By: Robert R. Sachs, Jennifer R. Bush
In March 2012, the Supreme Court ruled, in Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U.S. ___ (2012), that a patent claim on a method of determining whether a given dose of a particular type of medication was safe and effective was a “law of nature” and thus invalid. In the following two-part series, we will show that the Court’s characterization of Prometheus' patent as nothing more than a law of nature does not withstand critical analysis.
Prometheus Labs is the exclusive licensee of U.S. patents 6,355,623 and 6,680,302. The patents deal with a protocol to determine the safe and effective dosage for thiopurines, which are medications that treat autoimmune diseases such as Crohn’s disease and ulcerative colitis. These medications metabolize in the patient’s body into 6-thioguanine (“6-TG”). The different rates at which the medications metabolize make it difficult for physicians to determine whether a given dose is too high (and thus toxic) or too low (and thus ineffective). The Prometheus patents’ claims identify the range for safe and effective dosages as those which result in a concentration of 6-TG between about 230 pmol and about 400 pmol per 8x108 RBC red blood cells (“RBC”). The patents claim no underlying biological process that accounts for the differences in metabolization.
The Supreme Court ruled that Prometheus’ claims were not eligible for patent protection, holding that Prometheus did nothing more than identify a law of nature. However, The Court’s analysis of the claim as a law of nature is based on a superficial understanding of the scientific status of laws of nature.
The goal of the natural sciences, and of physics in particular, is to provide “comprehensive theories that recognize only a limited range of natural properties…Of course, the discovery of natural properties is inseparable from the discovery of laws.”[ii] To understand what counts as a law of nature, we must distinguish three types of statements:
Accidental generalization: All robin eggs observed to date are greenish-blue.
Universal truth: Every robin egg, both those observed in the past and those to be observed in the future, is greenish-blue.
Law-like statement: It is a law that robin eggs are greenish-blue.
An accidental generalization simply summarizes our past experience—that every robin egg found to date has been greenish-blue—and leaves open the possibility that some robin egg found in the future will not be greenish-blue. This is the weakest of these statements, as it is merely a summary of our past experience. Intuitively, it provides no guarantee that this state of affairs must be this way, as it leaves open the possibility that some robin egg could be found in the future that is not greenish-blue. It is the common sense equivalent of the cautionary rule that correlation does not imply causation. An assertion that something is a law of nature simply because every observation of relevance has certain properties (e.g., all the robin eggs observed are greenish-blue) is never sufficient.
A universal truth is stronger. Not only has every observed robin egg been greenish-blue—a universal truth predicts that every one in the future will be the same color. This assertion has more force, but it provides no reason that necessitates this result. Intuitively, a robin egg could be, for example, pale white, due to some genetic mutation. Merely stating the natural relationship in conditional form—“If x is a robin egg, then x is greenish-blue”—does not make it a law of nature.
A law-like statement means something more that just the accidental generalization or the predictions of the universal truth. Broadly speaking a law-like statement must have some level of necessity, the statement must be true for some intrinsic and/or extrinsic reason or property.
Without distinguishing among accidental generalizations, universal truths, and law-like statements, one cannot explain what makes a law of nature. “What is a Law of Nature” thus becomes “What properties are required of a law-like statement to make it a law of nature?” The philosophical literature discusses a number of such conditions:
None of these properties is entirely undisputed. But in general the more of these properties that are met by a given statement, the stronger the case for properly identifying it as a law of nature.
There are a number of different approaches to the question of whether there are laws of nature at all, and what properties they would have. Realists argue that laws of nature do in fact describe reality. Antirealists contend that there are no laws of nature, or at least that we cannot know what they are. I will very briefly summarize some of the theories in each camp. I feel no shame in only being able to set out the major approaches; one philosopher prefaced his 350 page book on the question with “I cannot possibly examine all extant accounts of laws, and [] new ones could spring up like toadstools and mushrooms every damp and gloomy night.”[iv]
Realists take a normative view of laws: one begins with a definition of what a law is (e.g., using various criteria such as above) and then tests various scientific claims against such definition. Among the realists, there are two main approaches: Regularity theory (or a “systems” approach) and “universals.” The regularity view is an essentially empirical view that “laws of nature are regularities among events or states of affairs, and law-statement, the linguistic expression of a law, is a description of a regularity that is a law.”[v] John Stuart Mill (1843) offers this elegant expression:
What are the laws of nature? may be stated thus: What are the fewest and simplest assumptions, which being granted, the whole existing order of nature would result? Another mode of stating it would be thus: What are the fewest general propositions from which all the uniformities which exist in the universe might be deductively inferred?[vi]
Using this framework, the question then becomes, “How do we identify which law-statements qualify as laws of nature and which are lesser accidental generalizations or universal (but still not lawful) truths?” One answer is that the “best system is the one that strikes as good a balance as truth will allow between simplicity and strength….A regularity is a law [of nature] if and only if it is theorem of the best system.”[vii] In short, a statement is a law of nature if it would be included in the “best” system of scientific principles and facts needed to explain a given phenomena. The criteria offered for “best” include simplicity and strength. One virtue of this approach is that it offers a satisfying application of Occam’s Razor. If statement A can be deduced from statements B and C, then A is not a law of nature. For example, assuming that “All humans are mammals” and “all mammals have hair” are laws of nature, then “All humans have hair” is not a law of nature because it can be logically derived from the former.
