Ross Winans took out a patent in 1847 “for making the body of a burden railroad car of sheet iron, the upper part being cylindrical, and the lower part in the form of a frustum of a cone, the under edge of which has a flange secured upon it, to which flange a movable bottom is attached” (see, description in Winans v. Denmead1). Defendants had made cars similar to Winans’, except that the form was octagonal instead of circular. In the infringement lawsuit, the U.S. Supreme Court determined that Winans’ patent was a new mode of operation, and what was claimed was merely a form separable from substance, and thereby covering various forms if the forms were “so near ... as substantially to embody the patentee's mode of operation, and thereby attain the same kind of result as was reached by [the] invention.” The analytical standard was later distilled into Function/Way/Result or F/W/R for the Doctrine of Equivalents in finding for patent infringement.
　　The rise of the Doctrine entailed profound impact on future patent practice. On the one hand, it effectively strengthened protection of the patentee’s rights, but on the other, it also created tremendous public inconvenience of uncertainty as to where the patent power stops.
　　Ⅰ.Definiteness of the range of patent protection
　　Speaking of the range of patent protection, a general analogy is the method of measuring one’s land property by using “metes and bounds.” In fact, the development of the patent system resembles that of the real estate system in many ways. In the pioneering days, an early settler came upon a new land. He built a new house, cultivated a new field, and raised a few chickens. He did not have neighbors, and worried little about intruders. He had no reason to build a fence or mark the boundaries of his property - “My property is all that I need, all that I can control, and all that is connected to the central point of interest - my home,” he would claim. And if this home becomes the center of town in the 21st century, the property can no longer be measured this way, but precisely by inches.
　　This is similar to the patent system development. Take the U.S. practice, for example. From the late 1700’s to the 1830’s, life was still difficult, and technical inventions were all but a few. There were understandably no formal requirements for patent documents2 except for being new and useful in a brief statement for the specification, sometimes accompanied by a drawing. Claims were no required, and when there was indeed one, it was usually an afterthought in the wording like “as shown in the specification.”3 The range of the patent protection is more or less based on the written description of the invention. This is what is widely referred to as the “central claiming” era. By the 1850’s, claims were officially required; but the modern definiteness of “clearly pointing out and distinctly claiming” requirements were not codified into the statute until the 1870’s when the transition from “central claiming” to “peripheral claiming” 4 was complete, thence, “the patentee knows what he has, and the public knows what he does not.”5
　　As a lawful monopoly, the range of a patent must be made clear, which is unexceptionally required by patent laws throughout the world, for, e.g., at least one independent claim, and the claim must be clear. In the Chinese Patent Law, e.g., Art. 26 mandates that the written description must give a clear and complete exposition of the invention….; and the range of protection shall be indicated on the basis of the written description; and Art. 56 further stipulate that the patent is protected in accordance with the range of the claim or claims, and the written description or drawings may be used for interpretation of the claims. The Implementing Regulations under the Patent Law further details that the claim shall point out the technical features, and claim the range of protection in clear and concise terms.
　　Ⅱ.The uncertainties of the doctrine of equivalents
　　As previously stated, the Doctrine is necessary, without which, a copyist could easily avoid infringement liabilities by making insubstantial alterations. However, because of the uncertainty in the “equivalents,” the Doctrine never squares well with the definiteness requirement for patent claims.
　　The criteria for the equivalents are for the “substantial function,” in “substantially the same way,” and to achieve “substantially the same result.” This simply begs the question: how substantial is substantial? What are standards this substantiality may be measured against? Can it be specified in some way?
　　In the famous (or maybe infamous) case of Corning Glass Works v. Sumitomo (1989), 6 the Doctrine was pushed to an extreme. Corning developed a product and a method for making fiber optics which involved reducing impurities in the core so as to increase refractive index (RI) which greatly increased the distance for telecommunication. It was commonly known that by increasing RI, the distance of fiber optic communications may be increased. Defendant Sumitomo achieved this result of increased RI by adding impurities into the layer around the core, which, in technical sense, could be easier to accomplish. The court found that adding impurities in the reflective layer and reducing impurities in the core was to “carry substantially the same function, in substantially the same way, to achieve substantially the same result.” Once handed down, the decision raised many brows in Japan, and critics believed that it was due to anti-Japanese sentiment. The author of this article asked several people on various occasions, “Do you believe that these two products are substantially the same?” The answers invariably were in the negative. Some even believe that the Sumitomo product could be a separately patentable technology. In the view of this author, Corning might have, perhaps, really overlooked what was considered an equivalent, which was evidenced by their filing of the process patent for reducing impurities in the core. In other words, if Corning never contemplated the Sumitomo solutions, it could not have made that invention which was determined as an equivalent, let alone encompassing it into the patent claims. Stated simply, Corning in effect embraced by way of the expansive Doctrine what was outside the scope of its invention. Such an expansive reach is indeed causing uncertainty as to the scope of the patent claims.
