Wesley C. Salmon, Four Decades of Scientific Explanation (Pittsburgh: Pittsburgh University Press, 2006).
The search for scientific understanding of the world has roots further back in Aristotle. Philosophers and scientists have tried to elongate the scientific quest of understanding this world, with the philosophical quest until recently had lagged behind; however, one of the central theses of Wesley Salmon’s book is that a substantial amount of progress is been made in this area recently. Notably, Wesley Salmond contends that there may be enough of a basis to date to build a new consensus in the philosophy of science in lieu of Hempel and Oppenheimer (1948).
INTRODUCTION: philosophers before the time of Aristotle at least, realized that there were two distinct types of scientific knowledge: that which seeks answers to why questions; and that which seeks answers to questions of how/what. The former is explanatory, and the latter is descriptive. Scientific knowledge, according to Wesley Salmon, is provided by the explanatory type of understandings not merely descriptive knowledge (3). According to Aristotle, scientific explanations or deductive arguments; he recognized, however, that not all deductive arguments give you properly so-called explanations. Philosophers of diverse persuasions agree that the fundamental aim of science is to explain natural phenomena (4). 20th century scientific philosophy arose in the context of post-Kantian and post-Hegelian German idealism, which was heavily influenced with transcendental metaphysics and theology. The early logical empiricist and positivists reacted against these influences and sought to eliminate them from science; one such influence was the concept of teleology, as it was seen to be a super-empirical entity, and it either appealed to a supernatural being who created and continues to direct nature, or to an empirically inaccessible agency like ientelechy or vital forces. In fact, in that era, many philosophers said that there could be no rule explanations without an appeal to extra-empirical; they thought in anthropomorphic ways that always appealed to “purposes” (4). In reaction to this type of thinking, many scientists ran away from explanation altogether, saying that science should not be concerned with that at all; those that did allow that science would offer explanations clarified that this did not entail anything beyond empirical knowledge.
In recent decades, however, there is been a general consensus that science can tell us not only what, but also why; by means of “why” explanations, science provides one with genuine understanding of the world (5).
The logical empiricists who were instrumental in forging the old consensus looked upon the task of philosophy as the construction of explications of fundamental concepts (see Carnap, 1962, chapter 1 for clear detailing of this; see also Coffa, 1973).Carnap (1962) specifies four criteria according to which explications are to be judged: one) by similarity to the explicandum–that is, the concept of confusion; two) exactness; three) fruitfulness; and four) simplicity. Wesley Salmon is quick to distinguish between offering an exclamation for some fact and providing grounds for believing it to be the cause (6). Moreover, evidence does not equal explanation according to Wesley Salmon. According to Carl Hempel, the deductive-nomological (DN) model is the formal scientific explanation, whereas the hypothetico-deductive (HD) schema is for scientific confirmation. It should be noted that in the HD schemata, the function of explanation is not to establish (or support) the truth of the conclusion, as that is already presupposed.
All Chapter 1:
logical empiricism arose in the context in which teleology, final causes, purposes, and ends played vital roles. Philosophers like Braithwaite, Hemple, and Nagel, do abandon the idea that science could provide legitimate explanations of various phenomena and were deeply concerned to provide accounts that would require “scientific explanations” to have empirical content, and also salt to eliminate appeals to entelechies or final causes (26). However, there appears to be cases in the empirical sciences in which there are veritable explanations that are framed in terms of ends or goals (26).
In 1953, Braithwaite addressed the problem of teleological explanation (1953, chapter 10); he saw no problem in teleological explanations of actions that are the result of conscious intention, but he also allowed for teleological activity in which conscious intent is seemingly absent. Notably, in these later cases, the main concern is whether a future state can legitimately be said to explain the present fact. Braithwaite points out to striking features of gold-directed behavior: one) plasticity––which refers to the fact that the agent has more than one way to achieve a goal; and two) variancy––which refers to the fact that the goal can be successfully reached under a wide variety of initial and background conditions (cross reference, for example, a homing torpedo or missile).
Discussions regarding teleological or functional explanations often employee either or both of two models: one) goal-directed behavior; or two) self-regulating (that is, homeostatic) behavior (cross reference John Canfield 1966, and Scheffer 1958). Ernest Nagel, for example, dealt mainly with teleological or functional explanation in physiology, focusing on the homeostatic (or self-regulating) type of explanations (cross reference Nagel 1956; Nagel 1961, chapter 12). In fact, Nagel argues the point that in many physiological cases, any teleological explanation can be replaced by an equivalent causal explanation––which I, BRADFORD, take to mean that in etiological explanation can be replaced with an equivalent efficient causal explanation instead (HOWEVER, I, BRADFORD, need to check on this assertion). For Nagel, no whiskers of goal-directed behavior is explainable only in terms of future results (27).
