Professional Documents
Culture Documents
Textbook What Makes A Good Experiment Reasons and Roles in Science Allan Franklin Ebook All Chapter PDF
Textbook What Makes A Good Experiment Reasons and Roles in Science Allan Franklin Ebook All Chapter PDF
https://textbookfull.com/product/the-geometry-of-musical-rhythm-
what-makes-a-good-rhythm-good-1st-edition-godfried-t-toussaint/
https://textbookfull.com/product/the-geometry-of-musical-rhythm-
what-makes-a-good-rhythm-good-second-edition-godfried-t-
toussaint/
https://textbookfull.com/product/wiser-the-scientific-roots-of-
wisdom-compassion-and-what-makes-us-good-1st-edition-dilip-jeste/
https://textbookfull.com/product/once-can-be-enough-decisive-
experiments-no-replication-required-allan-franklin/
Amateur A True Story About What Makes a Man Thomas Page
Mcbee
https://textbookfull.com/product/amateur-a-true-story-about-what-
makes-a-man-thomas-page-mcbee/
https://textbookfull.com/product/are-there-really-neutrinos-an-
evidential-history-2nd-edition-allan-d-franklin-author/
https://textbookfull.com/product/the-rise-and-fall-of-the-fifth-
force-discovery-pursuit-and-justification-in-modern-physics-2nd-
edition-allan-franklin/
https://textbookfull.com/product/what-makes-time-special-1st-
edition-craig-callender/
https://textbookfull.com/product/amateur-a-true-story-about-what-
makes-a-man-first-scribner-hardcover-edition-mcbee/
WHAT MAKES A GOOD EXPERIMENT?
ALLAN FRANKLIN
REASONS
& ROLES
IN SCIENCE
ACKNOWLEDGMENTS vii
INTRODUCTION 1
CONCLUSION 296
NOTES 307
REFERENCES 339
INDEX 359
ACKNOWLEDGMENTS
IN large high-energy physics groups there are criteria for who may be
an author of a paper. The criteria do not require that an author actually
participated in the writing of the paper, but rather that the scientist has
participated in a significant way in the production of the experimental
result. This may include design and construction of the experimental
apparatus, or the writing of computer programs to acquire or analyze the
experimental data, or analyzing the data, or writing the paper. Thus, the
paper announcing the discovery of the Higgs boson by the Compact Muon
Solenoid (CMS) collaboration at the Large Hadron Collider had 2,897
authors. If I used somewhat different criteria such as those whose work
has informed my work, colleagues and friends who have read my work at
various stages and offered constructive criticisms and suggestions, and
those who have provided valuable discussions, the number of authors of
this book, while not close to the number of CMS authors, would surely
exceed a hundred. A reasonable approximation to such an author list might
be obtained by looking at the acknowledgments sections of my previous
work, although I am sure it would omit a fair number of people who should
be on the list. The conventions of academic life outside high-energy physics
suggest that, as the person who has actually written this book, I should be
the sole author and receive all of the credit or blame for its contents.
I will, nevertheless, acknowledge those friends and colleagues who
have made specific contributions to this book. As those who have read my
vii
viii ACKNOWLEDGMENTS
previous work already know, and new readers of this book will learn, these
include the usual suspects, the members of the high-energy physics group at
the University of Colorado. They are always available to answer questions,
provide valuable discussions, and to inform me of recent developments
in the field. The experimentalists include Bill Ford, Kevin Stenson, Steve
Wagner, and Eric Zimmerman. In particular, I want to thank John Cumalat
and Alysia Marino, who read the chapters on the discovery of the Higgs
boson and on the Sudbury Neutrino Observatory (SNO) experiment that
solved the solar neutrino problem and checked them for any errors. Any
errors that remain are, of course, my responsibility. The theorists include
Shanta de Alwis, Tom DeGrand, Oliver DeWolfe, and Anna Hasenfratz.
I would also like to thank Gerald Holton and Michel Janssen, both of
whom read the chapter on the Michelson-Morley experiment and provided
helpful suggestions and whose work informed that chapter.
A significant portion of this work was done while I was the senior
visiting fellow at the Center for Philosophy of Science at the University of
Pittsburgh. I want to thank the center director, John Norton, and his staff,
Karen Kovalchick, Joyce McDonald, Cheryl Greer, and Yoichi Ishida, for
providing an atmosphere in which all I had to do was work on my own
research. John also put up with my numerous interruptions and puns and
provided very valuable discussions. I would also like to thank my fellow
fellows at the Center, Aristidis Arageorgis, Ori Belkin, Melinda Fagan,
Carrie Figdor, Lucia Foglia, Marco Giovanelli, Leah Henderson, Douglas
Kutach, Arnaud Pocheville, and Joshua Rosaler, for valuable discussions,
for their presentations and comments at our fellows’ meetings and at the
center seminars.
