Thursday, May 8, 2014

Laws of Nature: The Historical Background

Many economists think that their disciple discovers “laws,” and the belief in “laws of economics” stems from the idea of laws of nature in the natural sciences. The tendency is seen in both neoclassical and Austrian economics and Marxism.

But what is the background to the concept of “laws of nature” in physics and the natural sciences? What is the epistemological status of a proposition expressing a “law of nature”? How do “economic laws” compare with laws of nature?

This is a very complex issue, but the crucial point is this: there are reasonable – and certainly well known – traditions in modern science and philosophy of science that allow us to dispense with the very idea of a universal, necessarily true law of nature. But why does economics maintain the idea of universal economic laws? The answer is simply that economists and economic methodologists have failed to notice that they are using an old and (arguably) discredited philosophy of science.

In this post, I merely want to examine the background to the idea of physical “laws of nature.”

A “law of nature” has been defined as a fundamental regularity observed in nature or of natural phenomena that holds for all places and all times. In this sense, a “law of nature” can be defined as a true empirical statement of universal form that is necessarily true, not contingently true.

The earliest of the early modern philosophers/scientists who developed the notion of “laws of nature” and was influential in causing others to use the idea was Descartes: he combined the theological notion of god having “legislated” laws for nature with the view that these regularities are predictable by means of quantitative rules and mathematics, though scholars claim that earlier writers anticipated this usage (Ruby 1986: 342–344), and the idea of “laws of nature” in a general sense can be found even in the writings of some ancient Greek and Roman thinkers (Zilsel 1942: 260; Lehoux 2012: 55; Milton 1981: 174).

Descartes argued that the universe as created by god had an orderly and knowable nature described in terms of mathematics.

The rise of this view was part of the collapse of the prior Aristotelian Weltanschauung in which the universe was seen as organic, geocentric, and characterised by formal and final causes, and teleology.

Why the new worldview that displaced the Aristotelian one arose at this time is debated. Zilsel (1941; 1942) attempted a sociological Marxist explanation, and argued that the idea of natural law as determined by god arose from the rising environment of political absolutism in Europe and the emergence of experimenting capitalist artisans searching for quantitative rules of engineering and design. These explanations are not very convincing, given that the two necessary conditions have been present in many civilisations but these societies never developed the concept of “laws of nature” as defined above (Milton 1981: 178–181).

On the contrary, it seems that philosophers and thinkers in general were already moving toward the idea of “laws of nature” in the early modern period, and that discoveries in the sciences and theology assisted the development of the idea.

In particular, both the divine command theory of ethics and the natural law theory of ethics probably influenced the emergence of the doctrine of “laws of nature”:
“When Descartes spoke of God’s having imposed laws upon nature, all he really had to do, therefore, was to transfer from the moral order into the realm of natural philosophy the well-established theological doctrine of an omnipotent Legislator-God, whose sovereign will lies at the very heart, not only of the divine laws revealed in the Scriptures, but also of that natural law to which right reason is man’s unswerving guide. There can be little doubt that he was familiar with this tradition … .” (Oakley 1961: 441).
So too William of Ockham’s philosophy may have played a part (Oakley 1961: 443). And, above all, the new discoveries that were made in this era gave the fundamental impetus to the emergence of the belief in underlying “laws of nature.”

So the “laws of nature” in the classical scientific sense stemming from the early modern period and the work of Descartes and Newton are laws created by god to govern the behaviour of the universe (Giere 1995: 123), although many also believed that god could suspend these laws to effect miracles and that he could also have created a universe with different laws.

In general, the “theological” interpretation of laws of nature was very influential and was probably the dominant view until the 19th century, although by the 18th century deist and rationalist thinkers no longer thought that god caused any supernatural miracles in our universe, and that the universe was strictly governed by natural laws.

Gradually, a “secularisation” of scientific thought occurred with the triumph of the theory of Darwinian evolution, and there arose the view that laws of nature did not necessarily require god or any supernatural creator being (Giere 1995: 127).

Giere sums up the state of affairs in the 20th century:
“It is the secularized version of Newton’s interpretation of science that has dominated philosophical understanding of science in the twentieth century. Mill and Russell, and later the Logical Empiricists, employed a conception of scientific laws that was totally divorced from its origins in the theological climate of the seventeenth century. The main issue for most of this century and the last has been what to make of the supposed ‘necessity’ of laws. Is it merely an artifact of our psychological make-up, as Hume argued, an objective feature of all rational thought, as Kant argued, or embedded in reality itself?” (Giere 1995: 127).
In the 20th century, the very notion that laws of nature must have universal, physical or metaphysical necessity came under attack.

There was also an epistemological question too: how can we really know that a law of nature is necessarily true and valid in places and all times?

