converted the
information to his own possession. This requirement of an appropriation in the law of theft is not
unique to the Canadian statute, but appears in many theft statutes in both the United States and
Commonwealth countries. Compare with the Model Penal Code 223.2, which speaks in terms of
the defendant exercising control over the property.
One commentator has suggested that most of the problems with applying the criminal law could
be rectified by abandoning the requirement of a “taking” of property and focusing instead entirely
upon the conduct of the defendant (“Statutory Solution,” supra, note 37 at 384). The approach
proffered by this article would not go this far. In the view of the author, any reasoned approach
to the criminalization of industrial espionage must consider not only the conduct of the defendant,
but also the harm –
‘S0For more discussion of this feature of industrial espionage, see above text accompanying note
if any –
suffered by the victim.
167.
181As noted above (see text accompanying notes 146-50), the Court found that the Journal had
the right to make exclusive prepublication use of the information. Because the sole value of the
information to the Journal was the ability to be the first to publish it, it is difficult to see how a
right of prepublication use was of any appreciable value.
1991]
TRADE SECRETS
to monopoly value. The degree to which the information is available to compet-
itors has a direct effect on the profitability of that information to the original
holder. Therefore, the ability to control the dissemination of confidential infor-
mation is the ability to protect monopoly value, not use value. Because theft
extends only to use value, it would not apply to a case like Calpenter, where
the victim lost only the right to control dissemination of the information.
The conclusion that theft is generally unavailable is particularly appropri-
ate in cases like Stewart (C.A.), Oxford, and Offley. In those cases, the informa-
tion had no monopoly value whatsoever to the victim. The defendants’ acts
accordingly did not result in any loss of monopoly value to the holder of the
information.’82 Because the victim also retained the use value, it is difficult to
see how the acts of the defendants in those cases harmed the victims.
2.
Other Property Crimes
Most of the criminal law of property can be subsumed under the broad def-
inition of theft used in this article. There are, however, other criminal offences
that protect the rights of people in property. Many of these apply prima facie
only to tangible property, and are therefore difficult to use in the prosecution of
industrial espionage. As the Turner decision recognized, however, the property
offence of vandalism provides another possible approach.
Unlike theft, vandalism applies in the situation where the defendant dam-
ages or destroys the property of the victim, without necessarily taking posses-
sion. In theory, vandalism applies to cases in which the acts of the defendant
result in a loss of property value to the owner.
Vandalism’s emphasis on value makes the offence potentially useful in
cases of industrial espionage. The value of confidential commercial information
may be comprised of both use value and monopoly value. Most industrial espi-
onage does not affect use value. If the confidential information falls into the
hands of a competitor, the espionage may substantially reduce the monopoly
value –
of the information to the original holder.
Extending vandalism to all intentional acts which reduce the total value of prop-
erty would in theory allow it to be applied in many cases of industrial espionage.
Therefore, the offence of vandalism could apply in scenarios A, B, and C, sce-
narios in which the acts of the defendant have reduced the total value of the
and the total value –
18 2That the information could have had monopoly value in the hands of the defendant in all three
cases does not alter this conclusion. The offence of theft focuses primarily on loss to the victim,
not on any “unjust enrichment” of the perpetrator.
REVUE DE DROIT DE McGILL
[Vol. 36
information. 8 A decrease in either use value or monopoly value would be
sufficient.
Nevertheless, there are certain difficulties with this view of vandalism.
First, it is not clear that vandalism is intended to apply to situations, such as sce-
nario C, where only monopoly value is affected.” 4 One can argue that vandal-
ism, like theft, is designed to protect only use value. Vandalism arguably com-
plements theft by covering situations in which the defendant deprives the owner
of use of the property, but does not appropriate it.’
