The Critical Rationalist Vol. 01 No. 02 ISSN: 1393-3809 26-Nov-1996
(10) In dealing with the issue of the infinite versus finite nature of resources, we are dealing with a metaphysical issue. Expressed colloquially, metaphysics provides a view of the world as a whole. An example of such a view was Faraday's conception of the universe as a network of fields of forces.
(11) Methodologically, I think it useful to interpret metaphysical issues in terms of theories and their logical relationships with one another. Provisionally, we may say that a metaphysical theory is one that taken alone is empirically untestable by confrontation with basic statements. A basic statement is one that describes an observable event of definite space-time coordinates. "There is red ball of at least 8 cm diameter within one meter of the entrance to Bolton's central library" is a basic statement. An example of a metaphysical statement would be "There is a red ball which will give everlasting life to the person who touches it". To emphasise the fact that this distinction is one of logic and not between the everyday and the mystical, this last example can be changed to "There is an incompressible red ball". Another example would be "There is an inexhaustible barrel of oil". This definition of metaphysical conforms with the traditional sense in which metaphysics transcends experience, but conformity with tradition is not crucial. More importantly, it allows us to talk about interesting logical relationships between statements of quite differing character.
(12) All economic theories, even if they are scientific in being empirically falsifiable, contain explicit or implicit metaphysical assumptions or presuppositions. In this they are in good company as all the great scientific theories were not only inspired by, but also contained, metaphysics. In some cases the metaphysical elements are simply weak logical implications of the empirical theory. In some cases the metaphysical elements are more like adjuncts that can be dispensed with without diminishing the empirical content of the theory. In many cases, however, they play an important role not only in augmenting empirical content, but in guiding research. cf. (Watkins 1958). Also (Watkins 1975). See also (Agassi 1964). More recently (Zahar 1989).
(13) I have said that metaphysical statements are untestable by direct confrontation with basic statements, but things are not so simple. Popper has even shown that combining what are individually untestable metaphysical statements can sometimes yield a testable theory (Popper 1982b, Chapter III). This clearly shows that a statement may be only relatively metaphysical. This perhaps should not be so surprising as even scientific statements of the law-like variety only become fully testable in the presence of initial conditions and other theories, though some crude testable implications are derivable from the law-statement alone. In any case, it is clear that in the analysis of metaphysics we have moved a long way from the cavalier attitude of the Vienna circle, and that metaphysical elements in science require serious and discerning treatment. In assessing the empirical value of a statement that is metaphysical when taken alone, one has to investigate its role within a larger system of hypotheses, how it contributes to the empirical consequences of the system. The point here is that even though the assumption that resources are infinite may itself be untestable, the explanatory theory in which it occurs may be the more powerful (and more testable) for its presence.
(14) Examples of metaphysical doctrines are atomism, determinism, the irreversibility of time, and recently the idea of locality in physics.
(15) Let us become acquainted with the feel of metaphysical statements. I was struck recently by how common unacknowledged metaphysical assumptions are in every day conversations. One I encountered recently might be called failure-metaphysics: there are some people who have fallen into circumstances of poverty and misery from which it is impossible to escape. If any putative confirming candidate subsequently becomes rich and happy, then the advocate of failure-metaphysics can always say that he misidentified the example. Of course, the contrary (success-metaphysics), that there is always a way of escaping from any circumstances of poverty providing one uses the correct methods is also untestable by itself. Such metaphysical assumptions are more to do with self-help methods than economic theory as such. An example from economics might be the following:
A person is willing to sacrifice some bit of any desired thing if he can obtain a sufficient increase in the amount of some other desired goods. (Alchian & Allen 1964, p. 21)
(16) This is not empirically testable as it stands, for no limit is placed on what might be a "sufficient" increase in other goods: does it stop short of owning the Earth, the Solar system, the Universe? Nor does it specify a limit on the smallest "bit" of sacrifice allowed: does it include the sacrifice of one 1/1000th of a cup of coffee per year? However, I do suspect that this "postulate" does contribute to a theoretical system that is empirically testable when taken as a whole. The situation is analogous to Einstein's use of Lobachevskian geometry in his theory. By itself, Lobachevskian geometry cannot be empirically refuted, but it can contribute to the refutable empirical content of a theory. Assuming the economic postulate to be true, it does lead one on to interesting speculations about values and the structure of the world. For example, most people will not sacrifice any amount, however small, of their moral values such as "Thou shalt not kill". Is this a refutation of the postulate or could it be that the world is such that the "pay-off" just could not be arranged, perhaps because of the constraints of physical laws etc. Perhaps some might murder if the reward were everlasting life for them and their family--the Vampire option. But the world is such that everlasting life is impossible.