I also wish to thank Slobodan Perović and Kent Staley for carefully
reading the entire manuscript and for their valuable criticisms and
suggestions. I am also grateful to my editors at the University of Pittsburgh
Press, Abby Collier and Alex Wolfe. Their work made this book better than
my original manuscript.
Last, but certainly not least, I thank Cyndi Betts, my wife and best
friend. Without her encouragement and support, this work would not have
been possible.
WHAT MAKES A GOOD EXPERIMENT?
INTRODUCTION
One cannot take away integrity in the search for evidence and honesty
in declaring one’s results and still have science; one cannot take away a
willingness to listen to anyone’s scientific theories and findings irre-
spective of race, creed, or social eccentricity and still have science; one
cannot take away the readiness to expose one’s findings to criticism and
debate and still have science; one cannot take away the idea that the best
theories will be able to specify the means by which they could be shown
to be wrong and still have science; one cannot take away the idea that a
lone voice might be right while all the rest are wrong and still have sci-
ence; one cannot take away the idea that good experimentation or theoriza-
tion usually demand high levels of craft skills and still have science; and one
cannot take away the idea that, in virtue of their experience, some are
more capable than others at both producing scientific knowledge and at
criticizing it and still have science. These features of science are “essen-
tial,” not derivative.
—Harry Collins (2004, 156; emphasis added)
1
2 INTRODUCTION
ties; experiments that measure quantities that are of physical interest such
as Planck’s constant or the charge of the electron; and experiments that
have a life of their own, independent of high-level theory. An experiment
may also correct previous incorrect or misinterpreted results. Experiments
may also play a role in providing reasons for pursuit, the further investiga-
tion of a theory or a phenomenon. Yet another role for experiment is that
of an enabling experiment, an experiment that may give an incorrect result
but demonstrates that the quantity of interest can be measured.1 A related
role is that an experiment may demonstrate a successful new experimen-
tal technique. Following the suggestions by Pontecorvo (1960) and by
Schwartz (1960), a neutrino beam was constructed that led to the discov-
ery of two different kinds of neutrino (see chapter 10). The same beamline
technique is used in many neutrino experiments to this day. From this list,
it is clear that the roles of experiment include far more than only the testing
of theory. I do not, however, believe that this list of the varying roles that
experiments play is exclusive or exhaustive. As we shall see, a single exper-
iment can play several roles.
In that 1981 article, I also discussed various ways in which an exper-
iment can be good. I distinguished between “conceptually important”
experiments and “technically good” experiments. The former were classi-
fied primarily by their relationship to theory. Among the roles played by
conceptually important experiments are testing theory, deciding between
competing theories (crucial experiments), calling for a new theory, or
demonstrating that an existing theory is incorrect.
I might distinguish here between the ways in which an experiment can
be good and the role that it played from the attributes that it had. For ex-
ample, the Michelson-Morley experiment (1887; see chapter 16) was not
conceptually important in the genesis of Einstein’s special theory of rela-
tivity. The results did, however, at least in principle, call for either a new
theory or for a significant modification of existing theory. As Oliver Lodge,
a supporter of the ether theory remarked, “This experiment might have to
be explained away” (1893, 753). Similarly, Millikan’s (1916a) measurement
of Planck’s constant (see chapter 6) is, in retrospect and in textbooks, re-
garded as providing strong support for Einstein’s photon theory of light. At
the time, however, it was regarded as confirming Einstein’s photoelectric
equation.