A number of vandalism statutes support this more limited interpretation.’86
For example, the provisions of the Canadian Criminal Code, 1985 require, inter
alia, that the defendant either render the property “useless, inoperative, or inef-
fective,” or interfere with the victim’s “lawful use, enjoyment, or operation of
property ………
Read literally, these terms are more closely related to use value
than to monopoly value.’
However, in a number of jurisdictions, especially in the United States, the
statutory language is broad enough to allow vandalism to be applied to cases of
pure copying. Many statutes speak generally of “damaging” or “injuring” prop-
erty.8 9 A decrease in monopoly value represents damage to the property from
1
8 31n scenario B, the defendant has destroyed the use value of the information. In scenario C,
on the other hand, the monopoly value has been affected. In scenario A, both use value and monop-
oly value have been destroyed.
In scenario D, on the other hand, the acts of the defendant have in no way affected any com-
ponent of the total value of the information to the owner.
18’Even outside the realm of information, there are problems in extending the crime of vandal-
ism to all intentional acts which reduce the value of the victim’s “property.” Take, for example,
the case of a janitor who fails to show up to work for several days. The janitor’s intentional breach
of his contractual obligation will in all likelihood decrease the value of the building which he was
supposed to maintain. It is unlikely, however, that any Anglo-American court would convict the
janitor of “vandalism,” even under the authority of a broadly-worded statute.
185This view probably conforms more closely to the historical reasons for the development of
the offence of vandalism. At common law, a person could not be found guilty of larceny absent
the element of “taking and carrying away” (Criminal Law, supra, note 171 at 715).
‘8 6Some vandalism statutes limit the offence to acts which affect tangible property, thereby
clearly excluding the typical case of industrial espionage. See, for example, Utah Code Ann.
76-6-106 (1990); Wash. Rev. Code Ann. 9AA8.060 (1988).
’87Criminal Code, 1985, s. 430(1).
18STaken by itself, the term “enjoyment” could conceivably include monopoly value. If the
owner of information loses the ability to realize monopoly value, he has in a sense been deprived
of the ability to “enjoy” the full fruits of his property. Read in conjunction with the terms “use”
and “operation,” which flank the term “enjoyment” in s. 430(1), however, the term “enjoyment”
should be interpreted more narrowly as applying to the benefits that derive from use of the
property.
89The provisions of the Indiana statute explicitly cover acts which cause either “damage” or
“pecuniary loss” (Ind. Code Ann. 35-43-1-2 (Bums Supp. 1989)). The broad language used in the
statutes of a number of states are also potentially subject to this interpretation. See, for example,
19911
TRADE SECRETS
the owner’s perspective. It is nevertheless unclear whether these statutes can be
applied to losses of monopoly value which are not accompanied by a loss of
possession. The terms “damage” and “injure” are equally capable of application
to use value as monopoly value. In other words, the intent of the respective leg-
islatures in using these broad terms may have been to protect the utility of the
particular item of property held by the victim, not the ability of the owner to pre-
vent duplication of that property.
Case law unfortunately leaves it unclear as to whether vandalism extends
to cases of pure copying. But the lack of case law is understandable. Virtually
all cases in which the government has sought a conviction for vandalism
involved some sort of physical damage to tangible property. When physical
damage is involved, the property will also suffer a decrease in its use value.
Damaged property is clearly less useful than undamaged property. Because the
vast majority of prosecutions for vandalism or mischief involve physical dam-
age,’ 99 the courts have not addressed a situation where the only loss to the owner
is monopoly value.
Turner is the only major case to apply a vandalism-based provision to
industrial espionage.9 Unfortunately, the facts in that case were atypical.