On the issue of final causation, Hempel later suggests that subsequent events might sometimes explained earlier events (Hempel 1965, pg. 353-54).
Functional explanations can be often seen to explain biological evolution in which adaptation to the environment plays a crucial role, appearing to involve the attainment of goals. In biology, the term “function” is usually reserved for characteristics that are important to the health or survival of the organism (29).
According to Wesley Salmon, Nagel––neither in 1986 or 1961––has not succeeded in eliminating the problem of possibly equivalents for teleological or functional explanation in general (31).
In evolutionary biology, functional considerations play a crucial role, and––at least at the time of Charles Darwin––it has been apropos to deny that any and all appeals to functions involves the conscious purposes of the creator, or any other sort of final causation (31-32). Since Charles Darwin, according to Wesley Salmon, would you understand evolutionary process in terms of efficient causes in conjunction with, perhaps, chance occurrences. Evolutionary biology thus requires a causal explication of function, according to Wesley Salmon (32). In evolutionary biology, we need an adequate account of what has come to be called etiological explanations that encompass both genetic––that is, the stores that lead up to the event to be explained––and functional explanations, according to Wesley Salmon (32). In Wesley Salmon’s opinion, a real breakthrough regarding functional explanations was garnered by Larry Wright (1976), who advanced and explicitly causal account of functions as teleological explanation; Nagel took up this topic again in 1977.
Chapter 3:
largely in response to vitalism in biology, Braithwaite, Hempel, Nagel,and Scheffler explicitly addressed the problem of teleological and/or functional explanations (salmon, 111).
Philosophy of biology––properly so-called––picked up steam, so to speak, in the literature during the years of 1968–1977, according to Wesley Salmon (111). In fact, Morton Beckner’s classic, The biological way of salt (1968), opened this period with a proverbial bang. In the early 1970s, David Hull, Michael ruse, and William Wimsatt, entered the field with avengeance, and these people have not left the discussion even today. The problem of the teleological/functional explanation is central to biology (111). According to Wesley Salmon, Larry Wright’s 1976 book is fundamentally correct in its application to biology at least (111). In 1987, Bigelow and Pargetter offer a theory of possible explanations that may improve Wright’s (1976) account, but it is not fundamentally different from it.
Wright contends that geological explanations occur frequently in science, and––calendar Hempel and Beckner––many of them were sound; he regards to the logical explanations as causal in the sense that does not require that causes come after their effects; however, simultaneously, he considers teleological explanations as fundamentally future-oriented, as teleology by definition involves goal seeking behavior; capture both concepts above, Wright offers the term “consequence-etiology”. The basic idea of consequence-etiology (CE) is simple: a behavior, let’s say number one, has been causally efficacious in the past to achieve state number two, and therefore it occurs. In fact, such behavior has worked so often that it confers an evolutionary advantage on the members of the species (111). Such intentional behavior relies on a conviction that this action, number one, is suitable to achieving state number two. Roughly, the causal efficacy of number one in bringing about state number two in the past is an indispensable part of the cause of the occurrence of number one and the present. It is a CE because the consequences of doing number one are crucial part of the etiology of doing number one.
Larry Wright’s (1976) model teleology is, according to Wesley Salmon, particularly well-suited to evolutionary biology (112). Certain kinds of behavior become typical behavior for species of animals because it confers an evolutionary advantage with respect to the goal survival or reproduction. Wright notes, the behavior may have arisen in one of two ways: one) deliberate creation by a supernatural agency, or two) as a result of natural selection. The analysis of the concept of semiology does not tell us which of these two possible etiologies is the actual one, as empirical science must distinguish between the two types of etiology. According to Wright (1976), human conscious intentional behavior, human artifacts, and human behavior that results from natural selection, all can be properly called teleological, noting unit is subject/entity might do behavior number one if it tends to bring about state number two (Wright 1976, 39). The term “tends” is significant because action number one does not need always to bring about state number two, as it is sufficient for me is you have a disposition to do so under suitable conditions. Indeed, Wright notes: teleological behavior is behavior with the consequence etiology: and behavior with a consequence-etology is behavior that occurs because it brings about, is the type of thing that brings about, tends to bring about, is required to bring about, or is it some other way appropriate for bringing about some specific goal” Wright (1976, 38-39).