Although I certainly do not deny that there are still influential and prominent scientists and philosophers of science who regard laws of nature as necessary a posteriori truths (a view related to the return of metaphysics into later 20th century analytic philosophy), nevertheless the whole topic is open to debate, and old ideas are very much in doubt.

I provide only the briefest of sketches of this debate below.

The logical positivists were influential in questioning the traditional idea of a necessary natural law.

Ayer (1956: 145) pointed out that many think that the laws of nature have a kind of necessity attached to them, though this is difficult to define and defend. The view that laws of nature are inviolable laws established by god requires good reasons to believe in such a god, and it cannot be sustained if there are no good arguments for god’s existence.

Taking up David Hume’s notion of constant conjunction, the logical positivists argued that laws of nature are just empirical regularities to which we have found no exception, and need not have logical nor physical/metaphysical necessity.

Giere (1995: 130–130) proposed that many scientific laws such as Newton’s laws of notion can be given an abstract status.

Newton’s strict laws of notion, for example, can be construed as abstract laws, and be used to construct scientific models, in the same way mathematicians construct analytic a priori systems in pure mathematics. These models have necessary truth, but that logical necessity is a product of their non-empirical, analytic a priori character.

But, when we apply a model to reality or to some specific natural system or phenomenon, its usefulness consists in how well it “fits with” and describes reality, and even though the “fit” may not be perfect, the models do capture real ontological structures of nature (Giere 1995: 131). This “constructive realism,” which is a realist reformulation of van Fraassen’s instrumentalist view of science, does not, however, require strict, universal laws of nature at all.

In a moderate empiricist view of science, one could also argue that abstract laws and models get an empirical hearing when applied to the real world, so that any law asserted of the real world can only be true contingently and known as true a posteriori.

There might be physical reasons why laws of nature really are physically necessary in our universe, but the hypothesis that they are physically necessary cannot be known a priori and must remain an empirical proposition. That is, there are profound epistemological problems that prevent us from knowing whether laws of nature really are eternal, universal, and necessary.

In addition, the Popperian view of science as the process of creating falsifiable hypotheses and continuously subjecting these hypotheses to further testing and rejecting those that are falsified seems to not even require that science discovers irrefutable, necessarily true knowledge of the universe.

To conclude, one might note that a few years ago some remarkable evidence was reported that the “fine structure constant” (or what is called “alpha”) – a fundamental law of nature – may not be constant throughout the universe:
“Laws of Physics Vary Throughout the Universe, New Study Suggests,” Sciencedaily.com, 9 September, 2010.

Michael Brooks, “Laws of Physics may change across the Universe,” New Scientist, 8 September 2010.
You might ask: will science collapse and be thrown into chaos by the “fine structure constant” variation finding, if true?

Not at all. This is because many working scientists and philosophers of science have long since moved on from the view that science discovers immutable, eternal, and necessarily true laws of nature.

BIBLIOGRAPHY
Ayer, A. J. 1956. “What is a Law of Nature?,” Revue Internationale de Philosophie 36: 144–165.

Giere, R. 1995. “The Skeptical Perspective: Science without Laws of Nature,” in F. Weinert (ed.), Laws of Nature. de Gruyter, Berlin. 120–138.

Giere, Ronald Nelson. 1999. Science without Laws. University of Chicago Press, Chicago, Ill. and London.

Gribbin, John R. 2003. Science: A History 1543–2001. Penguin, London.

Harrison, Peter. 1995. “Newtonian Science, Miracles, and the Laws of Nature,” Journal of the History of Ideas 56.4: 531–553.

Henry, John. 2004. “Metaphysics and the Origins of Modern Science: Descartes and the Importance of Laws of Nature,” Early Science and Medicine 9.2: 73–114.

Inwood, Brad. 2005. Reading Seneca: Stoic philosophy at Rome. Clarendon Press, Oxford.

Lange, Marc. 2008. “Laws of Nature,” in Stathis Psillos and Martin Curd (eds.), The Routledge Companion to Philosophy of Science. Routledge, London. 203–212.

Lehoux, Daryn. 2012. What did the Romans Know?: An Inquiry into Science and Worldmaking. University of Chicago Press, Chicago and London.

Martin, Michael. 1990. Atheism: A Philosophical Justification. Temple University Press, Philadelphia.

Milton, John R. 1981. “The Origin and Development of the Concept of the ‘Laws of Nature,’” European Journal of Sociology / Archives Européennes de Sociologie 22.2: 173–195.

Needham, Joseph. 1951a. “Human Laws and Laws of Nature in China and the West (I),” Journal of the History of Ideas 12.1: 3–30.

Needham, Joseph. 1951b. “Human Laws and Laws of Nature in China and the West (II): Chinese Civilization and the Laws of Nature,” Journal of the History of Ideas 12.2: 194–230.