Because the defendant rendered the computer tapes inaccessible, he deprived
the victim of any use value that may have existed in the information stored on
those tapes. Accordingly, Turner is of little use in determining whether the
offence of vandalism applies to common cases of industrial espionage.’92
In addition to the problems of statutory interpretation and legislative intent,
there are policy reasons not to extend the current conception of vandalism this
far. Once vandalism is divorced from the concept of use, it becomes a poten-
Ala. Code 13A-7-21 (1975) (“damages property”); Ky. Rev. Stat. Ann. 512.020 (Michie/Bobbs-
Merrill 1985) (“intentionally or wantonly defaces, destroys or damages any property causing pecu-
niary loss”); N.Y. Penal Law 145.00 (McKinney 1988) (“intentionally damages property of
another person”); S.D. Codified Laws Ann. 22-34-1 (1988) (“intentionally injures, daiages, or
destroys… private property in which others have an interest”).
19The main reason that most vandalism cases are inapplicable to industrial espionage is because
they deal with situations where physical damage has occurred, not merely because tangible prop-
erty is involved. Tangible property may have monopoly value. For example, if X is the only firm
in the construction industry which owns a bulldozer, that bulldozer (regardless of its secrecy) will
have monopoly value. X could lose all or part of its monopoly value in one or two ways, namely:
(a) if someone damages X’s bulldozer, or (b) if one of X’s competitors acquires its own bulldozer.
Most vandalism cases involve the first situation, for example, where physical damage is involved.
A vandalism case involving the latter situation would be a useful precedent in the area of industrial
espionage, even though it involved tangible property.
191The Turner decision is discussed above at text accompanying notes 125-29.
192Even an expansive view of vandalism would not permit the offence to extend to acts falling
in scenario D. In this scenario, the owner retains the only type of value associated with the infor-
mation, its use value. Therefore, the defendant’s acts have in no way “damaged” the interests of
the owner in the property.
McGILL LAW JOURNAL
[Vol. 36
tially all-encompassing criminal offence. Unlike theft, vandalism does not have
a narrowly defined scienter element. Under most statutes, it is sufficient that the
prosecution demonstrates that the defendant intended to perform the acts which
damaged or destroyed the property. 93 The defendant’s intent need not be wrong-
ful or fraudulent, as is the case with theft-based offences. Once interpreted as
above, the lack of a scienter requirement gives the offence of vandalism a
broader scope in cases involving property with monopoly value. Thus con-
strued, vandalism could be used against otherwise socially acceptable activities.
For example, when a competitor discovers and implements a trade secret of
another through reverse engineering,”94 the monopoly value of that trade secret
will usually decrease. In effect, the competitor has intentionally damaged the
original owner’s property right in the trade secret. But few would argue that the
competitor should be subject to criminal sanctions. Nothing has been done
which violates a generally-accepted standard of conduct. It is interesting to note
that the above-described acts would not subject the competitor to any civil lia-
bility to the owner of the information for the loss of secrecy. Reverse engineer-
ing is a perfectly acceptable way of uncovering a trade secret, even when under-
taken to reap profits from the holder.9″ If the civil law views the actions as
acceptable, it would be paradoxical for those same actions to be subjected to
criminal penalties.
Because of these obstacles, the offence of vandalism should not be applied
in the typical case of industrial espionage. Vandalism, like theft, is designed pri-
marily to protect the benefits that the owner of property derives from use, not
the additional benefits attributable to the right of exclusive use. The elements of
the offence will only be satisfied in scenarios A and B, where the defendant has
affected the use value. Even if a statute could be crafted to cover a deprivation
of monopoly value alone, the lack of a scienter requirement in most vandalism
statutes makes the offence too broad to use in the special case of information.
I. The Future of the Criminal Law and Information
The above discussion demonstrates that although there are surface similar-
ities between the conduct proscribed by the property offences of the criminal
law and industrial espionage, certain difficulties prevent the extension of those
’93Criminal Code, 1985, s. 430(1)(a); statutes cited supra at note 189.
194Reverse engineering means “[s]tarting with the known product and working backward to
divine the process which aided in its development or manufacture” (Kewanee Vii Co. v. Bicron Co.,
416 U.S. 470 (1974) at 476.
195See Kewanee Oil Co. v. Bicron Corp., ibid. at 475-76; E.I. D11 Pont de Nemours & Co. v.
