faces, wedging devices, and other arrangements to facilitate removal of ice from the container." 101 Later as the home refrigerator came into use-inventors devised a variety of constructions particularly adapted to home use. These applied eccentric, cam, or wedging devices to lift the divider bodily out of the tray, divider constructions that could be deformed or flexed to free the cubes, flexible trays that could be twisted or peeled away from the ice, and many other arrangements.102 All of these patents have now expired, leaving the constructions freely available to the industry. At the present writing a tray construction using a two-part longitudinal divider enjoys considerable popularity 103 While this construction has been unusually successful, it shares the field with a variety of arrangements based directly on the principles of the expired patents-and the great bulk of currently live ice cube tray patents are directed to very specific and competitively unimportant variations on the expired art.10

Little need be said about patents to inventions not economically worthy of use. Such patents exist because it is impossible to predict which inventions will become commercially important and which will not. A patent may be directed to a product in use and of importance at the time of filing which has since been displaced by subsequent developments. Or a patent may relate to an invention having apparently great promise which has not materialized. Patents of this kind serve a useful purpose in encouraging the exploration of all avenues. In terms of economic power they are unimportant.

The inescapable over-all fact is that the practical economic reach of letters patent is limited. Even the patents of comparatively great commercial importance-such as those of the above examples-fall short of conferring "monopoly" in the antitrust law sense or in any odious respect. Rather, they operate in a market context within which product competition can and does take place. Far from being free to fix a monopolistic price, or otherwise ignore the activities of competitors, the owners of even these important patents must face the competition of alternatives-each having its own advantages and disadvantages in relation to the patented product or process as well as the all-important fact that competitors will devise superior noninfringing products and processes.

B. COMPULSORY LICENSING MYTH VERSUS FACT

The subject of compulsory licensing is a false issue that has long plagued discussions of the patent law. To some the subject has become a rallying point for criticism of the patent system and uninformed conjecture about "suppressed" commercially valuable inventions.1 105 Others have with equal vehemence regarded compulsory licensing of any kind and in any setting as a departure from all that

101 See, e. g., patents 529,346 (1894) and 961,781 (1910). For a partial collection of the patents in this field see Smith, Supplementary Material on Patent Law, Overbeck Co. (1953).

102 See e. g., Patents 1,675,599 (1928); 1,738,162 (1929); 1,830,260 (1931); 1,879,400 (1932); 1,932,731 (1933); 1,930, 680 (1933); 2,028,047 (1936).

103 The two-part longitudinal divider construction is shown in principle in patent 2,196,476, expiring in April 1957.

104 The automobile industry illustrates another broad field where narrow improvement patents predominate. See, e. g., the analysis of the patents of General Motors Corp. at pt. 2, ÎNEC hearings, pp. 691-697. Again, there is an occasional improvement patent of considerable commercial importance. See, e. g., Kesling v. General Motors, 164 F. 2d 824 (8th Cir. 1948).

105 See, e. g., Vaughan, Economics of Our Patent System (1925) p. 172 and Vaughan, The United States Patent System, Legal and Economic Conflicts in American Patent History (1956) p. 241.

has made the American patent system effective.106 The result is often a debate between theoretical extremes in complete disregard of the practical facts.

Two basic considerations bring the matter into perspective. The first is the fact that effectiveness of the patent system in certain areas-notably the small manufacturer or the newcomer-rests on freedom from compulsory licensing. The second is the fact that at the present time the patent right of exclusion is limited in various respects. Indeed, all the legal tools required to handle abuses of patent rights that might otherwise justify a compulsory licensing statute are presently available.

The difficulty with unqualified compulsory licensing is that it would reach phases of patent system operation that depend for their economic value on the patent right to exclude. The newcomer or small manufacturer needs more than the mere right to royalties— when and if the invention proves successful. Success with a new product or process entails investment in research, development, and marketing, and a willingness to move forward despite uncertainties and the ever-present risk of failure. In this area a patent subject to compulsory licensing would have little effect in promoting competition, for the possible license royalties would provide slight incentive and in most instances would not serve to justify the effort and risk involved.

