This Content Downloaded From 132.174.255.108 On Wed, 15 Feb 2023 14:05:00 UTC

Darwin's Theory of Evolution as a Science
Author(s): Raphael Falk

Source: Poetics Today , 1988, Vol. 9, No. 1, Interpretation in Context in Science and
Culture (1988), pp. 61-94
Published by: Duke University Press
Stable URL: https://www.jstor.org/stable/1772888
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Duke University Press is collaborating with JSTOR to digitize, preserve and extend access to
Poetics Today
This content downloaded from

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
All use subject to https://about.jstor.org/terms
Darwin's Theory of Evolution as a Science
Raphael Falk
Genetics, Hebrew University, Jerusalem
Introduction
The effect of Darwin's theory of evolution may best be epitomized by

the title that Th. Dobzhansky, a leading figure of modern research in
evolution, gave to one of his public talks (1973): "Nothing in biology
makes sense except in the light of evolution." But it is not only in
biology that Darwin's theory of evolution has had such a crucial im-
pact. Its effects are no less significant on disciplines as far apart as
geology and astrophysics on the one hand and sociology and linguistics
on the other.
Yet there have always been those who claimed that the theory of
evolution is not really a "scientific theory." Such claims were made not
only by creationists who maintained that Darwin's theory of evolution
was, at best, an article of faith, like the biblical creation story, but also
by philosophers like Popper who were profound adherents of an evo-
lutionary outlook. Among biologists themselves, there have been many
(e.g., G. G. Simpson, E. Mayr and F. Ayala) who insisted that biology in
general and the theory of evolution in particular should have unique
status among scientific theories, a status that would highlight their
basic difference from the physical sciences which have traditionally
been considered the paradigmatic model of a "Science."
This paper was prepared while I was a fellow at the Institute for Advanced Study in
Berlin. The many discussions with the other fellows of the Institute were invaluable
to me. But above all, it was my wife Ruma who helped me to crystallize my thoughts
and who was a constant critical reader of what I wrote and rewrote. Thanks to her
the paper does not include more mistakes and blunders than it actually does.
Poetics Today 9:1 (1988). Copyright ? The Porter Institute for Poetics and Semi-
otics. ccc 0333-5372/88/$2.50.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
62 Poetics Today 9:1
This paper examines some of the claims for the unique position
of Darwin's theory of evolution and argues that not only do physical
sciences not correspond to the utopian or normative model that many
scientists and philosophers of science still attach to them, but also that
biological research differs only in degree, and not in substance, from
physical research. More specifically, it will be shown that Darwin's
theory of evolution is one of the best examples of a successful scientific
theory.
The idea of evolution, i.e., the notion that groups of living organ-
isms change gradually over generations so that the species of today
stem from organisms of other species in the past, is not new. There
are claims that even the story of creation in the first chapters of Gen-
esis alludes to the idea of evolution with its sequential description of
the creation. Whether this interpretation is correct, there is no doubt
that the idea that forms might change and one species might be trans-
formed into another was discussed in the ancient world as indicated
by the evidence for evolutionary thinking in the writings of the Greeks
Anaximander in the sixth century B.C. and Empedocles in the fifth
century B.C. However, these abstract attempts of the ancients to ex-
plain the organization of the living world (and indeed of the nonlivin
world too) through some kind of developmental or transformational
processes were not significant in the scientific thinking of the ancient
world or in that of the Middle Ages.
Evolution as we understand it stems from the age of the scientific
revolution-at the end of the sixteenth century and especially at the
beginning of the seventeenth century. It was probably inspired by
the ideas of constant development or of "progress" of culture in gen-
eral and science and technology in particular which were the guiding
forces of the time (Bury 1920). Progress, it was believed, gradually
leads to a better understanding of nature, particularly the nature of
man and thus to change or to higher and more complete values, both
in matter and mind. Francis Bacon (1561-1626) extended these ideas
of cultural progress to the subject matter of the life sciences and it wa
probably due to him that the idea of evolution, as change in species
leading to perfection, was established in biology.
Evolutionary ideas must be contrasted with the notion that living
creatures not only form fixed and well separated entities but tha
intraspecific variation observed in nature merely represents deviation
from some ideal or characteristic types. Such typological thinking wa
deeply entrenched in the Middle Ages and left its mark on biologica
research, exerting a profound effect on the philosophy of biology even
now. The present classification of organisms, for example, is based
on the Systema Naturae published in 1775 by Carlus Linnaeus (1707-
1778) who believed in the stability of forms and hence catalogued the

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
Folk * Darwin's Theory of Evolution 63
whole living (and nonliving) world into a hierarchial system of well-

defined categories, each based on a type specimen. Species within each
group were identified, for convenience, by single diagnostic character-
istics (e.g., the distinguishing marker of mammals was differences in
dentation).
Other students of living beings even in Linnaeus's own time, at-
tached more importance to differences between creatures. His contem-
porary, Comte de Buffon (1707-1788) ignored the Linnaean classifica-
tion altogether and adopted a generalized system which considered the
variation of many traits simultaneously. Another contemporary, Pierre
Louis Moreau de Maupertuis (1698-1759), tried to formulate rules
allowing both for the stability of species and for their change. And
Johann Friedrich Blumenbach (1752-1840) argued that Linnaeus's
classification, based on arbitrarily chosen single characters, resulted
in correspondingly arbitrary and often accidental grouping of forms
(Lenoir 1980).
By the turn of the nineteenth century, the idea of evolution was quite
widespread. In 1809, Philosophie Zoologique by Jean Baptiste Pierre
Anton de Monet de Lamarck (1744-1829) was published. Lamarck
not only accepted evolution as fact but also made one of the most
important attempts to explain the mechanism by which it could have
occurred.
However, it was Charles Robert Darwin's (1809-1882) book, On
the Origin of Species by Means of Natural Selection, which appeared in
November 1859, that was a turning point in biology. His theory at
once attained worldwide attention and within less than ten years had
become a household item for many researchers in the life sciences.'
The influence of Darwin's theory was not less impressive on other
disciplines like the social sciences where it often won even more fame
and influence than in its original domain (see, e.g., Hofstadter 1944).
The immense impact of the theory cannot be explained with a claim
that the time was ripe and that Darwin's data and observations were so
convincing that any reasonable and unbiased person had to accept his
theory. The opposite is probably closer to the truth. As Allen indicated
(1968), even at the beginning of this century, several prominent biolo-
gists expressed serious doubts about the Darwinian theory of natural
selection. Despite the arguments of some historians the literature of
this period demonstrates that the idea of natural selection had by no
means gained wide or whole hearted acceptance even as late as 1915.
Data that accumulate in the drawers of scientists and on laboratory
1. As is well known, the biologist Alfred Russell Wallace (1823-1913) came to

a similar explanation of evolution during the very same years in which Darwin
formulated his ideas of evolution.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
shelves are not enough to produce changes in their views or to in-

duce them to accept something like "evolution" as a fact. Linnaeus
did not make his system of classification less stringent even when he
himself learned how difficult it was to maintain a system which puts
every species in a separate "box." Facts are just interpretations of our
reactions to sensory stimuli. Thus, it is appropriate that what seems
to be a "fact" to one person is utter nonsense to somebody else or is
accepted, at best, as a hallucination. Baron Leopold Chretien Fred-
erick Dagobert Cuvier (1769-1832) and Jean Louis Rodolphe Agassiz
(1807-1873) were no doubt two of the greatest anatomists of the nine-
teenth century and had access to the most extensive and detailed data
available at the time. Yet both categorically dismissed the "fact" of
evolution to the end of their lives-not to mention the evolutionary
explanation like Darwin's.
Darwin presented an impressive quantity of data to support the
claims of organic evolution but there is little doubt that his outstanding
and specific contribution was the "mechanism" he offered to explain
the phenomenon. Darwin was the first to propose an explanation that
did not rely on hidden forces or a superior will but which showed how
evolution could happen "by itself." This concept was incredibly liber-
ating, producing a revolution that is appropriately compared with the
Copernican revolution three centuries earlier. The latter moved our
planet out of the center of the universe, thus preparing the develop-
ment of the sciences from the perspectives of spectators of nature, as
part of it, rather than from the perspective of its purpose and mean-
ing. The Darwinian revolution completed the task, stripping man of
his unique position among living creatures and bending him to laws
of nature independent of him.2
For somewhat over a hundred years from its first publication, Dar-
win's theory of the origin of species by natural selection has itself
undergone an evolution. It is not only the many new and unex-
pected facts discovered in laboratories and in the field that demanded
changes; the views and attitudes of people have also changed. Like any
theory, Darwin's is also to a large extent the reflection of the opinions
and perspectives of his contemporaries.
The generality and breadth of its implications were no doubt de-
cisive factors in crystallizing the opposition encountered by Darwin's
2. In the first edition of The Origin of Species, Darwin used the term "evolution"
only once, at the end of the book, and then in a context completely different from
that of today. It seems that it was Herbert Spencer-one of the central figures
in the (over)interpretation of Darwin's theory beyond its biological contexts-who
introduced the term "evolution" in the sense known to us today.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
theory since its inception over one hundred years ago.3 Even now, it
seems that most resistance to the theory comes from those circles who
focus on what they believe to be the explicit and implicit consequences
implied by the acceptance of a theory of "evolution by natural selec-
tion" rather than on how far it can make sense of observations and
experimental data.
Principles of Darwinian Evolution

