doi:10.1130/2013.2502(03)Geological Society of America Special Papers 2013;502; 19-40 Gadi Kravitz principle in geohistorical explanationThe thermodynamics time arrow and the logical function of the uniformity Geological Society of America Special Papers
E-mail alerting services
this article
to receive free e-mail alerts when new articles citewww.gsapubs.org/cgi/alertsclick
Subscribe
Special Papers
to subscribe to Geological Society of Americawww.gsapubs.org/subscriptionsclick
Permission request
to contact GSA.www.geosociety.org/pubs/copyrt.htm#gsaclick
viewpoint. Opinions presented in this publication do not reflect official positions of the Society.positions by scientists worldwide, regardless of their race, citizenship, gender, religion, or politicalarticle's full citation. GSA provides this and other forums for the presentation of diverse opinions and articles on their own or their organization's Web site providing the posting includes a reference to thescience. This file may not be posted to any Web site, but authors may post the abstracts only of their unlimited copies of items in GSA's journals for noncommercial use in classrooms to further education andto use a single figure, a single table, and/or a brief paragraph of text in subsequent works and to make
GSA,employment. Individual scientists are hereby granted permission, without fees or further requests to Copyright not claimed on content prepared wholly by U.S. government employees within scope of their
on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from on October 5, 2013specialpapers.gsapubs.orgDownloaded from
The Geological Society of AmericaSpecial Paper 502
2013
The thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation
Gadi KravitzDepartment of Philosophy, University of Haifa, Haifa 31905, Israel
ABSTRACT
The issue of reductionism in geology has not yet been solved. The standard approach regards geology as a derived science, and therefore most modern philoso-phers are not particularly interested in it. On the other hand, in recent decades, interest in the philosophy of geology has grown, and growing numbers of modern philosophers oppose this approach. Some claim that geology and physics cannot be joined accord-ing to Nagel’s reduction model, while others claim that geology is an autonomous his-torical interpretative science. My argument in this chapter is that there is a geological principle that meets the requirements of Nagel’s reduction model, thereby enabling geology to function as an analytical science, deriving from the basic laws of physics, on the one hand, while also functioning as a geohistorical science, on the other. My argu-ment is based on a logical-conceptual analysis of the uniformity principle in geology, and on the exposition of its close link to the second law of thermodynamics.
INTRODUCTION
This chapter is concerned with the question whether, under proper boundary conditions, a possibility exists to “derive” the laws and theories of geology from the laws of physics. In order to answer this question, the paper will engage mainly in a philosoph-ical-conceptual analysis of the uniformity principle with special emphasis on its close link to the second law of thermodynamics. First, I will describe the issues and nuances of the realistic faith in the geological past of Earth, on which the geohistorical approach is based. I will continue with a description of the various elements of the uniformity principle, manifest in the history of geology and still affecting present geological research. Finally, I will argue that the causal explanations concerning the geohistory of Earth are based on and “derived” from the uniformity principle, which con-tains the second law of thermodynamics, connecting this law with the general geohistorical approach. In other words, the uniformity principle corresponds to the criteria of Nagel’s reduction model and indicates that it is possible to achieve a partial reduction of
geohistory and the causal explanations employed in geology, to physics. However, I maintain that the multidisciplinary nature of geology, based on obligatory integration between an analytical and a synthetic science, makes it impossible to “derive” geology completely from physics, and here the uniformity principle does not help. Thus, as long as geology continues to exist as a multi-disciplinary science—synthetic-historical-interpretative on the one hand and analytical-nomological-causal, on the other—we will never be able to reduce it fully to physics.1
1 It should be noted at the outset that this essay will be concerned with issues that most engage philosophers of science, including the certainty that can be associ-ated with logical argument and explanation in science. Therefore, I presume some knowledge of philosophy by the readers. Moreover, the chapter is con-cerned with history and the research of the past, and therefore it will deal with issues touching on wider metaphysical questions pertaining to concepts such as: the time arrow, the uniformity principle in nature, causality, induction, realism in relation to the past, etc. It is a philosophical article, and naturally it strives to expose the basic metaphysical assumptions of geology and raise philosophical questions concerning its conceptual basis. However, I would like to mention that I have no intention to cast doubt on the legitimacy of geological research or on the validity of geology as a full-fl edged scientifi c discipline.
THE UNIFORMITY PRINCIPLE AS A TOOL IN THE SERVICE OF REALISM IN RELATION TO THE PAST
Geology focuses mainly on geological phenomena and proc esses that occurred in the past.2 The preoccupation with geol-ogy as a historical science3 actually requires a belief in realism in relation to the past.4 The geologist must believe in the existence of geological events that occurred in the past and in his capacity to learn about them as they actually occurred. This is a basic as-sumption required of any person who undertakes to study past physical events. In other words, in spite of the fact that the past is not accessible to direct geological research, geologists believe in its existence, in its independence on the thought process of the geologist, and in his or her ability to understand it as it actu-ally existed. On the one hand, geologists distinguish between the study of present sources (evidence and geological observations in the present) and past geological events. On the other hand, they do not regard this as a problem: They do not see any problems with the transition from the present to the past. This realistic ap-proach interprets the expression “discovering the past” literally and entails the belief that the past can be discovered as it actually happened. Geologists usually accept this approach and regard it as intuitive and obvious.5
Geohistory without belief in realism in relation to the past is scientifi cally irrelevant, as there is no reason to attempt to study the past unless one believes in the possibility of discover-ing it and getting to know it.6 However, this attitude indicates that geologists entertaining a realistic approach consciously or unconsciously assume that the time arrow is one-directional and irreversible.7 This means that, contrary to the laws of physics
and like the second law of thermodynamics, geology, as the sci-ence of the history of Earth, assumes that the direction of the time arrow is from past to future, that is, asymmetric in rela-tion to the reversal of the direction of time (irreversibility of the time arrow).8
The realistic geologist claims that events that happened in the past belong to the past that cannot return, and that the geol-ogist has no causal infl uence on them. On the other hand, since he observes present events directly, he is able to understand and to explain the past, on the basis of causal relations and the uniformity principle. As a matter of fact, he claims that the past always precedes the future, and that the future events are causally infl uenced by past events—namely that causality is one-directional from past to future.9 The causes have occurred in the past, and the results occur in the present or in the future; therefore, the cause must always precede the result. One may argue that the assumption of asymmetry and irreversibility of the passage of time is a more basic assumption preceding the assumption of causality. In other words, a geological realist cannot present causal arguments concerning geological events that happened in the past without assuming the direction of the time arrow.
In conventional historical sciences (human history, archae-ology, etc.), realism in relation to the past assumes at least three time arrows, as follows:
1. the metaphysical time arrow, determining the order of events from past to future10;
2. the epistemic time arrow, according to which the past is closed, and there is no epis temic access to it except by means of assumptions referring to the present11; and
3. the causal time arrow, according to which every result in the present has causes rooted in the past and these causes pre-cede the results.
In addition to these time arrows, for historical sciences, such as geology, that deal with natural phenomena, the relevant sci-entists have to rely on the thermodynamics time arrow, in order to provide a scientifi c foundation to the historical explanation.12
On the face of it, this worldview is intuitive, naïve, and well adjusted to common sense. Geologists derive from it the basic assumptions on which their day-to-day work is based. However, close examination reveals its metaphysical aspect, and the dif-fi culty to realize it in the framework of geohistorical reasoning.
2 For more details, see, for example, Kitts (1978) and Kleinhans et al. (2005, p. 290–292).3 It should be noted that this essay mostly concerns geology as a historical sci-ence and the geologist as a reconstructor of past environmental conditions from specifi c outcrops and cores. The attributes of geology as a historical science will be displayed later herein. 4 Philosophic consideration of realism and the reality of past can be found in Sober (1988), Dummett (1978), Turner (2007), and Weinryb (1987).5 Here, I am referring to a particular type of truism. For more details, see Kitts (1978, p. 219). 6 This attitude is not confi ned to geology but is common to all historical sci-ences. However, before tackling the uniformity principle, the most fundamental principle of geology, I think it appropriate to describe the geologists’ conception of realism in relation to the geological past of Earth. This conception can be regarded as deriving from the uniformity principle itself, and it justifi es, in a cer-tain sense, referring to geology as a historical science. I would like to emphasize that defi ning the geologist as a realist in relation to the past does not mean that there is no chance of fi nding an idealist among geologists. My argument is not merely philosophical, but based on geological textbooks in which the realistic approach of modern geologists is evident. For example, a quotation from Hutton, regarded as the father of modern geology, in his book The Theory of the Earth (Hutton, 1795, p. 280–281) runs as follows: “A theory of the earth which has for object truth, can have no retrospect to that which had preceded the present order of this world; for, this order alone is what we have to reason upon.”7 Below, I maintain that this assumption is based, among others, on the sec-ond law of thermodynamics and on Boltzmann’s “past hypothesis,” i.e., my argument is that realism in relation to the geological past is partly based on statistical mechanics in general and on the second law of thermodynamics in particular. This fact indicates that, on principle, it is possible to reduce geology to physics.
8 Consideration about development of some conceptions of time in geology can be found in Overton (1994) and Gould (1987). 9 More in-depth discussion regarding this causality would be fulfi lled later.10 In this arrow, we can include the psychological arrow that relies on the psycho-logical feeling of the fl ow of events from the future to the past.11 In this respect, this arrow can be linked to the mutability arrow, according to which the future is subject to change, but the past is not.12 As I will specify later, modern geologists assume a progressive picture of the world (evolutionism) and the uniformity of law and process. These two assump-tions connect the geological perception to the second law of thermodynamics, or in other words to the thermodynamic time arrow.
20 G. Kravitz
The main diffi culty lies in the facts that the past is inaccessible and independent, and that there is no necessary logical connec-tion between the past and the present. The diffi culties cannot be solved in the physical world around us on account of the pre-viously mentioned existence of the time arrows (the thermo-dynamic, the causal, the epistemic, and the metaphysical). The past cannot be observed directly, and, although geologists talk of geological events that have happened in the past, they have never observed them directly. Moreover, there is an even deeper gap—the gap between past and present. Accessibility to sources and evidence is rooted in their availability in the present, but past events are inaccessible, since they are part of the past. Can this disparity be resolved? If the objects of geological inquiry are cognitively independent, is it at all possible to know them? On the other hand, if it is possible to know them, how can one argue that they are independent?
Bertrand Russell famously expressed his skeptical point of view about the gap between past and present (Russell, 1921, p. 159–160) as follows:
There is no logically necessary connection between events at different times; therefore nothing that is happening now or will happen in the future can disprove the hypothesis that the world began fi ve minutes ago. Hence the occurrences which are called knowledge of the past are logically independent of the past.
The geologist learns about the occurrence of natural proc esses in the past by means of observations conducted in the present. These are accessible on a perceptive level and can be perceived directly by his senses. For instance, he can map the rock in the fi eld, observe it from space, photograph the rocks at different wavelengths, drill in the depths of Earth, examine the magnetic attributes of the rock and its density, carry out a chemical analysis of the rock, carry out an isotopic analysis of the rock and the fossils in it, etc. The picture that emerges with regard to past geological events is quite different. On account of the one-directional time arrow and the fact that the past can never return, past events are not accessible and cannot be per-ceived directly by the geologist, and he is unable to know them for certain. Since this is the situation, geologists are content to learn about them indirectly, on the basis of an artifi cial timeless assumption (the uniformity principle), allegedly bridging the gap between arguments concerning past events and those con-cerning present events. In other words, the geologist must learn about the past indirectly by means of its remains—traces left by geological pro cesses that occurred in the past. Indirect learning is based on logical-causal arguments and a basic assumption of timelessness—the uniformity principle, on the basis of which the geologist makes inferences from observed events in the pres-ent to unobserved events of the past, i.e., from the known to the unknown.13
It can therefore be said that the geologists’ knowledge of the past is based on pretheoretical assumptions, often of a meta-physical nature, not susceptible to logical or empirical proof.14 In a certain sense, they are products of the geologists’ imagination (artifi cial). Thus, geologists are realists in relation to the past, both consciously (they believe that the events of the past are in-dependent of the geologist’s thinking; nevertheless, they are able to discover them), and latently (they do not actually discover the past, but rather construct it). To state this more exactly, by assuming the uniformity principle, the past becomes dependent on the geologist’s thinking, and in this sense, he can be said to construct it rather than discover it. This principle serves realism in relation to the past, and, by artifi cially abolishing the logical independence of both the past and the present, it enables a reduc-tion from the present to the past.15 Nevertheless, this creates an absurd situation: On the one hand, geologists believe in realism in relation to the past, but in practice (probably not consciously), they make metaphysical assumptions that are, in a certain sense, unrealistic.
In this situation, the problem of the inaccessibility of the past remains unsolved, and the uniformity principle, as I will show later, does not help. In spite of this, it should be remembered that the uniformity principle is a scientifi c method uniting geology with other natural sciences—physics, chemistry, biology, etc.—thereby creating a more sound and wider scientifi c basis for it. In other words, geohistorical thinking derives, in a certain sense, from the laws of physics,16 and this is confi rmed by the different elements of the uniformity principle (to be specifi ed later herein). This is probably one of the reasons that geologists do not worry about their realistic worldview, since the same worldview is also accepted in physics, and therefore the questions and diffi culties presented here must necessarily be solved in the context of basic theories of physics and other natural sciences.
In addition, it should be remembered that whoever seri-ously doubts realism in relation to the geological past, actu-ally casts doubt on human cognition in general. In principle, every kind of cognition, not only geological cognition, must overcome the gap between past and present. For instance, on the basis of a physical fact according to which light moves at a constant velocity, one might argue that direct visual observation does not exist in reality. Everything that we see around us is the past. Direct cognition continues for some time. This perceptual process has a natural time limit that at least takes time from the moment the rays of light pass from the object perceived through
14 We are referring, for instance, to the uniformity principle in nature, which, as Hume has already shown, is a specifi c principle that cannot be logically and empirically justifi ed. There is no logical obligation to assume the uniformity of nature, and this argument cannot be empirically confi rmed, since any such at-tempt would have to assume this very argument as a basic assumption. 15 The uniformity principle appears to bridge the gap between past and present and thus represents, in a certain sense, the reductionist approach regarded by some modern philosophers as an antirealistic approach (see, for instance, the discussion of the subject in Dummett, 1978, p. 359).16 This standard concept is accepted by most geologists and science philoso-phers. For consideration about this issue, see, for example, Schumm (1998, p. 2–3) and Kitts (1977, p. xi).
13 The use of causal explanations and logical arguments that enable one, by means of evidence from the present and universal timeless generalizations, to learn about the geological past and reconstruct it enhances the argument con-cerning geologists’ belief in realism in relation to the past.
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 21
our eyes until the stimulus reaches the brain. This means that there is a natural (physical) gap between the occurrence of the real event and its perception. This gap cannot be bridged within the limits of our physical nature, and it is therefore valid for any kind of human cognition and not particularly for geohistorical cognition. This is probably another reason why geologists do not attach great importance to it and see no diffi culty in the application of the uniformity principle and the passage from certain geological events to others that happened in a different geological time.
ELEMENTS OF THE UNIFORMITY PRINCIPLE
As stated already, the uniformity principle serves as an arti-fi cial methodological tool, the main function of which is to apply the realistic belief of the geologists concerning the geological past of Earth. Nevertheless, diffi culties in accurately describing it and defi ning it have cropped up throughout the history of geology.17 Many of the defi nitions suggested were too general or not sub-stantial enough, and these provided reasons for many arguments (Hubbert, 1967, p. 4). As a result, a lot of confusion arose around the concepts of actualism and uniformitarianism (Simpson, 1975, p. 271).18 To conclude, the concept of uniformitarianism was not properly understood by geologists, and this caused many errors in modern geological literature (Austin, 1979; Shea, 1982). In order to explain the confusion and clarify the concepts correctly, some philosophers/geologists distinguished between two central concepts of uniformitarianism (Hooykaas, 1959; Gould, 1965; Visotskii, 1961):19
1. substantive uniformitarianism, or simply uniformitarianism, and
2. methodological uniformitarianism, or simply actualism.