And when we consider the patent system in terms of its researchcompelling effect on established business, the threat of possible foreclosure from some activity is far more significant than the risk of penalty in the form of patent royalties. Compulsory licensing would greatly reduce or eliminate this risk. Established business could then relax with the assurance that it could promptly adopt new developments after others had made the investment and effort required to prove their value. This is the very kind of relaxation that a broad competitive public policy should prevent.

Nor can this difficulty with compulsory licensing be overcome by limiting the statutory scheme to cases where the patented invention is not being used. Nonuse of a patented invention is neutral, and should not be equated with any sinister or undesirable purpose. 107 To be sure, a superficially appealing argument can be made that the patentee ought either to use the invention or permit others to do so.10s Thinking of this kind lies behind the court of appeals decision in the Special

106 See, e. g., Separate Views on Patents of Congressman Cole in Report 2181, House of Representatives, 83d Cong., 2d sess., July 12, 1954, to accompany H. R. 9757, amending the Atomic Energy Act of 1946; Spencer, Thinking Ahead: Threat to Our Patent System, 34 Harv. Bus. Rev. 21 (1956).

107 Report of the Attorney General's National Committee to Study the Antitrust Laws, p. 229 (1955). 109 For a judicial expression of this thought see Hoe v. Knap, 27 Fed. 204, 212 (C. C. N. D. Ill. 1886). The unqualified expression of Hoe v. Knap was rejected by the Supreme Court in Continental Paper Bag Co. v. Eastern Paper Pag Co., 210 U. S. 405 (1908), where the defendant in a patent infringement suit contended that relief should be denied because the plaintiff was not using the patented invention. With respect to the argument that the patentee must use the invention or permit others to do so, the court stated:

"But, granting all this, it is certainly disputable that the nonuse was unreasonable, or that the rights of the public were involved. There was no question of a diminished supply or of increase of prices, and can it be said, as a matter of law, that a nonuse was unreasonable which had for its motive the saving of expense that would have been involved by changing the equipment of a factory from one set of machines to another? ***" (210 U. S. at p. 429).

The Court of Appeals for the First Circuit had reached a similar conclusion. Judge Aldrich in the Court of Appeals had dissented because he considered the evidence to show that "the complainant stands in the common class of manufacturers who accumulate patents merely for the purpose of protecting their general industries and shutting out competitors," (150 Fed. 741, 745 (1st Cir. 1906). For more complete discussion of this and other cases bearing on the subject see Frost, Legal Incidents of Non-Use of Patented Inventions Reconsidered, 14 Geo. Wash. L. Rev. 273, 435 (1946).

PART II—THE PATENT SYSTEM IN RELATION TO THE COMPETITIVE ECONOMY

In my opinion, it is a vital necessity that there be effective competition among the companies in research and development, as well as in market trade. The consuming public has much to gain from technological improvements, and the security of the country, as well as national prosperity, are largely dependent upon continued scientific advances. And market competition, itself, can be readily impaired if a single company secures patent control of the industry. * * * 68 American economic public policy is based on a general rule of freedom to compete in the manufacture of any product and in the use of any process. The assumption-generally valid-is that when restraints on competition are eliminated normal economic forces will give rise to the desired highly competitive order. The antitrust law proscriptions against restraint of trade and monopolization constitute the major general implementation of this policy. The right to exclude others incident to letters patent is superficially inconsistent with this broader economic public policy. Questions accordingly arise with respect to the place of these apparently unlike concepts within the framework of a single economic order.

The cue to resolution of these questions lies in recognizing that the patent system is a stimulus to competitive activity in research and development. Such competition demands a limitation on the general rule of complete freedom. Otherwise freedom to compete becomes freedom to appropriate the research accomplishments of others-and the value of development as a competitive effort is greatly reduced. In short, this is a case where some restraint is necessary to encourage the competitive activity desired. The patent right to exclude precludes appropriation of research efforts to the extent of the subject matter covered by patent rights-leaving all other areas open. In consequence competitive effort in research and development is made worthwhile and an increased pace of competition in such activity maintained. Over-all, the patent system serves the same broad public purpose of stimulating competition as the general rule of freedom. from restraint.

It follows that the patent system is an integral part of the competitive order. However since patents serve as a business asset to be used in competing-the permissible use of such rights must be defined. The following sections deal with these matters-all analyzed in the light of the practicalities of the patent system and the competition the patent system encourages.