Christopher, 431 F.2d 1012 (5th Cir.) (1970) at 1015-16, reh’g denied, cert. denied 400 U.S. 1024
(1970), reh’g denied 401 U.S. 967 (1970). But see Wiesner & Cava, supra, note 37 at 1126-27,
which challenges the notion that the “reverse engineering” of another’s secret comports with socie-
ty’s notions of commercial morality.
1991]
TRADE SECRETS
offences to confidential information. The two main categories of offences, theft
and vandalism, focus primarily upon use value. The statutes defining these
offences speak of the defendant taking away or damaging use value to the det-
riment of the person legally entitled to realize that value. Although either theft
or vandalism could be available in the relatively rare case in which a spy actu-
ally does deprive the owner of the information, neither is of much use in the
more typical case of industrial espionage, in which the spy merely duplicates the
information.
The inadequacy of traditional property offences means that Anglo-
American legislatures must create new specific offences if they intend to
achieve the goal of effectively criminalizing industrial espionage. As noted, 196
several states in the United States have enacted such specific statutes.1 97
Although there have been attempts to enact general criminal sanctions for
industrial espionage in the Commonwealth countries, no specific provisions
have yet been enacted.’ 9 However, even in the United States, there have been
few criminal prosecutions under these specific offences.
Virtually all of these state laws are limited to trade secrets. Accordingly
they do not attempt to cover the entire gamut of industrial espionage, which may
include both trade secrets and confidential information generally. The statutes
196See text accompanying notes 71-72.
197See, for example, Ark. Stat. Ann. 5-36-107 (1987); Cal. Penal Code 499c (West 1988);
Colo. Rev. Stat. 18-4-408 (1989); Conn. Gen. Stat. Ann. 53a-124 (West Supp. 1990); Fla. Stat.
Ann. 812.081(2) (West 1990); Ga. Code Ann. 16-8-13 (1988); Ill. Ann. Stat. ch. 38, 15-1 –
15-2 (Smith-Hurd 1989); Me. Rev. Stat. Ann. tit. 17-A, 352 (1983); Md. Ann. Code art. 27, 340
(Supp. 1989); Mass. Gen. Laws Ann. ch. 266, 30, 60A (West Supp. 1990); Mich. Comp. Laws
752.771 (1990); Minn. Stat. Ann. 609.52 (West 1990); N.H. Rev. Stat. Ann. 637:2 (1986); N.J.
Rev. Stat. 2C:20-1 – 20-3 (1990); N.Y. Penal Law 155.00, 165.07 (McKinney 1988); N.C. Gen.
Stat. 14-75.1 (1986); Ohio Rev. Code Ann. 1333.51 (1990); Okla. Stat. Ann. tit. 21, 1732 (West
Supp. 1990); Pa. Stat. Ann. 3930 (Purdon 1983); Tenn. Code Ann. 39-3-1126 (1982); Tex. Penal
Code Ann. tit. 7, 31.05 (Vernon 1989); Utah Code Ann. 76-6-401 (1990); Wis. Stat. Ann.
943.205 (West 1989).
The United States federal government has also enacted criminal statutes which could be applied
in certain types of industrial espionage cases. Examples are the Comprehensive Crime Control Act
of 1984, Pub. L. No. 98-473, 98[2] Stat. 1837 (1984), which prohibits the obtaining of confidential
financial or credit information from computers without authorization, and the Small Business Com-
puter Security and Education Act, Pub. L. No. 98-362, 98[2] Stat. 431 (1984), which contains pro-
visions governing illegal use of private computers.
191A summary of these various efforts can be found in Trade Secrets Report, supra, note 3 at
189-91, and in Coleman, supra, note 37 at 15.
