The issue of what can and cannot be patented will never receive a clear answer, as the bounds of patentable subject matter continuously evolve and change. This article will attempt to summarize the current state of US patent law in late 2009.
Patents are legal instruments, issued by the United States Patent and Trademark Office (USPTO), whose intent is to encourage innovation by granting a temporary monopoly on an invention in exchange for revealing how the invention works. Thomas Jefferson believed strongly that “Ingenuity should receive a liberal encouragement”. He saw to it that the requirement for a patent process was written into the US Constitution, and drafted the Patent Law of 1790. Article I, Section 8 of the US Constitution states, in part that Congress has the power:
To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.
Patent law, with very few exceptions, describes what can be patented, rather than what cannot. The issues encountered are quite complex, and often do not emerge naturally from the patent law [Title 35 of the United States Code paragraph 101, abbreviated as 35 U.S.C. 101]:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Most of the “conditions and requirements” concern such issues as who can apply for a patent, how to tell who are the legal inventors, how to settle issues of priority, and the like. We are going to concentrate on the subject of the patent claims, and assume that all other legal requirements are satisfied.
The Constitutional language requires that the claimed invention must be in the ‘useful arts’. The invention must therefore have some value in the real world, it cannot simply be an abstract idea or concept. This overlaps with the legal requirement that the patent disclosure must enable someone ‘skilled in the art’ to use the invention without too much experimenting. Both products and processes to make products can be patented.
To complicate matters, there are three different kinds of patents, utility, design, and plant patents. Utility patents are the ones with which people are most familiar, intended to cover inventions which can be used in themselves. A new mousetrap would be an example of an invention that would be the subject of a utility patent.
Design patents cover new, original, and ornamental designs for an article of manufacture. An example would be the design of a specific automobile hood ornament. If a design also conveys additional novel function to the article of manufacture, it might also be the topic of a utility patent.
Plant patents are a bit unusual, seeing that a plant found in nature cannot be patented. However, a plant patent can be obtained to protect a plant found in nature together with a method for asexually propagating that plant. The idea here is that it is the method developed for asexual propagation that raises the simple discovery of the new plant to the level of a patentable invention. The monopoly grant associated with a plant patent only extends to the sale or use of plant products which were grown by asexual propagation.
Plant varieties which are bred to have new properties can be protected under the Plant Variety Protection Act of 1970 even if they propagate sexually or by tubers. The new varieties are required to be genetically stable. This protection, however, is technically not implemented by the issuance of a patent.
There are some general guidelines as to what cannot be patented, although there are many exceptions. For example, the Supreme Court has determined that the laws of nature, natural phenomena, and abstract ideas (such as mathematical algorithms) are not patentable. However, a process, method, or product which uses laws of nature, natural phenomena, or abstract ideas to perform a real-world function might well be patentable. A theory of antigravity would not be patentable, while an antigravity generator would.
There are court decisions which have found that an invention that is illegal to practice or that would contravene public policy is not ‘useful’, and hence cannot be patented. This case law is very vague, and the boundaries established by these decisions are unclear.
A special provision appears in the Atomic Energy Act of 1954, which states, in part, that:
No patent shall hereafter be granted for any invention or discovery which is useful solely in the utilization of special nuclear material or atomic energy in an atomic weapon.
There is considerable pressure to introduce similar legislation to cover some portion of biological patents, but Congress has not been able to reach a consensus at this point (2009). A law was passed in 2004 prohibiting the patenting of human organisms, but this did not extend to parts of humans (for example, human stem cells can still be patented).
Returning to the topic of antigravity brings us to an interesting area, that of inventions which are generally thought to be inoperative, that is, the invention does not produce the results claimed for the invention. Such an invention would not be useful, and thus would not be patentable.
However, an invention by definition is a novel development providing new functionality. As such, it is quite possible that the utility of such an invention might not be obvious to someone who is at least nominally ‘skilled in the art’ (a phrase encountered often in patent law). Who is ‘skilled in the art’? Well, if a purported antigravity generator was composed of electronic equipment, an electrical engineer or a physicist would probably be considered ‘skilled in the art’, even if they had no experience with (or belief in) antigravity.
To make a clear illustrative example, we will consider the fate of a purported perpetual motion machine in the patent system. Contrary to common belief, there is no patent law, rule, or practice that automatically declares a perpetual motion machine to be unpatentable. However, such an invention must meet a stricter than normal demonstration of utility.
Let’s say that a patent application claiming the invention of a PMM is under consideration at the USPTO. Any patent application must provide evidence that the invention works.
The USPTO is required to presume that the described utility of the invention is valid. To challenge this presumption, they have to produce evidence that the invention cannot work. In the case of a perpetual motion machine, they might refer to a fundamental conflict between the claimed invention and the second law of thermodynamics (which is ‘well-known in the art’), and note that one ‘skilled in the art’ would consider the invention to be “incredible in view of contemporary knowledge”. The level of evidence here is not ‘beyond a reasonable doubt’, but rather the ‘preponderance of the evidence’. This would establish a prima facie case of inoperability, and the application would be rejected on that basis.