Scientific facts likewise are not laws in themselves because once the laws that describe the facts are expressed, the individual facts are not needed, and their addition only adds complexity to the scientific theory with no gain in strength. For example, given the Ideal Gas Law, PV=nRT, expressing the relationship in an ideal gas between pressure, temperature, volume and moles, it is unnecessary to hold that all of the facts covered by the law are also laws of nature, e.g., that one liter of argon contain 0.043 moles at one atmosphere of pressure, has a temperature of 10° C. Accidental generalizations are similarly weeded out because they are unnecessary additions to their underlying theory. For example, if we find a lawful explanation for the color of all eggs based on the presence of certain chemicals in eggshells, then we would no longer include the statement about the color of robin eggs as a putative law of nature.
The systems approach has certain intuitive appeal since it conforms well with our understanding of how science operates using empirical methods to explain natural phenomena: “Physics, and science more generally, is founded on regularities. Scientists study nature, find patterns, and codify these patterns into natural laws.”[viii] However, it is not without problems. First, the notions of balance, simplicity and strength are rather fuzzy: what does it mean to “balance” simplicity and strength? How does one know when the loss of simplicity resulting from the addition of candidate law-statements is beneficially outweighed by the increase in strength? Second, these very considerations are subjective and hence make the identification of laws of nature dependent on pragmatic concerns, contravening the “objectivity” criterion (g) above. Third, in what language do we evaluate a candidate law of nature? Statements that are simple in one language (natural or otherwise) may be complex in another, and it may be impossible to find or create a language that is not inherently biased toward certain forms of locution.
The primary realist alternative to the systems approach is the “universals” approach. A law of nature is one that states a non-logical, yet necessary relationship, between universal statements. Our statement about robin eggs would be a law of nature, for example, if there was a non-logical reason that essentially “forced” such eggs to be greenish blue. Universals theory identifies accidental generalizations because they are contingent on historical or circumstantial facts, rather than a priori necessary conditions. Similarly, universal truths are not laws because they are logically deduced from some set of prior statements, which are the laws.
For the regularity approach, laws are true summaries of the facts in the world; it is a “bottom up” explanation of laws of nature, and such laws derive their validity from the facts that they explain. Thus, laws are descriptive of the world, but they are not facts in and of themselves. By contrast, the universals approach requires that laws prescribe the facts, and govern what will happen in the world; this can be understood as a “top-down” model. Continuing the above example, under a universals approach, the statement “It is a law that robin eggs are greenish-blue,” would be a law of nature even if there were no robin eggs at all. Empiricists reject such asserts as vacuous. There are numerous critical treatments of the universals approach.[ix]
As noted above, one of the criteria often used for distinguishing accidental generalizations from laws of nature are counterfactuals. These are hypothetical situations that are like the present world in every relevant respect except for a difference in an assumed state of affairs—such a white robin egg—and in which a given statement is tested for being true. It is argued that an accidental generalization will be exposed as being false (or potentially false) in the counterfactual world, whereas a lawful generalization will be true in all such counterfactuals. For example, imagine a counterfactual world identical to ours in every respect except that no patent bar examination is required for patent attorneys. Thus, the statement “All patent attorneys are members of the patent bar,” while being a true universal statement, is not a law of nature. By contrast, there is (arguably) no counterfactual world that is identical to ours in every relevant respect and in which the speed of light is greater than c since this one change profoundly alters everything else. We will revisit the use of counterfactuals in the next section.