During the 1980’s, the Canadian Parliament was considering various amendments to the Crim-
inal Code, 1970 which would have criminalized the taking of confidential information. See “Elec-
tronic Crime in the Canadian Courts,” supra, note 42 at 148. Although these broad provisions were
never enacted into law, the Canadian Parliament, like the U.S. Congress, did enact provisions deal-
ing with the more specialized problem of computer espionage (see the Criminal Law Amendment
Act, 1985, S.C. 1985, c. 19, ss. 46 and 58).
REVUE DE DROIT DE McGILL
[Vol. 36
provide legislative recognition that confidential information possesses unique
characteristics. Although confidential information can in certain circumstances
be treated as property, it does not automatically mean that all property rules can
be extended without modification. Fundamental alterations must be made to
accommodate the unique characteristics of information.
For purposes of discussion, the statutes adopted by the various states can
be divided into two broad categories. The first is comprised of those states that
have simply amended the definition of property in their existing theft statutes
to include information.199 The purpose of this amendment, of course, is to allow
prosecutors to deal with industrial espionage under the general criminal law.
However, this approach to criminalization fails for the reasons already discussed
above.’
In short, the problem with theft is not primarily whether information
can bear the label property. The more pressing problem is the requirement of a
deprivation. Thus, attempting to mold existing theft provisions to the peculiar
rights inherent in confidential information creates more problems than it
solves.'”
The second category includes states that have drafted nonpareil offences in
an attempt to cover the unique features of industrial espionage.”u These states
have identified those aspects of industrial espionage that society deems repro-
bate, thereby allowing them to identify a specific actus reus and mens rea for
the newly-created offence. Although some statutes are limited to trade secrets,
others could conceivably extend to other confidential information. 3
199Although all of the states in this first category share a common approach to the issue, there
are significant differences in the coverage provided by the relevant statutes. In many states, for
example, only the tangible embodiment of the trade secret is considered to be property (see, for
example, Il. Ann. Stat. ch. 38, 15-1 (Smith-Hurd 1989); Md. Ann. Code art. 27, 340(h)(11)
(Supp. 1989)). A statute restricted to tangible items would be of little use in cases where the infor-
mation is transmitted orally or by means of electronic duplication.
The coverage of the North Carolina and Pennsylvania statutes, on the other hand, does not
extend to all trade secrets. Both of these statutes apply only to confidential scientific or technical
information. N.C. Gen. Stat. 14-75.1 (1986); Pa. Stat. Ann. 3930 (Purdon 1983).
v. Superior Court, 3 Comp. L. Serv. 206 (Super. Ct. 1972).
attempt to apply traditional theft provisions to cases of industrial espionage.
2See text accompanying notes 171-182.
20 Webber, supra, note 72 at 230-34, discusses some of the ambiguities that can arise from an
202There have been a few successful prosecutions under these statutes. See, for example, Ward
203Statutes limited to trade secrets, as that term is used in this article, include Cal. Penal Code
499c (West 1988); Fla. Stat. Ann. 812.081(2) (West 1990); and Wis. Stat. Ann. 943.205 (West
1989). Several of the other statutes, although using the term “trade secret,” define that term in lan-
guage broad enough to reach confidential information that has no monopoly value. See, for exam-
ple, Colo. Rev. Stat. 18-4-408 (1989); Ga. Code Ann. 16-8-13 (1988); Mass. Gen. Laws Ann.
ch. 266, 30, 60A (West Supp. 1990); Mich. Comp. Laws 752.771 (1990); Minn. Stat. Ann.
609.52 (West 1990); Ohio Rev. Code Ann. 1333.51 (1990); Okla. Stat. Ann. tit. 21, 1732 (Vest
Supp. 1990); Tenn. Code Ann. 39-3-1126 (1982); Tex. Penal Code Ann. tit. 7, 31.05 (Vernon
1989).