All is not lost, however. The inventor, unconvinced because he is sure his invention works, challenges the rejection. The burden of establishing utility now shifts over to the inventor, who must provide additional evidence which will convince one ‘skilled in the art’ that the invention provides utility as claimed. The USPTO may offer guidance as to what sort of additional evidence is needed, but the burden of shifting the ‘preponderance of the evidence’ toward the claimed utility rests squarely on the inventor.
More fantastic inventions would naturally require stronger evidence to overcome contradictions with well-established physical law. It is likely that receiving a patent on a perpetual motion machine would require thorough testing of a working model. However, there is nothing in the patent law which prohibits issuing a patent for a perpetual motion machine.
Computer-related inventions also raise a number of questions. The Supreme Court has clearly decided that the mathematics of a computer algorithm cannot be patented. Further, a computer program in itself cannot be patented. However, once a computer program is recorded in a computer-readable format, it is considered a “computer element which defines structural and functional interrelationships between the computer program and the rest of the computer which permit the computer program’s functionality to be realized”, and can therefore be patented [Lowry, 32 F.3d at 1583-84, 32 USPQ2d at 1035].
In practice, if a computer program is claimed as the instructions for a computer-controlled process, the computer program may be patentable as part of a process claim. If a patent claim covers a computer containing a computer program which will instruct the computer to perform a useful function, the computer program may be patentable as part of a product claim. Thus, a word processing computer program in itself cannot be patented. However, a computer programmed with a word processing program can be patented and protected as part of a device for word processing. Similarly, a process to do word processing using a computer programmed with the word processing program can be patented, again providing protection for the word processing program.
Computer-based methods also hold a central position in the current debate over the role of business method patents. Traditionally, it was held that business methods were unpatentable, under the argument that they did not represent technological inventions. However, as computer-based implementations of business methods became more common, the USPTO decided that they would no longer try to determine if a given invention was a business method, because the distinction between a business method and a technological invention was becoming blurry.
This line is still obscure, and a new case is before the Supreme Court, whose decision is hoped to clarify the situation. Since 1998, the case law declared that any business method that produced a ‘useful, concrete, and tangible result’ is potentially patentable. For example, the case involved in this decision turned on a new business method for establishing a price on assets held by a family of mutual funds. Since that time, avalanches of new business method patents have been issued.
The patenting of business methods is currently under attack in in Re Bilski, a case which is currently (November 2009) being decided by the Supreme Court. The lower courts found that, although business methods are potentially patentable, the ‘useful, concrete, and tangible’ test for patentability is not the correct test. Rather, they reintroduced a ‘machine or transformation’ test.
This test says that a claim to a process qualifies to be considered for patenting if and only if it either it is implemented using a machine that is specifically designed and adapted to carry out the process in a new and unobvious way or transforms an article from one state to another. The usual interpretation is that implementation of a business method on a general-purpose computer does not satisfy the first part of this test. As a result, very few existing business method patents would remain valid. There are many remaining questions concerning patenting business methods, which will remain in Limbo at least until the Supreme Court publishes a decision on Bilski.
Biological patents are a relatively recent development in patent law which is currently the center of a cyclone of controversy. Prior to 1980, living organisms were, with very few exceptions, held to be nonpatentable subject matter. Biochemicals purified from animals or plants could be patented, as were epinephrine (1903) and insulin (1921), but not the organisms which produced the raw materials.
Patenting of DNA extracted from living organisms, including genes and genomes, is largely justified by this same argument. The genetic material is extracted and purified from the living organisms, and, in the absence of new legislation, only differs from other biochemicals in chemical complexity. As a result, over 3 million patent applications have been filed in the United States for genetic materials. At present, human genetic material can, and is being, patented.
In the landmark case of Diamond vs Chakrabarty, the Supreme Court overturned centuries of patent practice and precedent by holding that:
A live, human-made micro-organism is patentable subject matter under [Title 35 U.S.C.] 101.
This case specifically concerned the patentability of a genetically modified bacterium which could eat oil spills, but impact of this 5-4 decision of the Supreme Court was extremely broad. In essence, the Court confirmed that ‘Congress intended statutory subject matter to include anything under the Sun that is made by Man’.
Since the Diamond vs Chakrabarty decision, the patenting of living organisms has proceeded apace. The first multicellular animal to be patented was a genetically modified strain of mice developed at Harvard Medical School to provide a new animal model for cancer studies. The patent was issued in 1988. It is now common for patents to be issued covering any mammal containing an extra gene inserted to produce an animal having a particular genetic attribute (humans, of course, are excluded as mentioned above).
Perhaps the leading edge of current controversies over lies in the area of patenting human stem cells. In particular, there is a class of human stem cell which is called totipotent. A totipotent embryonic stem cell can develop into a human being, so it is a bit disturbing to think that such a stem cell can be owned. Although the European Patent Office has decided that human stem cells should not be patented, the USPTO has granted a significant number of patents on such cells, some of which have survived strong legal challenge.
Patent law in the United States is perhaps as inclusive a system in terms of what can be patented as currently exists in the world. The precise location of the boundary between patentable and unpatentable subject matter will continue to evolve in response to advances in technology and changes in the public perception as to the appropriate role of patent protection.