1 Introduction PDF

CHAPTER ONE
INTRODUCTION
This book explores the general principles of scientific method that pervade
all of the sciences, focusing on practical aspects. The implicit contrast is with
specialized techniques for research that are used in only certain sciences. The
structure of science’s methodology envisioned here is depicted in Figure 1.1,
which shows individual sciences, such as astronomy and chemistry, as be-
ing partly similar and partly dissimilar in methodology. What they share
is a core of the general principles of scientific method. This common core
includes such topics as hypothesis generation and testing, deductive and in-
ductive logic, parsimony, and science’s presuppositions, domain, and limits.
Beyond methodology as such, some practical issues are shared broadly across
the sciences, such as relating the scientific enterprise to the humanities and
implementing effective science education.
The general principles that are this book’s topics are shown in greater
detail in Figure 1.2. These principles are of three kinds: (1) Some princi-
ples are relatively distinctive of science itself. For instance, the ideas about
Ockham’s hill that are developed in Chapter 8 on parsimony have a dis-
tinctively scientific character. If occasionally lawyers or historians happen to
use those ideas, they will not be reprimanded. Nevertheless, clearly those
ideas are used primarily by scientists and technologists. (2) Other princi-
ples are shared broadly among all forms of rational inquiry. For example,
deductive logic is squarely in the province of scientists, and it is explored
in Chapter 5. But deductions are also important in nearly all undertakings.
(3) Still other principles are so rudimentary and foundational that their well-
springs are in common sense, such as the principle of noncontradiction. Also,
science’s presuppositions, which are discussed in Chapter 4, have their roots
in common sense. Naturally, the boundaries among these three groups are
somewhat fuzzy, so they are shown with dashed lines. Nevertheless, the broad
distinctions among these three groups are clear and useful.
Downloaded from https://www.cambridge.org/core. The Librarian-Seeley Historical Library, on 17 Dec 2019 at 18:40:48, subject to the
Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511815034.003
2 Introduction
Figure 1.1. Science’s methodology depicted for five representative scientific disci-
plines, which are partly similar and partly dissimilar. Accordingly, scientific metho-
dology has two components. The general principles of scientific method pervade the
entire scientific enterprise, whereas specialized techniques are confined to particular
disciplines or subdisciplines.
There is a salient difference between specialized techniques and general

principles in terms of how they are taught and learned. Precisely because
specialized techniques are specialized, each scientific specialty has its own
more or less distinctive set of techniques. Because there are hundreds of spe-
cialties and subspecialties, the overall job of communicating these techniques
requires millions of instructional courses, books, and articles. But precisely
because general principles are general, the entire scientific community has
a single, shared set of principles, and it is feasible to collect and commu-
nicate the main information about these principles within the scope of a
single course or book. Whereas a scientist or technologist needs to learn
new techniques when moving from one project to another, the pervasive
general principles need be mastered but once. Likewise, whereas specialized
techniques and knowledge have increasingly shorter half-lives, given the un-
precedented and accelerating rate of change in science and technology, the
general principles are refreshingly enduring.
The central thesis of this book is that scientific methodology has two
components, the general principles of scientific method and the specialized
A Controversial Idea 3
Figure 1.2. Detailed view of the general principles, which are of three kinds: principles
that are relatively distinctive of science itself, broader principles found in all forms of
rational inquiry, and foundational principles with their wellsprings in common sense.
techniques of a given specialty, and the winning combination for scientists is

strength in both. Neither basic principles nor research techniques can sub-
stitute for one another. This winning combination can enhance productivity
and perspective.
A CONTROVERSIAL IDEA
The mere idea that there exist such things as general principles of scien-
tific method is controversial. The objections are of two kinds, philosophical
and scientific. But first, a potential misunderstanding needs to be avoided.
The scientific method “is often misrepresented as a fixed sequence of steps,”
rather than being seen for what it truly is, “a highly variable and creative pro-
cess” (AAAS 2000:18). The claim here is that science has general principles
that must be mastered to increase productivity and enhance perspective, not
that these principles provide a simple and automated sequence of steps to
follow.
Beginning with the philosophical objection, it is fashionable among some
skeptical, relativistic, and postmodern philosophers to say that there are
no principles of rationality whatsoever that are reliably or impressively
4 Introduction
truth-conducive. For instance, in an interview in Scientific American, the