Substantive uniformitarianism claims that geohistorical uni-formity or identity exists between present and past geological events.20 This is actually an ontological argument (that can be subjected to empirical examination), according to which there is uniformity (or identity) in the force, rate, and type of geological causes and effects throughout deep geological time—namely, the
phenomena (causes, processes, and effects) do not change their force, rate, and type in the course of time.
In contrast to substantive uniformitarianism, methodological uniformitarianism (actualism) is a methodological process that enables us, by means of an analogy,21 to know and explain the geological past of Earth based on geological events observed in the present. As a matter of fact, actualism is based mainly on empirical observations concerning the types of causes and the forces of the geological phenomena at present. On the basis of these observations, geologists make inferences about types of causes and their force in the past. This methodology is based on the uniformity of the laws of nature in time and space.22 In other words, in contrast to substantive uniformitarianism, which con-stitutes an ontological argument or a principle pertaining to the actual world, actualism is merely an epistemological methodol-ogy enabling geologists to study the geological past of Earth by means of geological observations in the present.23
The confusion of actualism with uniformitarianism had already begun at the time of Lyell, who, overtly or covertly, united methodological uniformitarianism and substantive uniformitari-anism24 and included four different elements in a single unifor-mity principle (Gould, 1987, p. 117–126):25
1. The uniformity law—The laws of nature do not change (static and uniform) in time and space. This is not a statement concerning Earth, but it is an a priori statement of the scientifi c method. With out this assumption, geological research would not have been possible.26
2. The uniformity of processes or actualism—Geological phenomena that occurred in the past may be explained by means of geological causes and factors operating in the present. Accord-ingly, the laws of geology act over time by means of the forces acting in the present. Thus, if we wish to reconstruct the geologi-cal past of Earth, we must rely on active processes forming the face of Earth in the present. This too is not a statement pertaining
17 Regarding diffi culties to defi ne the uniformity principle, see, for example, Austin (1979) and Shea (1982).18 As we shall see later herein, Lyell himself regarded the uniformity principle in a general and complex manner, including the consensus with regard to the method and ontological argument pertaining to complete uniformity in nature (Gould, 1987, p. 105). Apparently, this is what caused the confusion among geologists/philosophers in the past and in the present. 19 A similar terminology was suggested by Austin (1979), Gould (1984), and Rudwick (1971). Some scientists disagree with this terminology and claim that it is insuffi cient for modern geological research (Şengör, 2001, p. 38). I do not intend to discuss this argument in this paper but accept the terminology as presented by Gould (1984), which is accepted to this day by the majority of geologists. 20 The term “geological event” includes (1) the causes and factors that brought about the geological phenomena, (2) the geological process (the activating mechanism) of the phenomenon, and (3) the effects caused by the geological process.
21 As in other historical sciences (see, for instance, Gifford-Gonzalez, 1991, p. 215–226), the process of inference by means of analogies is often used in geology (for more in-depth discussion regarding this issue, see Baker, 2013). In this process, the researcher observes processes occurring in the present and, on the basis of direct observation, infers the causal relations between them. After that, he observes the historical objects—by direct observation of the remnants of the past—and compares them to contemporary objects. This comparison enables him to make inferences concerning causal relations of processes that occurred in the past and through them to reconstruct the causes of the effects observed in the present.22 About the role of methodology in Lyell’s theory, see Laudan (1982).23 About actualism as an epistemological argument and uniformitarianism as an ontological argument, see Gruza and Romanovsky (1975).24 Some geologists claim that Lyell united the concepts on purpose as part of his tactics. First, he united the concepts under the name “uniformity” and later argued that geologists must accept substantive uniformitarianism because their practice is based on methodological uniformitarianism (Gould, 1984, p. 10).25 For further discussion of this point, see Rudwick (1972, p. 164–217), Gould (1984, p. 11–12), Virgili (2007, p. 575–577), Camardi (1999, p. 537), Austin (1979, p. 32), and Anderson (2007, p. 451).26 Actually, this argument is valid for all scientifi c disciplines and not solely for geology. This is an a priori assumption, but the lack of it would preclude all science and any explanation of the physical reality. Lyell, an exemplary sci-entist, introduced this element into geology and placed it on solid scientifi c foundations.
22 G. Kravitz
to Earth, but an a priori statement pertaining to a process or a scientifi c method.27
3. The uniformity of rate or gradualism—The rate of change of geological processes is slow, stable, and gradual and contin-ues over time. This is an empirical statement pertaining to Earth (with regard to the uniformity of the rate, speed, and force of geological processes) that can be tested empirically.
4. The uniformity of state or nonprogressionism—The history of Earth does not have a defi nite direction (vector). The geological state of Earth changes in cycles and does not really progress in any direction. The adherents of this theory refer to the uniformity of state on the face of Earth. Earth always looks and behaves in the same way over long geological time; in other words, Earth is in a dynamic steady state. This is an empirical statement accord-ing to which the face of Earth has not changed substantially since its creation, and it has remained in the same dynamic balanced state. One moment is identical to any other moment. Species may change but the average complexity of life endures forever.
The disputes arose around these elements, especially around the degrees of uni formity, i.e., four different possible types of change refl ected in geological phenomena over time (Hooykaas, 1959, p. 3–32):
1. Geological phenomena that happened in the past differed in type and were more energetic than geological phenomena happening in the present.
2. Geological phenomena that happened in the past differed in type but not in energy from geological phenomena happen-ing in the present.
3. Geological phenomena that happened in the past differed in energy but not in type from geological phenomena happen-ing in the present.
4. Geological phenomena that happened in the past differed neither in type nor in energy from geological phenomena happening in the present.
The fi rst two elements of the uniformity principle, as shown previously, are methodological characteristics of methodological uniformitarianism that cannot be subjected to empirical proof and are actually versions of more basic principles—the principle of induction and the principle of simplicity.28 Without these prin-ciples, any geological explanation is not possible. In geology, as in any other scientifi c discipline, scientists must assume that
uniformity of the laws of nature and the laws of geology exists. This assumption enables one to make an analogy by stating that the geological causes active in the present are identical to those that were active in the past. Such an analogy makes it possible to construct the geological past of Earth, on the basis of geological observations of processes in the present. On the other hand, the last two elements deal with the actual state of Earth (substantive uniformi tarianism) and can be empirically proved or refuted, on principle. Therefore, most of the disagreements between geolo-gists from the time of Lyell until this very day refer to them.
In the course of the history of geology, most of the arguments of substantive uniformitarianism have been refuted (Gould, 1965, p. 226), or rather the elements of uniformity of rate and uniformity of state were refuted on theoretical and empirical grounds (Austin, 1979, p. 33–40). Lyell himself, at the end of his professional career, accepted Darwin’s theory of evolution on the basis of empirical-paleontological fi ndings and renounced the idea of uniformity of state.29 This renunciation introduced a historical element into his theory and emphasized the difference between himself and Hut-ton. On the basis of this renunciation, Lyell promoted the objective understanding of the history of Earth in terms of a series of chrono-logical events over a long time period. By renouncing uniformity of state, he actually rejected the notion of nonprogressionism and ac-cepted evolutionism (Gould, 1984, p. 14). Of no lesser importance is the fact that Lyell’s work was greatly infl uenced by the work of Kelvin30 concerning the laws of thermodynamics in general, and the second law of thermodynamics in particular. Kelvin is known to have rejected the theories of Hutton and Lyell concerning the uniformity of rate and the uniformity of state because they refuted the second law of thermodynamics (Dott, 1998, p. 17). This rejec-tion was in line with Darwin’s evolutionism, and on this account, Lyell renounced the element of uniformity of state, and, therefore, this element was abandoned by geologists in the nineteenth cen-tury, replaced by evolutionary geology (Austin, 1979, p. 33).31
However, Lyell never abandoned the element of uniformity of rate. He assumed that neither the internal energy of Earth nor the geological forces (causes) have changed and lessened over time. Therefore, all the disputes from the nineteenth century to this very day, concerning the uniformity principle, have been about this element. Those who referred to the age of Earth in terms of thousands of years required speedy processes and mighty forces in order to explain the present geological state of Earth. On the other hand, those who referred to the age of Earth in terms of millions or billions of years claimed that geological processes were gradual, extremely slow, and uniform over deep geological time.32 Finally, the dispute was resolved on an empirical basis;
27 This component is based on—and in certain respects derived from—the fi rst component of the uniformity of law. Without assuming uniformity in nature, one cannot talk of uniformity of geological processes, based on and explained by the laws of nature. The metaphorical sentence “the present is the key to the past” refers to the two components together. To state it differently, when Lyell mentions the uniformity of causes and forces, he was not referring only to physical and chemical uniformity of the material, but also to the uniformity of processes in the course of the interaction between material and energy that produces specifi c geological events. As a matter of fact, a certain hierarchy is discernible between the various components of the uniformity principle. The laws of physics occupy the top of the pyramid, and all the other components derive from them in an orderly way (Austin, 1979, p. 323).28 Explanation of this argument will come later.
29 For further discussion of this point, see Gould (1987, p. 167–178).30 Also, his work was infl uenced by Agassiz’s glacier theory.31 About geological evolution of the Earth, see, for example, Dott and Batten (1976) and Ozima (1987).32 Actually, it is an ancient debate between catastrophism and uniformitarianism. About this debate, see, for example, Cannon (1960), Hallam (1989), and Wilson (1967). For further discussion about development of catastrophism, see Ager (1993), Albritton (1989), Gretener (1984), Hooykaas (1970), Palmer (1993), Rudwick (1997), and Benson (1984).
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 23
the empirical data showed beyond doubt that the internal energy of Earth was reduced over time and that the geological phenom-ena—causes, processes, and effects—are not identical over deep geological time.33
The two central remaining elements of the uniformity prin-ciple are uniformity of law and uniformity of process. The mod-ern theory of actualism, governing the day-to-day work of the geologist, is based on these elements. One can, therefore, say that in order to employ the geological-evolutionary worldview (evolutionism) current among modern geologists, modern geol-ogy uses the uniformity principle as a methodological principle encompassing two major elements—uniformity of law and uni-formity of process. Although these elements are actually a priori assumptions (metaphysical in a sense), and cannot be subjected to empirical or logical proof, they enable geologists to realize their realistic belief concerning the past of Earth, in an artifi cial way. In this sense, these assumptions make a geohistorical expla-nation possible by means of logical-causal arguments—retroduc-tive and abductive inferences.34
THE ROLE OF THE UNIFORMITY PRINCIPLE IN THE GEOHISTORICAL EXPLANATION
The Uniformity Principle as a Principle of Simplicity
The practice of geology as a historical science is character-ized by three major attributes:
1. Geology as a historical science deals with specifi c (unique) events in a certain time and place and usually does not con-cern itself with universal laws, but it accepts them as given.35
2. Geology as a historical science deals mainly with past events.
3. Geology as a natural science deals mainly with active causes (efficient causes) and not with functional causes (final causes).
Items 1 and 2 distinguish geology from the other natural sciences, such as physics and chemistry, and categorize it as a historical science similar to conventional history.36 On the other hand, item 3 distinguishes geology from conventional histori-cal sciences and places it in the realm of natural sciences.37 The uniformity principle connects between these attributes, or rather, artifi cially unites unique events within a context of causal laws, characterized by active causes, making it possible to reconstruct the geological past of Earth. We may, therefore, conclude that the characterization we are dealing with is a methodological simplicity principle38 of a certain kind that geologists are obliged to assume because the past is inaccessible to them.39 They are obliged to make use of this principle because of their inabil-ity to examine all geological events that happened in the past and will happen in the future and must, therefore, content them-selves with a partial examination of the present, which serves as a foundation for universal generalizations about events that are not observed directly.40 By applying the uniformity principle (serving as the principle of simplicity), geologists can predict the future and reconstruct the geological past. On the basis of single geological observations in the present, they construct a
33 Actually, the component of uniformity of state is strongly connected to the component of uniformity of rate, and therefore refuting the uniformity of state also means refuting the uniformity of rate, in a certain sense. For empirical evi-dence on refuting uniformity of state and uniformity of rate, see, for instance, Austin (1979, p. 39–40) and Gould (1965, p. 226).34 The uniformity principle is an accepted principle in science and is not con-fi ned to geology. In most scientifi c disciplines, its function is to enable pre-dictions. In geology, as we will see herein, its main function is to enable the reconstruction of the geological past, and, therefore, in this paper, I will deal mainly with abductive inferences—enabling geologists to make inductions of the past (retrodiction)—and not with deduction. (Here, it should be noted that in this chapter, contrary to the view as presented and expressed by Peirce and others, the term “abductive inferences” is introduced in a simplest manner as a kind of inductive inference. In order to simplify, I defi ne here as “inductive” all inferences that are not “deductive.”) However, we should bear in mind that all types of inferences are used in geology (such as induction, deduction, ab-duction), as in other natural sciences. In this context, see Engelhardt and Zim-mermann (1988, p. 230, 233, 80–82) and Baker (2013). Generally speaking, geologists as scientists are particularly interested in researching causes, effects, and geo logical laws. Causes and laws allow them to predict effects (deduction). Joining causes and effects allows them to predict laws and generalizations (in-duction) and the knowledge of effects and laws enables them to formulate hy-potheses concerning causes (abduction). This last kind of inference (abduction) is unique for geology, making it, in a certain sense, a unique science.35 Certain scientists argue that there is no difference between generalizations based on singular events and universal laws, and, therefore, there is no substan-tial difference between a science such as physics and a historical science like geology. Limitations of space prevent me from addressing this argument in this paper. In this context, see Şengör (2001, p. 7).
36 About differences between historical science and experimental science, see Cleland (2001, 2002) and Dodick et al. (2009).37 This difference results from the fact that conventional history is concerned mainly with historical processes focused on humans. The engagement with hu-mans, by its very nature, requires it to focus on functional causes (fi nal causes) and other human processes (such as sociological, anthropological, economic, psychological, geographical, etc., processes). Geology, in contrast, deals with natural processes based on the laws of physics and chemistry, and therefore, as a rule, does not deal with functional causes but with active natural causes. Geo-logical reconstruction is not possible thanks to psychology and sociology, but thanks to the physics, chemistry, and physiology of the world of life (Hooykaas, 1959, p. 149). 38 Defi ning or describing the simplicity principle accurately is a diffi cult task, and still there are many philosophers who engage in it. It is not my intention to deal with it in detail in the present paper, but only to provide a general ex-planation in order to understand its connection with the uniformity principle. What I am interested in is to understand the methodological function that this principle performs in geological explanation without starting a discussion on its defi nition or structure. It is not necessary to understand the simplicity prin-ciple thoroughly and to know whether it really refl ects the simplicity of nature, but rather to understand its role in geological explanation. For more detailed under standing of the simplicity principle and its use in science and geology, see Sober (1975, 1988), Anderson (1963), Goodman (1967, 1964), Baker (2010), and Zellner et al. (2001).39 In this context see, for instance, Simpson (1975, p. 275–276), Shea (1982, p. 458), and Peters (1996, p. 70). Historically, the fi rst to connect uniformi-tarianism to the simplicity principle was the philosopher Francis Hutcheson (1694–1746) (Şengör, 2001, p. 38). In addition, Lyell too regarded uniformitari-anism as a kind of simplicity principle for the purpose of geological research on the past of Earth. Some claim that what began as a methodological assumption by Lyell gradually became a geohistorical theory of Earth (Dott, 1998, p. 16).40 From here, it is clear that the simplicity principle is related to the induction and curve-fi tting problems. For further discussion of these problems, see, for example, Sober (1988, p. 45–46) and Baker (2010, p. 21).