(17) Although the assumption of infinite resources and its contrary are metaphysical they play an important part in the empirical theories to which they belong. The assumption of infinite resources can be stated as follows: for every increase in resource-scarcity, there is at least one resource-augmenting invention or discovery that will more than compensate for this increased scarcity. The assumption of finite resources can be stated thus: at some point in the future the amount of resource-scarcity will increase irretrievably. Taken alone each of these statements is obviously metaphysical in the narrow sense of being unfalsifiable by confrontation with basic statements. The first places no limit on the delay between an increase in scarcity and its correction and it is not clear whether the compensatory correction takes account of time-preference. The resource finitist can sustain their gloom no matter how long humans live in abundance and luxury.
(18) The resource-finitists do try to include testable assumptions. The standard approach to estimating global resource quantity is first to estimate the presently-known physical quantity of the resource, secondly estimate the current rate of use, and finally predict a diminution of the first estimate over successive periods of time until the first estimate is exhausted. This approach, as Simon points out, is based on a confusion of the micro and macro contexts (Simon 1982). It is based on Hotelling's theory of the optimum rate of use of a spatially definite mine or well. The resource finitist's error can be put this way. The world's resources are seen as a large store-house which will be emptied in a finite period even if our technology improves the speed with which we empty it. But the world is not a store-house with pre-defined finite contents, since it is interpreted by our creative minds and theories.
(19) Prima facie, one might think (as Julian Simon does) that there are two ways of challenging the assumption of finite resources in the long run in a macroeconomic context. One is to point to empirical statistics, such as trends in resource prices and invention creation, and project from past data, a method pioneered by Barnet & Morse (1963). For example, price data for minerals over the past 200 years was adduced by Barnet and Morse (and more recently by Julian Simon) to show that the scarcity of minerals has actually been declining over the centuries, contrary to popular imagination. The argument is that if a resource were becoming more scarce, its price would be bid up by speculators, so reflecting its greater future level of scarcity. To reinforce this, statistics on the increasing rate of creation of inventions (some of which increase resources) is also adduced. When one projects this data into the future, the picture looks quite rosy. The supposed alternative is take a more theoretical approach.
(20) But such a division into empirical versus theoretical approaches is unsound. One can do theory without empirical research, for one can test a theory for internal inconsistency, simplicity, axiomatizablity and consistency with other background theories; but one cannot do empirical research without some, perhaps implicit, theory, as even singular statements describing definite events or objects contain universal terms that have dispositional (and counterfactual) implications that transcend any immediate (or indeed any finite set of) experiences or observations (Popper 1980). Even the seemingly trivial and untheoretical statement "There is petroleum oil in that barrel" is highly theoretical in that its implications go far beyond any experience. The statement implies that the liquid in the barrel can be refined into petrol and other useful organic substances.
(21) Simon thinks that the reason why economists have been slow to adopt the Barnet and Morse approach is that it is merely empirical or rule of thumb, without a theoretical justification. The resource finitist is interpreting the data in the light of a metaphysical theory. The resource finitist sees the world as a finite store-house. Humans may be getting the goods out faster and faster with improvements in the machines they use, but this will only empty the store-house sooner.
(22) To provide a theoretical interpretation of the data Simon introduces the idea of partial operationalism. Simon stops short of a wholesale operationalism, but insists that in order for any economic theory of finite resources to be testable, one has to construct an operational definition of the word "finite" in order that it may be measurable. The attempt to make any theory more testable is admirable. However, this aspect of Simon's approach is unsuccessful. Testability and measurability are, though related, quite different concepts. Strictly, introducing measurement-orientated concepts may increase the empirical information content of our theories, and as such is to be valued, but a theory may lack measurable concepts and still be testable. Moreover, a theory may be replete with measurable concepts and still be metaphysical, for example: "there exists a cylindrical iridium rod of precisely 1 centimetre in diameter and 1 metre long." One may consistently maintain such a theory no matter how much of space-time is surveyed. Even if some plausible candidate were found, the testability of the statement is confined to ranges of measurement outside the physical limits of measurement because of the qualification "precisely".
The Critical Rationalist Vol. 01 No. 02 ISSN: 1393-3809 26-Nov-1996
Copyright © 1994-96 All Rights Reserved.
TCR Issue Timestamp: Tue Nov 26 17:14:18 GMT 1996