noted philosopher of science Paul Feyerabend insisted that there are no
objective standards of rationality, so naturally there is no logic or method
to science (Horgan 1993). Instead, “Anything goes” in science, and it is no
more productive of truth than “ancient myth-tellers, troubadours and court
jesters.” From that dark and despairing philosophical perspective, the con-
cern with scientific method would seem to have nothing to do distinctively
with science itself. Rather, science would be just one more instance of the
pervasive problem that rationality and truth elude us mere mortals, forever
and inevitably.
Such critiques are unfamiliar to most scientists, although some may have
heard a few distant shots from the so-called science wars. Scientists typically
find those objections either silly or aggravating, so rather few engage such
controversies or bother to contribute in a sophisticated and influential man-
ner. But in the humanities, those deep critiques of rationality are currently
quite influential. Anyway, by that reckoning, Figure 1.1 should show blank
paper.
Moving along to the scientific objection, many scientists have claimed that
there is no such thing as a scientific method. For instance, a Nobel laureate
in medicine, Sir Peter Medawar, pondered this question: “What methods
of enquiry apply with equal efficacy to atoms and stars and genes? What is
‘The Scientific Method’?” He concluded that “I very much doubt whether
a methodology based on the intellectual practices of physicists and biolo-
gists (supposing that methodology to be sound) would be of any great use to
sociologists” (Medawar 1969:8, 13). In this regard, consider a little thought
experiment. Suppose that an astronomer, a microbiologist, and an engineer
were each given a grant of $500,000 to purchase research equipment. What
would they buy? Obviously they would purchase strikingly different instru-
ments, and each scientist’s new treasures would be quite useless to the others
(apart from the universal need for computers). By that reckoning, Figure 1.1
should show the methodologies of the individual sciences dispersed, with no
area in which they would all overlap.
What of these objections? Is it plausible that, contrary to Figure 1.1, the
methodologies of the various disciplines and subdisciplines of science have
no overlap, no shared general principles? Asking a few concrete questions
should clarify the issues and thereby promote an answer. Do astronomers use
deductive logic, but not microbiologists? Do psychologists use inductive logic
(including statistics) to draw conclusions from data, but not geologists? Are
probability concepts and calculations used in biology, but not in sociology?
Do medical researchers care about parsimonious models and explanations,
but not electrical engineers? Does physics have presuppositions about the
The AAAS Vision of Science 5
existence and comprehensibility of the physical world, but not genetics? If

the answers to such questions are no, then Figure 1.1 stands as a plausible
picture of science’s methodology.
THE AAAS VISION OF SCIENCE

Beyond such brief and rudimentary reasoning about science’s methodology,
it merits mention that the thesis proposed here accords with the official posi-
tion of the American Association for the Advancement of Science (AAAS).
The AAAS is the world’s largest scientific society, the umbrella organization
for almost 300 scientific organizations and publisher of the prestigious jour-
nal Science. Accordingly, the AAAS position bids fair as an expression of
the mainstream opinion.
The AAAS views scientific methodology as a combination of general prin-
ciples and specialized techniques, as depicted in Figure 1.1.
Scientists share certain basic beliefs and attitudes about what they do and
how they view their work. . . . Fundamentally, the various scientific disci-
plines are alike in their reliance on evidence, the use of hypotheses and
theories, the kinds of logic used, and much more. Nevertheless, scientists
differ greatly from one another in what phenomena they investigate and
in how they go about their work; in the reliance they place on historical
data or on experimental findings and on qualitative or quantitative meth-
ods; in their recourse to fundamental principles; and in how much they
draw on the findings of other sciences. . . . Organizationally, science can be
thought of as the collection of all of the different scientific fields, or con-
tent disciplines. From anthropology through zoology, there are dozens of
such disciplines. . . . With respect to purpose and philosophy, however, all
are equally scientific and together make up the same scientific endeavor.
(AAAS 1989:25–26, 29)
Regarding the general principles, “Some important themes pervade science,

mathematics, and technology and appear over and over again, whether we are
looking at an ancient civilization, the human body, or a comet. They are ideas
that transcend disciplinary boundaries and prove fruitful in explanation, in
theory, in observation, and in design” (AAAS 1989:123).
Accordingly, “Students should have the opportunity to learn the nature of
the ‘scientific method’ ” (AAAS 1990:xii; also see AAAS 1993). That verdict
is affirmed in official documents from the National Academy of Sciences
(NAS 1995), the National Commission on Excellence in Education (NCEE
1983), the National Research Council of the NAS (NRC 1996, 1997, 1999),
the National Science Foundation (NSF 1996), the National Science Teachers
6 Introduction
Association (NSTA 1995), and the counterparts of those organizations in

many other nations (Matthews 2000:321–351).
An important difference between specialized techniques and general prin-
ciples is that the former are discussed in essentially scientific and technical
terms, whereas the latter inevitably involve a wider world of ideas. Accord-
ingly, for the topic at hand, the “central premise” of one AAAS (1990:xi)
position paper is extremely important, namely, that “Science is one of the
liberal arts and . . . science must be taught as one of the liberal arts, which
it unquestionably is.” Many of the broad principles of scientific inquiry are
not unique to science, but also pervade rational inquiry more generally, as
depicted in Figure 1.2. “All sciences share certain aspects of understanding –
common perspectives that transcend disciplinary boundaries. Indeed, many
of these fundamental values and aspects are also the province of the humani-
ties, the fine and practical arts, and the social sciences” (AAAS 1990:xii; also
see p. 11).
Likewise, the continuity between science and common sense is respected,
which implies productive applicability of scientific attitudes and thinking
in daily life. “Although all sorts of imagination and thought may be used
in coming up with hypotheses and theories, sooner or later scientific ar-
guments must conform to the principles of logical reasoning – that is, to
testing the validity of arguments by applying certain criteria of inference,
demonstration, and common sense” (AAAS 1989:27). “There are . . . certain
features of science that give it a distinctive character as a mode of inquiry.
Although those features are especially characteristic of the work of profes-
sional scientists, everyone can exercise them in thinking scientifically about
many matters of interest in everyday life” (AAAS 1989:26; also see AAAS
1990:16).
Because science’s general principles involve a wider world of ideas, many
vital aspects cannot be understood satisfactorily by looking at science in
isolation. Rather, they can be mastered properly only by seeing science in
context, especially in philosophical and historical context. Therefore, this
book’s pursuit of the principles of scientific method will occasionally range
into discourse that has a distinctively philosophical or historical or sociolog-
ical character. There is a natural and synergistic traffic of great ideas among
the liberal arts, including science.
The brief remarks in this and the previous sections are not offered as a
rigorous defense of the (controversial) thesis that some general principles
are vital components of scientific reasoning. Only a whole book, such as
what follows, can aspire to such an ambitious goal. Rather, these preliminary
remarks are offered as evidence that the idea that there is a scientific method
has enough plausibility and backing to merit careful consideration, not breezy
dismissal.
Primary and Secondary Benefits 7
PRIMARY AND SECONDARY BENEFITS