Before considering the status of Darwin's theory of evolution as
scientific theory, I shall summarize its main premises. First, it is usefu
to distinguish between the arguments and evidence supporting the
occurence of an evolutionary process itself and the explanations of
how this process takes place.
Darwin presented an impressive amount of data based on his own
observations as well as on those of others about the variation between
individuals belonging to the same species and about the artificiality
of the demarcation lines between species. Along with the extensive
evidence from fossils about animals and plants related to but differ-
ent from those of today which apparently inhabited the earth in one
period after another in the past, such data were accepted as crucial
evidence for the existence of a process of evolution. This was sup-
ported by an impressive body of data from embryology and compara-
tive anatomy of the apparent traces of these evolutionary processes.
The accumulation of evidence which has not ceased since Darwin's
days has only supplied additional support to the notion and helpe
fill in and modify the details of the story with increasing accuracy.
But the main innovation of Darwin's theory was the description o
mechanism leading causally to the evolution of living creatures. Th
were two forces motivating the process of evolution:
(a) The production of inherited variation (mutations) between liv
organisms. The production of inherited variants is not necessa
"random," or due to chance processes. But, in principle, the proce
3. Darwin took special care not to formulate his theory about the origin of sp
in evaluative terms. The amoeba is adapted to its environment as well as we ar
ours. Who is to decide that we are more "advanced" living beings? Unfortuna
Darwin was nearly the only one who insisted on the fact that evolution leads
to a better adaptation of the organisms to their environment and is irreleva
any abstract ideal of "progress." If we had heeded Darwin's words, we could h
saved many confusions and misunderstandings between researchers of evolut
and nonexperts. Ever since Darwin, there have been those who saw in this th
support for the development of enlightenment and others who found in it jus
cation for social orders as different as Victorian liberalism on the one hand and
Soviet Marxism, on the other.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
that produce such variants act completely independent of their effect

on or their significance for the future of the organisms in which they
occur.
(b) Contingencies that control and check variation between

ganisms. Every living organism constantly encounters seve
and internal environments. Relationships with the environ
from climatic conditions to structural facts like hands ab
The capacity of living organisms to adjust to or withstand
nations of circumstances in which they find themselves is
a variety produces differential propagation of certain for
partial or complete elimination of others (or the change i
tive frequencies of the genes involved). Natural selection de
conditions for systematic change in the relative frequencies
among living organisms (see Grene 1974b).
It is the combination of both these forces, the producti
reditable variability and natural selection that are not onl
but also sufficient conditions for evolution to take place.4
the Darwinian theory of evolution proposes a procedure t
order and organization with mechanisms that produce dis
randomness simply miss the essential gist of the element o
lation inherent in the mechanism foreshadowed by Darwi
nature of self-regulatory systems (negative feedback systems)
are constructed of (basically) two devices which could act c
completely independent of each other but which have been
so that they counteract each others' products. The action
tails a situation that provides the necessary initial conditi
other, which, by its action, in turn, provides the condition for
the first. If both systems counteract each other, a predicta
equilibrium, or stable state ("goal"), will be obtained (Nage
It was largely the simplicity of these principles that gav
theory its sweeping power. Yet, the explicit formulation of
ciples also exposed the weak points of the theory which w
grasped by those who wanted to oppose it while others, r
give up a good theory, resolved to settle these discrepancies
framework of the theory itself.5 Although many of the sh
4. This insight was explicated by H. J. Muller in 1922: "It is not in

variation which bring about evolution, but the inheritance of variat
now, any material or collection of materials having this one unusu
istic and evolution would automatically follow, for this material wo
the accumulation, competition and selective spreading of the self
variations come to differ from the ordinary inorganic matter in inn
spects...."(Muller 1922: 35).
5. One of the most quoted stories is of the discussion in 1869 betwee
Samuel Wilberforce and the biologist Thomas Henry Huxley, who w

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
Falk * Darwin's Theory of Evolution 67
lations of the theory of Darwin's time have been reconstructed into

imposing props it would be wrong to claim that all the difficulties have
disappeared or that new ones have not cropped up.
In Darwin's time, for example, there were no clear and accepted
concepts about the inheritance of traits. Darwin himself offered a
theory of inheritance but only a few years after he published his book
it was demonstrated that, if Darwin's concepts about the mechanisms
of heredity were correct, his proposed evolution by natural selection
was impossible. Although the principles of heredity accepted today
were formulated byJohann Gregor Mendel (1822-1884) in 1865, they
were rediscovered only at the turn of this century. At first it seemed as
if there was no way to accommodate modern genetics with Darwinism.
It took thirty years to realize that Mendelian principles themselves re-
solved some of the most recalcitrant inconsistencies in Darwin's theory.
Another problem that soon arose was that, according to the prin-
ciple of natural selection, the better adapted creature was one that
would survive (or leave more offspring) in the long run. But the only
way to find out which was better adapted was to wait and see which
survived. This is a circular argument soon exposed by opponents of
Darwinism. It must be admitted that the avid adherents of biological
as well as social Darwinism often used this circular argument uncriti-
cally to explain orjustify nearly anything. Practically unlimited powers
were attributed to natural selection and every organ or trait claimed to
be in a living organism was considered evidence of its adaptive value
and their absence to its lack of adaptive value.
This trend continues to some extent up to the present and not only
among social Darwinists even though it became clear, for example,
that many genes in natural populations were often "neutral" in their
adaptive values and that the presence of traits in organisms may be due
to structural constraints in spite of their adaptiveness per se. Natural
selection does not act on traits in isolation. Its capacity to shape or
change also depends on the structural and on physiological properties
of other components in the system, rather than exclusively on those
of the trait under consideration (Gould 1977a, b).
Some investigators have even proposed adding a third principle to
the two basic principles of Darwin's theory of evolution which would
emphasize the role of the history of the organisms and the structural
and functional stresses on the constructive power of natural selection
(see, e.g., Webster and Goodwin 1982). However, since the term "envi-
ronmental circumstances" is not limited to the outer environment of
win's Bulldog" as a result. This verbal fight can hardly be noted for its thorough
arguments but rather for the eloquence of the discussants that was demonstrate
on that occasion.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
the organisms, these constraints may be included among the compo-

nents of the internal environment so there is no need to add another
principle to the two basic Darwinian principles.
What is a Scientific Theory?

In the opening pages of his The Structure of Science (1968), Nagel sug-
gests that the desire to explain is basic to scientific activity. But to be
scientific, the explanation should be organized and consistent as well
as anchored in factual data. In science, one endeavors to discover the
conditions where certain events occur and to formulate these condi-
tions in generalized terms. "To explain, to establish some relation of
dependence between propositions superficially unrelated, to exhibit
systematically connections between apparently miscellaneous items
of information are distinctive marks of a scientific inquiry" (Nagel
1968: 5).
Hempel also emphasizes the explanatory component of each scien-
tific description: "A class of phenomena has been scientifically under-
stood to the extent that they can be fitted into a testable and adequately
confirmed theory or a system of laws" (Hempel 1965: 329).
The elements of a scientific explanation are thus: (a) the formu
lation of laws or generalities that are (b) based on observations o
experimental data and, finally, that (c) the explanation can be tested
Indeed, it is the method that makes an activity scientific rather than
the desire to explain per se or the wish to expose the truth abou
the world around us and about ourselves. The only epistemic access
we have to the world is through our theories. As realists, we would,
perhaps, have preferred to construct a picture of the world from an
external standpoint and to compare theories from there. However,
the only way we have is to compare theories with each other and to
improve them in terms of accepted rational procedures. Since we, the
knowledgable subjects, are part of this world we want to know, what
makes an activity scientific is the methodology by which it strives to
improve theories rather than the acceptance of any given theory. Thi
should also be the meaning of Kant's assertion that "the purpose of
science is to produce a systematic experience of nature."
Bieri (1984) has pointed out that, as time-aware beings, "we must
distinguish between the happening of an event or the being of a thin
in the world on the one hand, and their being represented on th
other." Consequently, "there may be a difference between the way we
represent an entity and the way that entity is." To understand (one
past) one must be ready to appropriate it, i.e., to tell an explanatory
story about it. Although this is a requirement based on memories tha
are "for the most part correct, i.e., true," in telling the story, we "pro
ceed selectively." Bieri concludes: "I bestow on my life an inevitability
which it never had, by overlooking or refusing to acknowledge that

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
the cause of my past was also determined quite simply by countless

chances." It is the realization of this "human" nature that a scientist
has to face in devising methods to segregate, at least in the long run,
the "happening of an event or the being of a thing" as far as possible
from its representation.
Even when we limit ourselves to the operational definitions of "sci-
ence," that is, those definitions that state that a theory is scientific if
it is based on the methods of scientific investigation and discovery, we
will find that the difficulties have not been resolved.
Although we would no doubt agree that one of the basic principles
of the scientific method should be the unbiased collection of facts and
data, this is impossible. There is no abstract way to accumulate data:
from the moment we consider a problem we want to address, we form
opinions and ideas about what to look for and what we might find.
Popper stressed this when he "innocently" asked what a person who
was told to "Make observations!" or "Collect data!" would actually do.
If we do not tell him what data to collect and how to make observations,
he would have no choice but to apply his own views and theories to the
task. "Our drawing of the theoretical-observational line at any given
point is an accident and a function of our own physiological make-up,
our current state of knowledge, and the instruments that we happen
to have available and, therefore . . . it has no ontological significance
whatsoever" (Maxwell 1962). In fact, since Quine's "Two Dogmas of
Empiricism" (1953), we have known that we also have to abandon
the belief "in some fundamental cleavage between truths which are
analytic, or grounded in meaning independently of matters of fact,
and truths which are synthetic, or grounded in fact" (Quine 1953: 20).
We must agree with Jurgen Habermas that the statements that express
hypotheses are the function of our way of representing them; in other
words, our hypotheses are the way we grasp the world.
A common device which is supposed to overcome these difficulties is
the demand to formulate arguments that make sense to every rational
person. But what is rationality? Bieri (1984) suggests that rational be-
ings are those who "are acting in accordance with what they desire, feel
and believe." Hempel, in private discussion, was more explicit in sug-
gesting rationality to be goal-dependent: "It is the most efficient way
to achieve a goal, considering the knowledge available." This only em-
phasizes the relativistic value of the term. Under these circumstances,
I suggest accepting Carnap and Popper's instrumental view, that the
demand that scientific propositions be able to be "publicly" examined
and tested is a kind of guarantee for maintaining rationality. Yet do
we not all know of theories that were rejected for decades by every
rational scientist until the opinion of the single stubborn scientist was
the one rationally accepted?
Are we, then, really forced to accept that scientific theories can-