The anti-realists give no quarter to realist accounts. Anti-realists argue that to decide which set of statements is “best,” assumes that the criteria for what is best is correct. In other words, anti-realist believe that there is no objective way to determine which statements are the “best” (and thus laws) in terms of simplicity and strength without making some a priori assumption about precisely what statements are better or worse. Thus, to decide that “All robin eggs are greenish-blue” is not a law of nature, one has to assume beforehand that statements like this are undesirable. The universals approach ultimately suffers from having to rely upon some irreducible assumptions that cannot be otherwise demonstrated logically; that there is some universal (whether atoms, quarks or robin eggs) that have their properties simply because that’s the way they are.[x]
Van Fraasen offers an anti-realist view that science is not about the discovery of laws of nature: “You may open a scientific journal and read that some result was reached on the basis of considerations of symmetry—never that it was found through considerations of universality and necessity.”[xi]
This insight is particularly telling because it contradicts in a powerful, demonstrable way, the oft repeated assertion of the Supreme Court that “laws of nature” are the fundamental tools are scientific research. Using symmetry as his touchstone, van Fraasen offers the view that instead of strict laws, there are probabilistic explanations for which the relevant criteria is empirical adequacy[xii]
Laws of Nature in Biology
The question of whether there are laws in biology is one of the oldest questions in the philosophy of biology.[xiii] Biology is distinct from physics because biological systems are the result of evolutionary processes. Different biological outcomes are inherent in the operation of evolution and have several sources: 1) random mutation, which is necessary for any adaptation, 2) variances in the environments that present selection pressure, and 3) the existence of multiple different functionally equivalent adaptations. Stephen Gould, in Wonderful Life: The Burgess Shale and the Nature of History (1989), puts it vividly: “evolution is like a videotape that, if replayed over and over, would have a different ending every time.”
As one of the leading biologists of 20th century, Ernst Mayr contributed to the “Modern Synthesis” of evolutionary theory. In his masterwork, The Growth of Biological Thought: Diversity, Evolution, and Inheritance (1989), Mayr wrote:
The question has been raised in recent years whether or not laws are as important in biology as they seem to be in the physical sciences.…Biologists have paid virtually no attention to the argument, implying that this question is of little relevance to the working biologist….If one looks at a modern textbook of almost any branch of biology, one may not encounter the term “law” even a single time. [xiv]
Mayr then turns his attention to outlining eight positive requirements for “new” philosophy of biology, one that does not hold physics as the model of all sciences. He caps off his list with these admonitions:
I might even add a few ‘don’ts’…. It should not take as one of the existing philosophies of physics as a starting point….It should not focus most of its attention on laws, considering what a small role laws actually play in biological theory.[xv]
This view, by one of the foremost biologists of our time, is a devastating rebuttal to the Supreme Court’s assertion that “laws of nature” are “the basic tools of scientific and technological work.” If working biologists do not think that “laws” have much to do with their work, from what authority, other than perhaps degenerate Cartesian rationalism, does the Supreme Court derive this proposition?
From the foregoing is it clear that what laws of nature are is an unsolved and arguably unsolvable problem. John Carroll summarizes the state of the debate nicely: “Make no mistake, the divisions are serious ones: supervenience versus nonsupervenience, sparse ontology versus lush ontology, laws versus no laws, reducibility versus irreducibility. These are all major issues about which leading philosophers have made quite contrary judgments….New work will have to address these debates head-on.”[xvi]
Having now explored these various approaches toward defining a law of nature, in our next installment, we shall show that the Prometheus patent claim does not qualify as a law of nature.
[i] For a more detailed introduction to the question of laws of nature, see J. Carroll, ed. “Readings on Laws of Nature,” (2004).
[ii]D. Lewis, “New Work for a Theory of Universals,” Australasian Journal of Philosophy, 61 (1983), 343-377, at. 356-357.
[iii] B. van Fraasen, Laws and Symmetry, (1989), p. 26.
[iv] Supra, fn. 3, at 39.
[v] M. Lange, “Natural Laws and the Problem of Provisos,” in J. Carroll, ed. “Readings on Laws of Nature,” (2004), p. 161.
[vi] J. S. Mill, “A System of Logic,” Book III, Ch. 4., found at http://ebooks.adelaide.edu.au/m/mill/john_stuart/system_of_logic/complete.html#chapter19
[vii] D. Lewis, “Humean Supervenience Debugged,” Mind 103 (1994); 473-489, at 474.
[viii] Greene, “The Fabric of the Cosmos,” (2004) Random House: New York, p. 144.
[ix] Van Fraasen; Lange; Carroll, Laws of Nature (1994).
[x] Van Fraasen, “Armstrong on Laws and Probabilities,” in Carroll, supra, fn. 2, at p. 120.
[xi] Van Fraasen, supra, at p. 1.
[xii] Id., at p. 193.
[xiii] Brandon, “Does Biology Have Laws? The Experimental Evidence,” Philosophy of Science, vol. 64, Supplement (Dec. 1997), p. S444-S457.
[xiv] Mayr, “The Growth of Biological Thought: Diversity, Evolution, and Inheritance,” (1989), Cambridge: Belknap Press, p. 19,p. 37.
[xv] Mayr, fn. 14, supra, at. p. 76.
[xvi] Carroll, supra at p. 6.
*The perspectives expressed in the Bilski Blog, as well as in various sources cited therein from time to time, are those of the respective authors and do not necessarily represent the views of Fenwick & West LLP or its clients.