Whatever else may be controversial, one thing that is certain is that mastery
of the subject matter proposed and presented here, the principles of scientific
method, will require some time and effort. Accordingly, it is natural to ask
about the purposes and benefits that will result from this study.
Two general kinds of benefits are expected, namely, increased produc-
tivity and enhanced perspective. The primary benefit will be to help scientists
become better scientists, more creative and more productive, by providing
a deeper understanding of the basic principles of scientific method. A sec-
ondary benefit will be to cultivate a humanities-rich version of science, rather
than a humanities-poor version, so that scientists can gain perspective on their
enterprise.
Regarding the primary benefit, what a scientist or technologist needs in or-
der to function well can be depicted by a resources inventory, as in Figure 1.3.
All items in this inventory are needed. The first three items address the ob-
vious physical setup that a scientist needs. The last two items are intellectual
rather than physical, namely, mastery of the specialized techniques of a cho-
sen specialty and mastery of the general principles of scientific method.
A common concern is that frequently the weakest link in a scientist’s
inventory is an inadequate understanding of science’s principles. This weak-
ness in understanding the scientific method has just as much potential to
Figure 1.3. A typical resources inventory for a research group. The scientists in a given
research group often have excellent laboratory equipment, computers, infrastructure,
and technical training, but inadequate understanding of the general principles of
scientific method is the weakest link. Ideally, a research group will be able to check
off all five boxes in this inventory, and there will be no weak link.
8 Introduction
Figure 1.4. Perspective on the place and role of scientific method. The foundations
of scientific method are provided by the philosophy of science, which depends more
generally on philosophy, which is grounded ultimately by common sense. In turn,
scientific method supports scientific specialties and technology.
retard progress as does, say, inappropriate laboratory equipment or inade-

quate training in some research technique.
Moving along to the secondary benefit, an initial perspective on the place
and role of scientific method is offered in Figure 1.4. The scientific specialties
and technological accomplishments that emerge from applying the scien-
tific method are obvious. But a humanities-rich version of science will reveal
science’s roots in the philosophy of science and more generally in philoso-
phy, which is grounded ultimately in common sense. Accordingly, scientific
method will be better integrated and more interesting when presented in
its philosophical and historical context. Such perspective will also facilitate
realistic claims, neither timid nor aggrandized, about science’s powers and
prospects. A humanities-rich perspective will preclude imperialistic claims
about science’s domain, with all of the attendant false promises that could
only disappoint and alienate.
The topic of science’s basic principles has been around for millennia, even
before Aristotle (Losee 1993). So naturally many opinions have been ex-
pressed about the value of this topic, especially for a scientist’s ordinary,
practical, day-to-day work. Scientists themselves have written much, as have
philosophers, though most of that literature has been rather speculative or
anecdotal.
Fortunately, scholars in a different field, the science educators, have done
the work of conducting hundreds of careful empirical studies to character-
ize and quantify the benefits that can result from learning scientific method.
Many of those studies have involved impressive sample sizes and carefully
controlled experiments to quantify educational outcomes and scientific
Primary and Secondary Benefits 9
competencies for students who either have or else have not received instruc-
tion in science’s general principles. Consequently it has been educators, rather
than scientists or philosophers, who have provided the best information on
these benefits.
Incidentally, among educators, what here goes under such labels as “scien-
tific method” and “general principles” is most frequently termed the “nature
of science.” Because Chapter 11 will review the literature in science edu-
cation, here only brief remarks without documentation will be presented,
by way of anticipation. Empirical studies by educators have provided over-
whelming evidence for six specific claims.
(1) Better Comprehension. The specialized techniques and subject
knowledge that so obviously make for productive scientists are better com-
prehended when the underlying principles of scientific method are under-
stood – somewhat like the way that calcium is better absorbed by the digestive
system when accompanied by vitamin D.
(2) Greater Adaptability. It is facility with the general principles of
science that contributes the most to a scientist’s ability to be adaptable and to
transfer knowledge and strategies from a familiar context to new ones, and
that adaptability will be necessary for productivity as science and technology
continue to experience increasingly rapid and pervasive changes.
(3) Greater Interest. Most people find a humanities-rich version of
science, with its wider perspective and big picture, much more engaging and
interesting than a humanities-poor version, so including something of sci-
ence’s method, history, and philosophy in the science curriculum results in
higher rates of retention of students in the various sciences (and it especially
helps those ranked near the bottom, so that educational outcomes can become
more nearly equal).
(4) More Realism. An understanding of the scientific method leads
to a realistic perspective on science’s powers and limits, and more gener-
ally to balanced views of the complementary roles of the sciences and the
humanities.
(5) Better Researchers. Researchers who master science’s general princi-
ples gain productivity because they can make better decisions about whether
or not to question an earlier interpretation of their data as a result of new
evidence, whether or not there is a need to repeat an experiment, where to
look for other scientific work related to their project, and how certain or
accurate their conclusions are.
(6) Better Teachers. Teachers and professors who master science’s gen-
eral principles prove to be better at communicating science content, in part
because they are better at detecting and correcting students’ prior mistaken
notions and logic, and hence such teachers can better equip the next gener-
ation of scientists to be productive.
10 Introduction
The facts of the case are clear, having been established by hundreds of
empirical studies involving various age groups, nations, and science subjects:
Understanding the principles of scientific method does increase productivity
and enhance perspective. But why? Why does mastery of these principles
help scientists to become better scientists? The most plausible explanation
is simply that the central thesis of this book is true: It really is the case
that scientific methodology has two components, the general principles of
scientific method and the specialized techniques of a chosen specialty, and
the winning combination for scientists is strength in both. Therefore,
adequate understanding of scientific method is essential for an astrono-
mer, botanist, chemist, dietitian, engineer, floriculturalist, geologist, . . . , or
zoologist.
BEYOND THE BASICS