19911
TRADE SECRETS
An excellent and concise analysis of these state laws can be found in the
Trade Secrets Report of the Alberta Institute.2″ Although the analysis is con-
fined to the perspective of trade secrets, the observations are also relevant to the
general issue of confidential information. The Alberta Institute found none of
the existing state laws to be an adequate solution to the problem of industrial
espionage for two reasons. First, many of the statutes require that the defendant
affect a tangible object representing the trade secret.”5 This restriction limits the
reach of the criminal law by making the law useless in the situation in which
the defendant appropriates the information without taking a physical copy. It is
difficult to perceive why the misappropriation of confidential information is less
objectionable merely because it does not result in the loss of a physical object.
Second, and perhaps more fundamentally, the various states have had con-
siderable difficulty defining the actus reus of the new offence. Some states have
borrowed words such as “steal,”2 6 “embezzle, ”2 or “convert,”20 8 from the
existing criminal law. These terms, however, are by themselves independent
criminal offences. Thus, as the Trade Secrets Report recognized, 9 statutes
which utilize these terms combine a traditional property offence with language
designed to reflect the characteristics of information. As such, these new crimes
could only be applied if the underlying traditional property offence could itself
be applied.21 To the extent that the specific offences incorporate terms that are
2Supra, note 3 at 196-216. That report contains an in-depth discussion of the basic conceptual
difficulties with the state laws, for which this author is grateful. One caveat, however, is in order.
The discussion of individual state statutes in the report is of limited use, because many of the stat-
utes have been amended or repealed.
205See, for example, Cal. Penal Code 499c (West 1988); Fla. Stat. Ann. 812.081 (West 1990);
Mich. Comp. Laws 752.771 (1990); Minn. Stat. Ann. 609.52 (West 1990); N.Y. Penal Law
155.00, 165.07 (McKinney 1988); Tenn. Code Ann. 39-3-1126 (1982).
206Cal. Penal Code 499c (West 1988); Colo. Rev. Stat. 18-4-408 (1989); Fla. Stat. Ann.
812.081 (West 1990); Ga. Code Ann. 16-8-13 (1988); Mass. Gen. Laws Ann. ch. 266, 30
(Supp. 1990); Mich. Comp. Laws 752.772 (1990); Okla. Stat. Ann. tit. 21, 1732 (West Supp.
1990); Tenn. Code Ann. 39-3-1126 (1982); Tex. Penal Code Ann. tit. 7, 31.05 (Vernon 1989).
The exhaustive discussion set forth in M.A. Epstein, “Criminal Liability for the Misappropria-
tion of Trade Secrets” in Milgrim, supra, note 19, c. B5 at 1, is unfortunately no longer reliable
due to the amendment and repeal of several of the provisions.
in George, supra, note 10.
See also the discussion of state criminal provisions designed specifically to protect computers
207Fla. Stat. Ann. 812.081 (West 1990); Ga. Code Ann. 16-8-13 (1988); Mass. Gen. Laws
Ann. ch. 266, 30 (Supp. 1990); Mich. Comp. Laws 752.772 (1990); Okla. Stat. Ann. tit. 21,
1732 (West Supp. 1990); Tenn. Code Ann. 39-3-1126 (1982).
208Mass. Gen. Laws Ann. ch. 266, 30(4) (Supp. 1990); Minn. Stat. Ann. 609.52 subd. 2(8)
(West 1990); Ohio Rev. Code Ann. 1333.51(3); Pa. Stat. Ann. 3930(b)(2) (Purdon 1983).
2gSupra, note 3 at 206-207.
2raThe Colorado statute, for example makes it an illegal act to “steal” a trade secret with the
intent either to (i) deprive or withhold the control of that secret from the owner, or (ii) appropriate
the secret to the defendant’s own use (Colo. Rev. Stat. 18-4-408(1) (1989)). “Stealing,” however,
is synonymous with the offence of theft, which has its own actus reus and mens rea (a deprivation
McGILL LAW JOURNAL
(Vol. 36
burdened with meanings from the existing criminal law of property, they may
fail to cover a significant percentage of industrial espionage cases.