　　The Doctrine has been met with a warm welcome in this country. The Supreme People’s Court issued a judicial decree in 2001 interpreting Art. 56 of the Patent Law in connection with definiteness of patent claims by stating that “an equivalent refers to that which in substantially the same way, performs substantially the same function, to achieve substantially the same result as the claimed feature, and that equivalent may be achieved by a person having ordinary skill in the art without creative efforts.” These standards, obviously, substantially follow the U.S. judicially created F/W/R criteria, but the last one is really beside the question in that it is the standard for deciding obviousness of an invention. Does it mean that one has to practice any non patent technology, including all prior arts, in an unobvious way to avoid being accused of infringing a patent by equivalent? Or is it a leftover from the Winans mentality for protecting an inventive principle? Moreover, the Court failed to mention any counterbalance which would limit the expansive application of the Doctrine (such as, e.g., the prosecution history estoppels, prior art defense, whether the equivalent would be supported by what is disclosed in the written description, and whether such added equivalent would qualify under the patentability requirement for novelty and non-obviousness, etc.)
　　In practical areas, some judicial developments in employing the Doctrine can be no less worrisome. For example, the Intermediary Court of Shenzhen decided that a “bowl-shaped” heat radiator is equivalent to a “cylinder-shaped” one in the infringement case of Zalmantechco, Ltd. (S. Korea) v. PCCooler (of Shenzhen) (involving Chinese Pat. No. ZL02800004.8).
　　In another development in The Oriental Action General Factory of Ningbo v. Hardware Products, Inc. of Jiangyin, the Supreme Court determined that:
　　The Written Description which indicates the method of “the blind board being processed by the rotating blade in a mode other than being suspended” does not preclude other modes; it is merely a comparable mode.
　　Whereas, the accused device precisely employed the mode of having the blind board being suspended when processed by a rotating blade. When the testifying experts refused to give their opinions, the Court went ahead with its own finding of technical facts as one of typical equivalents where the patented product solved two technical problems with one device and the accused product solved the two problems with two devices. Moreover, the Court proffered an “inferiority” theory by which an equivalent would be found if the accused device is “inferior” to the patented product. It should be practically improper for a court, when contemplating legal issues involving patents, to offer a judgmental opinion on the technical superiority or inferiority of a product, for that is a matter of many considerations, including such factors as cost and effect, market conditions, etc. To make it worse, the Court commented that an accused product would sometimes be superior and sometimes inferior over the patented product. Then which one is the determining factor?
　　Compare the widely criticized Graver Tank v. Linde Air Products Co.7 case by the U.S. Supreme Court, in which it was found that magnesium silicate was an equivalent substitute for manganese silicate for soldering fluxes. Some commentators joked that two elements are both alkali earth metals, both begin with an “M”, and both have three syllables; other than that, they are in chemistry different metals with totally different properties.8
　　In this author’s view, the Doctrine of Equivalents, absent an objective standard for determination, could be used to extend patent coverage to a “substantially” non-equivalent which is clearly outside the range of the patent claims.
　　A more manageable term for equivalent would seem to be “a substitute with insubstantial difference.”
　　Ⅲ.How to handle the expansive assertion of patent claims
　　In some hundred and fifty some years since Winans v. Denmead, the U.S. patent statute has been revised several times, but the Congress always steered clear of the judicially created Doctrine,9 which pretty much explained its status; the Doctrine is not a principle; rather, it is more an exception to the principle.10
　　It is a generally accepted principle that the patent ought to clearly point out and distinctly claim the range of protection, which has been specifically provided in the patent statutes of various countries. However, woodenly applying this rule in claims construction would sometimes produce unfair result (as, e.g., where an infringer finds a loophole in the patent to evade liabilities), hence the Doctrine. But a closer look at Winans v. Denmead would reveal that the U.S. Supreme Court in 1853 when employing the Doctrine was still struggling with the central claiming mentality, which was to interpret the patent claims primarily according to the written description for a “principle” - a mode of operation - which was not limited by the claim that was merely an embodiment of the principle.
　　The Doctrine of Equivalents, as opposed to the general principle of clearly pointing out and distinctly claiming, is therefore an equitable exception, and as such, it has to be used very cautiously, so as not to swallow the general principle. If, for example, any patentee could extend the patent coverage at will to what is not literally covered by the patent under the Doctrine of Equivalents, the “controlled evil” of patent, then, will be out of control, and become a naked monopoly.