Do scientists typically have an adequate understanding of the scientific
method? Is it sufficient to yield the expected benefits of productivity and
perspective? Unfortunately, the current state of affairs seems rather dismal.
“Ask a scientist what he conceives the scientific method to be, and he will
adopt an expression that is at once solemn and shifty-eyed: solemn, because
he feels he ought to declare an opinion; shifty-eyed, because he is wonder-
ing how to conceal the fact that he has no opinion to declare” (Medawar
1969:11). Furthermore, countless recent studies by science educators have
confirmed that verdict.
The cause of the current situation is no mystery. Scientists are not born
already knowing the principles of scientific method, and neither are they
taught those principles in a vigorous and systematic manner. “The hapless
student is inevitably left to his or her own devices to pick up casually and
randomly, from here and there, unorganized bits of the scientific method, as
well as bits of unscientific methods” (Theocharis and Psimopoulos 1987).
Just where do scientists get what meager bits they do have? Because few
science majors ever take a course in scientific method, logic, or the history and
philosophy of science, their exposure to any focused attention to science’s
principles usually is limited to the occasional science textbook that begins
with an introductory chapter on scientific method. Figure 1.5 lists typical
contents for such chapters.
Despite the perhaps scandalous resemblance of such accounts to the anti-
quated view of science offered by Francis Bacon in the early 1600s (Urbach
1987; Peltonen 1996), it may well be that elementary ideas along those lines
provide the most suitable picture of science to convey to an eighth-grade
student working on a science-fair project. Also, it may well be that such
a rudimentary cartoon of science is much closer to the mark than is the
Beyond the Basics 11
Figure 1.5. Typical topics in an elementary presentation of scientific method intended

for college freshmen and sophomores. Introductory science texts often start with
several pages on scientific method, discussing the formulation and testing of hypothe-
ses, collection of data from controlled and replicated experiments, and so on. They
are unlikely, however, to include any discussion of parsimony or any exploration of
the history of scientific method beyond a passing mention of Aristotle.
currently fashionable postmodern take on science, as reviewed and criticized

by Koertge (1998). Whatever its merit in terms of simplicity, such an elemen-
tary view of scientific method is wholly inadequate for science professionals
and professors. In pursuit of increased productivity and sophisticated per-
spective, science professionals must go beyond the basics.
In some sense, the basic principles of scientific method, or at least their
wellsprings in common sense, are obvious and compelling. But these prin-
ciples are also difficult, challenging, and exacting. They yield their secrets
and benefits only to those who pay their dues and do their work. Surely it
is sobering to realize in historical perspective that civilizations rose and fell
around the globe for millennia before anything recognizable as the scientific
method emerged (say around a.d. 1200 to 1600 by various accounts). So there
must be some limit to that view about scientific principles being obvious.
Indeed, learning enough about deductive logic to avoid common fallacies
is no easy task. Learning enough about inductive logic and statistics to ana-
lyze data properly and vigorously is difficult, especially given the internal de-
bate in statistics between the frequentist and Bayesian paradigms. Mastering
the principle of parsimony or simplicity in order to gain efficiency and in-
crease productivity is anything but simple, requiring precise distinctions,
subtle concepts, and complex calculations. Acquiring some philosophical and
historical perspective on science is challenging. And developing the habit
of applying general principles to daily scientific work with creativity and
effectiveness requires considerable mentoring and practice.
12 Introduction
Moreover, these principles work in concert, so to understand their con-