24 G. Kravitz
kind of timeless function enabling them to predict the future and reconstruct the past.41 This function is continuous and connects the single observations, and in this sense constitutes a simplifi ca-tion enabling geologists to predict the geological state between observations, before and after them, i.e., it bridges between ob-servations and theoretical hypotheses concerning the geological state of Earth at different times (Sober, 1988, p. 45), or, on the basis of single observations and the uniformity principle, geolo-gists interpret and artifi cially imagine reality with the help of analogies,42 extrapolations, abductive and inductive inferences. We are actually referring to an artifi cial principle not necessarily refl ecting reality as it is—it does not necessarily describe na-ture, but rather it describes what we think about nature and what we think about the methodology we wish to apply in order to study it (Goodman, 1967, p. 94–95). Although it serves as an instrument of realism in relation to the past, it is itself a method-ological-artifi cial principle based on the antirealistic uniformity principle, imagining reality and not necessarily presenting it as it is in reality.43
Usually science is concerned with an effort to simplify theo-ries in order to describe them simply and elegantly. Indeed, it is easier to assume that all the factors, processes, and geological phenomena seen today are similar to those that happened in the past and are bound to happen in the future. Geologists are look-ing for theories to explain a large gamut (as large as possible) of geological phenomena by means of the smallest possible number of assumptions (Anderson, 1963, p. 177). This goal is attained by incorporating empirical data into a broad theoretical framework, which is what geologists are trying to do when they adopt the
uniformity principle.44 They incorporate observations in the pres-ent in a timeless system, enabling them to make inferences about the geological past or future of Earth.45 In other words, this prin-ciple simplifi es reality by artifi cially integrating empirical fi nd-ings and observations into a larger system of laws (Hooykaas, 1959, p. 163–164).46 Thus, the uniformity principle can be re-garded as a pragmatic ahistorical abstraction serving as an inter-pretative tool (hermeneutic to a certain extent)47 used to decipher historical phenomena. Any kind of understanding that recognizes relevant connections involves the capacity to make abstractions (Lorand, 2010, p. 93). Connections between objects and infer-ences from fi ndings, i.e., between what is given and known and what is not given and unknown, cannot be made without a certain degree of abstraction, enabling orderly connections between ob-jects. Abstraction essentially means paying attention to a certain collection of attributes of the object (Lorand, 2010, p. 93–97). The uniformity principle as a historical abstraction (in contrast to individuality) assumes similarities between objects and ignores the differences between them. By means of this principle, geolo-gists abstract the object of its unique attributes and turn it into something similar, in certain respects, to other objects.48
From the previous discussion, we understand that inductive thinking is based on the uniformity principle, and that it is com-mon in all natural sciences. Its contribution is vital, and without it, as I will show later herein, the modern geological explanation breaks down. Geology is concerned mainly with the past, and, therefore, its explanations rely on abductive inferences49 and on observations carried out in the present. Unless based on the uni-formity principle, these inferences cannot be made, and the his-tory of Earth cannot be reconstructed. Although not exclusive to geology, it has many practical advantages in the context of the method of actualism. Its greatest advantage lies in the fact that it establishes the actualist methodology as a basis for observations of geological phenomena conducted in the present, and it makes inferences concerning geological phenomena that occurred in the past. What are the advantages of geological observations
41 Geology, as a historical science, deals with events that occurred in different periods and, therefore, had to be united within a timeless system of laws, in order to derive them logically from each other. Without this—geological expla-nation would be impossible. In this respect, the uniformity principle is actually an ahistorical principle, supporting historical explanations of events that oc-curred in the course of the geological history of Earth.42 Research methodology by analogy is common, not only in geology and his-torical sciences, but also in the experimental natural sciences. For instance, the laws of physics are based on an analogy between causes and effects in several cases. Gravity (the cause) produces the pull (effect) of one body, just as it pro-duces the same effect on another body in a different place and time. Therefore, it is possible to make an analogy from one body to another (a specifi c analogy in sense and context to gravity). In geology, we are dealing with similar but not identical phenomena, and therefore the geologist is obliged to choose the at-tributes he wants to in clude in the analogy, which could be causes, processes, or geological effects. Since we are dealing with similar processes and phenomena, the act of choosing is prone to be very diffi cult and subject to errors. The dif-fi culty lies particularly in the fact that there are no clear-cut rules, subject to em-pirical testing, for the legitimacy of analogies. Some geologists even maintain that analogy is inappropriate in geology and suggest using homology instead (Schumm, 1998, p. 19). For lack of space, I will not address this argument and not go into any more details than those already mentioned with regard to the analogical research methodology. For additional details, see Hooykaas (1959, p. 154–162) and Baker (2013).43 This is my opinion and it does not necessarily refl ect the opinion of other geol-ogists. As stated, most geologists are realists in relation to the past, and, there-fore, I assume that they believe that their work methods are adjusted to their ontological conception concerning the true structure of the world. Apparently, they believe that their methodology refl ects the real world and is, therefore, able to expose the true geological structure of Earth. In other words, they actually believe that this method cannot be artifi cial, but must refl ect the real world.
44 In this respect, geologists behave just like most scientists, who search for law-fulness and simplicity in theory. However, some maintain that nature is much more complex than the abstractions that we make (Baker, 2000, p. 6). Accord-ing to them, nature is so complex that it cannot be explained by the simplicity principle. This is also the reason that it cannot be explained solely by physics but requires theories from a variety of scientifi c disciplines. See in this context Cartwright (1999) and Dupré (1993). I agree with this attitude in certain re-spects and will discuss it later herein.45 In a certain sense, this principle provides methodological justifi cation for the use of inductive and abductive inferences. Naturally, this justifi cation is not meant to be a logical justifi cation of the induction process.46 A similar argument can be seen in Goodman (1967, p. 96–97).47 About natural science as hermeneutic of instrumentation, see Heelan (1983).48 This artifi cial procedure eliminates the individuality of the object and thus, apparently, enables connection between objects. In other words, this abstrac-tion enables a verbal explanation connecting the individual case to a general principle, a pattern or a theory. It affords signifi cance to the singular case at the expense of relinquishing its individuality.49 As mentioned in footnote 34, the abductive inference is a particular type of inductive thinking, enabling geologists to make inductions of the past—infer-ence from knowledge of effects and laws to the hypotheses con cerning causes. The logical structure of this inference will be displayed later herein.
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 25
in the present? The most obvious advantages are the following (Frantsuzova and Pavlinov, 1968, p. 9):
1. Geological processes and phenomena in the present are directly accessible to the geologist. This enables the researcher to study all the details, taking the dynamics and all the factors and products into consideration. This enables the geologists to introduce models and quantitative methods into their research.
2. Processes and phenomena can be continuously studied in the present and their interrelations identifi ed. This is not always easy when the sources from the geological past are based on in-consistent “remnants” and pieces of information concerning geo-logical processes and phenomena that occurred in the past.
3. The study of geological processes and phenomena in the present can, to a certain extent, serve as laboratory experiments.
4. Products of processes that occurred in the past should not be compared to products of processes occurring in the present only if they happen to be completely identical, but also if cer-tain differences exist between them. This enables the geologist to learn about specifi c geological conditions existing at the time in the past in which the processes or phenomena occurred.
In spite of these advantages, actualism (based on the unifor-mity principle) should not be regarded as a universal method pro-viding complete knowledge about the geological past of Earth. Actualism is based on inductive (abductive) and analogic infer-ences enabling geologists to make extrapolations concerning the geological past and future of Earth. Assuming that the geological state is changing constantly (a progressive process), we can say that the longer the time that has elapsed between the geological phenomenon in the present and the one that occurred in the past, the less is the chance of similarity between them. When compar-ing relatively young periods of time, we assume that similarity between the geological phenomena exists from the point of view of their causes, processes, and effects, whereas in comparing an-cient periods with the present, similarity is usually retained only in the mechanism, or, under the most favorable circumstances, between causes as well. In other words, reliability of inductive inferences is reduced for longer elapsed time periods. In some radical cases, it is impossible to make any inductive inferences from geological events in the present about those that occurred in the past (Şengör, 2001, p. 5).
Another problem lies in the fact that the present is only a small window to the past and does not provide geologists with a full set of analogies (Frodeman, 1995, p. 965). This is true in two respects:
1. Sometimes geological events that occurred in the past are not seen in the present. This may result from the rarity of these events (perhaps even uniqueness) and from the fact that we have not had the opportunity to see them again, or that they still occur today but are hidden from us.
2. The geological processes that form the face of Earth are usually very slow and take a long time. Sometimes the “life span” of these periods exceeds that of humans. Therefore, in these cases, it is diffi cult to identify the geological effects (outcomes) and make inferences about their causes.
It should also be mentioned that the uniformity principle is manifest in qualitative similarity (not in identity) of a rudimen-tary and general nature. Since geologists do not have direct ac-cess, they must content themselves with a partial and imprecise reconstruction, based on similarity between events in the present and events in the past by means of abstractions and analogic in-ferences. Unless geologists consider the possibility that similar outcomes could result from different causes,50 analogic inferences could lead to mistaken conclusions,51 for instance, geological re-search of processes occurring in the present in deserts, teaches us that in hot and dry environmental conditions, there is no vegeta-tion. However, it is possible that in the remote geological past, the lack of vegetation resulted from the different chemical com-position of the atmosphere and not necessarily from hot and dry environmental conditions.
Geologists should, therefore, be aware of the limitations of the uniformity principle and the actualist method, and make use of alternative geological methods according to their needs. In this situation, some modern philosophers who support actualism suggest that geological research should be conducted, on prin-ciple, according to the following four phases (Frantsuzova and Pavlinov, 1968, p. 11):52
1. They should look for signs (evidence) of geological phe-nomena from the past. They should study such signs and evi-dence and describe them from the morphologi cal aspect, their chemical composition, and stratigraphic structures.
2. The evidence mentioned in item 1 requires a primary theoretical explana tion supported by models based on up-to-date geological theories.
3. The actualist method (the uniformity and simplicity prin-ciple) should be used for comparing causes, processes, and ef-fects in the present, and causes, processes, and ef fects in the past.
4. On the basis of comparison in items 1 and 3, reconstruc-tion of the past should be conducted on the basis of similar causes, processes, and outcomes observed in the present. In this stage, the geologist actually improves his/her primary explanation (as he/she did in item 2) on the basis of an empirical comparison car-ried out in item 3 (by means of the actualist method).
The Uniformity Principle and the Second Law of Thermodynamics
As mentioned already, modern geologists have adopted, on the one hand, the geological worldview of evolutionism. On the other hand, they use the actualist method based on the uniformity principle, consisting of two main elements, uniformity of process and uniformity of law. Uniformity of law refers mainly to the
50 Further discussion about the connection between causality and uniformity principle will come later.51 With regard to errors in the process of geological interpretation, see Schumm (1998).52 In principle, I also support this method, and therefore the description of the stages includes my interpretations and amplifi cations that I consider to be rel-evant for contemporary research methodology. For more in-depth discussion regarding this issue, see Engelhardt and Zimmermann (1988, p. 233–331).
26 G. Kravitz
basic laws of physics and chemistry, while uniformity of process refers to various geological processes.53
Which law of physics can explain developmental geological phenomena on a macroscopic level? Actually, there is only one such law, the second law of thermodynamics.54 Naturally, in geol-ogy, various physical laws are assumed and applied, but there is only one historical law that defi nes geology as a historical sci-ence. Geologists rely on this law and use it consciously or uncon-sciously in all their geohistorical explanations. In other words, if geology, as a historical science, assumes the uniformity of law, it must rely, among others, on the second law of thermodynamics, the only one that enables a historical explanation. Since geology deals mostly with the past (Kitts, 1978, p. 218), reliance on the second law of thermodynamics is an essential condition for any geohistorical thinking.55
As already mentioned, the fi rst scientist who explicitly used the second law of thermodynamics to explain geological phenomena and to renounce the uniformitarianism of Lyell and Hutton, was Kelvin, who applied it in his theory of the reduc-tion of the temperature of Earth and its loss of energy (Hubbert, 1967, p. 25). In contrast to Lyell and Hutton, Kelvin maintained that Earth was a developing system that loses energy and, there-fore, cools over time. He rejected the theories of Lyell and Hut-ton concerning the uniformity of rate and the uniformity of state, because they confl ict with the second law of thermodynamics (Dott, 1998, p. 17).56 In this sense, Kelvin assumed the thermo-dynamic, one-directional and irreversible time arrow, and laid the foundations for the geological developmental orientation that is accepted by most geologists until today.57
Most modern geologists assume, like Kelvin, that the uni-formity principle applies according to the second law of thermo-
dynamics, which, by its very nature, promises the rejection of the uniformity of rate and the uniformity of state. Moreover, this uniformity principle based on the second law of thermodynam-ics, by encompassing the uniformity of law and the uniformity of process, is not just an inductive principle (the principle of sim-plicity) but a geohistorical principle without which geology can-not exist as a historical science.
What distinguishes the second law of thermodynamics and why is it so important for the geohistorical explanation? Most basic theories in physics do not distinguish between past and fu-ture, i.e., they are symmetrical with regard to time—they have no clear preferred direction of the time fl ow. The equations rep-resenting these laws are symmetrical in relation to the rever sal of the direction of time. As for the description of the physical process, the reversal of the sign of the variable time in these equa-tions makes no difference; for instance, we can sense this process if we make a fi lm photograph of the process of elastic clashing between two billiard balls in a table lacking friction, and then run the fi lm from the end to the beginning. We will not notice the dif-ferences between the real state of clash and that appearing in the reversed fi lm, nor will we be able to defi ne the “correct” direction of the movement of the billiard balls.
In contrast to the other laws of physics, the second law of thermodynamics describes the macroscopic world of nature as a world governed by an asymmetric and irreversible time arrow—in other words, it describes a series of unique events, joined to each other and developing in one direction (law of historical succession). Therefore, the majority of physical phenomena, ir-reversible in the time dimension, are explained by this law. In sta-tistical mechanics, the second law of thermodynamics is a law of probability based on a basic concept in physics, that of entropy. Entropy is a measure of disorder in physical systems. The second law of thermodynamics states that in closed isolated physical sys-tems, entropy can be maintained or can even grow in the course of time up to a point of thermodynamic equilibrium, which is the most probable fi nal state from a statistical point of view (as an as-pect of microscopic dispersion of states) in which the system can exist (a state of maximal entropy and disorder). The law enables the inference that every closed and isolated physical system has a time arrow with a clear direction, from a state of low entropy (in the past) to a state of thermodynamic equilibrium (in the future), in which the system receives the highest value that it can sustain.
To state it differently, any isolated system is bound to de-velop over time—from past to future—from a state of low en-tropy (a certain degree of order) to a state of greater entropy (disorder in relation to the primary state of the system). This is the reason that it is impossible to avoid making use of the sec-ond law of thermodynamics in geohistorical explanation, since it is the only historical law in physics. Therefore, if geologists are anxious to rely on physical laws58 and provide geohistorical ex-planations as well, they cannot refrain from accepting the second
53 We should bear in mind that the uniformity of law is a generic principle rel-evant for all the natural sciences and not exclusively for geology. Uniformity of process, on the other hand, is a unique principle for geology. In the past, geologists distinguished between these variables, by calling the laws of nature “primary causes” and the geological processes “secondary causes.” They as-sumed that the secondary causes derive, in certain respects, from the primary causes. In this context, see, for instance, Simpson (1975).54 It should be noted that the application of the second law of thermodynam-ics advocated in this chapter is not affected by the strict limitation to closed systems. This means that even if we assume that Earth is an open system, it is still subject to the second law of thermodynamics. (Actually, as a part of the entire universe, it cannot be really an open system.)55 It is generally assumed that a good historical narrative contains two kinds of connections: a causal connection between events described in the narrative, and the connection of a part to the whole, uniting all the singular events into one complete narrative (Hull, 1975, p. 260). In this respect, a historical narrative based on the uniformity principle and on the second law of thermodynamics meets the criteria of a good historical narrative.56 Although Kelvin erred in his calculations concerning the age of Earth, he was the fi rst to assume the evolution of Earth on the basis of the second law of ther-modynamics. He did not err in that he based his theory on the laws of thermo-dynamics, but rather because in his days, the phenomenon of the decomposition of radioactive material (radioactive decay) inside Earth was not known. With regard to Kelvin’s calculation of the age of Earth, see Burchfi eld (1975, 1998).57 In his calculations of the age of Earth, Kelvin wanted to reevaluate Darwin’s theory of evolution, and particularly of the dominance of physics over geology. He calculated the age of Earth on the basis of the second law of thermodynam-ics and thus cast doubt on Darwin’s simplistic and inaccurate calculation in the Weald valley (Dodick and Orion, 2003, p. 198–199).