Oakley, Francis. 1961. “Christian Theology and the Newtonian Science: The Rise of the Concept of the Laws of Nature,” Church History 30.4: 433–457.

Psillos, Stathis. 2007. “Laws of Nature,” in Philosophy of Science A–Z. Edinburgh University Press, Edinburgh. 135–140.

Rosenberg, Alexander. 2012. Philosophy of Science (3rd edn.). Routledge, New York.

Ruby, Jane E. 1986. “The Origins of Scientific ‘Law,’” Journal of the History of Ideas 47.3: 341–359.

Van Fraassen, Bas C. 1989. Laws and Symmetry. Clarendon Press, Oxford.

Zilsel, Edgar. 1941. “Physics and the Problem of Historico-Sociological Laws,” Philosophy of Science 8.4: 567–579.

Zilsel, Edgar. 1942. “The Genesis of the Concept of Physical Law,” The Philosophical Review 51.3: 245–279.

Zilsel, Edgar. 2000. “The Genesis of the Concept of Physical Law,” in Diederick Raven, Wolfgang Krohn, and Robert S. Cohen (eds.), The Social Origins of Modern Science. Kluwer Academic Publishers, Dordrecht and London. 96–122.

14 comments:

  1. "The view that laws of nature are inviolable laws established by god requires good reasons to believe in such a god..."

    It also requires a positive theology. Negative theology is a big part of theology and it would deny the existence of knowable lows of nature. Keynes, I think, can be read as a negative theologian. Marx, a positive theologian.

    These questions are not really epistemological at all. Ultimately they are ontological and, I increasingly suspect, theological (whether you choose to worship God or Nature matters little).

    ReplyDelete
  2. LK: I get the sense that this was directed in some capacity towards the very brief discussion in your previous post on value theory (I responded to your queries by the by), but I'm iffy on how, so please bear with me.

    Whether we wish to call them "laws" or "empirical regularities," we're still describing the powers or tendencies of particular underlying generative mechanisms. When we discover behavior that thwarts previous observations, the immediate assumption is generally not that some aspect of the universe's causal catalogue has fundamentally changed since the last time the test was performed ("now when water freezes, it expands to nearly eight times its previous volume!"), but rather that said object of study has properties we had not previously witnessed, or that what we have established about it displays domain-dependency (as with the alpha example, or with water having different boiling points at different air pressures). Insofar as we keep returning to gravity as a source of analogy: when we see a cat bound up to a shelf, it does so quite contrary to gravity, but this does not imply a suspension thereof -- rather, it's the operation of other powers (in this case, biological) providing countertendential effects. Etc.

    Put another way, I am aware of no reason to suppose that falsification of a theory -- even an otherwise reliable one -- suggests that the intransitive object of knowledge (reality) has changed so much as to the transitive objects of knowledge (the complexes of theories we use to understand it) need to change. This holds as well for the realist (that our understanding of the material world has improved) as for the Berkleian (that our thoughts are brought more into accord with those of the ubermind).

    I agree fully that an excellent way to obtain powerful and reliable knowledge is by determining how things do not work, vis-a-vis falsification, or what Philip has compared to apophatic theology. However, this is also not a "magic bullet" to understanding (especially when closure is impossible, as in the social sciences). There's a reason Popper proposed a conjectural approach, and even "metaphysical research programmes," in conjunction.

    The claim that a particular law is completely domain-independent and universal in the fullest sense of the word is epistemologically untenable. I don't think we can illustrate Marx claiming anything that strong, especially in light of the importance of the historicization of social phenomena to his entire body of work. Thus, the "law of value" is a tendency he associates specifically with capitalism, and so on.

    As for "the law of development of human history" you discuss in the previous comments section, this is probably in reference to historical materialism. But this isn't a "science" as such; it's an analytical lens. Popper criticized it for being unfalsifiable, but ironically it is in fact much closer to the aforementioned research programs he himself propounded. Of course, it's trivial to note that production, conceived broadly, is a condition for social life, though it can't be proven that it is necessarily the determining one. As Bhaskar notes of it: "And so, like any other fundamental conceptual blueprint or paradigm in science, historical materialism can only be justified by its fruitfulness in generating projects encapsulating research programmes capable of generating sequences of theories, progressively richer in explanatory power. Not the least of the problems facing historical materialism is that, although considerable progress has been made in particular areas of explanation, the blueprint itself still awaits adequate articulation."

    Hope that clarifies it a bit. (If I'm wrong about drawing this connection to the previous post, then I suppose that's egg on my face.)

    ReplyDelete
    Replies
    1. "The claim that a particular law is completely domain-independent and universal in the fullest sense of the word is epistemologically untenable. "

      You contradicted yourself.