nections and interrelations in a functioning whole, they must be gathered
together and presented in a book or in a course in an integrated fashion.
Because the principles of concern here are general principles, they do indeed
emerge repeatedly in science and technology, but that alone is not sufficient
to guarantee that scientists will perceive and grasp their generality. For in-
stance, the case studies in Chapter 9 will reveal parsimony at work in diverse
applications in genetics, agriculture, drug design, and electrical engineering.
The great risk is that a neophyte may see the material involving parsimony as
being just one more of the technicalities needed to accomplish some special-
ized task. Therefore, it is imperative to present general principles as being
truly general principles!
If the wide generality and applicability of a principle are taught explicitly
and near the beginning of a scientist’s training, then subsequent instances of
that principle at work will reinforce the lesson and will promote adaptability
and productivity. But if relevant instances involving that principle are merely
encountered repeatedly, but with no larger story ever being told, then only
the rare student can be expected to assemble the big picture without proper
mentoring.
However, despite the deficiencies of the current situation and despite
the inherent challenges of this topic of science’s principles, two factors are
encouraging. They imply that rapid and dramatic improvement certainly is
possible:
First, students are receptive, finding this subject matter quite interesting.
For example, Albert Einstein observed that “I can say with certainty that the
ablest students whom I met as a teacher were deeply interested in the theory
of knowledge. I mean by ‘ablest students’ those who excelled not only in
skill but in independence of judgment. They liked to start discussions about
the axioms and methods of science” (Frank 1957:xi). Likewise, Machamer
(1998) remarked, “Now part of the fun of science, as in most interesting
human activities, lies in thinking about how and why it is done, and how it
might be done better.” Also recall the story about the graduate student in
this book’s Foreword who eagerly wanted to know “the logical underpinnings
and processes of science at the level of basic principles” so that she could
graduate with a doctorate in science and feel “worthy of more than a technical
degree.” Again, countless studies by science educators have shown decisively
that students are interested in a humanities-rich account of the principles of
scientific method.
Second, the AAAS vision for science includes a vigorous and sustained call
for students and scientists to understand these principles well, and it is rea-
sonable to suppose that this leadership will be influential. So, on two counts,
there are good prospects for scientists to acquire their winning combination,
A Timely Opportunity 13
the principles and techniques that can enhance productivity and perspective.
The bottom line is that if this book’s central thesis is true, then it will not be
possible to keep this winning combination a secret for much longer.
A TIMELY OPPORTUNITY
Again, the central thesis of this book is that science’s methodology involves
both general principles and specialized techniques, and these principles and
techniques together constitute the winning combination for scientists that
will enhance their productivity and perspective. However, such views have
suffered considerable neglect during the past century or so.
A major cause of that neglect has been the common perception that even if
there are such things as the principles of scientific method, the study of those
principles would confer no benefit to scientists. For instance, writing in his
usual witty and engaging style, Medawar (1969:8, 12) mused, “If the purpose
of scientific methodology is to prescribe or expound a system of enquiry
or even a code of practice for scientific behavior, then scientists seem to
be able to get on very well without it. Most scientists receive no tuition in
scientific method, but those who have been instructed perform no better as
scientists than those who have not. Of what other branch of learning can it be
said that it gives its proficients no advantage; that it need not be taught or, if
taught, need not be learned?” (Medawar 1969:8; also see p. 12). In fairness to
Medawar, he also remarked that “Of course, the fact that scientists do not
consciously practice a formal methodology is very poor evidence that no
such methodology exists” (Medawar 1969:9), and he did go on to offer some
positive comments.
In any case, the sentiment that scientists get along just fine without probing
science’s philosophical and methodological foundations is at least common.
Such sentiments are mistaken. They bespeak a lamentable and dangerous
complacency. Indeed, on three counts, it is time for serious consideration of
this book’s central thesis.
(1) Science Education. The AAAS has stated with confidence and en-
thusiasm its vision that a humanities-rich understanding of science makes
for better scientists. During the past decade, science educators have gener-
ated an enormous literature that provides a wealth of compelling empirical
evidence in support of the AAAS vision, as will be reviewed in Chapter 11.
Because this literature is so recent, however, one cannot blame scientists
and philosophers in the past for not having taken into account the findings
of educators when they offered their anecdotes and speculations about the
relevance or irrelevance of science’s principles for day-to-day research. But
this does mean that earlier assessments, such as that by Medawar, now need
to be taken with a grain of salt. More important, any future opinions and
14 Introduction
reflections from scientists and philosophers about the value of mastering