58 Apparently they are interested; otherwise, they would not have assumed the uniformity of law as a basic assumption.
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 27
law of thermodynamics as a basic assumption.59 This enables them to distinguish between geological events that occurred in the past and those occurring in the present, and to join them in a framework of cause and effect, the cause always being situated in the past, and the effect always being situated in the present, and not the other way round.60 As we have seen, this is the most basic assumption on which the geologist bases his belief in real-ism in relation to the past. This means that the second law of thermo dynamics and Ludwig Boltzmann (1844–1906)’s “past hypothesis” (from which the thermodynamic time arrow derives) play a central role in geohistorical explanation, without which the modern explanation would break down.61 Their role is especially evident in the context of the uniformity principle, an ahistorical principle that enables a historical explanation of events that oc-curred in the course of the geohistory of Earth. On the face of it, there is a contradiction here (how is it possible that the same principle is ahistorical, on the one hand, and contains a historical law, on the other?), but in reality, such an explanation is possible by assuming the second law of thermodynamics—a physical law, ahistorical in nature—and the thermodynamic time arrow, which makes it historical in this sense.
Nevertheless, it should be mentioned that the second law of thermodynamics is not without diffi culties and disagreements. On the macroscopic level, the law explains the irreversible phe-nomena and the asymmetry of the time arrow, as perceived in-tuitively by us, but this is not the picture obtained if we want to explain macro scopic events based on physical laws on the micro-scopic level.
As already mentioned, most of the theories in physics do not distinguish between past and present or, rather, are symmetrical with regard to the reversal of the direction of time. Equations rep-resenting these laws are symmetrical with regard to the descrip-tion of the physical process; the reversal of the sign of the time variable in these equations makes no difference, and, therefore, if you assume that the world is composed of microscopic particles (atoms), the reduction of the second law of thermodynamics to Newton’s mechanic laws is impossible. For example, if we fol-low the reaction of gas on the microscopic level, we will fi nd that the motion of the gas molecules is governed by Newton’s laws and equations of motion, which are symmetrical with re-gard to the direction of time. If we place t+ and t– in these equa-tions, we will obtain identical descriptions of the process from the physical point of view in both cases. On the other hand, if we examine the same gas on the macroscopic level, we will notice that the gas reacts according to the second law of thermodynam-ics—the gas spontaneously aspires to achieve a state of thermo-dynamic equilibrium. In other words, when we examine the gas on the macroscopic level, we identify an asymmetrical and ir-reversible direction of time, and when we do so on the micro-scopic level, the process is neutral (symmetrical) with regard to the direction of time. Therefore, the main diffi culty lies in the fact that the second law of thermodynamics is irreversible and asymmetrical with regard to the direction of time and cannot be derived fully from the laws of motion of Newton’s mechanics (or from the laws of motion of quantum mechanics), which are sym-metric and reversible in time.62
If the world of atoms described by Newton’s laws of mo-tion is symmetrical and reversible in the aspect of the time arrow , then it is diffi cult to explain why, in our daily observations, we encounter processes occurring in one direction (i.e., governed by the second law of thermodynamics). Physicists and philosophers attempted to solve this diffi culty in the nineteenth century, and the subject is still open to discussion and disputes. The outstanding physicists who have tackled these problems are Boltzmann and James Clerk Maxwell (1831–1879). Both of them reached the conclusion that the second law of thermo dynamics is a statistical law, very plausible, but not mandatory, i.e., under certain excep-tional conditions, it can be renounced without its universality suffering severe harm. In order to explain the irreversibility of the time arrow in reality, Boltzmann argued that natural phe-nomena are not explained solely by laws but also by initial con-ditions. Such states can occur in initial conditions of phenomena occurring in the direction of time, as observed today, but also in initial conditions in the opposite direction of time, which is possible but has never been observed in reality. Thus, Boltzmann attempted to solve the problem of reduction between Newton’s laws of mechanics and the second law of thermo dynamics by
59 Actually the second law of thermodynamics is well adapted to common sense and to the physical world in which we live our everyday lives (on the macro-scopic level). The assumption of an asymmetrical and one-directional time arrow is an intuitive assumption. In everyday macroscopic processes, we iden-tify a one-directional fl ow of physical processes in time. We are present in our ageing process and always remember the past and never the future. Even simple processes like that of ice cubes melting into water are subject to the second law of thermodynamics. Ice always melts in water, and we have never seen an op-posite process, in which ice is produced in a glass of warm water. Heat always fl ows from a hot to a cold body until their temperature reaches an equilibrium (the stage of thermodynamic equilibrium). This fl ow is a spontaneous process, and we have never witnessed an opposite process of the fl ow of heat from a cold body to a hot body. All these are spontaneous processes testifying to a clear and irreversible direction of the time arrow. Nevertheless, geology as a scientifi c discipline cannot rely only on human intuitions and must obtain support for its assumptions from other scientifi c disciplines, i.e., as a discipline belonging to the natural sciences, it must also explain the assumption concerning the one-directional nature and irreversibility of the time arrow by means of the laws of physics; otherwise, the geohistorical explanation will be scientifi cally incom-plete. As already mentioned, the most appropriate physical law for a geohistori-cal explanation is the second law of thermodynamics, which is a law of physics on the one hand and a kind of historical law on the other.60 As mentioned, the second law of thermodynamics describes a system develop-ing in time from a state of low entropy (in the past) to a state of thermodynamic equilibrium (in the future), in which the state of entropy of the system reaches the highest value it is capable of. These thermodynamic states change in a con-tinuum, and, therefore, we can speak only of similar states and certainly not of identical states. Besides, one state is causally affected by another, and this is the basis of the causal arrow, which I will discuss in the following. The thermo-dynamic and causal arrows enable us to discover the geological causes in the past on the basis of the effects observed by geologists in the present.61 I will discuss later herein the special status of the second law of thermodynam-ics and Boltzmann’s past hypothesis in the geohistorical explanation.
62 In other words, there is a fundamental diffi culty in reducing the second law of thermodynamics, which is irreversible and asymmetrical in the aspect of the direction of time, to the laws of motion in Newton’s mechanics, which are symmetrical and reversible in time. For a detailed discussion, see Sklar (1993), Price (1996), Callender (2001), and Albert (2000).
28 G. Kravitz
assumptions about the initial conditions of the system and not necessarily about the laws themselves. In his opinion, initial con-ditions are usually asymmetrical and irreversible from the time perspective. The rare state that appears to refute the second law of thermodynamics, Boltzmann explained by means of his defi ni-tion of entropy. He claimed to have succeeded in proving that any macroscopic state can be described by means of a large number of microscopic states equaling it. Boltzmann defi ned entropy as a physical size meas uring several existing possible microscopic states for the purpose of describing a certain macroscopic state. He showed that the macroscopic state with the highest degree of entropy (thermodynamic equilibrium) is identical to the state describing the greatest number of possible arranged microscopic states. Moreover, a state of thermodynamic equilibrium is the most probable state in which a system will exist, and it, therefore, spontaneously aspires to achieve it. This is the reason that we observe the asymmetrical and irreversible processes that occur according to the second law of thermodynamics.
The problem is that the statistical mechanics of Boltzmann, as presented here, are based on Newton’s laws of mechanics, which are symmetrical with regard to time direction. Therefore, one may argue that, as Boltzmann considered that entropy could grow in the direction of the future, it could also grow the other way round—in the direction of the past. Everybody can agree that entropy is bound to grow, but why should we assume that it grows only toward the future, and not toward the past? If the increase in entropy does not have a clearly defi ned direction from the past toward the future, it might just as well fl ow from the fu-ture to the past. In this way, one could argue that, as energy fl ows from a warm body to a cold body, it could just as well fl ow from a cold body to a warm body.
This embarrassing problem was raised by the physicist Joseph Loschmidt (1821–1895). Boltzmann suggested a solution to Loschmidt’s problem by posing his famous “past hypothesis,” thereby providing an answer to the question why entropy grows in the direction of the future, or rather, why we witness asymmet-rical and one-directional physical processes of the time arrow. By means of this hypothesis, Boltzmann placed the responsibility for asym metry of the time, as observed nowadays, on the initial con-dition of the universe. At the starting point, the universe was in a state of low entropy, which grew continuously in the direction identifi ed by us at present. This hypothesis enables us to maintain the symmetry of Newton’s laws of motion on the microscopic level while also explaining why, in spite of this symmetry, we observe asymmetrical and irreversible thermodynamic processes from the point of view of the direction of time. Nevertheless, it should be remembered that Boltzmann’s “past hypothesis” is no more than a hypothesis. Boltzmann never managed to prove it, and it remains a metaphysical assumption to this very day. The controversy concerning this hypothesis has not ceased, and many physical and philosophic questions still remain unsolved. Most of them are concerned with the initial state of the universe. Why was it in a state of disequilibrium? If the state of low entropy is so unreasonable, why should the universe be in such a state? These
questions and many others are still discussed by physi cists and philosophers, and I do not intend to deal with them or to resolve the subject of the time arrow in this paper but only to emphasize that the problem of reduction between the second law of thermo-dynamics and Newton’s laws of motion is still not solved.63
This state of affairs has implications for geology as a sci-ence dealing with the past in general, and with realism in re-lation to the geological past in particular. Realism in relation to the geological past of Earth is based on the assumption of the one-directional and irreversible thermodynamic time arrow. This is based on Boltzmann’s past hypothesis, which cannot be derived from the motion equations of mechanics, as these are invariant with regard to the reversal of the direction of time. In a certain sense, this important fact exposes the metaphysical nature of geology as a historical science, and its methodologi-cal principle—the uniformity principle. In the last resort, as I showed previously herein, this principle is based on the second law of thermodynamics, which determines the direction of the time arrow on the basis of Boltzmann’s metaphysical hypothesis concerning the initial state of the universe. Until the problem of Boltzmann’s “past hypothesis” is solved, the geohistorical explanations will not be conclusive; however, it should be men-tioned that this reservation does not destroy the legitimacy of geology as a science. Physics serves as a scientifi c support for geology, and, therefore, the fate of geology will be the same as that of physics.
The Uniformity Principle, Causality, and Geohistorical Inferences
As we have seen, the uniformity principle is, in a certain sense, a principle of simplicity, the main function of which is to serve as a logical-methodological “tool” for coping with the time arrows and the historical-evolutionary view of the world. To be more specifi c, its main role and goal are to enable geolo-gists to provide geohistorical explanations by means of logical-causal explanations (such as: retrodiction, abduction) in order to overcome the diffi culty inherent in relying on a geological-evolutionary worldview, based on the assumption of the second law of thermodynamics, or the thermodynamic time arrow. In practice, this principle makes it possible to bridge between em-pirical observations in the present and theoretical hypotheses concerning the past and future of Earth, and this is its function. It is made possible by means of logical-causal arguments used by geologists to explain geological phenomena. The structure of these arguments resembles Hempel’s famous model—the covering-law model.64 As in Hempel’s model, these arguments contain an explanans part and an explanandum part. The ex-planandum part is the conclusion of the argument and describes events that occurred in the past. The explanans part contains two
63 For more in-depth discussion regarding time arrow and Boltzmann’s past hy-pothesis, see Albert (2000, p. 1–21, 71–96).64 About the covering-law model and its role in historical explanation, see Weinryb (1987, p. 77–121) and Kitts (1977, p. 6–12).
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 29
kinds of introductions—those describing events in the present (initial conditions) and those concerned with timeless universal generalizations. Timeless universal generalizations are based on the uniformity principle, and without them, no arguments con-cerning the past are possible.65 They provide a kind of logical coverage to the introductions describing events in the present and enabling them to make logical inferences concerning events that occurred in the past. As in physics and other natural sci-ences, bridges by means of general laws and generalizations contribute to the understanding of the geological explanation. The explanation of specifi c events in accordance with a general law or generalization places it as part of a larger general system behaving in accordance to a known system of laws. This system of laws provides the general system with the order and cohesive-ness that make them understandable. When an event occurs ac-cording to a certain law, it is not extraordinary but expected and understood as a detail in a larger system.
Geologists show that a geological event was bound to occur, not only on account of their familiarity with the specifi c facts, but also as a result of their knowledge of general laws that can be put to test in other cases and in other times as well. Such ex-planations are used in all natural sciences and not just in geology. The special contribution of the uniformity principle to the geo-historical explanation based on abductive inference (abduction) is unique for geology. Peirce claimed that explanations in the earth sciences differ substantively from explanations in other ex-perimental sciences, such as physics and chemistry. Peirce, like Whewell, attacked Lyell’s induction.66 In contrast to this induc-tion based on a search for fi ndings (facts) that could support his a priori conceived theory, Peirce suggested a thinking process that used data (facts) in order to create a theory. He called this process “retroduction” and based it on abductive inference, from a his-torical outcome, of the causes and states that caused this outcome to occur (Dott, 1998, p. 16).67 Usually, in the empirical natural sciences, explanations are based on logical inferences of the de-duction/induction type. In contrast, the earth sciences naturally deal with the history of Earth, and therefore most of their expla-nations are based on an inference of the abduction type, inferring the causes that occurred in the past from the out comes (geologi-cal effects) in the present (Engelhardt and Zimmermann, 1988, p. 82, 210).68
The uniformity principle plays a decisive logical role in inferences of the abductive type, allowing geologists to apply
realism in relation to the past. In order to understand how the uniformity principle integrates in the abductive inference, and its logical role, one must fi rst understand its logical structure. This logical structure is usually defi ned as follows:
If H, then II ________H
where H represents the geological causes that occurred in the past, and I represents the geological results observed in the present.69 It is obvious that the argument is not logically valid, and therefore result I cannot be assumed to be caused with certainty by the cause H. One of the greatest diffi culties in retrodiction70 in geology derives from the fact that the same outcome can result from different causes (Kitts, 1978, p. 222).71 Finding the exact cause (as far as possible) in the remote past requires massive interpretative work, and even then we are not sure to fi nd the true cause of the outcome (that we observe in the present).72 In order to overcome the problem and increase the validity of the argument presented here, geologists assume
65 About this point, see Kitts (1977, p. 5).66 For further discussion of Lyell’s geological principles, see Lyell (1830, 1832, 1833) and Rudwick (1998).67 Of course, this explanation of Peirce’s logic is too simplistic, but it meets the requirements of this paper. Peirce considered logic to be an ideal for productive inquiry. According to Peirce, abduction is not concerned with results caused with certainty by the causes; abduction is concerned with a continuing process of fruitful inquiry, not with an explanation that can be immediately justifi ed with certainty. For further discussion of Peirce’s logic and abduction inference, see Fann (1970), Josephson and Josephson (1994), Magnani (2001), Flach and Kakas (2000), Kruijff (1998), Bird (1959), and Burch (2009).68 This argument is valid for all historical sciences and not only for geology. About this point, see Simpson (1963, p. 45).