Technically good experiments are those that measure a quantity of
physical interest with greater accuracy and precision than had been done
previously. My use of “technically good” is meant to apply either to an ex-
periment in which there have been previous measurements of the quantity
of interest and/or to those experiments in which there has been a signifi-
INTRODUCTION 3
[the standard cosmological model] in the range 30 < 1 < 150, incon-
sistent with the null hypothesis at a significance of > 5σ” (1). If correct,
this is an important result that confirms both inflationary cosmology and
the existence of gravitational waves “Although highly successful, the in-
flationary paradigm represents a vast extrapolation from well-tested re-
gimes in physics. It invokes quantum effects in highly curved spacetime
at energies near 1016 GeV and timescales less than 10 –32 s. A definitive
test of this paradigm would be of fundamental importance. Gravitational
waves generated by inflation have the potential to provide such a defini-
tive test” (Ade et al. 2014, 2). The experimenters remarked that inflation
theory predicts the existence of gravitational waves that would produce a
polarization pattern: “The detection of B-mode polarization of the CMB
[Cosmic Microwave Background] at large angular scales would provide a
unique confirmation of inflation” (2). Furthermore, the “observed B-mode
power spectrum is well-fit by a lensed-ΛCDM + tensor theoretical model
with tensor-scalar ratio r = 0.20+0.07–0.05, with r = 0 disfavored at 7σ” (1).13
A discussion of the five-sigma criterion also formed a significant part of
the discussion as to whether gravitational waves had been observed by the
Laser Interferometer Gravitational Wave Observatory (LIGO). The discus-
sion was complicated by the fact that the observed signal might have been
either a real signal or a blind injection, a simulated signal injected into the
data stream to test whether the analysis procedures would detect such a
signal (see Collins 2013 for the fascinating details).
I think it’s worth stating again that, despite the increasing presence of
the five-sigma criterion in physics, I do not believe that any fixed statistical
criterion should be a necessary requirement for a good experiment.
In this book I will revisit the question “What makes a good experi-
ment?” and will provide more extensive discussions of various exemplars
of the types of good experiment discussed above. These discussions will in-
clude the stated purpose of the experiment, how the results were used, and
the arguments given for the correctness of the results. My aim is to provide
a better and more extensive answer to that question than I did in 1981.
The set of experiments included in this book is not intended as a com-
pilation of the eighteen best experiments. One of them, Peter Thieberger’s
(1987a, 1987b) experiment on the Fifth Force (see chapter 17), produced a
result that is generally regarded as incorrect. Nevertheless, I will argue that
it is still a good experiment. These experiments were selected because they
illustrated both the various ways in which an experiment can be good and
because they also illustrate the different roles that experiment can play.
In presenting the experiments chosen, I have organized them into five
parts, according to what I believe is their primary role in science. These
8 INTRODUCTION
11
12 GREGOR MENDEL, “EXPERIMENTS IN PLANT HYBRIDIZATION”
(Mendel 1865). His stated purpose was to investigate whether there was a
general law for the formation and development of hybrids, something he
noted had not yet been formulated. He stated:
Those who survey the work in this department will arrive at the conviction
that among all the numerous experiments made, not one has been carried
out to such an extent and in such a way as to make it possible to determine
the number of different forms under which the offspring of hybrids appear,
or to arrange these forms with certainty according to their separate genera-
tions, or definitely to ascertain their statistical relations.
It requires indeed some courage to undertake a labour of such far-
reaching extent; this appears however, to be the only right way by which
we can finally reach the solution of a question the importance of which
cannot be overestimated in connection with the history of the evolution
of organic forms. (qtd. in Bateson 1909, 318) 2
Mendel proposed to remedy the situation and did so. His paper is a
model of both clarity and organization. As R. A. Fisher remarked, “Men-
del’s paper is, as has been frequently noted, a model in respect of the order
and lucidity with which the successive relevant facts are presented” (1936,
119).
In order to carry out such experiments successfully, Mendel required
that “the experimental plants must necessarily: 1. Possess constant differ-
entiating characters. 2. The hybrids of such plants must, during the flower-
ing period, be protected from the influence of all foreign pollen or be easily
capable of such protection. The hybrids and their offspring should suffer no
marked disturbance in their fertility in successive generations” (Bateson
1909, 319). He further noted that “in order to discover the relations in
which the hybrid forms stand towards each other and towards their pro-
genitors, it appears to be necessary that all members of the series developed
in each successive generation should be, without exception, subjected to
observation” (319).
TABLE 1.2. MENDEL’S RESULTS FOR THE FIRST TEN PLANTS IN THE EXPERIMENTS ON SEED SHAPE AND
SEED COLOR (Mendel 1865, in Bateson 1909, 327)
EXPERIMENT 1 EXPERIMENT 2
shape of seeds coloration of albumen
plant round wrinkled yellow green
1 45 12 25 11
2 27 8 32 7
3 24 7 14 5
4 19 10 70 27
5 32 11 24 13
6 26 6 20 6
7 88 24 32 13
8 22 10 44 9
9 28 6 50 14
10 25 7 44 18
Note: The fact that the number of seeds in each plant differs for each numbered
plant shows clearly that the plants for Experiments 1 and 2 are different plants. Thus,
plant 1 in Experiment 1 has 57 seeds, whereas plant 1 in Experiment 2 has 36 seeds.