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
not be demarcated from nonscientific theories and, presumably, as a

result that what are considered scientific theories are just irrational
constructs? Do we have to agree with Kuhn that there is no logic of
discovery but only (social) psychology of discovery, that scientific theo-
ries are essentially metaphysical and irrational or with Feyerabend's
conclusion that, since science is irrational, "anything goes"? I do not
think so. Although Popper's efforts to delineate science and non-
science failed as any other attempt probably would, this does not mean
that science is irrational. Following Bieri's assertion that "rationality
is a parameter in the ascription of intentional states and in this sense
is a condition of personhood" (1984), the rational response to our
limitations (including our often irrational behavior) is to be constantly
aware of them as part of the effort "to achieve the goal, considering
the knowledge available." In other words, to admit the impossibility of
a perfect solution is a rational step toward understanding the meaning
of the knowledge we obtain about the world and thus can also be a
guide to possible rational action.
If Kuhn's and Feyerabend's picture of science is correct, it "vindi-
cates, no doubt, unintentionally, the basic political credo of contem-
porary religious maniacs" (Lakatos 1970: 93). Lakatos is referring to
"student revolutionists," but it is equally true of creationists of vari-
ous shades. Nevertheless, given such premises, what, indeed, makes
a "metaphysical theory of evolution" different from the metaphysical
story of the bible?6
Lakatos (1970) indicates that science is doing better than that. His
methodology of scientific research program, although basically descriptive,
assumes a normative aspect by showing that scientists "intuitively"
tend to develop what amounts to rational programs. Just as no iso-
lated proposition (even an observational one) can be verified, so no
theory can be falsified. Science, however, is not concerned with sin-
gle propositions and indeed not with single theories either but rather
with research programs. "A typical descriptive unit of great scientific
achievement is not an isolated hypothesis but rather a research pro-
gram" (Lakatos 1978: 4). Such a research program has room for nu-
merous factors to act on and counteract each other: first there is the
"hard core" "tenaciously protected from regulation by a vast 'protec-
tive belt' of auxiliary hypotheses" (p. 4). Then, such a program has
the "positive heuristic" as well which can act as a "powerful prob-
lem solving machinery" (Lakatos 1978: 4). Faced with the inherent
inability to achieve more than "a systematic experience of nature,"
6. Notice that this argument differs from that of Popper, who suggested that the
theory of evolution is a metaphysical theory because it does not qualify as a science,
while the argument here is that all scientific theories are, in reality, metaphysical.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
this is a remarkably rational system, keeping the program in constant

flux or shift and endowing scientific behavior with its specific flavor
or hallmark, i.e., "a certain scepticism even towards one's most cher-
ished theories" (Lakatos 1978: 1). Lakatos, however, goes further in
his analysis of the rationality of research programs, asserting that
"problemshift is progressive by the degree to which the series of theo-
ries leads us to the discovery of novel facts" (Lakatos 1970: 118).
The two laws of Mendelian genetics, of the segregation of traits
and of the independence of different segregating traits, would proba-
bly qualify as the "hard core" of the (transmission) genetic program.
Around this core, there are vast protective belts so that no experiment
and no observation can refute the laws. (Cf., for example, Castle's at-
tack on the noncontamination of coloration-factors in hybrid rats, in
the first decade of modern genetics [Carlson 1966; Falk 1986] and its
rejection by Muller on the ground that "it violates one of the most fun-
damental principles of genetics-the nonmixing of factors-in order
to support a violation of another fundamental principle, the constancy
of factors" [Muller 1914:573]). Similarly, its powerful problem-solving
machinery allows it to change deviant results from refuting evidence
into an important progressive frameshift. Thus, the irregularities in
the law of independent segregation of markers (Mendel's second law),
which geneticists already encountered in the first decade of the cen-
tury, did not lead to the refutation of Mendel's principles. On the con-
trary, they stimulated investigators to find an explanation within the
framework of Mendel's theory and finally culminated in the theory of
the arrangement of the genes in linear "linkage maps," isosequential
with loci on the chromosomes in the cell nuclei.
Today it is clear that there is no exhaustive and exclusive definition
of scientific hypotheses or theories (i.e., a definition that includes all
scientific theories on the one hand and excludes all nonscientific ones
on the other). But this does not mean that the construction of scientific
theories is an irrational act. Once the limitations are realized, the way
is opened to cope rationally, i.e., to adopt the most efficient way to
achieve the goal with the knowledge available. Thus, we may list the
conditions for defining a theory as "scientific." In a recent talk, Carl
Hempel suggested a list of such "desiderata":7
1. A scientific theory should be consistent not only with data within
its own framework but also with those observations obtained
within the framework of other theories and hypotheses.
theory should accommodate, or at least not be contradictory
"adjacent" scientific theories. This is the Principle of Coherenc
7. In his talk at the Technische Universitat in Berlin in the fall of 1983, He

specified five or six "desiderata;" I have condensed them into three.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
2. A scientific theory should be rich, i.e. it should be able to suggest

new observations or experiments and accommodate connections
and relations that have not been envisaged before (or that could
not be explained properly in the past). It must be able to offer
new theoretical considerations and it should be consequential in
its explanatory power. This is the Principle of Fertility.
3. A scientific theory should be simple and have an "aesthetic" struc-
ture. As an example of the importance of this principle, which
might seem "nonscientific" in character, in a series of experi-
ments, the data suggest a relation between a factor x and an effect
y, as shown in the following representation:
x x
x
It would be natural to propose a theory that suggested a simple line

extending through all these points, thus:
However, objectively, that theory is as justified as another that con-

nects the points like this:

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
Yet, reasonable people prefer the simple graph theory, even though
there is nothing in the data to indicate logically that, on more detailed
observation, the more complex theory would turn out to be "correct."
This is the Principle of Simplicity.
Physics versus Biology

In 1623, Galileo Galilei claimed: "The book of nature is written in
the language of mathematics." Although an illusion, Galileo's claim
that the laws of nature could all be formulated as mathematical for-
mulae proved to be fruitful. It certainly had a decisive effect on
unique status that physics has acquired among the scientific di
plines. Biologists and philosophers of science may be divided even n
into those who claim that biological research is simply an extension
the model of research in physics to living organisms and those wh
claim that biology is unique and that "evolution rescues the uniquen
of biology" (see Grene 1974a).
The systems that a physicist studies are often well defined and rather
isolated. In such "semiisolated" systems, it is often possible to identify a
limited number of factors which affect the system causally. Hence,
fate of the system can be predicted with a reasonably high degree
confidence once the quantitative values of the initial conditions and
rules governing these factors are explicated. This is essentially wha
formally presented by Hempel and Oppenheim in their nomologic
deductive model of explanation (Hempel 1965). Strictly speaking, t
formulation holds only in closed systems. While systems studied b
the physicist may approximate this situation often enough to negle
the incomplete isolation, this is not the case in most studies in the l
sciences and in the social sciences.
Some time ago the BBC broadcast a scientific program dramatically
illustrating the problem of the precision of prediction by a comparison

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
of predictions made in astronomy, meteorology and the social sciences.