69 There is a substantial difference between logical derivation and causal deriva-tion. In contrast to logical arguments, causal arguments are concerned with facts pertaining to the state of affairs in the world, and their examination requires an examination of things in the world. In our case, we engage in logical arguments, but these contain causal (factual) arguments concerning the geological state of Earth in the present and the past. Geologists use logical arguments on the one hand, whereas on the other, their explanations are causal explanations based on factual arguments.70 For further discussion about retrodiction process, see Engelhardt and Zim-mermann (1988, p. 212–218) and Kitts (1977, p. 39–47).71 In this respect, an asymmetry exists between the process of prediction and the process of retrodiction. The diffi culty in making predictions is that the same cause can have several different effects. On the other hand, the diffi culty in retrodiction is that the same effect can have many causes. 72 Sometimes the work of the geologist is likened to the work of a detective (Frodeman, 2000, p. viii; Baker, 2000, p. 7) or a physician. The detective ar-rives at the scene of the crime and fi nds the effect (the body of the victim, a house broken into, etc.). On the basis of signs in the area and evidence, the detective attempts to construct the event and fi nd the cause of the crime. Simi-larly, the physician searches for the illness in the symptoms manifesting in the patient. The geologist, like the detective and the physician, looks for the causes of geological events and phenomena. As a matter of fact, such a type of re-search fi rst offers a variety of possible hypotheses (the possible causes for the appearance of a certain effect) that explain the effect, out of which the geolo-gist tries to fi nd the inference to the best explanation or, in other words, to the one that is most confi rmed (since one effect can have many causes, there might be several hypotheses that explain the same effect). The hypothesis chosen is sometimes called the ruling hypothesis, as it offers the best explanation for a certain phenomenon. Such a hypothesis can be chosen by eliminative induc-tion—the less promising hypotheses are discarded, and the best is retained—or by the combination method (in light of the complexity of geological processes, sometimes several hypotheses are required to explain a certain phenomenon). This hypothesis guides the geologist in his daily work up to the stage when he obtains data that refute it. As soon as the hypothesis is re futed, the geologist begins to look for another hypothesis (using the same method). This process is extremely complex and sometimes termed “the method of multiple working hypotheses.” (See the development of the process in Chamberlin [1897] and Kitts [1977, p. 102–114].) I do not intend to discuss the matter further in this chapter. What is important to understand is that this complex process is based on an abductive inference, as described earlier. For further details concerning the process of choosing, confi rming, or refuting hypotheses in geology, see Schumm (1998, p. 10–34).
30 G. Kravitz
the uniformity principle as a kind of “covering law.” To be more specifi c, the assumption “if H, then I” is a hypothesis of a kind, and it appears as a law or a causal generalization. This general-ization does not assure the one-to-one causal relation between the cause and the effect, and therefore many possible causes for the same effect are feasible. In order to strengthen the causal rela tion, a law or generalization with a stronger connection is required. Such a generalization is often called biconditional law,73 and its role in argument is to assure that the cause (H ) will be a suffi cient and necessary cause74 for the occurrence of the effect (I ). Only if the cause (H ) occurs will the effect (I ) certainly occur.75 As a matter of fact, geologists assume the uni-formity principle as a cover law, thus attempting to improve (artifi cially) the hypothesis “if H, then I,” bringing it to the fol-lowing form: If H and only H, then I. On account of this as-sumption, the argument obtains the following improved form:
If H and only H, then II _ _______________H
Here, one can see that the logical-causal connection between H and I is stronger, and therefore H can be inferred from I with greater probability. In a certain sense, the abductive inference is based on an inductive statistical generalization, and therefore such a generalization is often called “probabilistic postdictive in-ference” (Kitts, 1977, p. 14–16). Inductive inference is accepted in science and is based on a thought process by which, on the ba-sis of singular cases, generalizations, including all such cases, are made. In this process, on account of the fact that we have found a large number of objects with the attribute A, we generalize that all objects of this kind (with no exception) possess the same attri-bute. By means of a similar process, the generalization “if H, then I” was made, and when geologists add the uniformity principle to it, they artifi cially “improve” the force of the inductive inference, as a result of which a stronger causal generalization ensues—if H and only H, then I.76 Thanks to this generalization, the abductive inference from the present to the past becomes, so to speak, a
valid deductive inference.77 In this sense, the uniformity principle bypasses the gap between the present and the past (which, as men-tioned already, is one of the main problems of realism in relation to the past), enabling the scientist to present causal arguments, thus expanding the classical principle of causality, which states “everything necessarily has a cause.” This argument, to which the uniformity principle is attached, becomes a more focused argu-ment: “the same causes necessarily have the same effects.” This argument formed the bases of the improved argument: “If H and only H, then I.” If so, the uniformity principle expresses the unity of nature in terms of orderly and uniform natural laws. These laws can be applied only if we assume that there is a uniform principle of causality, namely, that the same causes will neces-sarily produce the same effects. In this sense, when geologists use causal explanations, they start from the assumption that the same causes will necessarily produce the same effects, and hence the realistic belief that when we observe a phenomenon in the present, we are able to discover its cause78; i.e., in geohistorical explanations, causes are suffi cient and necessary conditions for the occurrence of the effects. It is enough that event A (the cause) occurs, for event B (the effect) to occur as well. This is the basis for the possibility of inferring “backwards” (retrodiction).
Why does the uniformity principle require the obligatory relation between a certain cause and a certain effect? As shown earlier herein, the uniformity principle is based on the second law of thermodynamics. In statistical mechanics, this law describes an evolutionary process progressing from a state of low entropy to a state of growing entropy. In this process, the geological phe-nomena and events are singular events in a certain sense, and every one of them is infl uenced by the one that preceded it in time (Engelhardt and Zimmermann, 1988, p. 323), and, there-fore, it is possible, on principle, to infer from every state, what
73 See, for example, Engelhardt and Zimmermann (1988, p. 208).74 I will hereby clarify some of the concepts, such as “suffi cient condition” and “obligatory condition.” Suffi cient Condition—Event H is a suffi cient condition for the occurrence of event I only, and only if when event H occurs, event I too occurs, i.e., it is enough that H occurs for I to occur. Obligatory Condition—Event H is an obligatory condition for the occurrence of event I, only, and only if without the occurrence of event H, event I cannot occur, i.e., if H does not occur, I too cannot occur.75 Namely, the effect I can occur only as a result of cause H, and effect I cannot be caused by any other cause. In terms of logic, one could say that geologists exchange a truth function called material implication (material conditional) for another truth function called logical equality (biconditional).76 Here, I should mention Lyell, who also based uniformitarianism on inductive geology and on vera causa. Some geologists claim that Lyell’s uniformitarian-ism, like that of Playfair and Herschell, is based on inductive inference for which the validity and justifi cation are based on vera causa. On the other hand the catastrophism of Whewell and Sedgwick is based on abductive inference and retrodiction (in this context see Baker, 1998). The abductive inference, as I regard it, along with the uniformity principle, integrates abductive and inductive inference into one whole.
77 We are referring to a kind of transformation that enables us to infer geological events that occurred in the past from arguments concerning geological events occurring in the present. However, I must stress that although adding the uni-formity principle to the argument improves the reliability of the inductive infer-ence, it does not justify the process of induction from a logical point of view. As for the justifi cation of the process of induction, see Black (1964).78 We should remember that I made this inference on the basis of the structure of the improved abductive argument based in the uniformity principle, which, as mentioned be fore, is a methodological and artifi cial principle, in a certain sense. This is where the central problem of modern actualism is exposed. Apparently, geologists are not aware of this philosophical conclusion and are confronted with a multitude of practical problems concerning geological causality. For example, they often claim that the same cause can have various effects (diver-gence) or that various causes could produce the same effect (convergence). In addition, they claim that a certain effect is sometimes caused by a combination of various causes (multiplicity). With regard to practical problems pertaining to geological causality, see Schumm (1998, p. 58–75, 95–119). These argu-ments are correct from a practical point of view but contradict the philosophical conclusion deriving from the logical integration of abductive inference and the uniformity principle (assumed by the geologists themselves). From a pragmatic point of view, their actions in the practice of their profession are perhaps justi-fi ed, but this does not cancel the diffi culty rising out of the logical-conceptual analysis of the uniformity principle. As a matter of fact, since geologists cannot renounce the uniformity principle as a method of work, since its absence would preclude the geohistorical explanation, they use it by default. At the same time, they adopt the multiple working hypotheses (as described earlier), in order to increase the probability of fi nding the true reasons for the effects that they are observing.
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 31
its predecessor was. For instance, the temperature of the water fi ve minutes ago can be calculated if the present temperature, the quantity of water, and the characteristics of their heat convection are all known.79
The second law of thermodynamics assures the one-to-one causal relation between cause and effect, thus enabling the geohis-torical explanation in geology in its most basic sense. To be more specifi c, when geologists assume the uniformity principle based on the thermodynamic time arrow, they actually secure causality by means of the following three deterministic assumptions:80
1. Every effect necessarily has a cause (classical principle).2. Every cause necessarily has the same effect (derives from
the second law of thermo dynamics stating that causes are suffi cient conditions for effects to happen).
3. The cause always precedes the effect (the causal time arrow, probably arising from the thermodynamic time arrow).
To sum up, we can say that actualism is based on the uni-formity principle, which is essentially based on the principle of causality, which in turn obtains its justifi cation from the second law of thermodynamics. The actualist maintains that only the causes observed in the present can explain the events that occurred in the past, but how can he or she know that the causes that occur in the present are the same as those that occurred bil-lions of years ago. In order to answer this question, the actualist assumes the uniformity principle. Only if we assume that the laws of nature and geology do not change over time can we assume that the geological causes observed in the present can explain the geological effects observed in geological evidence (rocks) from the past, but how can the actualist really know that the laws of nature and geology really do not change in the course of such deep geological time? In order to answer this question, the actual-ist adopts another universal principle, the principle of causality. In his opinion, the causal chain is a universal eternal attribute of the universe ensuring uniformity in nature and in various geologi-cal processes. The question arises—how can the actualist be sure that such a causal chain really exists? The answer can be found in the second law of thermodynamics, which ensures the universal validity of the principle of causality, i.e., the causes as observed in the present can explain the effects that occurred in the past.
On the basis of this explanation, we might think that the uni-formity principle succeeds in resolving the problem of the gap between past and present, which is one of the main diffi cul-ties encountered by realism in relation to the past, but does this
principle really resolve the problem of the inaccessibility to the events of the past? This question raises several diffi culties.
Since geologists do not have direct access to the past, they are obliged to make inductive and abductive inferences (relying on causality, among other assumptions), reasoning from acces-sible events occurring in the present to inaccessible events that occurred in the past. In order to justify this method of inference, it is important that at least for once, the geologist should be able to verify his inference on the basis of direct observation without resorting to various logical inferences. If the geologist were able to observe geological events in the past directly, he/she would be completely justifi ed81 in using inductive/abductive inferences in similar cases inaccessible to him/her. However, the geologist cannot carry out such an undertaking, even for once. Since the time arrow is one-directional, the past is inaccessible. More accu-rately, these inferences actually involve a kind of circularity—we learn about the past from the present, in order to know the pres-ent better. The abductive approach contains a hidden argument according to which the present is the key to the past, but we will only prove this assumption if we succeed in proving that there is a one-to-one relationship between the causes in the past and the effects in the present (or in proving that the past is the key to the present).82 The main problem here is that geologists do not have access to the past, and therefore such a proof is patently im-possible, so they will have to content themselves with inductive/abductive inference based on an artifi cial methodological prin-ciple—the uniformity principle.
As we have seen, this principle is based on the principle of causality, the justifi cation of which is the second law of thermo dynamics, but this does not really solve the problem of past-present , since it is based on Boltzmann’s metaphysical hypothesis concerning the past of Earth. Moreover, geology is a historical science and as such deals mainly with singular geological global events. Even without the problem of directly observing the past, one must consider the fact that Earth is a unique entity, and, as far as global events that occurred on it are concerned, each one of them was a unique one-time event, which limits the possibility of making inferences based on the inductive thinking process—universal causal laws cannot be de-rived by observing a singular event. Standard statistical theory teaches us that the wider the scope of evidence, the greater is the legitimacy of statistical inference.83 Therefore, there is a
79 Here I am referring to a substantial argument, and the example of water is a simple illustration. There is no doubt that such inferences are extremely com-plicated and not easily applied in practice. For example, calculating the entire quantity of the entropy of Earth one million years ago, accurately, is an enor-mous task and perhaps an impossible one.80 This argument concerning causality indicates that reduction of geology to physics is possible. An important element strongly connected to reductionism is the determinism based on causality (Dupré, 1993, p. 99). Causality as presented earlier herein clearly proves a particular kind of determinism and thus supports the claim that substantial reduction of geology to physics is possible.
81 Here I am talking about the practical justifi cation (methodology) and not about the logical justifi cation.82 For further discussion about the interesting question “Is the present the key to the past or the past the key to the present?” see Şengör (2001). Actually, here I am talking about the ancient debate between historicism and uniformitarianism.83 Here, again, I am not referring to general logical justifi cation of the induction process, but to a kind of methodological justifi cation based on the extent of reli-ability of the inductive inference. There is a substantial difference between the issue of logical justifi cation of the inductive inference (can it be logically justi-fi ed on a general basis?) and the issue of the methodological use of the inductive process as a scientifi c research tool (does the reliability of the induction process suffi ce to obtain a high degree of certainty required for the inference based on it?). With regard to various types of inductive justifi cations and problems inher-ent in them, see Black (1964, p. 1–8).
32 G. Kravitz
major problem in justifying generalizations and predictions on the basis of one-time events that occurred in a different time. The classical problem of induction is increased fourfold in this case. The meaning of the classical problem of induction lies in the notion that the shift from particular (individual) observations to general laws is not justifi ed deductively and is, therefore, not essential, i.e., predictions are concerned with what has not been observed yet and, therefore, cannot be inferred from what has already been observed in the past. What was observed in the past does not place any logical restrictions on what is to be ex-pected in the future. From a logical point of view, the inference of universal arguments from singular arguments is not an obvi-ous procedure.84 As far as geology is concerned, the problem mentioned earlier is added to this problem of logic. In a situation in which a singular observation cannot be carried out, even with-out the classical problem of induction, a shift from a singular case (or in the case when even one observation is lacking) to generalizations would have been hard to justify. The memory of one event (under the most favorable circumstances) makes it ex-tremely diffi cult to justify generalizations and predictions about future events. A legitimate inductive process must be based on the memory of many events.
Another problem closely connected with the diffi culties raised so far is inherent in the diffi culty to grant validity to mem-ory. As human beings, we tend to make logical inferences “back-wards” (from effects to causes). For instance, we infer from the wet road that it has rained before, from the fragments of a broken cup—to a whole cup, from smoke to fi re, etc. Such inference seems natural to us because in our individual or collective mem-ory (textbooks, etc.), we recollect similar events that occurred in the past. The main diffi culty in such inferences is to establish cognition on the basis of causal relations relying on memory. In order to connect events in causal relations, memory is required. Without memory, it is impossible to remember the specifi c cause and connect it to a specifi c effect. As we have seen, such mem-ory does not exist in geology—the geologist does not have any memory concerning geological events that occurred in the distant past even before humans existed on Earth. Even if there was such a “mysterious” ability to remember events from the distant past, the problem of memory would still remain unsolved on account of the substantial diffi culty to establish memory. In order to es-tablish causal laws, one must fi rst establish the validity of mem-ory, and how can we establish such validity without making use of memory itself? Such a circular proof would be unavoidable in this case. To be able to prove the existence and validity of mem-ory, we will have to activate our thoughts through means of our memory, for instance, we will be obliged to remember the laws of logic that we learnt in the fi rst years of studying philosophy. In any case, even if the problem of circularity had not existed, it is extremely diffi cult to establish the validity of memory. All of us
can prove that memory is often confusing, and we cannot always distinguish between memory and fantasy.
In short, the uniformity principle does not resolve the prob-lems arising in the context of realism in relation to the past, nor those arising from the assumption of the one-directional time ar-row. Nevertheless, it should be remembered that the uniformity principle is a scientifi c method connecting geology with the rest of the natural sciences (physics, chemistry, biology, etc.), placing it on a wider and more stable scientifi c basis. Therefore, the dif-fi culties and open questions must be resolved in the wider context of physics and the other natural sciences. Until then, geology will continue to be a legitimate science like any other branch of mod-ern science.