A. PATENTS AND ECONOMIC POWER

An understanding of the actual economic power conferred by letters patent lies at the root of any meaningful discussion of the place of the

68 Per Judge Knox in U. S. v. Aluminum Co. of America, 91 F. Supp. 333, 410 (S. D. N. Y. 1950).

patent system in a competitive economy. The economic and legal literature abounds in discussions touching on this subject, much of it abstract and centering about application of terms such as "monopoly" to the rights associated with an issued patent.69 The power to exclude conferred by a patent is meaningful in a market context only to the extent that it gives a practical commercial advantage to the patentee. The focal point of the matter is not whether one word or another more aptly describes what the patent may be theoretically said to confer. Rather, our attention should be directed to the competitive impact of particular patents in the particular market settings of which they are a part. We will therefore look to some of the recent judicial decisions shedding light on this subject, bearing in mind that it is only in the context of actual events that analysis of the economic power associated with letters patent can have significance in relation to a competitive economy.

A dramatic illustration of the vigor of product competition incident to the patent system is found in the leakproof dry cell litigation and its aftermath.70 In 1940 Anthony patent 2,198,423 issued to Ray-OVac Co., then a comparatively small manufacturer of dry cells for flashlight and other uses. The patent included broad claims to a dry cell with a "protecting sheet-metal sheath *** tightly embracing" the cell proper to prevent leakage." Despite its simplicity, the battery was a marked break from past thinking. Previously both the industry and users had accepted the propensity of exhausted cells to swell and leak-as indeed most persons know from sad personal experience. Most manufacturers had applied notices to their flashlight batteries warning of this effect. The Anthony patent construction went far to overcome this problem. Indeed, Ray-O-Vac was able to advertise the new battery as "leakproof" and to guarantee replacement of the entire flashlight if damaged by a leaky cell. The battery was an immediate commercial success. When a competitor adopted the patented construction, Ray-O-Vac brought a patent infringement suit which ultimately resulted in 1944 in affirmance by the Supreme Court of the lower court decisions of patent validity and infringement.72 Ray-O-Vac issued ro licenses, and realized its profits from the patent by way of manufacture of the patented cell, both for sale under its own name and for private brand sale.

The net effect of the Supreme Court decision and the Ray-O-Vac policy was to create a crisis in the industry. The Ray-O-Vac competitors the leading battery manufacturers-faced an important patent to which they could not gain access. Hard pressed, they turned their attention to the development of noninfringing batteries which they could advertise as leakproof. By 1948 one of the defendants in the

69 For a collection of the materials see Oppenheim, Cases on Federal Antitrust Laws (1948), p. 464 et seq. On the futility of analysis based on the term "monopoly" compare the following passages: "Though often so characterized, a patent is not, accurately speaking, a monopoly, for it is not created by the executive authority at the expense and to the prejudice of all the community except the grantee of the patent. The term monopoly connotes the giving of an exclusive privilege for buying, selling, working or using a thing which the public freely enjoyed prior to the grant. Thus a monopoly takes something from the people. An inventor deprives the public of nothing which it enjoyed before his discovery, but gives something of value to the community by adding to the sum of human knowledge. ***"' Justice Roberts in U. S. v. Dubilier, 289 U. S. 178, 186 (1933).

per Mr.

"It is the protection of the public in a system of free enterprise which alike nullifies a patent where any part of it is invalid *** and denies to the patentee after issuance the power to use it in such a way as to acquire a monopoly which is not plainly within the terms of the grant. The necessities or convenience of the patentee do not justify any use of the monopoly of the patent to create another monopoly. **** per Mr. Justice Douglas in Mercoid v. Mid-Continent, 320 U. S. 661, 665 (1944).