The classical system of physical research is that of the motion of the
planets which may be considered as completely closed and isolated as
feasible. Indeed, in this system, given the initial conditions, predictions
of extremely high accuracy are made: the position of the planets can
be predicted accurately not only tomorrow or in a year but also at a
given date tens of thousands or millions of years in the future. Even
though there are factors of uncertainty and error, due to the action of
lesser-known factors and probably of some unknown ones as well, over
short periods, their effects are negligible for all practical purposes. Yet,
the further we go the more inaccurate our predictions become. If we
go far enough, the result of the cumulative effect of these "negligible"
factors one day, perhaps tens of millions of years from now, will be
that our prediction about the position of the planets is no better than
pure guess work. Such a point, where the value of a prediction is no
better than pure guess, exists in every system and, in most cases, we
do not have to go tens of millions of years to achieve it.
For example, the number of factors significantly determining the
weather is higher than that of the movement of the planets. I.e., the
weather system is less closed than that of the planets. Still, it is pos-
sible to give an accurate description of several initial conditions and
the rules for combining these are reasonably well known, so that we
can predict the weather in a couple of hours or even in a day or
two with some accuracy. If, however, we were to try to predict the
weather for longer periods, we would find that we had to increase the
number of initial components in our formulae at an alarming rate.
More and more factors that are negligible in short term predictions
assume significance as the range of predictions grows longer. It has
been estimated that, to make a meaningful weather forecast eighteen
days ahead, every wing-beat of every lark hovering over the ocean
should be considered.
When an economist offers politicians a new program based on th
initial conditions known when he prepared it, it turns out that the
time necessary to compute the predictions or the time necessary to
read the report (or both) is usually longer than the viable life span
of these predictions. (No wonder politicians often mistrust their eco
nomic advisors.) A similar situation exists in many biological system
The number of nonnegligible factors that may affect a biological sy
tem is so enormous that, even in short time ranges, it is often impo
sible to make very useful predictions. In systems where it is possib
to draw a fairly clearcut demarcation line between a few significan
components of the system and all the rest (this is the rationale in th
design of many experimental situations), the latter may all be lumpe
together under the ceteris paribus clause as reasonably constant back

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
ground affecting the results only slightly. When, however, the relia-
bility of the initial conditions is not very high or when not enough
meaningful initial conditions are available, obviously, the possibility of
arriving at reliable predictions is also low.8
Thus we face a continuum of explanations, from those for which
the assumption of a closed system holds rather well within wide limits,
like astronomic ones, to those in the social and life sciences which
are so "open" that it is practically impossible to describe their initial
conditions adequately. Furthermore, there are good reasons to believe
that, in many of these open systems, a full description of the initial
conditions is not even possible since, in principle, the effect of so-called
undeterministic factors must be considered.
In such circumstances, it seems that, in open systems, a priori dec
sions on what are "main factors" and what are "negligible" ones often
depend on the subjective relative values of the investigators. The best
that a rational scientist can do is to be aware of this and to constantl
remain alert to the possibility of change in the weights given differen
factors in the research program, according to the degree of success o
the initial predictions. Such a procedure may increase the degree of
objectivity of the weights allocated to various factors but these are als
relative decisions for which no stringent procedural prescriptions can
be given, and whose success in achieving the desired objectivity is no
guaranteed.
The description of scientists as people who pursue their work with-
out a shade of prejudice or bias not only errs as a description of scien-
tists and their society but also does a great disservice to the successful
development of research either in the social sciences, the life sciences
or physics. No doubt, the repeated emphasis on the demands for ob-
jectivity, for strict and logical analysis of data, for the performance
of controls and repeated attempts to verify the results is extremely
important. But to insist on these demands without also stressing the
impossibility of achieving these noble goals is bound to lead to dog-
matic and regressive research and to create conceited people. Hence,
I cannot accept the claim that the public image of scientists, still widely
current, as "rational, open minded investigators proceeding methodi-
cally, grounded incontrovertibly in the outcome of controlled experi-
ments, and seeking objectively for the truth, let the chips fall where
8. In a talk at the Freie Universitat in Berlin in the spring of 1984, Hempel

introduced the term "proviso" to denote the factors and conditions that must be
considered in addition to the theory to allow the bridge between the initial data and
the consequences of an experiment or series of observations. These provisos are
not part of the theory and cannot be inferred from or correlated with it. They are
the reality that we have to surrender to, rather than the "ceteris paribus" clause,
since "every thing else being equal" is not a realistic situation.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
they may" (Brush 1974: 1164), is an illusion vital for scientific work.
As with every illusion, it is a distraction from reality. As such, it is
dangerous for science as well as for the status of scientists in society.
The days of the heroic stories of the life and deeds of scientists like
those in Paul DeKruif's Microbe Hunters may be over but the images
are still with us. In 1974, the title of a paper in Science magazine asked
"Should the history of science be rated X?" (Brush 1974). Considering
the stringent standards that many people expected from a scientist,
the author asks ironically if most modern studies of the life histories of
prominent scientists should not be purged at least from the curricula
of future scientists.
As one illustration of this, the paper cites the case of an eminent
scientist who, in spite of his better "objective judgment," rationally
applied his "subjective intuitions." During the 1850s, the laws of pres-
ervation of energy and thermodynamics were formulated but scien-
tists were not sure if one could neglect the "ether-theory" in favor
of the new, fashionable "kinetic-theory." When James Clerk Maxwell
(1831-1879) wrote his first paper on the kinetic theory of the gases, he
was quite cautious and for good reason: It turned out that two of the
deductions from the kinetic theory conflicted with known experimen-
tal facts about gases. Concerning one of these deductions, Maxwell
boldly told the British Association for the Advancement of Science
in 1860: "This result of the dynamical theory, being at variance with
experiment, overturns the whole hypothesis, however satisfactory the
other results may be" (Brush 1974: 1169). Fortunately, in spite of his
declaration, Maxwell ignored the "critical experiment" and continued
to develop the new kinetic theory and inspired others to follow him.
Later, when a new interpretation of the old results was advanced, the
discrepancy between "facts" and theory disappeared ("Facts," too, are
theory dependent!). Even though the other discrepancy could not be
brushed aside so easily, Maxwell refused to abandon his simple, co-
herent theory, which agreed with so many observations and suggested
explanations and meaningful connections not seen before for one or
a couple of contradictory results.
No one can honestly respect all the demands made of the perfect,
utopian scientist. Taking such demands seriously must lead investi-
gators to despair or to bigotry and, in fact, the demands themselves
change constantly since they are socially determined by the commu-
nity and are context dependent. Indeed, we do expect a scientist to
use his "intuition" and thus we expose our flanks to subjectivity; but
this cannot be helped. The rational response is to be aware of the
situation and to try to build alarm systems and controls so that, in the
long run, unavoidable mistakes will cancel each other out. Paradoxical

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
as it may sound, the most rational thing we can do is to admit that we

are irrational.
Darwin's Theory of Evolution as a Scientific Theory

If we examine Darwin's theory of evolution by natural selection in ligh
of what has been discussed thus far, it becomes clear that it is a most
successful scientific theory. There are not many theories that can boas
of their degree of coherence, of their success in explaining so many
apparently independent phenomena, of correlating observations and
theories from so many different and varied aspects of the sciences
and of corroborating them into one whole, as does Darwin's theory
of evolution. And all this is based on only a few astonishingly simpl
core-principles:
-The principle of the production of inherited variation.

-The principle of natural selection.
I do not intend to ignore the difficulties that these core-principles pos

and I will discuss them later. But this is a long way from a claim like
Popper's that the theory of evolution cannot be considered a scientifi
theory. It may not conform to a principle that, for a theory to be
considered "scientific," it should include an explicit specification of
the conditions or situations which, if applicable, would falsify it. As
we have seen, however, there are scarcely any scientific theories that
actually fulfill this criterion. Quite the contrary, the better establishe
and entrenched a theory is, the less such a criterion applies to it (se
Lakatos 1970). Furthermore, I would suggest that the criterion o
falsification would never have been invented if life sciences, rather
than physical sciences, were the model for scientific theories. Biologica
systems, which are only rarely closed enough to specify a finite number
of factors that provide adequate initial conditions and laws that govern
them, would never have suggested such a criterion.
Much of the achievement of biological research, beginning with the
middle of the last century, stems from biologists' skills in the study
of isolated parts of living beings. Under such conditions these parts
could be treated as more or less closed systems with the methodologie
of chemistry and physics. The flourishing of disciplines like physi-
ology depended on the possibility of isolating whole organs from livin
creatures (presumably without disturbing their function), and later on
the possibility of studying living isolated cells under the microscope
Modern developments in biochemistry and later in molecular biology
also depended, to a large extent, on the possibility of isolating and
studying, as more or less closed systems, not only cells but even cell
components.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
Still, we must admit that adopting the mechano-reductionist frame-

work in problems of biology and especially of evolution often leads
to an impasse. One of the hallmarks of living systems is that they are
organized complexes. Disentangling this complexity is one thing but dis-
regarding the effects of organization derived from it is another. In
physics, we found that we can succeed in reducing the problem to a
few components and concentrating explanatory efforts on these, at
least as long as we do not sail to distances of millions of years, or at
speeds approaching that of light. In biology, we encounter the limita-
tions of this approach at every step. The need to refer to a problem on
the level of the whole organism or the whole population of organisms
is as essential to the elucidation of biological organization as that of
the levels of the cell, the gene or the atoms of the DNA molecule.
Organized complexity imposes constraints which may be due to
architectural or functional limitations in the development of individu-
als (Webster and Goodwin 1982) or to events that have had a causal
history in the species (Gould 1977a,b). Gould and Lewontin (1979)
illustrated the danger of missing the significance of the data when
there is too much eagerness to reduce one level of organization to
another. The ornaments adorning some of the corners of the great
cathedral of San Marco in Venice, corners formed by the architectonic
constraints of building a round cupola on a square structure, are so
impressive and attain such levels of artistic completeness that it seems
as if the causal sequence of events is reversed and that the corners
were designed specifically for the ornaments rather than vice-versa.
Similarly, in our reductionist enthusiasm, we may often see the whole
as nothing but the sum of those details that in fact we only constituted
in our minds. Quine (1969) discussed the ontological relativity of our
terms in translating native primitive languages: "This much transla-
tion can be objective, however exotic the tribe. It recognizes the native
expression as in effect a rabbit-heralding sentence. But the linguist's
bold further step, in which he imposes his own object positing pattern
without special warrant, is taken when he equates the native expres-
sion or any part of it with the term 'rabbit' . . . Given that a native
sentence is true when and only when a rabbit is present, it by no means
follows that the so-and-so are rabbits. They might be all the various
temporal segments of rabbits. They might be all the integral or unde-
tected parts of rabbits. In order to decide among these alternatives we
need to be able to ask more than whether so-and-so is present" (1969:
2). If there were a "language of nature," to translate the terms of that
language properly would be a similar, though much more difficult task
than translating a tribe's language.
The declared aim of physical research is to find and formulate uni-
versal laws whose extension covers all bodies in the universe (living or