THE UNIFORMITY PRINCIPLE AND REDUCTIONISM IN GEOLOGY
From the point of view of most scientists and philosophers, geology is considered a derivative science from other sciences such as: physics, chemistry, biology, astronomy, etc.,85 and not an autonomous science in its own right.86 According to them, all geological phenomena can be explained by the laws of phys-ics plus known initial conditions (Kleinhans and Buskes, 2002, p. 1). This is probably one of the reasons that philosophy of sci-ence only rarely shows interest in geological thinking and its basic assump tions (Kleinhans at al., 2005, p. 289–290).87 This reductionist viewpoint is supported by several schools of analytic philosophy, all of which claim that scientifi c methodology must adhere to three basic conditions: empiricism, objectivity, and epistemological monism (the possibility of reducing all sciences to physics). The adherents of this approach are convinced that geology is a “derived” science and are not interested in it. As a matter of fact, very few efforts have been made to show that geol-ogy could actually be reduced to physics and chemistry, because, from the analytical philosopher’s point of view, this was obvious.
Nevertheless, some modern philosophers reject the no-tion that geology is a “derived” science.88 They regard it as a historical-interpretative science (Frodeman, 1995) with its own laws, explanations, and unique multidisciplinary methods (Dott, 1998, p. 15). In this sense, geology is different and more compre-hensive than other branches of science.89 The main argument here
85 Geology is particularly considered a derivative science of physics.86 This standard concept is accepted by most modern geologists and science philosophers. For further discussion of this point, see Scriven (1959), Pantin (1968, p. 24), Cartwright (1999, p. 7), Schumm (1998, p. 3), Kitts (1977, p. 60), Goodman (1967), and Engelhardt and Zimmermann (1988). 87 About the neglect of geology as an autonomic science, see Frodeman (1995, p. 960; 2000, p. ix) and Dodick and Orion (2003, p. 197–206).88 Therefore, in their opinion, the educational system that trains geologists in geol ogy and the history of geology should be changed (Dodick and Orion, 2003; Greene, 1985).89 On account of the complexity of geology as a multidisciplinary science, its explanations are often much more complex than explanations in physics. Per-haps this is the reason that geology is often called an unrestricted science and physics is called a restricted science. For this distinction, see Pantin (1968, p. 1–25, 123–128).
84 Actually, I am referring to Hume’s classical induction problem, which raises the problem of our inability to justify the metaphysical assumption concerning the uniformity of nature.
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 33
is that since geology is a complex and multidisciplinary science, it cannot be “derived” from physics, which means that geology cannot be reduced to physics.90 In this line of argument, the main problem in geology lies in the diffi culty of discovering universal laws and bridging principles between it and the laws of physics, according to the criteria of Nagel’s traditional model of reduction (Kleinhans et al., 2005, p. 295).
In light of these two approaches, I will attempt, in this paper, to provide an answer to a central issue: Is a complete reduction of geology to physics possible?91 More accurately, can we “derive” or infer the laws and theories of geology from the laws of physics and chemistry?
I see a problem in the two approaches described, namely, that both of them do not make a suffi ciently clear distinction be-tween the analytical parts and the synthetic parts of geology. The analytical approach almost completely ignores the synthetic ap-proach. On the one hand, it regards the laws of physics as the major tools for explaining geological events in the past, present, and future, and, on the other, it ignores the historical-interpreta-tive character of geology. Therefore, is clear why this approach supports reductionism in geology. The continental approach does the opposite—it is concerned mainly with the synthetic aspect of geology, allocating a secondary role to the analytic aspect. It em-phasizes the historical-interpretative role of geology and regards the laws of physics as one of the methodological tools by which geologists interpret the history of geology. Therefore, is clear why this approach contradicts reductionism in geology.
The dispute between the two approaches has never been re-solved because each party sticks to its opinion, and except for one attempt by Kleinhans et al. (2005),92 no comprehensive attempt has been made to answer the question whether it is at all possible to “derive” geology from the basic laws of physics.
In this situation, if we are interested in answering this ques-tion, we must fi rst distinguish between the analytic aspect of geol-ogy and it synthetic aspect and explain the signifi cance of such a distinction. In the second stage, on the basis of the nature of the uniformity principle (as described earlier) and in contrast to the argument presented by Kleinhans et al. (2005), I will argue that on principle, there is a way to “derive” the analytic part of geol-ogy from the basic laws of physics.
In this discussion, I am following Baker’s (2000) distinc-tion between the two types of science that comprise the science of geol ogy—analytical science and synthetic science—which to-gether form the epistemological basis of geology. These are really two strata of cognition in which geological research is conducted. The fi rst is based on analytical-causal explana tions emphasizing quantitative data, empirical testing, predictions, and abstractions of nature, while the second emphasizes qualitative, historical-probabilistic data based on circumstantial evidence (descriptions and verbal arguments and graphs) emphasizing the complexity of nature (Dott, 1998, p. 15).93 Some philosophers and scientists distinguish between these sciences by means of concepts such as time-bound and timeless94 or historical and causal. Timeless knowl-edge (causal) is concerned with the search for general attributes and behavioral patterns, such as laws of nature that characterize be-havioral patterns of geological objects and their interaction, while time-bound knowledge (historical) refers to specifi c objects, such as mountains, rivers, layers of rock, and the changes undergone by them in the course of time. For example, crystallography, geo-physics and geochemistry are analytical sciences of a causal and timeless nature. On the other hand, paleontology and stratigraphy are synthetic sciences of a historical or time-bound nature. The ma-jority of the earth sciences are located between these two extremes. These distinctions derive from the objectives that geology has set for itself. It is interested in the geological structure of Earth, in geological phenomena and processes, and their history. Therefore, it has at least two major goals (Kleinhans et al., 2005, p. 290–291):
1. to describe the historical direction of events, processes, and geological phenomena, i.e., to describe the historical devel-opment of Earth from its creation until today; and
2. to explain the causes forming Earth in the past and continu-ing to do so in the present.
The two goals show that geologists apparently attempt to an-swer two major questions:95
1. Why do geological processes occur the way they do (the “why” question)?
2. How and under what circumstances do geological processes occur (the “how” question)?
90 This argument is usually raised along with the approach that opposes reduc-tionism in science and regards the world as much more complex and varied and, therefore, cannot be simplifi ed and contained in a single scientifi c explanation. In order to explain all the aspects of the world, explanations, other than physical explanations, from other scientifi c disciplines, are required; however, a system of laws uniting all of them does not exist. For reasons concerning these argu-ments, see Dupré (1993) and Cartwright (1999). 91 It should be mentioned that we are not dealing with practical reduction, but with a type of reduction supporting the derivation of geology from physics and not the substitution of physics for geology. Moreover, I am not referring to reduction in the sense of historical connections between geology and physics (clearly physics precedes geology from a historical point of view). With regard to the types of reduction, see Dupré (1993, p. 94–99).92 Historically, the view that geology reduces from physics has also been argued by Watson (1966, 1969), who pointedly disputed the arguments made by Simp-son (1963). A more nuanced position about this issue was taken by Kitts (1963), who fi nally generated a whole approach to philosophy of Geology (Kitts, 1977).
93 Similar distinction can be seen in Simpson (1963, p. 34).94 In another distinction, historical research is regarded as the analysis of com-plex phenomena with low recoverability of the same phenomenon, and physical research is regarded as dynamic investigation and the analysis of phenomena with a high rate of recoverability, leading to the formulation of universal laws. For such distinctions, see Schumm (1998, p. 4).95 In spite of the closeness of geology to biology, geologists attempt to answer the questions “why” and “how” and do not ask “for what purpose,” as does biology. To state it differently, geologists are usually interested in functional explanations and do not ask what is the purpose or the end of one or another geological process. In spite of that, we must remember that like evolutionary biology, geology is also a historiographic science dealing mainly with geologi-cal events that occurred in the past. Therefore, the main objective of geology is to provide a correct description of the order of events that occurred in the past and also to explain the causes of this particular order of things.
34 G. Kravitz
Both questions/goals are connected to each other, and, there-fore, in practice, geological explanations are a combination of causal and narrative explanations. On the one hand, geology is a historical-synthetic science dealing with geohistorical descrip-tions and geological reconstruction of Earth, and, on the other hand, it is an analytical-nomological science providing causal explanations and predictions.
At this point, it should be mentioned that the distinction be-tween the analytical and the historical-hermeneutical aspects of geology is not clear-cut and is even artifi cial to a certain extent. Actually, the main branches of geology are located between the two extremes (Frodeman, 2000, p. ix–x), and sometimes the ana-lytical and synthetic aspects mix and complete each other (Dott, 1998, p. 17).96 In other words, a good geological explanation must contain both aspects, and in practice it is diffi cult to sepa-rate them.97
Let us now return to the main question: Can we “derive” geology from the basic laws of physics? In order to analyze and answer this question in a consistent and comprehensive way, fol-lowing Kleinhans et al. (2005), I will adhere to Nagel’s tradi-tional model of reduction,98 in which he argues that on principle it is possible to reduce geology to physics and chemistry if we conform to the following goals (Kleinhans et al., 2005, p. 292):
1. to fi nd a bridging principle between terms included in the physical law and terms included in the geological law/gen-eralization; and
2. to fi nd a geological law or generalization that meets the cri-terion of lawfulness in the sense of universality.
According to Nagel’s model, we are looking for a bridging principle (between a geological law or generalization and physi-cal law) that meets the criterion of lawfulness.99
If we were to put most geological laws and theories accord-ing to Nagel’s model of reduction to test, we would have diffi -culty in fi nding a bridging principle between the laws of physics and the laws of geology. Besides, most of the laws and theories of geology refer to local global events on Earth and, therefore, can-not fully meet the criterion of lawfulness.100 For instance, theories concerned with the “impact hypothesis” on Earth, which caused the elimination of profuse species of life and vegetation, are theo-ries describing a geological state in the past in certain locations. Such theories are theories in the sense that they present hypoth-eses about events in the past that explain observations in the pres-ent. Nevertheless, they are not universal theories and do not meet the criteria of lawfulness, as stated by Kleinhans et al. (2005, p. 293). These authors maintain that even the plate-tectonics theory is a local theory that does not meet the requirement of the universality of lawfulness (Kleinhans et al., 2005, p. 293–295). According to them, even though this theory explains the majority of geological phenomena occurring on Earth, it is not a universal theory in the sense of a law of physics. Like a biological theory, it is a theory that describes contingent events, dependent on certain structural confi gurations of Earth or other planets that resemble it. This means that although this theory can actually predict what is bound to happen in planets resembling Earth from a geological point of view, it is not appropriate to explain geological phenom-ena occurring in other celestial bodies, such as gas giants. More-over, it is impossible to fi nd a bridging principle in this theory, because as soon as one tries to translate the terms of the theory (e.g., mantle, tectonic plate, oceanic crust, earthquake, etc.) into the language of physics or chemistry, diffi culties crop up. These terms usually refer to a random distribution of material, like the tectonic plates on the face of Earth, or to other processes related to the specifi c structure of Earth (Kleinhans et al., 2005, p. 294). In other words, the plate-tectonics theory is an autonomous theory that cannot be “derived” globally from the laws of phys-ics. According to this conclusion, these authors maintain that it is not possible to reduce geology to physics according to Nagel’s traditional model (Kleinhans et al., 2005, p. 301).101
Is it really impossible to reduce geology to physics accord-ing to Nagel’s traditional model of reduction? Could we not, on the one hand, fi nd in geology a law or principle that could serve as a bridging principle between the laws of physics and those of
96 An indication justifying this argument can be found already in the most ba-sic principle of geology—the principle of uniformity. As already mentioned, this principle contains the uniformity of law and the uniformity of process. The fi rst is connected to the analytical-nomological science and assists geolo-gists to answer the question of “why,” and the other is connected more to the synthetic-historical science, enabling geologists to answer the question “how.” The fi rst attempts to answer the question “why do geological processes occur as they occur?” and therefore requires a universal law (on the basis of the law and its context, one can explain why the processes occur as they occur). On the other hand, the other attempts to answer the question “how do the geological processes act?” and it must assume that similarity exists between geological processes in the course of time. I am referring here to the similarity in the sense of various kinds of geological processes, such as volcanic processes, tectonic processes, etc. At fi rst, these processes must be defi ned on a conceptual level, and then a description of how they operate follows. The similarity is expressed mainly by the mechanism of the processes, and this, in itself, is a description and explanation of the way in which the process itself operates.97 These aspects merge, and it is sometimes diffi cult to separate them. Some-times, the historical-hermeneutic aspect is explained by means of the analytical aspect and vice versa. This mixture sometimes creates diffi culties of commu-nication and interpretation in geological research. In this context, see Schumm (1998, p. 35–94). For a discussion of the connection and integration of an ana-lytical-empirical science and a historical science, see McAllister (1997).98 For details of Nagel’s reduction model, see Nagel (1961, p. 336–401). I de-cided to adhere to this model because it is considered a traditional, conservative, and exact model. It should be mentioned that other modern and perhaps more fl exible models exist, but I do not intend to deal with them in detail, because they are still subject to dispute, and their acceptance by most of the philoso-phers engaged in this fi eld is still in doubt. With regard to additional models, see Kleinhans et al. (2005, 2008). About reductionism in general, see Dupré (1993, p. 87–106).
99 The intention is to fi nd a law/generalization/principle that meets both these requirements simultaneously.100 These theories are actually local (not universal) generalizations and therefore are often called low-level generalizations. In this context, see McAllister (1997, p. 245–246).101 In order to soften this argument, these philosophers suggest rejecting Nagel’s traditional reduction model and basing the reduction of geology to physics on other models, more fl exible than Nagel’s traditional model. On account of limi-tations of space, I will not discuss these models in this paper, especially since there is no consensus about them among the majority of modern philosophers. With regard to these models, see Kleinhans et al. (2005, p. 294–297).
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 35
geology, and on the other hand, fi nd a universal principle that meets the criterion of lawfulness according to Nagel’s model? Contrary to Kleinhans et al., I maintain that in geological re-search, there is a bridging principle that meets the severe criteria of Nagel’s reduction model to a certain extent—the uniformity principle. Although this principle does not solve the problem of the past-present gap, it constitutes a scientifi c method joining geology to the other natural sciences—physics, chemistry, biol-ogy, etc.—thereby providing a wider and more stable scientifi c basis for it. In this sense, I regard the uniformity principle as a bridge between geology and physics, an indication that it might be possible to reduce geology partially to the laws of physics. As we have seen, this principle combines the uniformity principle of law with the uniformity principle of progress under the same roof, and can thus be regarded as a bridging principle between the laws of physics and the laws/generalizations of geology.
The unity of geology and physics in the framework of one methodological principle substantiates the viability of reducing geology to physics.102 Such a basis of reductionism provides the ultimate requirement for uniting all the sciences under the um-brella of physics. This requirement assumes a certain unity in nature, and, therefore, the uniformity of all the sciences is sine qua non (a matter of necessity). According to this conception, there is only one reality that, on principle, can be explained by the laws of physics. The logical structure of the uniformity principle, as presented here, is in line with this worldview. As an illustra-tion, in the geohistorical sense, the uniformity principle connects the second law of thermodynamics to the laws/generalizations concerning the similarity of geological processes. The terms it contains actually belong, on the one hand, to the fi eld of geology (e.g., the eruption of hot lava from the depth of Earth to the face of Earth, convection current, geothermal gradient, etc.), and on the other, to the fi eld of thermodynamics (the increase of entropy, heat transfer, thermodynamic equilibrium, etc.). Here, the uni-formity principle acts as a bridging principle, as it contains both the physical law and the geological generalization/law. The con-tainment itself displays bridging of a kind, at least in the meth-odological sense.103 As such, the uniformity principle also meets the second criterion of lawfulness. The uniformity principle of process cannot be assumed before fi rst assuming the second law of thermodynamics (uniformity principle of law). Therefore,
from the point of view of universality, the validity of the uni-formity principle of process with regard to geological processes is at least as applicable as the second law of thermodynamics is to physical processes with regard to heat transfer. Thus, contrary to other geological laws, the uniformity principle presumes to be a universal methodological principle; otherwise, it would be impossible to make use of it to explain geological processes that occurred in the past (retrodiction), occur in the present, and will occur in the future (prediction).