science’s principles can gain in realism and interest by incorporating the
factual findings of science educators.
(2) Recent Developments. In many respects, the topic of science’s gen-
eral principles is an ancient story with refreshingly enduring content. In some
other respects, however, it is a living and growing topic that includes exciting
recent developments.
The foremost instance of recent advances is that wonderfully subtle but
surprisingly practical application of parsimony, as will be explored in Chap-
ter 8. Briefly, experiments and investigations in many scientific specialties
produce large amounts of relatively noisy or inaccurate data. This situa-
tion is especially common in some applied sciences, such as agriculture and
medicine. Remarkably, parsimonious models of such data can yield find-
ings considerably more accurate than are indicated by the original data.
Frequently a parsimonious model can improve accuracy, prompt better
decisions, and increase productivity as much as can the collection of several
times as much data. Yet the cost of a few seconds of computer time to fit
a parsimonious model is minute compared with the cost of collecting much
more data, so parsimonious models offer a remarkably cost-effective means
to increase productivity.
Sadly, however, that opportunity to increase productivity is one of sci-
ence’s best-kept secrets. Apart from scientists and technologists in a few
specialties, such as signal processing, that option is all but unknown. Nor is
such an application of parsimony something so simple and obvious that scien-
tists are likely to stumble across it without intentional study and mentoring.
The underlying concepts are unfamiliar and even somewhat counterintu-
itive. Though in typical applications the calculations require only seconds
on an ordinary computer, that represents the millions or billions of arith-
metic steps in a highly structured algorithm that earlier someone mastered
and programmed. Obviously such powerful methods were unknown prior to
the general availability of computers beginning in the 1960s, as well as the
advent, at around the same time, of some critical developments in statistical
theory.
Recent decades and even recent years have seen great advances in deduc-
tive logic, probability, and inductive logic (or statistics), as will be explored
in Chapters 5–7. So, on many fronts, the recent advances in understanding
and implementing science’s general principles have been so spectacular that
earlier accounts are outdated, and nothing but a contemporary evaluation
of the possibilities merits serious consideration.
(3) Appropriate Focus. The AAAS vision of a humanities-rich version of
science has breathtaking sweep. The history, philosophy, and methodology of
science compose an enormously broad and involved interdisciplinary field.
Personal Experience 15
Consequently, when approaching this field for an audience of science stu-

dents and professionals, one needs to focus on those specific portions of the
information that are of greatest interest and benefit to scientists. Otherwise
the result could be more dissipative than beneficial.
Too often the curriculum in science’s principles that has reached scientists
has been developed along the path of least resistance by borrowing wholesale
from the literature in the philosophy of science. But the goal of that literature
has been to make philosophers better philosophers, not to make scientists
better scientists, and similar remarks could be made about the literature in
the history of science or the sociology of science.
Obviously, any book or course for scientists regarding science’s principles
must seriously engage the companion literatures from historians, philoso-
phers, sociologists, and educators, but the borrowing must be selective and
focused if the goal is to help scientists become better scientists. Conse-
quently, any assessment of the value of having scientists study science’s basic
principles will be inaccurate if that study was based on materials without
the proper focus. A scientist could study an enormous number of pages
directed at making philosophers better philosophers or making historians
better historians and still not become a better scientist. The real issue is
whether or not a scientist will benefit from mastering material properly
focused for scientists.
There has been little such focus in the past. However, now is the time to
put the AAAS vision to the test in a manner that is properly focused and fair.
Besides the AAAS (1990) vision, relatively recent position papers that set
forth specific recommendations for curricula in the nature of science include
those by the AAAS (1989), NAS (1995), NRC (1996, 1999), and NSF (1996).
In all of those reports, the principles of scientific method hold a prominent
position in the proposed curricula.
So, on three counts, there is now a new day for the thesis that mastering
science’s general principles can help scientists. The time for complacency is
past. The time is right for rapid progress.
PERSONAL EXPERIENCE
Thus far, this introductory chapter has drawn on the insights of others,
especially those of the AAAS and science educators, to illustrate and support
this book’s central thesis. As this chapter approaches its close, perhaps some
readers would be interested in the personal experience that has prompted
my interest in the principles of scientific method.
My research specialty at Cornell University during the past three decades
has been the statistical analysis of ecological and agricultural data. A special
focus in this work has been agricultural yield trials. Worldwide, billions of
16 Introduction
Figure 1.6. A soybean yield trial conducted in Aurora, New York. The soybean varie-
ties here varied in terms of numerous traits. For example, the variety in the center
foreground matured more quickly than the varieties to its left and right, making
its leaves light yellow rather than dark green as the end of the growing season ap-
proached. Yield is a particularly important trait. (Reprinted from Gauch, 1992:3, with
kind permission from Elsevier Science.)
dollars are spent annually to test various cultivars, fertilizers, insecticides, and
so on. For instance, Figure 1.6 shows a soybean yield trial to determine which
cultivars perform best. The objective of yield-trial research is, of course, to
increase crop yields.
From studying the philosophy and method of science, but not from reading
the agricultural literature, I came to realize that a parsimonious model could
provide a more accurate picture than could its raw data. So I tried that
concept on yield-trial data and found that the resulting gain in accuracy could
be assessed empirically and exactly by data splitting using replicated data
(Gauch 1988). It worked. The parsimonious model, which required but a few
seconds of computer time, typically produced findings as accurate as would
have been achieved using averages over replications based on two to five
times as much data. Such additional data would have cost tens to hundreds
of thousands of dollars, in various instances, so those gains in accuracy and
efficiency were spectacularly cost-effective. Furthermore, statistical theory
was able to explain that surprising phenomenon, which was demonstrated
repeatedly for many crops in diverse locations and agroecosystems.
Personal Experience 17
Accordingly, I submitted a manuscript to a prestigious statistics journal.