These two experiments are important for the determination of the average
ratios, because with a smaller number of experimental plants they show
that very considerable fluctuations may occur. In counting the seeds, also,
especially in Expt. 2, some care is requisite, since in some of the seeds
of many plants the green colour of the albumen is less developed, and at
first may be easily overlooked. . . . In luxuriant plants this appearance was
frequently noted. Seeds which are damaged by insects during their devel-
opment often vary in colour and form, but, with a little practice in sorting,
errors are easily avoided.” (327–28)
TABLE 1.3. MENDEL’S RESULTS FOR THE HETEROZYGOUS-HOMOZYGOUS EXPERIMENT (THE 2:1
EXPERIMENT) (Mendel 1865, in Bateson 1909, 330)
experiment dominant hybrid
3. seed coat color (grey-brown or white) 36 64
4. pod shape (smooth or constricted) 29 71
5. pod color (green or yellow) 40 60
6. flower location (axillary or terminal) 33 67
7. stem length (long or short) 28 72
8. repetition of Experiment 5 35 65
total 201 399
nal Experiment 5 as well as its repetition. The sum totals for the six plant
characteristic experiments, including the repetition of Experiment 5, was
399 (hybrid) to 201 (dominant) or 1.99 to 1. Mendel’s conclusion was quite
clear:
The ratio of 3 to 1, in accordance with which the distribution of the domi-
nant and recessive characters results in the first generation, resolves itself
into a ratio of 2:1:1, if the dominant character be differentiated according
to its significance as a hybrid-character or as a parental one: “Since the
members of the first generation [F2] spring directly from the seed of the
hybrids [F1], it is now clear that the hybrids form seeds having one or the other
of the two differentiating characters, and of these one half develop again the hy-
brid form, while the other half yield plants which remain constant and receive
the dominant or the recessive characters, [respectively], in equal numbers.”
(330–31)
generations, “and no departure from the rule has been perceptible. The off-
spring of the hybrids separated in each generation in the ratio of 2:1:1 into
hybrid and constant forms [pure dominant and pure recessive]” (331). He
did not, however, present his data for the experiments on the subsequent
generations.10 He went on to state, “If A be taken as denoting one of the two
constant characters, for instance, the dominant, a, the recessive, and Aa the
hybrid form in which both are conjoined, the expression A + 2Aa + a shows
the terms in the series for the progeny of the hybrids of two differentiating
characters” (331).
OÙ L’ON VA…
CHAPITRE I.
CONSULTATION 5
CHAPITRE II.
LES DÉBUTS DE PAMPHILE 11
CHAPITRE III.
L’AMATEUR 17
CHAPITRE IV.
LA PROFESSION « SECONDE » 22
CHAPITRE V.
PREMIERS ESSAIS, PREMIERS ÉCHECS 28
CHAPITRE VI.
EXPÉRIENCES PERSONNELLES 34
CHAPITRE VII.
LE CONTE 39
CHAPITRE VIII.
DU JOURNALISME 44
CHAPITRE IX.
TYPES DE JOURNALISTES 50
CHAPITRE X.
POLÉMIQUES LITTÉRAIRES CONTEMPORAINES 55
CHAPITRE XI.
UNE OPINION POLITIQUE POUR L’ÉCRIVAIN 59
CHAPITRE XII.
ESPOIRS ET REGRETS 64
CHAPITRE XIII.
VACHES GRASSES ET VACHES MAIGRES 69
CHAPITRE XIV.
PUBLICITÉ LITTÉRAIRE 75
CHAPITRE XV.
LA CRITIQUE 80
CHAPITRE XVI.
PRIX LITTÉRAIRES 86
CHAPITRE XVII.
L’ÉCRIVAIN ET L’ARGENT 91
CHAPITRE XVIII.
LE MARIAGE DE L’ÉCRIVAIN. L’ÉCRIVAINE 97
CHAPITRE XIX.
SALONS LITTÉRAIRES 101
CHAPITRE XX.
L’ÉCRIVAIN ET L’ACADÉMIE 107
CHAPITRE XXI.
OÙ L’ON VA… 118
IMPRIMERIE CRÉTÉ
CORBEIL (S.-ET-O.)
5527-25
*** END OF THE PROJECT GUTENBERG EBOOK L'ÉCRIVAIN ***
Updated editions will replace the previous one—the old editions will
be renamed.
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside the
United States, check the laws of your country in addition to the terms
of this agreement before downloading, copying, displaying,
performing, distributing or creating derivative works based on this
work or any other Project Gutenberg™ work. The Foundation makes
no representations concerning the copyright status of any work in
any country other than the United States.