Since the uniformity principle meets the criteria of Nagel’s model of reduction and thus enables the reduction of geohis-tory to physics or the reduction of causal explanations of geol-ogy to physics and chemistry, it also serves as an important aid (methodological principle) to making possible hermeneutic explanations, interpretations, and analogies (thereby enabling geologists to make inductive and abductive inferences with re-gard to the past). If we could not make assumptions regarding uniformity in nature (unchangeable over time) and similarity between various kinds of geological processes, it would have been impossible to explain the past by any kind of interpreta-tion based on observations/evidence in the present. In contrast to Frodeman’s (1995) argument, I maintain that the uniqueness of geology cannot be attributed solely to the fact that it uses in-terpretative tools. It is also necessary that geology’s geohistori-cal assumptions be based on the thermodynamic time arrow.104 I can agree with Frodeman that geology is a hermeneutic-historical science, but I also argue that it is based mainly on the uniformity principle, which can be derived, essentially, from the second law of thermodynamics. In this sense, geology is a unique science, but it is not an autonomous science in the full sense of the word.
The geologist, as a hermeneutic interpreter, deciphers105 and reconstructs the geological past of Earth. For the deciphering to
102 Namely, in my opinion, the fact that geologists make use of the uniformity principle exposes their belief in a certain kind of determinism and the unity of all the sciences under the large umbrella of physics. In contrast to this approach, the opponents of reductionism reject uniformitarianism and the natural order of things and endorse scientifi c pluralism, or rather, multiple theories from vari-ous disciplines (epistemological diversity) that explain various aspects of real-ity. With regard to these arguments, see Cartwright (1999, p. 1–19) and Dupré (1993, p. 221–264). 103 I presume that this can create a bridge between concepts included in a physi-cal law and concepts included in a geological law/generalization; otherwise, unity or containment would make no sense. For instance, in order to explain the similarity of geological processes in the course of time, the use of concepts derived from the second law of thermodynamics is required. For a discussion of the signifi cance of the bridging principle and the principles of reductionism in the context of unique sciences (including the problems of reduction of these sciences to physics), see Fodor (1974).
104 In this respect, geology is a unique science differing from other sciences. For example, in conventional historical research, such an assumption is unneces-sary. In historical research, historians assume various time arrows (the causal arrow, the epistemic arrow, the metaphysical arrow, etc.), but they do not refer to the thermodynamic arrow and are not interested in it.105 Some philosophers distinguish clearly between the term “explaining” and the term “deciphering” (Lorand, 2010, p. 115–116). The explanation bridges be-tween the singular case and the general principle (theory) and places it (by surrendering its individuality) in a system with regular relations between its elements. The explanation is oriented to the general principle, while decipher-ing is oriented to the singular story of the case. Deciphering is concerned with a concealed object, and the person doing the deciphering attempts to discover the truth behind the hidden signs. Therefore, deciphering is adapted to geology and to any other historical science. The geologist, as a realist in relation to the past, is interested in the true story, and therefore the concept of deciphering is relevant for him. He constructs narratives and hypotheses by means of relevant signs (evidence), but cannot guarantee certainty. (For a discussion of semiotics in geology, see Baker [2000, p. 7–9] and Baker [1999].) Since he is interested in a true narrative, he cannot be content solely with hypotheses in harmony with the evidence, but must con struct hypotheses and support them on the basis of harmony between the fi ndings, analogies to similar cases, etc., by which he attempts to decipher the true narrative among all the other narratives that also harmonize with the facts. In this paper, I do not consider this distinction impor-tant and will not discuss it. In the last resort, the geologist is concerned with explanations and deciphering of geological phenomena, and, therefore, as far as I am concerned, they are equivalent terms.
36 G. Kravitz
be part of a context, the geologist requires preconceptions and foresight, without which nothing can be accomplished.106 Ac-tually, we are referring to the hermeneutic circle.107 The geol-ogist as an interpreter attempts to understand the geological object as a whole on the basis of the parts observed. Under-standing the whole depends on understanding the parts, but in order to understand the parts as components of the whole and not as isolated elements, he must understand their role in the whole, as this is what gives them signifi cance. The circularity is created by the interdependence between the whole and its elements—understanding the whole depends on understanding the components, and understanding the components depends on understanding the whole (Lorand, 2010, p. 61). In practice, because geologists lack complete understanding of the whole, they are obliged to proceed in the following way: In the fi rst stage, according to the actualistic method, they have direct ac-cess to the components of the whole and study it by observing its parts, and not the other way round. In the second stage, they reconstruct the geological past of Earth on the basis of these observations in the present and the preconception of the uniformity principle pertaining to geological and physical processes .108
Thus, the uniformity principle serves as a bridging principle (albeit an artifi cial one) that enables geologists to understand the whole by observing its parts—uniting the parts under a common principle (or a common system), thus making it possible to under-stand the whole (unknown) by understanding the parts (known). This principle does not solve the philosophical (logical-concep-tual) problem concerning the paradox of the hermeneutic circle, but it offers a practical solution for the geologist. This solution creates a connection between physics (the second law of thermo-dynamics) and geology as a historical-hermeneutical science. In this respect, one can say that the uniformity principle establishes the connection between the analytic and the synthetic parts of geology. By means of the components of the uniformity prin-ciple, the analytic part is represented by the uniformity principle of law, and the synthetic part is represented by the uniformity principle of process. The fi rst explains why geological processes occur the way they occur, and the second answers the question of how and under what circumstances they occur. This proves that it is possible to explain geological phenomena that occurred in the past by means of hermeneutic interpretations based on the laws of physics. Thereby, geology places itself on a sound scientifi c basis, and, accordingly, in my opinion, reductionism in geology is justifi ed.
We can summarize by saying that the causal explanations concerning the geohistory of Earth are based and “derived” from the uniformity principle containing the second law of thermo-dynamics, which bridges the gap between this law and the whole geohistorical way of thinking. In other words, geologists assume that the second law of thermodynamics is completely valid in geology, and, in this respect, they derive the geohistorical think-ing from thermodynamics, which is one of the most important branches of modern physics (although it is not a basic theory). Therefore, in my opinion, reductionism in geology is justifi ed. Moreover, this principle enables us to regard geohistory as an analytic-nomological science based on the second law of thermo-dynamics on the one hand, and, on the other, as a historical sci-ence “derived” from the thermodynamic time arrow, on which the modern evolutionary worldview describing the historical de-velopment of Earth is based.
Despite what was stated earlier, we should bear in mind that geology is a complex multidisciplinary science contain-ing both elements of an analytical science and elements of a synthetic science. This level of complexity arouses questions concerning the issue of full reduction of geology to phys-ics, and here the uniformity principle does not help, since it contains both these elements. Although geology possesses an alytical elements (such as crystallography, geophysics and geochemistry) that can be clearly “derived” from physics and chemistry, it also contains synthetic elements (such as paleon-tology and stratigraphy) that cannot be derived from physics or chemistry. This means that the interdisciplinary character of geology limits the possibility of making full use of reduc-tionism in geology. As a matter of fact, since the uniformity principle establishes the connection between the analytic and the synthetic elements in geology, and since geology is ba-sically a historical-interpretative science, geological expla-nations must contain the two complementary elements that comprise an essential condition for a full geological explana-tion. This fact challenges the position of full reductionism in geology.
The synthetic element in geology cannot be reduced to physics, which is a causal-analytical science. The diffi culty lies mainly in the methodological difference and in the difference of objectives, which cannot be bridged, between a causal-ana-lytical timeless science and a time-bound synthetic science, the former emphasizing mainly quantitative, empirical data, predic-tions, and abstractions of nature, and the latter emphasizing the complexity of nature and mainly qualitative data of a historical probabilistic nature base on circumstantial evidence. While the former is geared to explain the causes that formed and are still forming Earth, the second is geared to describe the direction of the historical development of geological events, processes, and phenomena on Earth. Physics, as an analytical science, deals mainly with the “why” question and uses logical arguments in its explanations, whereas the synthetic science dealing with the “how” question describes the narrative and does not require logi-cal arguments. The geohistorical narrative cannot be explained
106 About this point, see Frodeman (1995, p. 964).107 About the hermeneutic circle, see Frodeman (1995, p. 963–964) and Lorand (2010, p. 60–70).108 A good example of such thinking appears in plate-tectonics theory. On the ba-sis of observation of continents (parts) in the present, and the assumption of the uniformity principle concerning geological and physical processes, geologists succeeded in reconstructing the supercontinent “Pangea” (the full assembly). In due course, they make inferences, concerning geological single events occur-ring in the present (parts), based on their understanding of the entire situation (whole).
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 37
by logical arguments based on the laws of physics, which ex-plain why events occur the way they occur but are not suited to explain the geological narrative of Earth.109 The problem be-comes even more severe when the synthetic part of geology con-tains areas concerned with functional causes (fi nal causes), such as geobiology and geosociology. These disciplines can never be derived from physics or chemistry, which deal only with active causes (effi cient causes). On principle, one cannot reduce func-tional explanations provided in biology and sociology to causal explanations based on universal laws, and the conclusion is that full reduction of geology to physics and chemistry is an impos-sible undertaking.
Besides the fundamental diffi culty of bridging the gap be-tween the synthetic and the analytic elements, there is another diffi culty pertaining mainly to the analytical-causal element of geology. Although this is, in a sense, a practical problem, it constitutes a major impediment for full reductionism in geol-ogy. The diffi culty stems mainly from the problem of under-determination.110 The geological system is chaotic to a certain extent, and its initial conditions cannot always be determined accurately, and therefore future geological events cannot be pre-dicted accurately. More accurately, since geological processes are chaotic or random by nature and therefore particularly sensi-tive to initial conditions, it is diffi cult to re construct events that occurred in the past from a large and complex collection of ob-servations in the present. Theoretically, there are endless laws and possible initial conditions that might impede physical ex-planations relying mainly on knowledge of these physical laws and initial conditions. These problems are especially relevant to causal explanations requiring a determined (clearly defi ned) set of observations and initial conditions. The gravest problem in this context is the problem of complex geological phenomena developing over time out of a variety of simple basic compo-nents and relatively simple physical interactions (emergent phe-nomena).111 These phenomena acquire different attributes from those of their components, and hence the diffi culty to predict them from their basic components.
In the last resort, we can say that geological phenomena are not the exclusive domain of the unchangeable laws of physics and chemistry, but they are dependent on changeable factors of
place, time, and material, affected by the geological process.112 Moreover, since geological nature is varied, geological theories are also varied, and not all of them can be derived from the basic theories of physics.113
CONCLUSION
While I support the statement that geology is a complex his-torical-interpretative science using unique methodologies, on the basis of a conceptual-logical analysis of the uniformity principle, my argument is that fundamentally it is possible to make a partial reduction of geology to physics, at least in the methodological sense, for those aspects of the geohistorical research that con-tain analytical-nomological-causal elements. Geology, though it makes use of interpretative tools from a variety of disciplines, can still be regarded as partially derived from physics, and ulti-mately the interpretative methods manifest in the geological ex-planations can be partially based on the laws of physics. On the other hand, as long as geology continues to be a multidisciplinary science—synthetic-historical-interpretative on the one hand, and analytical-nomological-causal on the other—we will never be able to reduce it completely to physics and chemistry. Geological nature is varied and so are geological theories, and not all of them can be reduced to basic physical theories.
These arguments are not contradictory and do not undermine the uniqueness of geology. When it provides analytical-causal ex-planations, it can be a science “derived” from physics, and when it is concerned with historical-descriptive explanations, it can use special research methods and through them establish its uniqueness.
REFERENCES CITED
Ager, D.V., 1993, The New Catastrophism: The Importance of the Rare Event in Geological History: Cambridge, UK, Cambridge University Press, 252 p.
Albert, D., 2000, Time and Chance: Cambridge, Massachusetts, Harvard Uni-versity Press, 192 p.
Albritton, C.C., 1989, Catastrophic Episodes in Earth History: London, Chap-man and Hall, 224 p.
Anderson, C.A., 1963, Simplicity in structural geology, in Albritton, C.C., ed., The Fabric of Geology: London, Addison-Wesley Publishing, p. 175–183.
Anderson, O., 2007, Charles Lyell, uniformitarianism, and interpretive principles: Zygon, v. 42, no. 2, p. 449–462, doi:10.1111/j.1467-9744.2007.00449.x.
Austin, S.A., 1979, Uniformitarianism—A doctrine that needs rethinking: The Compass of Sigma Gamma Epsilon, v. 56, no. 2, p. 29–45.
Baker, A., 2010, Simplicity, in Zalta, E.N., ed., Stanford Encyclopedia of Philos-ophy: http://plato.stanford.edu/entries/simplicity (accessed summer 2011).
Baker, V.R., 1998, Catastrophism and uniformitarianism: Logical roots and cur-rent relevance in geology, in Blundell, D.J., and Scott, A.C., eds., Lyell:
109 Even though a narrative contains an order of events in time and its state-ments are clear, it is not a logical inference because it lacks assumptions and a conclusion. A narrative is a collection of arguments that describe reality with connections of continuity, relevance, and completion, but not of logi-cal derivation. However, such a statement does not reject the use of logical arguments in the course of a descriptive process of the narrative. Possibly the geologist, in various stages of his work, will use several logical arguments in order to explain certain statements in the narrative, but these arguments are not important for the narrative in general. In the last resort, the narrative in general is not a logical inference, but a story. On the other hand, an analytical science must deal with logical inferences because it demands from itself to explain why a certain fact is what it is and not something else. In other words, an analytical science strives to know the cause and, therefore, requires logical arguments.110 For further discussion of this problem, see Kleinhans et al. (2005), Turner (2005), and Kitcher (1993).111 About this phenomenon, see Kleinhans et al. (2005).
112 For instance: All volcanic eruptions adhere to the principles of physics concern-ing the interactions among heat, pressure, density, viscosity, and the dynamics of fl ow. These eruptions can consist of slow basaltic fl ow or reach formidable eruptions/explosions. As a result, the outcomes and products will differ. Although both scenarios can be explained by the laws of physics, they cannot be predicted only on the basis of these laws. Similarly, the characteristics of the sediments cannot be accurately determined on the basis of the laws of fl ow. Particles of sediments in water, air, or snow sink according to the laws of physics, which are linked to the dynamics of fl ow. Nevertheless, a change of factors such as the size of particles, their shape, depth of fl ow, speed, etc., does not enable us to predict the exact characteristics of the sediment produced by the sediment process. 113 This argument can be supported also by Cartwright (1999) and Dupré (1993).
38 G. Kravitz
The Past Is the Key to the Present: Geological Society of London Special Publication 143, p. 171–182.
Baker, V.R., 1999, Geosemiosis: Geological Society of America Bulletin, v. 111, no. 5, p. 633–645, doi:10.1130/0016-7606(1999)111<0633:G>2.3.CO;2.
Baker, V.R., 2000, Conversing with the Earth: The geological approach to under standing, in Frodeman, R., ed., Earth Matters—The Earth Sciences, Philosophy, and the Claims of Community: Upper Saddle River, New Jersey, Prentice-Hall, p. 2–10.
Baker, V.R., 2013, Terrestrial analogs, planetary geology, and the nature of geological reasoning: Planetary and Space Science, doi:10.1016/j.pss.2012.10.008 (in press).
Benson, R.H., 1984, Perfection, continuity, and common sense in historical geol ogy, in Berggren, W.A., and Van Couvering, J.A., eds., Catastrophes and Earth History—The New Uniformitarianism: Princeton, New Jersey, Prince ton University Press, p. 35–75.
Bird, O., 1959, Peirce’s theory of methodology: Philosophy of Science, v. 26, no. 3, p. 187–200, doi:10.1086/287674.