Notably, even though they are closely related, Wright does not identify teleology with functional acriptions or explanations. It seems that the reason for this is because only behavior, according to Wright, is truly theological or goal-directed––in many cases, something may fulfill a question just by “being there”. The function of an entity is distinguished from all sorts of other things that might happen because it is “they are” by the fact that its location has a consequence-etiology (cross reference Wesley Salmon, 113). Wright offers the following schematic regarding functions: the function of action number one is state number two if and only if state number two is a consequence (result, that is) of actionable ones being there and action number one is there because it does (or results in) state number two (cf. Wright 1976, 81).
Functional ascriptions, Wesley Salmon notes, are frequently made in evolutionary biology (113). In evolution, certain attributes first arise by chance, and this first occurrence does not have to be a function because it has no consequence-etiology at that first appearance; the function is established when the recurrence or persistence of the attribute occurs because of a consequence of its previous presidents (114). Wright asserts explicitly that his account of teleological explanation and functional ascription is not inconsistent with a completely mechanistic explanation of goal-directed behavior (Wright 1976, 57-72). Wright maintains, moreover––Calvert Charles Taylor 1964 and others––that even if it should turn out to be possible to provide a completely physico-chemical account of the behavior of plants and animals, that would not eliminate its teleological character. Teleological and functional explanations are on as firm a ground as any in science (114). The proposal of Bigelow and Pargetter next next seven is a “propensity theory” that they claim to be more forward-directed than Wright’s consequence-etiology theory; Bigelow and Pargetter’s (1987) theory relies heavily on the tendency of a given item to do its function. According to Wesley Salmon, Bigelow and Pargetter (1987) fail to note that Wright’s analysis of function supplements his analysis of teleological behavior––they seem to run teleology and function together (114).
According to Wesley Salmon, Wright’s analysis of teleological behavior is as much a propensity theory and is also as much forward-looking as is the theory of functions offered by Bigelow and Pargetter (1987) (Wesley Salmon, 114). Wesley Salmon posits that on Wright’s account, every case of a function presupposes some bit of teleological behavior. In any case, Bigelow and Pargetter’s (1987) theory is not very much different than Wright’s (1976)––the main difference would be a disagreement over whether or not an item has a function the first time it occurs, or––as Wright maintains––it only has a function subsequently. As Wesley Salmon notes, if this is in fact the main difference, is really miniscule indeed (114).
Walt Wesley Salmon “wholeheartedly” agrees with Wright’s complete causal grounding of teleological and functional explanations, does have one major philosophical disagreement with him: Wright––following his teacher, Michael Scriven, maintains that we have perfectly objective ways of recognizing causal relations, and therefore does not offer any characterization of causality itself; Wesley Salmon thinks this is simply an evasion of the problem (114-115). After all, Wesley Salmon has spent much time on excavating causality because he does not believe the underlies notion of it is merely as unproblematic as Scriven and Wright have climbed.
Nagel (1977) again addressed the issues of teleological and possible explanations and criticizes the account of Hempel and Wright (1977, 305-309, 296-301); note that Wesley Salmon does not find Nagel’s criticism of Wright ” particularly weighty” Wesley Salmon, 193FN48). Nagel’s main response to Hempel is a challenge to the notion of functionally equivalents, noting that if one specifies with sufficient precision the nature of the organism or system within which an item has a function, it is in many cases necessary for the fulfillment of the function (115).
Nagel maintains that many sciences, including biology in particular, provide legitimate teleological or functional explanations; he allows that various kinds of facts can be explained by appeal to subsequent conditions (115).
During the onset of the. That Wesley Salmon hearing investigates, 1947-1987, scientific philosophers had extremely negative views toward Hans Driesch’s attempts to explain biological phenomena in terms of entelechies and vital forces (Wesley Salmon, 179). During the same era, according to Wesley Salmon, Karl Popper––on the other proverbial end of the spectrum––was severely critical of Freudian psychoanalytic theory and of Marxian economics because they were empirically vacuous to the extent that they were and are devoid of scientific explanatory import (179).
According to Wesley Salmon, there is widespread agreement among current scientific philosophers that science can teach us not only that but also why insomuch as one of the most important fruits of modern science is understanding the world––we don’t have to go outside of the world to find it (181).
Wesley Salmon notes in passing––just for my purposes––a CS Peirce stall walls as those generalizations that end up as fundamental principles in the ideal limit of scientific investigation.
Note Nagel (1961) lists 4 types of explanation: one) deductive; two) probabilistic; three) functional/teleological; and four) genetic. Nagel does not indicate, then, in any way, that the last two can be reduced into one, or if they can collapse into the previous two.
John Canfield (1966) is a clear and perceptive introduction to many of the same issues discussed herein by Wesley Salmon, note.
END OF REVIEW OF SALMON’S TEXT.