      Your statement asserts the "domain-independent and universal" law that there are no "domain-independent and universal" laws.

      Your statement is "epistemologically untenable".

      Delete
    2. Haha. While my statement does not actually make the assertion claimed of it (look up the "problem of induction" for an overview), I do appreciate what you're trying to do.

      Delete
    3. "Anonymous" sounds like some Austrian idiot, frankly.

      Delete
  3. "there are reasonable – and certainly well known – traditions in modern science and philosophy of science that allow us to dispense with the very idea of a universal, necessarily true law of nature."

    There are well known traditions that deny the existence of economic laws. However, these traditions are not reasonable.

    These traditions claim "the only economic law is that there are no economic laws"

    A person who says there are no economic laws is contradicting himself.

    The claim that "there are no economic laws" is in an economic law. One cannot deny the existence of economic laws without asserting an economic law. Traditions that deny the existence of economic laws are not reasonable traditions.

    ReplyDelete
    Replies
    1. The sentence "the only economic law is that there are no economic laws" is nothing but a pithy or facetious phasing -- or mere rhetoric -- of a serious epistemological view: that many of the postulated economic laws are not in fact laws.

      To discuss this issue seriously:

      (1) you define what you mean by "law" for me.

      (2) are laws -- and the economic laws you postulate -- analytic or synthetic?

      (3) are laws -- and the economic laws you postulate -- known a priori or a posteriori?

      (4) do the laws -- and the economic laws you postulate -- have a physical/ontological necessity?

      Delete
    2. Anon: Properly speaking, "there are no economic laws" (something claimed by no one here, incidentally) would probably be better classed as a "meta-economic" law than an economic one. Just sayin'.

      Delete
    3. "many of the postulated economic laws are not in fact laws."

      I agree. For example, Keynes's Fundamental Psychological Law is not an economic law.

      However, the view that "there are no economic laws" cannot be correct.

      The contradiction becomes obvious when we restate the view in more honest language: 'the only economic law is there are no economic laws'.

      Delete
    4. You didn't answer my questions, anonymous.

      Is it because you can't (probably because your knowledge of philosophy is too shoddy) or if you did we'd quickly see you're an internet Austrian trying defend Austrian economic laws and synthetic a priori knowledge?

      Delete
    5. Anonymous, don't bother wasting any more of my time with any more comments unless you answer these questions clearly and explicitly and succinctly:

      (1) you define what you mean by "law" for me.

      (2) are laws -- and the economic laws you postulate -- analytic or synthetic?

      (3) are laws -- and the economic laws you postulate -- known a priori or a posteriori?

      (4) do the laws -- and the economic laws you postulate -- have a physical/ontological necessity?

      Delete
    6. Economic laws like Keynes's Fundamental Psychological Law (p. 96 GT) are synthetic a priori statements with no physical necessity.

      Now answer these question explicitly and succinctly:

      (1) Is the statement "all economic knowledge is a posteriori" an economic law?

      (2) Is the statement "all economic knowledge is a posteriori" an analytic or synthetic statement?

      (3) Is the statement "all economic knowledge is a posteriori" an a priori or an a posteriori statement?

      (4) Is the statement "all economic knowledge is a posteriori" an ontological statement?

      Delete
    7. I'll give you one last chance:

      (1) if Keynes thought that there was real synthetic a priori knowledge, then he was simply wrong.

      (2) Actually, I do not think that "all economic knowledge is a posteriori" -- this is not my view -- because some of the fundamental economic terms and their definitions are ultimately matters of convention and are analytic a priori: this is proven when see that different economic schools define some terms in different ways: e.g., "money" is defined in different ways between schools.

      So general economic knowledge includes both (1) analytic a priori propositions and (2) synthetic a posteriori knowledge.

      Since I do not assert as true the statement "all economic knowledge is a posteriori" -- and your questions have no relevance to my own position on this subject.

      But -- just for the sake of argument -- if someone really believed that, then I presume they would answer them in this way:

      "(1) Is the statement "all economic knowledge is a posteriori" an economic law?"

      If one defines a "law" as an empirical and necessarily true proposition, then no.

      It would be a synthetic a posteriori statement and its truth would be asserted as probabilistic.

      (2) Is the statement "all economic knowledge is a posteriori" an analytic or synthetic statement?

      synthetic

      (3) Is the statement "all economic knowledge is a posteriori" an a priori or an a posteriori statement?

      a posteriori

      (4) Is the statement "all economic knowledge is a posteriori" an ontological statement?

      It is primarily an epistemic/epistemological statement, but could be interpreted as a statement about what exists.

      It would not have physical/ontological necessity.

      Delete
    8. So are you going to answer my questions clearly, explicitly and succinctly?

      Delete