One reviewer flatly rejected my manuscript, complaining that my results were
“magical” and too good to be true – the ideas involving parsimony, one of the
principles of scientific method, were just too unfamiliar. But fortunately the
editor understood my work better and accepted the paper. Subsequently I
published a paper in American Scientist that provided a broad philosophical
and scientific perspective for understanding the relationship between par-
simony and accuracy (Gauch 1993). Meanwhile, the groundbreaking idea
(within the agricultural literature) that parsimonious models could increase
accuracy and efficiency has now become rather common, and it has made no
small contribution to yield increases for many crops in many nations.
The salient feature of that story is that the requisite parsimonious models
and computers had been available to agronomists and breeders for a couple
of decades, but no one had capitalized on that opportunity until 1988. What
has been the opportunity cost? Standard practices in agricultural research
today are increasing the yields for most of the world’s major crops by about
0.5% to 1.5% per year. An exact projection is impossible, but a conservative
estimate is that parsimonious models of yield-trial data often can support an
additional increment of about 0.4% per year (Gauch 1992:184–185).
In other words, for a typical case, if ordinary data analysis supports an av-
erage annual yield increase of 1%, whereas a parsimonious model supports
1.4%, then something like 30% of the information in the data is wasted when
researchers fail to put parsimony to work. As will be reviewed in Chapter 8,
over the past several years there has been compelling evidence from numer-
ous plant-breeding projects that parsimonious models can routinely support
that additional yield increment of 0.4% per year.
If an additional increment of 0.4% per year had been achieved after that
window of opportunity was opened around 1970, then today’s crop yields
would be 12.7% higher. But to be conservative, suppose that just half of
that advantage had been transferrable from research plots to farmers’ fields.
Even then, putting parsimony to work could have increased crop yields by
6% over the past 30 years. That may not seem like much, but given that the
world’s population today is about 6 billion people, that 6% increase would
feed 360 million people, more than the population of North America.
But tragically, that window of opportunity from 1970 to 2000 has come and
gone. Thus far, only a fraction of agricultural researchers have learned to use
parsimonious models to gain accuracy, and even that limited application did
not take place until the last of those three decades. The resultant loss of 6%
is now irretrievable, because breeding is an incremental process. Each year’s
efforts begin where the last year left off as breeding stocks are gradually
improved. Even though parsimony can be put to work in greater measure in
the future, that does not change the historical fact that the ongoing process
18 Introduction
of plant breeding already has built into it a 6% opportunity cost, caused by

neglecting parsimony from 1970 to 2000, and that loss can never be erased.
Opportunities come and go; they do not linger forever.
What was missing? What caused that now permanent 6% reduction in
crop productivity? It was not the lack of specialized techniques. Nor was
it inability to easily perform billions of arithmetic steps. Nor was it lack of
funding. It was lack of understanding, or, better yet, mastery, of parsimony,
one of the principles of scientific method. What was missing was the last of
the critical resources listed in Figure 1.3.
Needless to say, during the past three decades, countless additional mea-
sures could have been taken to strengthen agricultural research and thereby
to boost farm productivity. The lost 6% could have been regained in many
different ways. What is so remarkable about that particular lost opportunity
to exercise parsimony, however, is how cost-effective it would have been.
Besides having low cost, it also would have involved low risk, because that
approach to data analysis had already been tried and proved, unlike many
other possible measures that were unproved and risky. Nor would any new or
different or expensive requirements have been imposed on data-collection
procedures. Rather, the missed opportunity was failure to apply parsimo-
nious models to extract more accurate and more useful information from
data already in hand using computers already in place.
That loss is analogous to buying a bag of oranges and then squeezing out
only half of the juice. Such waste doesn’t make sense. Regardless of which
particular opportunities could have been used to change agricultural research
for the better, one factor remains the same: Whatever data are obtained, it
makes sense to extract all or nearly all of the useful information from the
data. Getting only half of the juice is deplorable. And the only way to get all
or nearly all of the juice is to master not only research techniques but also
general principles.
The bottom line is that for lack of mastering the principles of scientific
method, crop yields worldwide are now considerably less (about 6% less)
than they could have been had the value of parsimony been appreciated
three decades earlier. The principles of scientific method matter.
The larger issue that this example raises is that many other scientific and
technological specialties present us with tremendous opportunities that can-
not be realized until some specialist in a given discipline masters and applies
a critical general principle, be it parsimony or another principle in a given
instance. Precisely because these are general principles, my suspicion is that
my own experience is representative of what can be encountered in countless
specialties, as the diverse case studies in Chapter 9 will clearly indicate.
Finally, these reflections on my own experience have focused thus far on
only one of the two proposed benefits, namely, productivity. On balance,
something might be added about the other benefit, namely, the perspective
Summary 19
gained from a humanities-rich perception of science. My own experience

