What is a Law of Nature?

[Originally published 29 January 2013]

We’ve all heard references to ‘laws of nature’, but have you ever wondered what is really meant by that? Let us think about this a bit more deeply.

The Regularity Theory

Perhaps you support this theory of laws, which says that the world possesses certain fundamental truths, or facts, or things, and that a law of nature explains why there are such regularities (examples of regularities include the boiling point of water being 100 degrees, gravitational attraction, or how a species produces offspring).

Yet, there are other views, for depending on the scientist or philosopher you talk to, laws of nature can be understood in many different ways:

Let us reconsider the Regularity theory of laws of nature (as being understood as regularities of universals (*) that occur in every instance). Water doesn’t always boil at 100 degrees (pressure can change the boiling point). When it comes to species propagation, at some point a descended organism may become a member of a different species from its distant ancestor. So it would seem that regularity is less than universal.  A regularity view can probably only be said to be true ceteris paribus (other things being equal).  Can the regularity theory still be defended?

Best Systems Theory

The ‘Best Systems’ theory says that whether something is a law is not because of some purely intrinsic feature that it possesses, but instead, something is a law when it is part of a system which can provide part of a systematic account of the world’s history.  The laws must cohere with each other in a systematic unity.  This view doesn’t rely on just regularities, but rather on the whole history of the world, which includes the regularities and anything else.  Of course, the problem with this type of coherence theory, is that an element of subjectivism is brought into the idea of laws, it is not entierly an objective matter, for we might very well find other systematizations of the same history which can be considered ‘best.’

Constructive Empiricism

The philosopher Bas van Frassen has a view called ‘Constructive Empiricism’ which says that “science aims to give us theories which are empirically adequate.” And that “the belief involved in accepting a scientific theory is only that it ‘saves the phenomena’, that is that it correctly describes what is observable.”  He is saying that laws are nothing more than a set of important features of some model that has been adopted.

Consider this elaboration of van Frassen by Stephen Mumford (who doesn’t necessarily support this view): “Science tells us what the particular, observable patterns in the world are, usually expressed in mathematical form.  It does not tell us in more general terms what it is to be a law of naure, what a law of nature does and by what means it conducts its business.  In ‘saving the phenomena’, science gives these metaphysical concerns a wide berth.  What matters, for science, is the epistemic integrity of the discovered patterns explicated in terms of explanation, sustaining of counterfactuals and supporting inductive inference.”  What do you make of van Frassen’s theory?


The last view I will present (just to keep this note short, for Philosophy of Science and Metaphysics has produced many considerations of what constitutes a ‘law of nature’!) is the ‘Essentialist’ theory.  This view – best put forth by Brian Ellis – says that there are natural kinds (**) in nature, and these are what they are irrespective of what our thoughts are about them.  Sometimes it is said that nature has ‘joints’, or ‘real cleavages’, essentialism says that science discovers what these are, rather than invents them.  Ellis structures these ‘real’ natural kinds in a hierarchical structure, with the most basic at the bottom (something like physics), and the most general kinds at the top (think of the world).  The laws of nature spell out for us the essential properties of these natural kinds.

But there is more to this, because for Ellis, an understanding of some particular entity is to be understood entirely by its essential or dispositional properties (think of the negative charge of an electron, chemical reactions, or energy transfer processes).  This means for Ellis that the essential or dispositional properties are the laws of nature.  Which leads to one possible conclusion which might concern some people: If this is true, then what purpose is there in even speaking of laws of nature?  Is there any reason to bring them up at all if everything can be explained by the activity and dynamism of the particulars themselves?

I will leave you with some final thoughts regarding Laws of Nature: Are the law of nature we are considering just the pattern we see in some natural occurence, or are they something which underlies and produces the pattern? What do you think?

Mumford says in regards to laws of nature that: “What we lack are clear and uncontroversial necessary and sufficient conditions for lawhood.”

* Universals in philosophy are defined as a property or relation that can be instantiated by several particular things (eg. a red object is an instance of redness).  But some metaphysicians raise concerns with this concept, for how do we perceive the general property as well as the particular instance of it?  Can sharing the same property be analyzed in terms of resemblance?

** Natural Kinds are also controversial in philosophy, for example: if nature can be divided into different categories or species, is this ‘carving of nature at the joints’ something that can be considered a real division, or is it more historical or theoretical based on the scientific taxonomic system in use at the time.

[The discussion for this note drew in part on Stephen Mumford’s book “Laws in Nature.”]