　　Maximization of one’s patent right, of course, is understandably the hope of every patent owner, especially by means of some expansive interpretation of the patent claims when an accused product does not fall within the literal boundary of the patent claims. But this has to be done on basis of clear legal support, and this happens only as an exception to the general rules. The Patent Law has unequivocally mandated in Art. 56 that “The range of protection for invention or utility model patents is determined by the contents of the claims; the written description and drawings may be used in claim interpretation.” What is meant in Art. 56 (1) “determined by” is clearly to set the demarcation against which the rights are measured; and (2) when [and only when] ambiguities arise as to the meaning of the words in the claims, the written description and the drawings may be used for proper interpretation. In other words, the law in the first clause would be nonsense if the second can override it. By the same token, a motorbike could have become an equivalent of a bike (suppose the “bike” discloses the features of rear wheel drive and front wheel steering).
　　The Doctrine of Equivalents in defining the boundaries of the monopolistic powers has in practice brought more ambiguities and uncertainties than to provide effective protection of the patent rights. In the view of this author, such practical failure came as a natural result from the failure to appreciate the nature of the Doctrine, as well as other existing available mechanisms.
　　Ⅳ.Available mechanism for expanding patent rights
　　The Doctrine of Equivalents is a non-definable concept. Theoretically, the equivalent must be substantially the same as the claimed features, i.e., without expanding the claims. But the equitable concept of “substantially” in reality has always been used in expansive interpretation of the claim. There’s no denying of the equitable value in the Doctrine, but the question remains as to where the expansion would stop. The current limitations of the Doctrine (e.g., prosecution history estoppels, which creates more problem than it solves in light of the prosecution history), works neither for the protection of the patent rights nor for clarifying the protection range. It should in fact, in this author’s view, be not too difficult to find a reasonable solution from the currently existing mechanisms.
　　Jurisprudentially, the range of patent protection may not be expanded, which theory has been widely adopted in various national patent practices. But it is a fact that the claim-expanding Doctrine has also been widely accepted as well. Some may say that it does not in theory expand the claims, which, since it so conflicts with the general principle of “clearly pointing out and distinctly claiming,” is not much worth of commenting.
　　Other scholars suggest: for the defense based on prior art in equivalent infringement cases, to ascertain what is claimed by equivalent, and to avoid uncertainties, a “hypothetical claim” examination could be carried out,11 as:
　　1. Starting from the current claims, make a proper interpretation;
　　2. Modifying the claims (sometimes by one limitation would be sufficient) so that it literally reads on the equivalent portion of the accused infringing device;
　　3. Taking the expanded claims in light of the prior art, and evaluating patentability (novelty and non-obviousness); if it fails, the defense would stand, and no infringement under the Doctrine of Equivalent may be found, and vice versa.
　　This solution, being worthwhile of consideration, has its shortcoming in that it overlooked whether the expansion had been sufficiently disclosed in the written description.
　　In the view of this author, a measure, somewhat similar to “reissue” used in American patent practice would be highly recommendable.
　　The reissue scheme is primarily practiced in the United States, which dates back to the beginning of American patent regime. The reissue is used mainly to correct defective claims, and could even be used to expand the claims with the following features:
　　For equitable reasons, an inventor should be permitted to correct his unintentional errors;
　　Because of the errors, the inventor has claimed more or less than he had right to claim;
　　Reissue can be requested by the patent owner, and none other;
　　If request for expanding the claims is involved, it must:
　　Be made within two years after issuance (a Statute of Limitation)
　　Be supported by the disclosure of the written description (inventive scope)
　　Not include what has been abandoned in the initial examination (prosecution history estoppels)
　　Not be used against third parties who have obtained rights outside the originally issued patent (equitable estoppels)
　　For reissue, the original patent must be surrendered, and the amended claims would be examined like any regular claims (i.e., full examination for patentability);
　　A reissued patent takes the remaining part of the original patent term;
　　Like any patent, a reissued patent can also be challenged for validity.
　　The practice of reissue has caused little controversy in the U.S., and has been legislated into the statutory law.12
　　Reissue does not replace the Doctrine of Equivalents, because:
　　1. Expansion of claims through reissue can only be done within two years of the original issuance; the Doctrine could be asserted at any time during the patent;
　　2. In litigation involving patent infringement, the patentee may request the court to interpret the current claims, but it may never ask the court to rewrite the claims, let alone for patentability evaluation of a rewritten portion, which is the professional realm of the patent administration.