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
not). Biologists have more modest goals; all they want is to explain the
phenomena of life and life, as we know, constitutes only a tiny part of
the universe. Does this call for a "uniqueness of biology" as a research
discipline?
Even if it were feasible to describe life by writing a long list of
formulae for the time-space coordinates of each molecule in a living
organism (as well as those molecules surrounding the creature and
participating "meaningfully" in its metabolism) and further to specify
all the rules and laws for causal interactions between them, this would
be utterly uninteresting (except perhaps for the final demonstration
that the laws of physics and chemistry are valid for the components of
living organisms as well as for the nonliving material).
It is futile to force a method of solving problems that worked for
one type of question on another group of problems. One of the basic
premises for a scientist's success is his ability to be judicious, not only
in the selection of the goals of the investigation but also in the method
of research to ensure attention to those factors that are relevant for
him or her.
-There is no sense discussing a problem at the level of the com-

ponents of atomic nuclei when the issue is the functioning of a
computer.
- There is no sense discussing a problem at the chemico-physical
level of cell structure when the issue is the functioning of the organ-
ism as a whole.
-There is no sense discussing the details of the functioning of a

vehicle's engine when the issue is the flow of traffic in rush hour.
- There is no sense discussing the details of the structure of the
alimentary canal of some animal in a population of living things
when the issue is the dynamics of penetration of a population into
a new ecological feeding niche.
Scientific research may thus be envisaged as consisting of a hierarchy

of levels of investigation, beginning with the most basic and universa
and rising to more complex and specific levels. Obviously the laws and
rules of the lower, more comprehensive level also apply to the higher
and more specific level. The laws of atomic physics also apply to the
level of chemical interactions and both the laws of atomic physics and
of intermolecular chemistry apply on the level of the physiology of the
living cell as well as on the level of the behavior of organisms. And o
the level of the dynamics of populations, nothing occurs to defy th
laws of the lower levels.
But what about the other way round-how far can the laws of the
higher level be reduced to those of the levels below? Or, in other
words, to what extent do new laws emerge in a shift from a more

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
generalized to a more specific level? This much discussed problem of

reduction and emergence appears to consist of two problems which I
shall call ontological reduction and epistemological "reduction."
On the ontological scale, the entities of the lower level are also
present on the higher levels. Neutrons and protons also apply on the
level of atoms as well as on that of molecules; molecules constitute
parts of cells and cells are parts of organisms, which, in turn, compose
populations. The rules concerning ontological entities of one level may
be completely reduced to those of the level below and those of the up-
per level could be formulated on the basis of those of the lower level.
In Waddington's words: "Throughout its history biology has found
itself faced with a dilemma as to how far its reliance on physics and
chemistry should be pushed.... It was Whitehead who [pointed out]
that ... this ancient dilemma arises essentially from seeing the situa-
tion upside down. It is not the case that we begin by knowing all about
the ultimate constituents of the inorganic world, and can then ask
whether they can account for the observable phenomena of biology.
Always ... it is from observable phenomena that we have to start....
When [a new] compound is formed, it is not that some new 'emergent'
properties appear; it is simply that a new avenue is open to us for
discovering a little more about [its constituents]" (1963: 19-20).
What life scientists (and social scientists) mean when they insist that
biology cannot be reduced to physics and chemistry (and psychology to
biology) is "reduction" on the epistemological scale. The rules govern-
ing the coding of information in the DNA strands of the chromosomes
of living cells do not contradict the laws of physics and chemistry.
They are irrelevant. The information coding problem is insensitive to
whether the chemistry of information transfer happens to be DNA or
silicon chips. Similarly, the rules of dynamics of a population entering
a new ecological niche are basically refractive to the question of which
organisms are concerned. The difference between the levels is an epis-
temological one, the way we organize the data in our minds, the way
we constitute the world, not the organization of the real world. We
may "look at the world" from a different "point of view" and it is not
necessary for one viewpoint to be reducible to another. On the con-
trary, as long as ontological reducibility is maintained, as long as entities
present at a lower level are still there when we are dealing with the
higher level, it is just this independence of epistemological reducibility
that makes the notion of emergence possible.
Hierarchical structures are not confined to relations within biology
or between biology and chemistry and physics, as illustrated, for ex-
ample, by the relation between the structure of computers and their
functioning or that of the thermodynamics of a vehicle's combustion
engine and the dynamics of traffic flow.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
Folk ? Darwin's Theory of Evolution 81
The mechano-reductionist approach, which proposed that the struc-

ture of a complicated system and its function may be understood by
knowing its components and their causal relations, was extremely suc-
cessful not only in physiology, biochemistry and molecular-biology. It
also became one of the basic strategies of Mendelian (transmission)
genetics. Since the heredity of the whole organism was too compli-
cated to follow, attention focuses on single traits or functions that
were meaningful to the investigators and ignored the structure and
function of all others. There is, however, a significant methodological
difference between the reductionist approach introduced by the logic
of physico-chemical sciences and that of the biologists. The first en-
deavors to hold constant as many of the irrelevant variables as possible
or assumes that their contribution to the phenomenon under study
is negligible. The latter does not make such assumptions. Well aware
of the impossibility of holding "background variation" constant in a
biological system, the strategy is often to increase variability to the
maximum and thus to find (often with the aid of statistical methods)
the common denominator to which the system can be reduced, in spite
of the variability contributed by other factors. Gregor Mendel con-
structed a classification criterion, a "trait," like seed color, by which
he could group the peas into two categories, e.g., plants with yellow
seeds versus those with green seeds, disregarding all other character-
istics of these plants. When this reductionist approach was extended,
for example, to the categorization of abnormal forms and diseases,
it juxtaposed attention to the complementary categories, those of the
standard, normal, healthy individual. Thus it was the deviation or dis-
ease that became a key to the understanding of the normal and the
healthy in biological research.
But the dazzling successes of the reductionist approach led many
biologists and nonbiologists alike to conclude that, ultimately, it would
be possible to reduce biological research to physical research. This
attitude has sometimes been explicitly expressed in such assertions as
that there was no good biological research apart from physics. Such
assertions can only be due to confusing ontological reductionism with
epistemological "reductionism." While the former holds in biology just
as in physics, the latter need not hold in either.
We may be justified in concluding that there is no reason to see
the different levels of scientific research as anything but our way of
concentrating on given aspects in the continuum of the phenomena.
Just as there is no rigid and principle demarcation between biology
and physics, so there is no place for demarcation in biology at various
levels of research notwithstanding the great value attached to viewing
entities from different angles and emphasizing different aspects on
different occasions.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
The theory of evolution is strongly dependent on reductionist re-

search within biology and the adjacent sciences but is itself the oppo-
site of a reductionist science. When we refer to a unique event, like
hereditary change, a mutation, it may be reduced to a chemical change
in a given atom of a given base at a given site on the DNA molecule or
it may be viewed synthetically within the framework of the theory of
evolution, as a causal factor, at ever higher levels of organization, of
the survival chances of the individual gene, the cell, the whole organ-
ism and, finally, the interbreeding population. The theory of evolution
is essentially a synthetic rather than an analytic theory.
One of the characteristics of a scientific theory is the ability to per-
ceive generalizations. A scientific explanation is not interesting when
it is directed to a specific event, unless that event may be viewed as one
of many, as an instantiation of a generality. Science is interested only
in explanations that can be tested, i.e., in obtaining the same results
repeating the same initial conditions. To be convinced and to be con-
vincing that we are not dealing with miracles, an experimenter repeats
his experiment several times to demonstrate that the result was not a
unique event but one of many.
Here, it is often claimed, a difficulty exists with the theory of evo-
lution: if one assumes that the combination of initial conditions is so
complex that it cannot be repeated, how can tile explanation be tested?
But even here, the difference between disciplines is only one of
degree. A distinction between those aspects that may be called the
ascertainment of the historical "facts" of the changes in life on earth
and the aspects that endeavor to explain the mechanisms that pro-
duce these changes may be useful. For historical reconstruction, an
historical methodology is appropriate, even though history normally
extends only over hundreds or thousands of years and evolution is
concerned with millions and hundreds of millions of years. On the
other hand, the methods for investigating the mechanisms that could
lead to these evolutionary changes are mainly experimental and are
often performed even under semi-controlled laboratory conditions.
The experimenter in the theory of evolution investigates the factors
leading to hereditary variation or to differential survival, as general-
izations, cases that hold beyond the limits of any given species. In this
respect, he is no different from the physicist who studies the free fall
of bodies, all free falling bodies, not specific ones.
However, when the experimental approach of reasonable control on
initial conditions was not possible in biological research, as in physical
research, other methodologies had to be developed. At the beginning
of the twentieth century, the philosopher C. S. Peirce indicated that
the theory of evolution does not allow predictions in the sense of New-
tonian scientific methodology. There is no possibility of predicting the

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
fate of a given individual. But individual organisms or even individual

species are not the objects of the theory. In this sense, the theory
of evolution by natural selection is no different from the theory of
gases established at the same time. The latter was also statistical in
that it allowed only predictions concerning a great number of parti-
cles (atoms, molecules) over a relatively long time span. Similarly, the
theory of evolution allowed predictions concerning whole populations
or even species of living organisms over very long periods.
In physical science, as in the biological sciences, we pay for the
demand for generalizations with the impossibility of making precise
predictions about specific events.
Is an Explanation also a Prediction?