Several states, recognizing that most industrial espionage involves duplica-
tion of information, have drafted their trade secret statutes to cover copying. 2 ‘
Because copying by its very nature does not result in the deprivation of posses-
sion, these provisions at first glance suggest a workable actus reus. Unfortu-
nately, the copying provisions often create new problems which hinder their
effectiveness. A statute which simply prohibits copying may be too broad. In the
absence of a valid patent or copyright, Anglo-American legal systems generally
encourage the copying of creative activity. Copying is objectionable only when
the means used are themselves objectionable. Use of the term copy in the crim-
inal statutes without an adequate scienter requirement could result in the crimi-
nalization of conduct which would not give rise to civil liability.212
Legislative attempts of the various states to criminalize industrial espio-
nage provide a first step towards a workable criminal statute. For example, the
use of the term “control” in the Colorado statute,2″3 although overbroad, focuses
upon the injury suffered by the victim of industrial espionage.2″4 If the informa-
tion is a trade secret, the industrial spy does wrest a form of control from the
original owner. Once the information is made available to others, the owner is
no longer able to reap the full profits. A legislative definition of the term control
of possession with the intent to deprive). As discussed above, the acts of most industrial spies do
not meet this element of the offence of theft. The additional provisions in the Colorado statute do
not serve to expand the concept of “stealing”; to the contrary, they narrow the offence even further
by placing additional limitations on the defendant’s intent. Because most industrial espionage does
not involve “stealing,” then, this portion of the Colorado statute, even with the additional modi-
fying language, will be inapplicable in most cases of industrial espionage.
21’A number of the specific state statutes include “copying” within the proscribed conduct. See,
for example, Ark. Stat. Ann. 5-36-107 (1987); Cal. Penal Code 499c(b)(4) (West 1988); Colo.
Rev. Stat. 18-4-408(1) (1989); Fla. Stat. Ann. 812.081(2) (West 1990); Ga. Code Ann.
16-8-13(b)(2) (1988); Mass. Gen. Laws Ann. ch. 266, 30(4) (West Supp. 1990); Mich. Comp.
Laws 752.772(2) (1990); Minn. Stat. Ann. 609.52 subd. 2(8) (West 1990); N.Y. Penal Law
165.07 (McKinney 1988); Okla. Stat. Ann. tit. 21, 1732(A)(b) (West Supp. 1990); Pa. Stat. Ann.
3930(b)(2) (Purdon 1983); Tenn. Code Ann. 39-3-1126 (1982); Tex. Penal Code Ann. tit. 7,
31.05(b)(2) (Vernon 1989); Wis. Stat. Ann. 943.205(l)(b) (West 1989).
2121n a few states, it is enough to show that the defendant intentionally copied the secret (see,
for example, Minn. Stat. Ann. 609.52 subd. 2(8) (West 1990); Tex. Penal Code Ann. tit. 7,
31.05(b)(2) (Vernon 1989)). Most states, however, have established a higher standard. A common
formula is to require a showing that the defendant acted with the intent either to (1) deprive the
owner of control of the information, or (2) appropriate the secret to the defendant’s own use (Colo.
Rev. Stat. 18-4-408(1) (1989); Okla. Stat. Ann. tit. 21, 1732(A)(b) (West Supp. 1990); Tenn.
Code Ann. 39-3-1126 (1982); Wis. Stat. Ann. 943.205(1)(b) (1989)). Neither of these formula-
tions capture the essence of what is morally objectionable about industrial espionage. See Trade
Secrets Report, supra, note 3 at 204-209.
213Colo. Rev. Stat. 18-4-408 (1989).
214For a discussion of why the term “control” is overbroad see text accompanying note 181-82.
