1.1.1. Laws in physics and biology
As a philosopher of biology, you have obviously considered the scientific specificity of biology. In your article âLa biologie entre loi et histoire (Biology, from laws to history)â3, you spoke of biology as a âscience without lawsâ. What do you mean by this?
Philosophers more or less universally agree that biology, and evolutionary biology in particular, has presented significant challenges to a number of traditional concepts in the general philosophy of science, including the notion of natural law, at least in terms of its application to the life sciences.
The statement that âthere are no laws in biologyâ refers to the concept of law promulgated by the neopositivist philosophers of science (particularly Carl Hempel and Ernest Nagel). For the neopositivits, laws are empirically true statements with a universal logical form. However, this formulation is insufficient, as, applied in its strictest sense, it implies that all accidental generalizations (e.g. âall of the coins in my pocket are one-euro coinsâ, a phrase which has the logical form of a universal statement) are laws, which would be absurd. It was therefore necessary to clarify that enigmatic property of the laws of nature which philosophers, starting with David Hume in the 18th Century, referred to as ânatural necessityâ. The result was the formulation according to which a law is a statement of unlimited universal scope, that is containing no spatio-temporal limitations, whether explicit or implicit. Nelson Goodman4 modified this formulation by adding a modal criterion. For Goodman, a nomological statement is a statement which supports (i.e. justifies or permits) counterfactual statements, that is statements which are contrary to fact. For example, when we say that sugar is water-soluble, we implicitly accept the following counter-factual statement: âif I put this sugar cube in my cup of coffee, it would dissolveâ. The two propositions in this phrase are counterfactual, as the specific sugar cube involved in our thought experience is not in my coffee, and is therefore not dissolved hic et nunc in this world. The strength of a nomological statement lies precisely in the fact that it is not only valid in this world, but in any other possible world resulting from a different sequence of events and obeying the same laws. Thought experiments based on counterfactuals offer a powerful tool for identifying generalizations which are not laws but are simply accidental, that is de facto generalizations limited to a particular portion of space and time. Returning to the example of the coins, the phrase âall of the coins in my purse are one-euro coinsâ does not allow us to infer that âif this ten-cent piece in my hand were in my pocket, it would be (or would become) a one-euro coinâ.
Nelson Goodman developed a linguistic criterion which is remarkably effective in detecting statements which take the form of universal logic, but which are inextricably linked to particular temporal conditions (i.e. accidental generalizations). This proposal has attracted considerable attention within the field of philosophy. It has been criticized on the grounds that the notion of âpossible worldsâ it uses is too vague. David Lewis suggested a more restrictive formula, according to which universal statements which are empirically true may only be considered to be laws if they are true in all nomologically-accessible worlds5. The restriction to ânomologically-accessible worldsâ excludes possible worlds governed by laws which are completely different to those which we know, or ânon-nomologicalâ worlds, that is worlds without order.
This modern understanding of the laws of nature is easy to apply in the field of physics, but problematic in biology. Within the life sciences, it is hard to find generalizations with unlimited universal scope. Most biological generalizations appear to be limited to a small portion of the history of the universe, that corresponding to the history of life on our planet.
As far back as the 19th Century, the French philosopher and mathematician Antoine-Augustin Cournot noted that the life sciences are faced with collections of singular entities (species), each of which appears to be governed by its own laws. According to Cournot, instead of universal laws, each species seemed to be subject to unique decrees, the result of coups dâĂtat, which the author compared to Napoleonâs coup of 18 Brumaire, a historical event. In biology, therefore, there are no (timeless) laws, but rather a series of unique decrees, each resulting from a coup dâĂtat6.
But the term âlawsâ is still used in biology. Is a term such as âMendelâs lawsâ still epistemologically relevant?
Not if we understand the concept of laws in the way I just described. Lewisâ criterion may be usefully applied in determining whether or not âMendelâs lawsâ are genuine laws of nature7. The reason âMendelâs lawsâ cannot be considered to be genuine in the same way as Newton, Maxwell or Einsteinâs laws is that it is possible to imagine worlds governed by the same laws (i.e. all existing laws with the exception of Mendelâs) in which Mendelâs âlawsâ would not be respected. For example, there is nothing to prevent us from imagining that sexually reproducing animals may have evolved on an exoplanet, independently of the history of life on Earth. These animals would exist in a physical environment obeying the same laws found on Earth and throughout the whole universe, and would also be structurally and functionally similar to terrestrial animals; for example, they would be made up of macromolecules such as nucleic acids, proteins and carbohydrates. They would have an articulated internal skeleton and would share many physiological properties with animals on earth. However, meiosis, that is the mode of division which produces haploid gametes (n chromosomes) fro...