Logically, there is symmetry between explanation and prediction, as
can be seen from the Hempel-Oppenheim model:
-from basic principles or laws

-and known observations and data
-a conclusion may be deduced
When all details above the line are given, the conclusion belo
line can be predicted. When two systems of data or observatio
known, one above the line and the other below it, specifyin
connecting the two systems so that one set of data could be de
from the other provides an explanation.
Strictly speaking, the deductive prediction presented here is
possible when the basic laws and the initial conditions are not
necessary but also sufficient. And it has been amply demonstrat
this is possible only rarely. De facto, unequivocal conclusions f
basic laws and initial conditions can be deduced only in very si
situations in physics and in biology. In many biological situation
best we can do is to express our initial conditions in probabilistic
Obviously, the conclusions would also have to be only in proba
terms. The need to be satisfied with statistical predictions is esp
compelling in biological research since, often, some of the "ini
conditions are simply impossible to give. As we noted earlier, fo
theory of evolution, the data are often stochastic and the stochastic
nature of initial conditions makes the conclusions contingent rather
than logical necessities, thus relieving us from the uneasy conclusion
of a deterministic sequence of events (Grene 1974b).
As for a posteriori explanations, i.e., explanations that reconcile
sets of data or observations already available to the investigator, these,
it may be claimed, should be most suspect. With post factum expla-
nations, everything can be explained and explanations that explain

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
everything and eliminate every difficulty do not leave room for com-
parisons and controls and thus render the research barren, preventing
any further investigation.
Even though, logically, it makes no difference if a specific event
occurred in the past or is about to occur in the future, we see it differ-
ently. To explain events in the past, we can know that, of all possible
contingencies, only one occurred. When, however, we consider a fu-
ture event, we may at best enumerate all the possible contingencies
and perhaps also attach relative probabilities to each. The situation
is like asking an expert about the probability of winning the lottery.
Applying the basic laws of probability theory and detailed initial data
(the number of tickets sold, the number of tickets I've bought, the
policy of the distribution of the prizes, the honesty of the people in-
volved, etc.), the expert will be able to calculate the probability of my
winning. If it was one in fifty million, the rational prediction would
be that my chances of winning are practically nil. If, next morning I
were to come to the statistician and tell him that I did win the prize, it
would be ridiculous for him to explain to me that my winning is highly
improbable since the chances were one in fifty million. All that can be
"explained" with great accuracy is that my win is one of fifty million
such events, each of which has a probability, tiny as it may be but still
more than zero-i.e., it could have happened.
An illuminating demonstration of such a confusion between an ex-
planation and a prediction is the criticism often raised against the
claim that the origin of living matter is from nonliving matter. It is
not necessary to go into details to be convinced that the probability of
such an event is extremely low. If somebody (something?) had been
asked a priori to state the chances that living matter as we know it was
produced from nonliving matter on our planet, the answer would be
that, beyond any reasonable doubt, there is no chance of such an event
occurring. Yet, once we know that this improbable event did occur (at
least according to the explanation given by many investigators), the
claim that the a priori probability that such an event would occur is
negligible does not hold.
As for the claim that a posteriori explanations lead to "explanations"
of practically any set of data and observations by appropriate adjust-
ments within the theory, no matter what the "real" situation was, it
should be asked what the "real" situation was. As we already noted, this
is a question that a scientific theory does not at all pretend to attend to.
All it strives for is to increase both the coherence of the observations
and our capacity to explain them in a way that might suggest new and
fruitful theories. Not every theory does this and, thus, there are not
many theories that do it better than Darwin's theory of evolution. And
if we extend predictions not only to single experiments but also to
research programs, as should actually be done (see Lakatos 1970), we

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
may agree with Jacques Monod's statement that "the predictive power
of Darwin's theory of evolution" is unique in its efficiency. At least
on two occasions did Darwin's theory "predict" or anticipate major
developments of new theories, which both considerably increased our
understanding of nature (Monod 1975).
Shortly after Darwin published his book on the origin of species,
he had to confront one of the most eminent physicists of his time.
Lord Kelvin, the founder of modern thermodynamics, calculated that
our planet could not be more than twenty-five million years old. He
assumed that the sun was a huge coal furnace and when he applied
the most liberal estimates available to him to the prevalent theories, he
concluded that the amount of coal in the sun could not suffice for a pe-
riod exceeding twenty-five million years. Darwin was left with an acute
feeling of failure, since it was obvious to him that, under these cir-
cumstances, the time for evolution as he envisioned it was insufficient.
Monod suggested that Darwin inadvertently "predicted" the inade-
quacy of Kelvin's explanation as well as the formulation of a new physi-
cal theory whose explanatory power would supercede the theory of the
preservation of energy. It can be argued that Darwin's theory actually
included premises that would have refuted Einstein's theory if they
had not been answered by it, as Popper demanded of scientific theo-
ries. Eventually, the falsifying argument was turned into an impressive
corroboration.
Similarly, Darwin "predicted" or anticipated the theory of particu-

late entities of inheritance. Already in 1867, the mathematician Fleem
ing Jenkins demonstrated that Darwin's theory collapsed if it were
built on the basis of Darwin's own theory of heredity. Jenkins indicated
that a precondition for Darwin's theory of evolution was the existenc
of a theory of inheritance of discrete particles that keep their individu-
ality and hereditary potential from one generation to the other. Th
Mendelian theory of inheritance is such a theory. Formulated in 1865
and rediscovered only in 1900, it eventually "saved" the Darwinia
theory of evolution from refutation.
These examples testify more to the fertility of Darwin's theory than
to its predictive power as formulated by the hypothetico-deductive
model. But, as discussed here, the predictive value of a theory, as wel
as its explanatory value, can be evaluated only in relative terms of its
coherence, fertility and simplicity. If we refer to the ability of a theory
to explain new connections that have not been foreseen, there are few
theories that would compete with Darwin's theory of evolution.
Functional Explanation
A scientific explanation is largely a search for "causes." The assump-
tion that there is an antecedent cause (or causes) for every event is
fundamental to scientific explanation. Modern physics started when

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
scientists got rid of explanations that argued that the stone falls down
"because the stone strives for the ground." While such teleological ex-
planations were by and large eliminated from science, it is claimed
that such explanations, or at least one kind-functional explanations
-are still common in biology. A functional explanation refers to the
purposes that a thing or a process has, rather than to its causes: "The
function of chlorophyll in plants is to enable them to perform photo-
synthesis (or to assimilate carbon-dioxide and water to sugars)"; or,
"The function of the blood is to provide oxygen to the tissues" and
similar phrases that refer to the goals rather than to the causes. The
term "goal" as a premeditated intention is foreign to a system lacking
cognition or conscience; it derives from our tendency for introspec-
tion. Being conscious and cognitive creatures, we tend to impose our
concepts on nature. "To 'explain' the presence of the heart in ver-
tebrates by appeal to what the heart does is to 'explain' its presence
by appeal to factors that are causally irrelevant to its presence ... to
explain why it is there, why such a thing exists in the place (system,
context) it does-this does require specifying factors that causally de-
termine the appearance of that structure or process" (Cummins 1975:
745). Barring the possibility of the intervention of an all-knowing and
all-powerful god, we cannot assume a mechanism that intentionally
directs or chooses the necessary means to achieve the goal.
This is not to say that there are no goal-directed processes. As
we have indicated previously, components of a system often interact
in an autoregulative manner which may be described as functioning
in a "goal-directed" fashion (Nagel 1977). The homing torpedo and
the steam-engine governor are well-known examples. The causal fac-
tor that designed the goal-directedness in these cases stems, however,
from the will and intentions of their god-the human designer. Living
organisms are even more complex interacting entities, exhibiting goal-
directed structures and functions, i.e., autoregulation, in their embry-
onic development and physiology which may be explained causally.
The question is, however: "How did it come to be, how did it arise?"
(Grene 1974a: 210). Kant already saw this very clearly, as Lenoir
(1981) has shown. Kant realized that "reciprocal effects due to the
dynamic interaction of matter" were not limited to the organic world.
But, whereas in the inorganic world the phenomena could often be
simplified so that they were "capable to be analyzed in some way as
a linear combination of causes and effects, A -> B -> C etc.," (p.301)
this was not the case in the organic realm. In the organic world, "cause
and effect are so mutually interdependent that it is impossible to think
of one without the other, so that instead of a linear series it is much
more appropriate to think of a sort of circular series A -> B -> C -- A.
This is a teleological model of explanation, for it involves the notion of