1991]
TRADE SECRETS
that focuses on the monopoly value inherent in the trade secret, could be used
to prosecute industrial spies. Similarly, the Ohio statute,” 5 although subject to
criticism for its possible application in cases of honest reverse engineering, nev-
ertheless focuses upon the perceived wrongs committed by the industrial spy.
Although not yet enacted, the Trade Secrets Report of the Alberta Institute
has also culminated in a set of proposed criminal law reforms to deal with
industrial espionage.”16 Too complex to discuss in this article, the proposals
come closer than any existing law to isolating and defining those types of indus-
trial espionage that should be subject to criminal sanctions. The proposal would
create two separate offences. The first, which focuses upon the misappropriation
of a trade secret, would be analogous to existing theft offences. It would allow
prosecution of anyone who “fraudulently and without colour of right acquires,
discloses or uses the trade secret of another person” with the intent to deprive
that other person of either control of the trade secret, or of an “economic advan-
tage associated with the trade secret. ‘217 Reverse engineering would be explic-
itly included as a defense.218 The second offence would cover a situation where
a defendant fraudulently induces an owner to disclose a trade secret. The pro-
posal also includes a lesser offence under each of the above provisions in cases
where the defendant did not know, but should have known, that the information
was a trade secret.219
The proposals are not perfect. For example, by speaking in terms of a dep-
rivation of control, they fall prey to the same problems of interpretation that
plague existing American statutes.2 In addition, they are limited to trade
secrets, and are therefore of little use in protecting confidential information that
does not have monopoly value. Nevertheless, the proposals represent a willing-
ness to admit that traditional property offences cannot be used effectively
against industrial espionage. Instead of speaking in terms of theft and vandalism
–
the proposals create parallel offences that
protect the major values associated with trade secrets. Once the value to be pro-
tected is isolated, the mens rea of the offence –
is
roughly the same. This would lessen the burden on courts and prosecutors when
changing over from the existing criminal law to the new specific offences.
crimes that focus on possession –
the “intent to deprive” –
It is the opinion of the author that the Alberta proposals are fundamentally
sound, and merely need to be fine-tuned to resolve some of the problems. The
proposals could then serve as an impetus for further discussion and refinement
2150hio Rev. Code Ann. 1333.51.
216Supra, note 3 at 262-64.
2171bid. at 262.
21Slbid. at 263.
2191bid. at 264.
220See text accompanying note 189.
REVUE DE DROIT DE McGILL
[Vol. 36
of the means to punish industrial spies as criminals in Anglo-American
jurisdictions.
Conclusion
The debate that has ensued as to whether information is property has been
reflected in a series of recent decisions. These cases have attempted to apply the
criminal law of property to defendants who have wrongfully revealed or discov-
ered confidential information held by another. They demonstrate some of the
conceptual difficulties involved in applying laws created to protect tangible per-
sonal property to information.
While confidential information is in some respects similar to tangible prop-
erty, this article has attempted to demonstrate that concluding that information
is or is not property is not essential. The term “property” is shorthand for saying
that society will in certain contexts protect current owners of information from
the claims of others. Therefore, it becomes apparent that most industrial espio-
nage should not be categorized under either of the traditional criminal offences
of theft or vandalism, which are designed to protect either the possession or use
of property. As seen, the typical case of industrial espionage does not result in
the owner losing either possession or use of the information.
It is clear that an owner may be injured by an industrial spy. If the infor-
mation is a trade secret, the loss of confidentiality may deprive the owner of a
large portion of the value of that information, namely, the ability to reap addi-
tional profits because of the edge that the information gives the owner over
competitors. A new set of statutes is needed that recognize the unique values of
confidential information and define mens rea and actus reus in terms designed
to protect these values. Although several states have attempted to draft such
statutes, their attempts to date rely too heavily on concepts borrowed from the
traditional criminal law of property.
Although the final decision to protect all confidential information under
criminal law rests with the legislatures, this article has identified the prevalent
conceptual difficulties involved with the task and offered direction with respect
to applying criminal law to industrial espionage.