　　In fact, reissue has been widely used as a remedy for saving a defective patent when possible infringement is detected, the distinction between reissue and the Doctrine being:
　　1. In both the original patent and the reissued patent, the range of the patent claims remains clear and distinct, whereas the Doctrine makes it uncertain;
　　2. A reissued patent claim cannot exceed what has been disclosed in the written description; whereas, an equivalent can “substantially” be extended to what is not within the inventive scope;
　　3. The cost of reissue is more or less the same as filing the original application, if not less; whereas, cost of justifying infringement by equivalent is much greater in light of litigation expenses;
　　4. Reissue is ex parte in nature, while litigation for infringement by equivalent necessarily involves the alleged infringer as defendant;
　　5. Reissue may be carried out on claims in light potential infringing device so that the modified claims can read on the devices; whereas the Doctrine is asserted for failure to read on the accused device, and its effect is severely limited by various defenses including estoppels, making it a weak assertion;
　　6. A reissued patent stands on a much better footing than an ordinary patent, for any patent that has been twice affirmed by the patent authorities would theoretically have cleared all possible defects; whereas, an unchallenged patent would have to be scrutinized relentlessly in the invalidation process, again in a much weaker position.
　　Taking full advantage of reissue mechanism in the U.S. is not a problem at all, and by doing so, a patentee can still resort to the Doctrine on a reissued patent. Most countries, however, including China, would have to first resolve the ideological problem, i.e., whether the patent claims can be expanded on equity basis, if they really want to settle the difficulties of applying the Doctrine and patent claim certainties. To put it more clearly, expansion of the patent claims is already a reality through applying the Doctrine of Equivalents; the point is whether it can be codified into law through some mechanism like “reissue.”
　　The reissue mechanism is premised on the equity basis, or more precisely, the practice of exceptions to the general rules. A reissue may not be requested unless some defects have been detected; and such defects are the result of unintentional error of the inventor, because of which, the inventor has claimed more, or less, than he has right to.
　　It is the belief of this author that a “hypothetical” examination, if it cannot be immediately adopted by legislation for understandable reasons, could at least provide some working models in the judicial process, such as the trial hearing somewhat like the Markman Hearing, taking testimony from experts like Patent Office examiners so as to achieve similar result of reissue, and to minimize the uncertainties resulting from relying on the Doctrine.
　　Ⅴ.Conclusion
　　It is a fundamental principle that the range of patent rights must be clearly defined, without which, the monopolistic powers of the patent would be unleashed for misuse and extensive waste of judicial resources would result, making the patent a naked tool for constraining competition in the marketplace, and the legislative goal of “promoting science and technology” a nullity. On the other hand, the claims should not be a prison for the patentee; neither should it be a vehicle for intentional copyist to evade punishment of the law. The Doctrine of Equivalent is effective to the extent of closing the loophole so that the patent is not just a “paper patent” for the inventor’s right, but equally important, it should not be overlooked for the effect of causing uncertainties as the range of patent protection.
　　In handling judicial disputes, it would be wise to follow the established principles in the existing framework, and avoid making excessive exceptions. If uncertainties of patent claims could be avoided by available legal remedies to suppress infringement, it would be unwise to resort to other means (such as the Doctrine) that would unnecessarily result in making the range of patent protection less certain.
　　The patent reissue scheme has been a successful practice, and fully operational for about two hundred years in the United States. It has proved to be effective and legislated into the patent statute. Such a scheme is worth studying by patent authorities in all countries, and could be adopted first in the judicial process, and eventually, into the patent law.
　　Endnotes:
　　1. Winans v. Demead, 56 U.S. (15 How.) 330 (1853).
　　2. All U.S. patent documents prior to 1836, the so-called “X-Patents,” were destroyed in the famous 1836 Patent Office fire. Some have been retrieved from various private collections and local libraries since then. So, for this reason, we can hardly have any specific example of a real patent from then.
　　3. See, generally, Janice M. Mueller, An Introduction to Patent Law, CTIIC Pub’g House, 2003.
　　4. It ought to be noted that a general misunderstanding exists as to “central claiming” and “peripheral claiming.” These are primarily techniques for patent claim drafting, not claims construction theories. It has been so misinterpreted that even Patent Office official spoke about a “compromise” method between “central” and “peripheral” in claims construction, which could have been resulted from mistranslation.
　　In U.S., in fact, a court handling patent infringement must first take the patent claims as it finds them, construe them according to their plain meaning to determine the scope of the protection; and when ambiguities arise, they may be resolved against the background of the written description and the drawings, if any.
　　5. Festo v. Shoketsu, Id.
　　6. Corning Glass Works v. Sumitomo Electric U.S.A., Inc., 868 F.2d 1251 (Fed. Cir. 1989).
　　7. 339 U.S. 605 (1950).
　　8. See, Adelman, Rader, Thomas, Wegner, Patent Law, West Group, 1998, at 897 (Notes: Criticizing Graver Tank).
　　9. See, Festo v. Shoketsu, supra (“Congress could legislate the doctrine of equivalent out of existence at time it chooses.”)
　　10. Some commentators use the expression of “the principle of equivalent,” which, in principle, should be a mistake.
　　11. Janice M. Mueller, An Introduction to Patent Law, Chap. 9.
　　12. See, 35 U.S.C. §§ 251, 252.