Black, M., 1964, The Justifi cation of Induction: Washington, D.C., U.S. Infor-mation Agency, The Voice of America, Forum Lectures, Philosophy of Science Series 15, 8 p.
Burch, R., 2009, Charles Sanders Peirce, in Zalta, E.N., ed., Stanford Ency-clopedia of Philosophy: http://plato.stanford.edu/entries/peirce (accessed spring 2013).
Burchfi eld, J.D., 1975, Lord Kelvin and the Age of the Earth: New York, Sci-ence History Publications, 278 p.
Burchfi eld, J.D., 1998, The age of the Earth and the invention of geological time, in Blundell, D.J., and Scott, A.C., eds., Lyell: The Past Is the Key to the Present: Geological Society of London Special Publication 143 p. 137–143.
Callender, C., 2001, Thermodynamic asymmetry in time, in Zalta, E.N., ed., Stanford Encyclopedia of Philosophy: http://plato.stanford.edu/archives/win2001/entries/time-thermo (accessed 15 November 2001).
Camardi, G., 1999, Charles Lyell and the uniformity principle: Biology and Philosophy, v. 14, p. 537–560, doi:10.1023/A:1006504910017.
Cannon, W.F., 1960, The uniformitarian-catastrophist debate: Isis, v. 51, no. 1, p. 38–55, doi:10.1086/348838.
Cartwright, N., 1999, The Dappled World—A Study of the Boundaries of Sci-ence: Cambridge, UK, Cambridge University Press, 260 p.
Chamberlin, T.C., 1897, The nature of multiple working hypotheses: The Jour-nal of Geology, v. 5, p. 837–848, doi:10.1086/607980.
Cleland, C.E., 2001, Historical science, experimental science, and the sci-entifi c method: Geology, v. 29, p. 987–990, doi:10.1130/0091-7613(2001)029<0987:HSESAT>2.0.CO;2.
Cleland, C.E., 2002, Methodological and epistemic differences between histori-cal science and experimental science: Philosophy of Science, v. 69, no. 3, p. 474–496, doi:10.1086/342455.
Dodick, J., and Orion, N., 2003, Geology as an historical science: Its percep-tion within science and the education system: Science & Education, v. 12, p. 197–211, doi:10.1023/A:1023096001250.
Dodick, J., Argamon, S., and Chase, P., 2009, Understanding scientifi c method-ology in the historical and experimental sciences via language analysis: Science & Education, v. 18, no. 8, p. 985–1004, doi:10.1007/s11191-008-9146-6.
Dott, R.H., Jr., 1998, What is unique about geological reasoning?: GSA Today, v. 8, no. 10, p. 15–18.
Dott, R.H., Jr., and Batten, R.L., 1976, Evolution of the Earth (2nd ed.): New York, McGraw-Hill Book Company, 504 p.
Dummett, M., 1978, Truth and Other Enigmas : Cambridge, Massachusetts, Harvard University Press, 528 p.
Dupré, J., 1993, The Disorder of Things, Metaphysical Foundations of the Dis-unity of Science: Cambridge, Massachusetts, Harvard University Press, 320 p.
Engelhardt, W., and Zimmermann, J., 1988, Theory of Earth Science: Cam-bridge, UK, Cambridge University Press, 400 p.
Fann, K.T., 1970, Peirce’s Theory of Abduction: The Hague, Netherlands, Mar-tinus Nijhoff, 62 p.
Flach, P.A., and Kakas, A.C., 2000, Abductive and inductive reasoning: Back-ground and issues, in Flach, P.A., and Kakas, A.C., eds., Abduction and Induction—Essays on Their Relation and Integration: Dordrecht, Nether-lands, Kluwer Academic Publishers, p. 1–27.
Fodor, J.A., 1974, Special sciences (or: the Disunity of science as a working hypothesis): Synthese, v. 28, p. 97–115, doi:10.1007/BF00485230.
Frantsuzova, N.P., and Pavlinov, V.N., 1968, Actualism and its place in geologi-cal investigations: Geologiya i Razvedka, no. 3, p. 3–12.
Frodeman, R., 1995, Geological reasoning: Geology as an interpretive and historical science: Geological Society of America Bulletin, v. 107, no. 8, p. 960–968, doi:10.1130/0016-7606(1995)107<0960:GRGAAI>2.3.CO;2.
Frodeman, R., 2000, Shifting plates: The new earth sciences, in Frodeman, R., eds., Earth Matters—The Earth Sciences, Philosophy, and the Claims of Community: Upper Saddle River, New Jersey, Prentice-Hall, p. vii–xiii.
Gifford-Gonzalez, D., 1991, Bones are not enough: Analogues, knowledge, and interpretive strategies in zooarchaeology: Journal of Anthropological Ar-chaeology, v. 10, p. 215–254, doi:10.1016/0278-4165(91)90014-O.
Goodman, N., 1964, Science and Simplicity: Washington, D.C., U.S. Informa-tion Agency, The Voice of America, Forum Lectures, Philosophy of Sci-ence Series 16, 8 p.
Goodman, N., 1967, Uniformity and simplicity, in Albritton, C.C., ed., Uni-formity and Simplicity: A Symposium on the Principle of the Uniformity of Nature: Geological Society of America Special Paper 89, p. 93–99.
G ould, S.J., 1965, Is Uniformitarianism necessary?: American Journal of Sci-ence, v. 263, p. 223–228, doi:10.2475/ajs.263.3.223.
Gould, S.J., 1984, Toward the vindication of punctuational change, in Berggren, W.A., and Van Couvering, J.A., eds., Catastrophes and Earth History—The New Uniformitarianism: Princeton, New Jersey, Princeton University Press, p. 9–34.
Gould, S.J., 1987, Time’s Arrow, Time’s Cycle—Myth and Metaphor in the Discovery of Geological Time: Cambridge, Massachusetts, Harvard Uni-versity Press, 240 p.
Greene, M.T., 1985, History of geology: Osiris, 2nd Series, v. 1, p. 97–116.Gretener, P.E., 1984, Refl ections on the “rare event” and related concepts in
geology, in Berggren, W.A., and Van Couvering, J.A., eds., Catastrophes and Earth History—The New Uniformitarianism: Princeton, New Jersey, Princeton University Press, p. 77–89.
Gruza, V.V., and Romanovsky, S.I., 1975, The principle of actualism and logic in understanding the geologic past: International Geology Review, v. 17, no. 2. p. 167–174, doi:10.1080/00206817509471713.
Hallam, A., 1989, Great Geological Controversies: New York, Oxford Univer-sity Press, 244 p.
Heelan, P., 1983, Natural science as a hermeneutic of instrumentation: Philoso-phy of Science, v. 50, no. 2, p. 181–204, doi:10.1086/289105.
Hooykaas, R., 1959, Natural Law and Divine Miracle—A Historical-Critical Study of the Principle of Uniformity in Geology, Biology and Theology: Leiden, Netherlands, E.J. Brill, 237 p.
Hooykaas, R., 1970, Catastrophism in geology, its scientifi c character in rela-tion to actualism and uniformitarianism, in Albritton, C.C., 1975, Phi-losophy of Geohistory: 1785–1970: Stroudsburg, Pennsylvania, Dowden, Hutchinson & Ross, Benchmark Papers in Geology, v. 13, p. 310–356.
H ubbert, M.K., 1967, Critique of the principle of uniformity, in Albritton, C.C., ed., Uniformity and Simplicity—A Symposium on the Principle of the Uni-formity of Nature: Geological Society of America Special Paper 89, p. 3–33.
Hull, D.L., 1975, Central subjects and historical narratives: History and Theory, v. 14, no. 3, p. 253–274, doi:10.2307/2504863.
Hutton, J., 1795, Theory of the Earth with proofs and illustrations: Edinburgh, William Creech, 2 vols: v. I, 610 p.; v. II, 567 p. [Facsimile reprint, 1959, Weinheim/Bergstr., Germany, H. Engelmann, (J. Cramer) Codicote, Herts; Wheldon & Wesley, Ltd.]
Josephson, J.R., and Josephson, S.G., 1994, Abductive Inference—Computation, Philosophy, Technology: New York, Cambridge University Press, 320 p.
Kitcher, P., 1993, The Advancement of Science—Science without Legend, Ob-jectivity without Illusions: New York, Oxford University Press, 432 p.
Kitts, D.B., 1963, Historical explanation in geology: Journal of Geology, v. 71, p. 297–313, doi:10.1086/626903.
Kitts, D.B., 1977, The Structure of Geology: Dallas, SMU Press, 199 p.Kitts, D.B., 1978, Retrodiction in geology, in Proceedings of the Biennial Meet-
ing of the Philosophy of Science Association, Volume 2: Symposia and Invited Papers: Chicago, Published by The University of Chicago Press on behalf of the Philosophy of Science Association, p. 215–226; http://www.jstor.org/stable/192470 (accessed 14 October 2010).
Kleinhans, M.G., and Buskes, C.J., 2002, Philosophy of Earth Science; Just Sloppy Physics?: Nijmegen, Netherlands, Netherlands Centre for River Research, p. 1–3.
Kleinhans, M.G., Buskes, C.J., and De Regt, H.W., 2005, Terra incognita: Expla-nation and reduction in earth science: International Studies in the Philoso-phy of Science, v. 19, no. 3, p. 289–317, doi:10.1080/02698590500462356.
Thermodynamics time arrow and the logical function of the uniformity principle in geohistorical explanation 39
Kleinhans, M.G., Buskes, C.J., and De Regt, H.W., 2008, Philosophy of earth science, in Allhoff, F., ed., Philosophy of the Sciences: New York, Black-well, p. 1–16.
Kruijff, G.-J.M., 1998, Peirce’s Late Theory of Abduction: A Comprehensive Account: Dordrecht, Netherlands, Kluwer Academic Publishers, p. 1–29.
Laudan, R., 1982, The role of methodology in Lyell’s science: Studies in His-tory and Philosophy of Science, v. 13, no. 3, p. 215–249, doi:10.1016/0039-3681(82)90009-7.
Lorand, R., 2010, On Interpretation and Understanding: Tel-Aviv, Israel, The Haim Rubin Tel Aviv University Press, 288 p.
Lyell, C., 1830, 1832, 1833, Principles of Geology: London, John Murray, 3 vols.: v. I, 1830, 511 p.; v. II, 1832, 330 p.; v. III, 1833, 398 p. plus ap-pendices and index.
Magnani, L., 2001, Abduction, Reason and Science—Processes of Discovery and Explanation: New York, Kluwer Academic/Plenum Publishers, 224 p.
McAllister, J.W., 1997, Laws of nature, natural history, and the description of the world: International Studies in the Philosophy of Science, v. 11, no. 3, p. 245–258, doi:10.1080/02698599708573569.
Nagel, E., 1961, The Structure of Science: London, Routledge & Kegan Paul, 618 p.
Overton, W.F., 1994, The arrow of time and the cycle of time: Concepts of change, cognition, and embodiment: Psychological Inquiry, v. 5, no. 3, p. 215–237, doi:10.1207/s15327965pli0503_9.
Ozima, M., 1987, Geohistory—Global Evolution of the Earth: Berlin, Springer-Verlag, 176 p.
Palmer, T., 1993, Catastrophism, Neocatastrophism and Evolution: Notting-ham, UK, The Society for Interdisciplinary Studies in association with Nottingham Trent University, 110 p.
Pantin, C.F.A., 1968, The Relation between the Sciences: Cambridge, UK, Cambridge University Press, 218 p.
Peters, E.K., 1996, No Stone Unturned—Reasoning about Rocks and Fossils: New York, W.H. Freeman and Company, 237 p.
Price, H., 1996, Time’s Arrow and Archimedes’ Point—New Directions for the Physics of Time: New York, Oxford University Press, 320 p.
Rudwick, M.J.S., 1971, Uniformity and progression: Refl ections on the struc-ture of geological theory in the age of Lyell, in Roller, D.H.D., ed., Per-spectives in the History of Science and Technology: Norman, Oklahoma, University of Oklahoma Press, p. 209–237.
Rudwick, M.J.S., 1972, The Meaning of Fossils—Episodes in the History of Palaeontology: London, Macdonald, 287 p.
Rudwick, M.J.S., 1997, Georges Cuvier, Fossil Bones and Geological Catastro-phes: Chicago, Illinois, University of Chicago Press, 318 p.
Rudwick, M.J.S., 1998, Lyell and the principles of geology, in Blundell, D.J., and Scott, A.C., eds., Lyell: The Past Is the Key to the Present: Geological Society of London Special Publication 143, p. 3–15.
Russell, B., 1921, The Analysis of Mind: London, Allen and Unwin, 310 p.Schumm, S.A., 1998, To Interpret the Earth—Ten Ways to Be Wrong: Cam-
bridge, UK, Cambridge University Press, 144 p.Scriven, M., 1959, Explanation and prediction in evolutionary theory: Science,
v. 130, no. 3374, p. 477–482, doi:10.1126/science.130.3374.477.Şengör, A.M.C., 2001, Is the Present the Key to the Past or the Past the Key to
the Present?—James Hutton and Adam Smith versus Abraham Gottlob Werner and Karl Marx in Interpreting History: Geological Society of America Special Paper 355, 51 p.
Shea, J.H., 1982, Twelve fallacies of uniformitarianism: Geology, v. 10, p. 455–460, doi:10.1130/0091-7613(1982)10<455:TFOU>2.0.CO;2.
Simpson, G.G., 1963, Historical science, in Albritton, C.C., ed., The Fabric of Geology: London, Addison-Wesley, p. 24–48.
Simpson, G.G., 1975, Uniformitarianism—An inquiry into principle, theory, and method in geohistory and biohistory, in Albritton, C.C., ed., Phi-losophy of Geohistory: 1785–1970: Stroudsburg, Pennsylvania, Dowden, Hutchinson & Ross, Benchmark Papers in Geology, v. 13, p. 256–309.
Sklar, L., 1993, Physics and Chance—Philosophical Issues in the Foundations of Statistical Mechanics: New York, Cambridge University Press, 456 p.
Sober, E., 1975, Simplicity: London, Oxford University Press, 200 p.Sober, E., 1988, Reconstructing the Past—Parsimony, Evolution and Inference:
Cambridge, Massachusetts, Massachusetts Institute of Technology Press, 288 p.
Turner, D., 2005, Local underdetermination in historical science: Philosophy of Science, v. 72, p. 209–230, doi:10.1086/426851.
Turner, D., 2007, Making Prehistory—Historical Science and the Scientifi c Real ism Debate: New York, Cambridge University Press, 238 p.
Virgili, C., 2007, Charles Lyell and scientifi c thinking in geology: Comptes Ren-dus Geoscience, v. 339, no. 8, p. 572–584, doi:10.1016/j.crte.2007.07.003.
Visotskii, B.P., 1961, Problema Aktualizma i Uniformizma i Sistema Metodov v Geologii: Voprosy Filosofi i. Akademia Nauk S.S.S.R, v. 3, p. 134–145.
Watson, R.A., 1966, Is geology different? A critical discussion of “The Fabric of Geology”: Philosophy of Science, v. 33, p. 172–185, doi:10.1086/288088.
Watson, R.A., 1969, Explanation and prediction in geology: Journal of Geol-ogy, v. 77, p. 488–494, doi:10.1086/628374.
Weinryb, E., 1987, Historical Thinking: Issues in Philosophy of History: Tel-Aviv, Israel, Everyman’s University, 497 p.
Wilson, L.G., 1967, The origins of Charles Lyell’s uniformitarianism, in Albrit-ton, C.C., ed., Uniformity and Simplicity—A Symposium on the Prin-ciple of the Uniformity of Nature: Geological Society of America Special Paper 89, p. 34–62.
Zellner, A., Keuzenkamp, H.A., and McAleer, M., 2001, Simplicity, Inference and Modelling: Cambridge, UK, Cambridge University Press, 316 p.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 FEBRUARY 2013