resonates with the AAAS (1990:xi) expectation that broad experience of
science as a liberal art is worthwhile for “the sheer pleasure and intellectual
satisfaction of understanding science.”
Like the graduate student mentioned in this book’s Foreword, I also had a
restless curiosity and deep interest regarding the basic principles of scientific
thinking. And also like her, that spark of curiosity had received no stimulus
or encouragement whatsoever from the courses and ideas presented in my
university education. Nevertheless, such curiosity is normal and common,
as Aristotle observed in the opening words of his Metaphysics: “All men
by nature desire to know” (McKeon 1941:689). More recently, the AAAS
(1990:xi) has reaffirmed Aristotle’s observation that there is great satisfaction
and pleasure in “the human desire to know and understand.”
In a campus bookstore, I stumbled across a book by Burks (1977) not
long after it was first published in 1963. Arthur Burks was a professor of both
philosophy and computer science. His book was quite long, about 700 pages,
and frequently was rather repetitious and tedious. But it had the content
that I had been seeking and had not yet found anywhere else. There at last
I had found an intellectually satisfying account of the underlying principles
and rationality of scientific thinking. That book immediately became a great
favorite of mine. Subsequently I sought and occasionally found additional
books to nourish my continuing interest in the principles of scientific method,
most notably that by Jeffreys (1983), which was first published in 1961.
Thus my interest in science’s principles dates to about 1965. For the next
two decades, my primary motivation for that interest was – to echo the
AAAS – the “sheer pleasure” that accompanies “the human desire to know
and understand.” Grasping the big ideas that are woven throughout the fabric
of the entire scientific enterprise generates delight and confidence.
Because of youth and bad company, however, the idea that mastery of
those principles could also promote productivity was an idea that would
slumber in my mind for a couple of decades! It was not until 1982 that some
scattered thoughts began to be reawakened and to coalesce (Gauch 1982),
eventually resulting in the Biometrics article mentioned earlier (Gauch 1988).
Since then, I have been keenly aware that the principles of scientific method
can enhance not only perspective but also productivity. Whether at present
my interest in these principles is motivated more by a desire for intellectual
perspective or for scientific productivity I am not able to say.
SUMMARY
This book takes as its subject matter the general principles of scientific
method that pervade all of the sciences, as contrasted with specialized tech-
niques that occur only in some sciences. These basic principles include
20 Introduction
hypothesis generation and testing, deductive and inductive logic, parsimony,

and science’s presuppositions, domain, and limits.
The primary benefit to be expected from understanding these principles is
increased productivity. A secondary benefit will be a humanities-rich version
of science that will promote a wider perspective on the scientific enterprise.
To obtain these benefits, however, scientists must go beyond the basics of
scientific method. They must master the principles of scientific method as an
integrated, functioning whole.
The central thesis of this book is that scientific methodology has two
components, the general principles of scientific method and the specialized
techniques of a chosen specialty, and the winning combination for scientists
is strength in both. Neither basic principles nor research techniques can
substitute for one another. This winning combination will enhance both
productivity and perspective.
On three counts, this thesis merits serious consideration. First, that science
has a scientific method with general principles and that these principles can
benefit scientists is the official, considered view of the AAAS (and the NAS,
NRC, NSF, and other major scientific organizations in the United States,
as well as similar entities in numerous other nations). Second, science edu-
cators have demonstrated in hundreds of empirical studies, often involving
sizable samples and controlled experiments, that learning science’s general
principles can benefit students and scientists in several specific, quantifiable,
important respects. Third, my own research experience, primarily involv-
ing agricultural yield-trial experiments, confirms the two expected benefits.
Mastery of the principles of scientific method promotes vital scientific pro-
ductivity and technological progress. In addition, there is intellectual pleasure
in gaining a humanities-rich perspective on how scientific thinking works.

1 Introduction PDF

Uploaded by

Copyright:

Available Formats

1 Introduction PDF

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

1 Introduction PDF

Uploaded by

Copyright:

Available Formats

CHAPTER ONE

There is a salient difference between specialized techniques and general

techniques of a given specialty, and the winning combination for scientists is

truth-conducive. For instance, in an interview in Scientiﬁc American, the

existence and comprehensibility of the physical world, but not genetics? If

THE AAAS VISION OF SCIENCE

Regarding the general principles, “Some important themes pervade science,

Association (NSTA 1995), and the counterparts of those organizations in

PRIMARY AND SECONDARY BENEFITS

retard progress as does, say, inappropriate laboratory equipment or inade-

BEYOND THE BASICS

Figure 1.5. Typical topics in an elementary presentation of scientiﬁc method intended

currently fashionable postmodern take on science, as reviewed and criticized

Moreover, these principles work in concert, so to understand their con-

reﬂections from scientists and philosophers about the value of mastering

Consequently, when approaching this ﬁeld for an audience of science stu-

Accordingly, I submitted a manuscript to a prestigious statistics journal.

of plant breeding already has built into it a 6% opportunity cost, caused by

gained from a humanities-rich perception of science. My own experience

hypothesis generation and testing, deductive and inductive logic, parsimony,

You might also like