70 The Goodyear Tire & Rubber Co. v. Ray-O-Vac Co., 321 U. S. 275 (1944).

71 Claim 1, patent 2,198,423. See also claims 2, 4, and 5.

72 Footnote 70, supra.

litigation had such a battery in quantity production.73 Another competitor had a battery on the market in 1950 for which it gave the same flashlight replacement guarantee that had been emphasized by RayO-Vac in obtaining the patent from the Patent Office and before the courts in securing the favorable decisions.74 Throughout there has been vigorous competition and effective selling by all manufacturers. When Ray-O-Vac contended that the term "leakproof" was distinctive of its batteries, competitors satisfied the Patent Office that this was not true.75 Ray-O-Vac has moved up in the industry, but it has not enjoyed any broad range freedom from competition in the sale of flashlight batteries. And now-with the patent expiring early in 1957-the company faces not only competition in the manufacture of the patented construction, but also competition based on improvements made by others in efforts to design around the Anthony patent. When we turn to the Linde case 76-the most recent Supreme Court decision upholding a patent-we find the same pattern within the confines of the case itself. The patent related to a welding flux particularly desirable and unusual in that it permitted welding of plates of great thickness and produced no visible arc during the welding process." The District Court, the Court of Appeals, and the Supreme Court, all held the patent valid and infringed.78 And in decision on rehearing the Supreme Court in sweeping language adhered to the "doctrine of equivalents" to place a broad construction on the patent claims.79 Yet, by the time the case was again before the district court on remand, the defendant had devised an "unreacted" flux which it sold in competition with the patented flux. This "unreacted" flux was ultimately held to be outside the scope of the patent.80 While we cannot here pass judgment on the merits of the "unreacted" flux vis-a-vis the patented flux, the fact that it was sold in competition with the patented flux and was charged as an infringement on the basis that it did the same thing, indicates that the practical differences, if any, were minor.

The recent Cellophane decision 8-though not a patent infringement action-brings out even more forcefully the limitations on the

73 General Dry Batteries, Inc. et al v. Ray-O-Vac Co., 104 U. S. P. Q. 347 (Comm'r of Patents, 1955). The General Dry Batteries case gives a particularly interesting picture of the developments in the industry after the Supreme Court decision. It arose on an opposition filed by four competitors of Ray-O-Vac to a trademark application to the term "leakproof." Assistant Commissioner Leeds found that the term "leakproof" had acquired no secondary meaning as indicating dry cells made by Ray-O-Vac. In partial support of this conclusion the decision lists a great number of patents directed to "leakproof" dry cells (104 U. S. P. Q.349-350). Burgess Battery Co. rested its position in the opposition in part on the fact that "Burgess has been for some 10 years engaged in the development, manufacture, and sale of truly leakproof dry cells and batteries" (104 U. S. P. Q. 349).

On May 7 and 8, 1956, the writer purchased flashlight batteries at a number of stores in downtown Chicago. The batteries bore seven different trademarks. Four (Ray-O-Vac, RCA, Sears, and Wards) were manuactured by Ray-O-Vac under the Anthony patent. Three (United States Electric Manufacturing Co., National Carbon Co., and Burgess Battery Co.) were not made under the patent. The United States Electric battery bore patent number 2,410,826, issued in 1946. All of the batteries except the RCA battery bore the same guaranty that a complete new flashlight would be supplied if damaged by the battery. 74 General Dry Batteries, Inc. et al v. Ray-O-Vac Co., 104 U. S. P. Q. 347, 353 (Comm'r of Patents, 1955). 75 Footnote 73, supra.

76 Graver Tank & Mfg. Co. v. The Linde Air Products Co., 366 U. S. 271 (1949), on rehearing (339 U. S. 605 (1950).

77 See 86 F. Supp. 191, 192. The patent in suit also involved a number of claims to a method of welding. These were held invalid by the district court (86 F. Supp. 191, 196, et seq.); held valid by the Court of Appeals (167 F. 2d 531); and held invalid by the Supreme Court (336 U. S. 271).

78 Footnotes 76 and 77, supra.

79 339 U. S. 605 (1950).

80 The district court held that the new flux was an infringement. Union Carbide and Carbon Co. v. Graver Tank & Mfg. Co., 106 F. Supp. 389 (1951). The Court of Appeals reversed at 196 F. 2d 103 (1952). The district court concluded that the new fluxes "are substantially the same with respect to silicates as required by the patent; that such fluxes infringe because the elements in the compositions are, first, substantially the same thing as required by the Jones et al. patent; second, operate substantially the same way; and, third, they substantially accomplish the same results" (106 F. Supp. 393). The Court of Appeals decision rests on disclaimers made earlier by the plaintiff to distinguish the prior art and not on any conary conclusion as to the similarity of operation.

United States v. E. I. du Pont de Nemours and Co., 76 Sup. Ct. 994 (June 11, 1956).