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
a 'final cause'" (Lenoir 1981: 301). Blumenbach and his successors in

nineteenth-century biology had to rely on a "Bildungskraft" to meet
these difficulties, even though they apparently realized that the dif-
ference between the inorganic and organic world was more of degree
than of principle. As Lenoir (1981) puts it, they realized that, even
in the inorganic world, systems were only "capable to be analyzed in
some way as linear combinations of causes and effects." Anyhow, theirs
was a "Teleology Without Regrets" (Lenoir 1981: 293) and, applying
Darwin's theory of evolution, we may even be relieved of the need for
such a regulative concept as a "creative force."
If we ask a biologist who says that "the function of the heart is
to provide oxygen to the tissues" what he means, he will no doubt
claim that the phrase is not intended to express anything about the
idea that the heart exists or was designed to provide oxygen. He will
probably claim that all he wanted to say was that a complex system
like a vertebrate can function properly only when there are efficient
means for oxygen supply to the tissues and that the heart does that
efficiently. This statement is not a causal-teleological explanation but
a different point of view of a problem. A teleological statement of
this kind may have a heuristic value or a descriptive importance but
not an explanatory meaning (Grene 1974a). The biologist needs the
information that such an organ exists that provides defacto for oxygen
supply to the tissues as a starting point for other questions. Therefore,
he is usually content to express this starting point with a concise "in-
strumental teleology" or functional phrase. This is fine as long as he
does not think of his statement as an answer to the question "Why is
X there?" but only as a statement of an observational or experimental
fact, i.e., the system would collapse (die) if X, doing something causally
(functions properly), were not there. Hence, Nagel (1968: 405) appro-
priately called the teleological statement a telescoped argument: "The
difference between a teleological explanation and its equivalent non-
teleological formulation is thus comparable to the difference between
saying that Y is an effect of X, and saying that X is a cause or condition
of Y. In brief, the difference is one of selective attention, rather than
of asserted content."
Yet the functional statement also maintains important regulative

value (Grene 1974a). From the assertion that oxygen supply for the
vertebrate's tissues was needed, no "heart" can be deduced logically.
An artificial pump may serve (though not too efficiently, as yet) instead
of a flesh-and-blood heart. For millions of years, insects have been
provided with an efficient alternative for oxygen supply to the tissues
in the form of a system of tracheae traversing all their organs. Thus,
the heart is a sufficient condition for the supply of oxygen to the tis-
sues (provided other conditions have been fulfilled), but it is certainly

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
not a necessary solution to the problem. "We do reason from the per-
formance of a function to the presence of certain specific processes
and structures ... it is a species of inference to the best explanation
[but] inference to an explanation has been mistaken for an explanation
itself" (Cummins 1975: 748).
To turn a functional necessity from a mere causally sufficient ex-
planation into a necessary explanation as well, we need something
else which could be another argument, independent of the one given,
that would convince us that, in the given circumstances, the functional
explanation was not only a possible explanation but the only possible
explanation.
This is exactly what the theory of evolution does. It gives us the
"historic factor" that enables us to demonstrate that, under the given
circumstances, our functional explanation was, if not the only possi-
ble one, at least the most probable or the optimal explanation. When
we specified in the functional statement that we were talking about
a vertebrate, we mentioned the existence of a complex and compli-
cated system of interactions that have actually been performed for
millions of years, each upon the previous ones. Thus, they left only
a very narrow and limited range of possibilities and constraints for a
satisfactory answer to the demand of efficient oxygen supply to the tis-
sues. Once we specify that we are dealing with certain organisms, i.e.,
with vertebrates whose ancestors were provided with something that
could act as a "potential heart" and not with a system that could func-
tion as tracheae, we assert, with our functional statement, that a heart
was not only a sufficient answer but, under the circumstances, also a
necessary answer to the need for oxygen supply to the tissues. Once
the Darwinian theory of evolution was formulated, there was no place
anymore for a teleological explanation. It is not that "the function (or
purpose) of the heart is to supply oxygen to the tissues" but rather:
"since vertebrates are equipped with a heart, it serves (or functions)
to supply oxygen to the tissues." As for the historical constraints, we
conclude, albeit post factum, that there was only one alternative. Had
this alternative not been utilized, the creatures we discussed would
not have survived. Evolution is not an a priori necessity. "The privilege
of living beings is not to evolve but on the contrary to conserve....
[However,] the privilege of living beings is the possession of a struc-
ture and of a mechanism which ensures two things: (i) reproduction
true to type of the structure itself, and (ii) reproduction equally true
to type, of any accident that occurs in the structure" (Monod 1973; see
also footnote 4).
In fact, taking this interpretation of the theory of evolution strictly,
"it apparently forces us to interpret the whole history of life in a

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
thoroughly deterministic manner" (Grene 1974b: 231-232). Consid-

ering, however, the conditions given at each step of evolution, the
necessity of the consequences was a physical rather than a logical ne-
cessity. Furthermore, as a rule, it is not the case that only one set of
given conditions is open at each evolutionary step. There may be many
but the crucial point is that they vary stochastically in relation to the
causal processes (Grene 1974b).
The Fit Survives or Survives Who Fits?

When a researcher of evolution claims that, in the course of natural
selection, the fitter individual will survive, how is "the fitter" defined?
Is it not that individual that de facto survived who is by definition the
fitter one? The very fact that it survived implies that he was fit. Thus,
presumably, as long as we do not know a priori who will survive (or
what the a priori probability of the survival of an individual is), we
cannot really define "fitter." It has been indicated that lacking a better
definition, the "survival of the fitter" really means simply "the survival
of the survivor." We could argue that "the survival of the fitter" is
axiomatic. There is nothing wrong with a scientific theory based on
axioms: Newton's theory is based on axioms and is still very produc-
tive. However, even if such a solution avoids the difficulty, it does not
eliminate it. Physicists seem to be less disturbed than biologists about
the "truth" of their theories as long as they are helpful and interesting.
As is well known, there is an old dispute about whether a circularity
similar to that presented above does not also exist in Newton's defini-
tions. Newton's first law states that a body maintains its rest condition
or its uniform movement as long as there is no force exerted on it. On
the other hand, Newton referred to this law when he tried to define
"force" as a factor that causes the state of a body, either at rest or in
uniform movement in a straight line, to change.
In biology, we may have to rely on operational definitions of "fit-
ness." An operational definition of the fitness of an individual (or a
group of individuals) will be of relative fitness; the (mean) number
of progeny of an individual (group of individuals) which reached the
same age (developmental stage) as the individual (group of individu-
als) concerned, in relation to the corresponding numbers of progeny
of another individual (group of individuals). There is of course no
escape from circularity in this definition and it also depends on the
future development of the individual whose future we wish to define.
Darwin was a material realist who believed that his description fits
the real situation better than the descriptions that preceded it. His con-
viction was that "fitness to environmental conditions" is a real thing,
whether we are capable of identifying it or not. The circularity dif-

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
ficulty is therefore ours and not in the concept. Thus Darwin con-
structed his argument in a different way from the one that is adopted
now, arguing by analogy. He compared natural selection to artificial se
lection. In the first chapters of The Origin of Species, Darwin explains in
great detail how pigeon breeders select pigeons to maintain the stock.
The breeder knows what he wants to achieve and thus he provides the
"environmental conditions" conducive to the survival of the fitter ones
according to his criteria. The "fitter" ones are those that survive since
the breeder decided that he wanted them to survive. In nature there
is no one who makes such decisions but the case of artificial selection
is an experimental analogy of natural selection (Schweber 1977). In
artificial breeding, the breeder expresses his wishes by "changing the
environmental conditions" of the pigeon population so that only some
individuals survive, those whom the breeder is interested in. These
survive because their "fitness" is desired by the breeder. Natural selec-
tion is nothing more than the response of the population to changes in
the environment, even if nobody specifically "desired" these changes.
Certain traits, structures or behaviors are a priori preferential for life
in a new environment and will survive because they fit the given condi-
tions, even when-contrary to the situation with the pigeon breeder-
we cannot know ahead of time what characteristics the environmental
conditions would "choose" (Gould 1977a: 39-45). According to this
kind of argument, there is in principle a possibility for a definition of
"fitness" even though we still lack the criteria that allow us to do so.
A promising path to break such a circular argument and to achieve
a "real" definition is to try to form a definition outside the theoretical
framework where the problem is defined. Some years ago, the philoso-
pher and biologist Henry Atlan suggested to me that such a solution
may also be offered to the problem of the definition of the "survival of
the fitter." This solution is based on the experimental system in which
scientists endeavor to replicate in vitro the conditions for the origin of
life from nonliving matter. Alfred Eigen and his colleagues were able
to define in pure chemico-physical terms the conditions of systems in
which molecules capable of self-replication were present with limited
resources of raw materials ("food" molecules and energy-supplying
molecules) and to predict the outcome of the system (Eigen, Gardiner,
Schuster and Winkler-Oswatisch 1981). Under such limiting condi-
tions, a situation like Darwinian evolution ensues: Since the process
of self-replication of these molecules is not very accurate, a "halo" of
self-replicating molecules, similar to but not identical with the ones
originally put into the vial, is produced ("mutants"). Thus, heritable
variability has been produced and competition for resources starts
(see footnote 4). A detailed description of the initial chemical and

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
physical characters of the molecules, applying the laws of chemistry

and physics, allows deductions as to the molecules that will "repro-
duce" and with time dominate the contents of the experimental vial.
Hence, the statements provide definitions of "survival of the fitter"
independent of postfactum assertions of which one survived.
Whether we accept Eigen's system as a solution to the problem of
reconstruction of prebiological evolution or not, these experiments
provide the basis for defining "the survival of the fitter" without re-
course to circular arguments or axiomatic assumptions. Whether such
a solution is necessary to save the theory of evolution by natural selec-
tion is another matter.
Conclusions
To answer the question of whether Darwin's theory of evolution is a

scientific theory, it was first necessary to examine the wider question
of what a scientific theory is. Apparently, there is no simple answer
to this question. The answers we usually give or hear often include
inaccuracies and misconceptions rooted in our illusions about science,
rather than in what science is and how it works. Even the science of
physics, often quoted as a paradigm of scientific theory, does not c

form to the demands that both scientists and philosophers of scien
would like to confer on it. Once we concede that the best one can do
is to formulate "desiderata" for scientific theories and that we can en-
visage only rational procedures for scientific programs, it is clear that
Darwin's theory of evolution should be accepted as an appropriate
scientific theory.
Examining the characteristics of the systems studied by the physicist
and by the biologist, we argued that differences were mainly of degree.
There is no problem that can be studied in isolation as a completely
closed system. Yet, while many problems studied by the physicist may
be treated as partly isolated or semiclosed systems, this is hardly ever
possible with living systems, which must be treated as open systems. In
semiclosed systems, attention may be focused on a few main factors for
making predictions which will be confirmed with high probability and
the action of minor factors assumed to affect all components equally
(the "ceteris paribus" clause) may be disregarded. In open systems, the
possibility of defining a small number of major factors and ignoring
all other components never exists. Furthermore, it is often impossi-
ble, even in principle, fully to define the "initial conditions" of such
systems. Thus, in addition to the theory, we must always deal with
innumerable provisos that cannot be derived from the theory itself.
Such differences primarily require variations in the methodology of
research. But it must be stressed that the demarcation between meth-

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
odologies applied to open systems and to semiclosed ones (if such a

demarcation exists) does not necessarily coincide with that between
biology and physics.
Like every scientific theory worth its name, Darwin's theory of evo-
lution has never been stagnant but undergoes constant evolution, fol-
lowing the growth of knowledge both in its own sphere and in that
of adjacent spheres of science. Even though many of the difficulties
that plagued the theory in its early days have been resolved, others
still exist. Among the latter is that of functional explanations and of
the limited capacity to predict results, as compared with the theory's
ability to explain them. An attempt was made to show how these kinds
of difficulties can be faced in a system that cannot provide fully satis-
factory answers.
Darwin's theory of evolution is an impressive witness to the capacity
of human rationality to organize an enormous amount of observa-
tions and experiences into a simple and coherent theory. It is a theory
capable of explaining a multitude of new and unexpected findings on
the one hand and of accommodating many theoretical developments
and changes on the other. These characteristics of Darwin's theory of
evolution have not diminished with time. Of course, it is not a "True
Theory," in the sense that every word is finite and unchallengeable nor
is it free of errors and misconceptions. But these are precisely some of
the major hallmarks of a real scientific theory.
References
Allen, G. E.
1968 "Thomas Hunt Morgan and the problem of natural selection," Journal of
the History of Biology 1: 113-139.
Bieri, P.
1984 "Time-awareness and personhood." Unpublished manuscript.
Brush, S. G.
1974 "Should the history of science be rated X?" Science 183: 1164-1172.
Bury, J. B.
1920 The Idea of Progress (London: McMillan).
Carlson, E. A.
1966 The Gene: A Critical History (Philadelphia and London: Saunders).
Cummins, R.
1975 "Functional analysis," TheJournal of Philosophy 72: 741-765.
Dobzhansky, Th.
1973 "Nothing in biology makes sense except in the light of evolution," American
Biology Teacher 35: 125-129.
Eigen, M., W. Gardiner, P. Schuster, and R. Winkler-Oswatisch
1981 "The origin of genetic information," Scientific American 244: 88-96.
Falk, R.
1986 "What is a gene?" Studies in the History and Philosophy of Science 17: 133-173.
Gould, S. J.
1977a Ever Since Darwin (Harmondsworth: Penguin Books).

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC
1977b Ontogeny and Phylogeny (Cambridge, MA: Harvard University Press).

Gould, S. J., and R. C. Lewontin
1979 "The spandrels of San Marco and the Panglossian paradigm: a critique of
the adaptionist programme," Proceedings of the Royal Society, London, Series B
205:581-598.
Grene, M.
1974a "Explanation and evolution," in The Understanding of Nature. Studies in
the Philosophy of Science 23 (Boston: Reidel), 207-227.
1974b "On the nature of natural necessity," in The Understanding of Nature. Stud-
ies in the Philosophy of Science 23 (Boston: Reidel), 228-242.
Hempel, C. G.
1965 [1959] "The logic of functional analysis," in Aspects of Scientific Explanation
(New York: The Free Press); repr., with some changes, from Symposium of
Sociological Theory, edited by L. Gross. (New York: Harper & Row).
Hempel, C. G., and P. Oppenheim
1965 [1948] "Studies in the logic of explanation," in Aspects of Scientific Explana-
tion, edited by C. G. Hempel. (New York: The Free Press).
Hofstadter, R.
1944 Social Darwinism in American Thought, 1860-1915 (Philadelphia: University
of Pennsylvania Press).
Lakatos, I.
1970 "Falsification and the methodology of scientific research programmes,"
in Criticism and Growth of Knowledge, edited by I. Lakatos and A. Musgrave,
91-196. (Cambridge: Cambridge University Press).
1978 "Introduction: Science and pseudoscience," in The Methodology of Scien-
tific Research Programmes, Philosophical Papers, Vol. I (Cambridg bridgeambridge
University Press), 1-7.
Lenoir, T.
1980 "Kant, Blumenbach, and vital materialism in German biology," Isis 71:
77-108.
1981 "Teleology without regrets. The transformation of physiology in Germ

1790-1847," Studies in the History and Philosophy of Science 12: 293-354.
Maxwell, G.
1962 "TFhe ontological status of theoretical entities," in Minnesota Studies it the
Philosophy of Science, edited by H. Feigl and G. Maxwell, 3: 3-27.
Monod, J. L.
1975 "On the molecular theory of evolution," in Problems of Scientific Revolution.
Progress and Obstacles to Progress in the Sciences, edited by R. Harre. (Oxford:
Clarendon Press).
Muller, H. J.
1914 "The bearing of the selection experiments of Castle and Philips on the
variability of genes," The American Naturalist 48: 567-576.
1922 "Variation due to change in the individual gene," The American Naturalist
56: 32-50.
Nagel, E.
1968 The Structure of Science (London: Routledge & Kegan Paul).
1977 "Teleology revisited," TheJournal of Philosophy 74: 261-301.
Quine, W. V. 0.
1953 "Two dogmas of empiricism" in From a Logical Point of Viezw, 20-46. (Cam-
bridge, MA: Harvard University Press).
1969 "Speaking of objects," in Ontological Relativity and Other Essays, 1-25. (New
York: Columbia University Press).

132.174.255.108 on Wed, 15 Feb 2023 14:0an 1976 12:34:56 UTC
Schweber, S. S.
1977 "The origin of the Origin revised," Journal of the History of Biology 2: 229-
316.
Waddington, C. H.
1963 The Nature of Life (London: Unwin Books).
Webster, G., and B. C. Goodwin
1982 "The origin of species: a structuralist approach," Journal of Social and Bio-
logical Structure 5: 15-47.

132.174.255.108 on Wed, 15 Feb 2023 14:05:00 UTC

This Content Downloaded From 132.174.255.108 On Wed, 15 Feb 2023 14:05:00 UTC

Uploaded by

Copyright:

Available Formats

This Content Downloaded From 132.174.255.108 On Wed, 15 Feb 2023 14:05:00 UTC

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

This Content Downloaded From 132.174.255.108 On Wed, 15 Feb 2023 14:05:00 UTC

Uploaded by

Copyright:

Available Formats

Darwin's Theory of Evolution as a Science

Author(s): Raphael Falk

Stable URL: https://www.jstor.org/stable/1772888

This content downloaded from

The effect of Darwin's theory of evolution may best be epitomized by

This content downloaded from

This content downloaded from

whole living (and nonliving) world into a hierarchial system of well-

1. As is well known, the biologist Alfred Russell Wallace (1823-1913) came to

This content downloaded from

shelves are not enough to produce changes in their views or to in-

This content downloaded from

Principles of Darwinian Evolution

This content downloaded from

that produce such variants act completely independent of their effect

(b) Contingencies that control and check variation between

4. This insight was explicated by H. J. Muller in 1922: "It is not in

This content downloaded from

lations of the theory of Darwin's time have been reconstructed into

This content downloaded from

the organisms, these constraints may be included among the compo-

What is a Scientific Theory?

This content downloaded from

the cause of my past was also determined quite simply by countless

This content downloaded from

not be demarcated from nonscientific theories and, presumably, as a

This content downloaded from

this is a remarkably rational system, keeping the program in constant

7. In his talk at the Technische Universitat in Berlin in the fall of 1983, He

This content downloaded from

2. A scientific theory should be rich, i.e. it should be able to suggest

It would be natural to propose a theory that suggested a simple line

However, objectively, that theory is as justified as another that con-

This content downloaded from

Physics versus Biology

This content downloaded from

of predictions made in astronomy, meteorology and the social sciences.

This content downloaded from

8. In a talk at the Freie Universitat in Berlin in the spring of 1984, Hempel

This content downloaded from

This content downloaded from

as it may sound, the most rational thing we can do is to admit that we

Darwin's Theory of Evolution as a Scientific Theory

-The principle of the production of inherited variation.

I do not intend to ignore the difficulties that these core-principles pos

This content downloaded from

Still, we must admit that adopting the mechano-reductionist frame-

This content downloaded from

-There is no sense discussing a problem at the level of the com-

-There is no sense discussing the details of the functioning of a

Scientific research may thus be envisaged as consisting of a hierarchy

This content downloaded from

generalized to a more specific level? This much discussed problem of

This content downloaded from

The mechano-reductionist approach, which proposed that the struc-

This content downloaded from

The theory of evolution is strongly dependent on reductionist re-

This content downloaded from

fate of a given individual. But individual organisms or even individual