13
CHAPTER
Argumentation
in Science
KEY TERMS
science, p. 204
hypotheses, p. 206
theory, p. 206
scientific law, p. 206
natural order, p. 207
claims of fact, p. 207
peer review, p. 209
empirical, p. 209
replicate, p. 209
hypothetico-deductive method, p. 210
positionality, p. 210
generalization, p. 211
retroduction, p. 211
conditional cause, p. 213
homology, p. 215
specific instances, p. 217
decision rule, p. 218
testimony, p. 219
operationalizing, p. 220
For many people, scientific methods stand as the most competent way to understand
what is going on in the world. Scientific standards for evidence and argument are
held up as the way to understand what the natural world is like. Arguments that
fail such tests are easily disregarded, not only by the scientists who work in the sphere but
by lay persons as well. The sphere of science has great credibility in our society, and an
examination of its understanding of evidence and argument will provide insight into the
standards people frequently seek in public arguments.
WHAT IS SCIENCE?
M13_RIEK0596_08_SE_C13.indd Page 204 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 204 1/3/12 3:26 PM user-f501 F-401F-401
CHAPTER 13 Argumentation in Science 205
tells itself about the cosmos.” It you think back to Chapter 3 , you may recall our statement
that stories, or narratives, are important forms of argument. The science narrative provides
a supposedly “objective account of the material world based upon measurement and quan-
tification so that structure, process, movement, and transformation can be described math-
ematically in terms of fundamental laws” (208). Most of the scientific endeavors that fall
within this account are quantitative, using physical science as a model and mathematics as
Some postmodern and feminist scholars have challenged the scientific narrative de-
scribed in the previous paragraph. Their critique argues that, because scientists are mem-
bers of society, scientific knowledge cannot be the outcome of completely neutral and ob-
jective rational thought. Therefore, the understanding gained through science is no more
accurate than understanding gained through other approaches such as astrology or palm
reading. From this perspective, science is a game with a set of rules created by scientists,
and apparent successes of science in understanding the universe would not be defensible
if society did not accept the rules of the scientific game. Postmodern critiques of science
sometimes question whether a natural world exists outside of the mental constructs that
humans erect. According to these critiques, science is no more than an elaborate social
construct dedicated to maintaining existing patterns of hegemony.
Less extreme forms of postmodernism and feminism shift from challenging whether
objective reality exists to challenging “magical notions of scientific objectivity” (Peterson
et al. 2007, 75). This line of argument claims that everyone’s—including scientists’—in-
terpretation of reality is partial, and is influenced by their perspective. They have argued
for more rigorous self-examination by scientists, an activity that is in harmony with sci-
ence’s basic tenets. For example, feminist critiques of science have argued that scientific
theory and practice marginalize women. In response to this claim, the National Science
Foundation has developed educational programs that affirmatively encourage young girls
to study science and mathematics, has developed grant programs exclusively for women
scientists, and has hosted seminars designed to encourage feminist scholars and traditional
scientists to engage in critical discussions about the philosophy of science. The story of
This does not mean science has given up its search for understanding, nor does it
translate into science as social construct. If you fall down the stairs, you are likely to get
at least a few bruises. If you spend all of Saturday and Sunday partying, you will be less
able to comprehend the 8:00 a.m. lecture on Monday morning than if you had taken time
to sleep over the weekend. Notice that the expected results in these two examples include
uncertainty—as does all scientific prediction. Although this chapter focuses on the use of argu-
mentation in science as defined by the traditional markers of objectivity and quantification
M13_RIEK0596_08_SE_C13.indd Page 205 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 205 1/3/12 3:26 PM user-f501 F-401F-401
206 PART 3 Applications
as a means of discovering fundamental laws, it is important to realize that the values of
objectivity and quantification are merely markers, and that discovery is the goal.
The sphere of quantitative science can sometimes be identified by academic depart-
ments (e.g., physics, chemistry, geology, biology) in the natural sciences. Many social
sciences departments (e.g., anthropology, communication, psychology) include both
quantitative and qualitative science. The science that we discuss in this chapter ranges over
a wide variety of fields with the physical sciences as its model.
In this chapter, we examine not the political and legal roles of scientists, but what
kind of argument and evidence they use in scientific journals, research papers, and grant
applications. Although scientists frequently argue in the public sphere as experts, we look
here at how they argue where other scientists are the decision makers. We will examine
how science is integrated into political argument in Chapter 16 .
SCIENTIFIC VALUES
One reason science presents such a persuasive narrative (as noted in Chapter 3 ) is that it
appears consistent with many of society’s core values, which we discussed in Chapter 8 . The
values of scientific study include knowledge, order, prediction, rationality, and usefulness.
The individual values in this suite work together within scientific argumentation, although
many of the values remain unstated. Sir Francis Bacon (1561–1626), who was one of the
leading figures in the development of scientific philosophy and method, saw science as a
means of constructing a better world for humans, through understanding natural truths.
Although much has changed in the ensuing centuries, science in the twenty-first century
remains consistent with this goal. Science begins from the premise that, where knowledge
is concerned, more is better. This knowledge enables people to discover natural order .
Once we understand the proper order of things, it becomes easier to predict future events,
which should enable us to make more rational choices. All of this is important because it
is useful for improving the human condition.
M13_RIEK0596_08_SE_C13.indd Page 206 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 206 1/3/12 3:26 PM user-f501 F-401F-401
CHAPTER 13 Argumentation in Science 207
Even if the natural order cannot be directly observed (i.e., you cannot observe a quark,
a neutron, or an attitude), scientists still require that there be empirical adequacy . That is,
the signs of the phenomenon must be observable. The procedures for finding these claims
of fact must be clearly defined so that they may be replicated or questioned. These claims
of fact are linked together to provide theoretical propositions of explanation. Such propo-
sitions are in turn used to predict another specific situation that has not been observed.
That is, the signs of the phenomenon must be either directly or indirectly observable.
From this perspective, a theory about climate change is built from careful observa-
tion of the geological record. The case of global climate change is a useful place to begin
our discussion of how the scientific method functions argumentatively because there is
widespread interest in this issue. In this chapter, we will limit the discussion to argument
within the scientific community. In Chapter 16 , we will show how the argument changes
when it moves into the sphere of government and politics.
Global climate change is a complex phenomenon that is planetary in scope and oper-
ates on a timescale that exceeds seasons, political terms, and the human life span. Global
climate change differs from weather in both its spatial and temporal expanse. During the
twentieth century, the earth’s annual mean temperature increased by about 2°F. Given this
small number, you may be wondering what all the fuss is about. After all, the temperature
fluctuates more than that between noon and midnight, and between winter and summer.
Physiological mechanisms have evolved that enable living things to adjust to short-term
(easily up to a year) and localized changes in temperature. The problem is, however,
that a small change is a serious matter to the global climate system. Temperatures that
THE TRADITION OF ARGUMENTATION IN SCIENCE
There are several ways that the climate change debate helps to define the tradition of
argumentation in the science sphere, four of which serve as a preliminary definition of sci-
entific argumentation. They are that science (1) deals in claims and propositions of fact,
(2) searches for truth over personal gain, (3) reveals results that are complete enough to
test, and (4) establishes theory that changes slowly.
Claims of Fact
First, traditional scientific argument, in its central concern, argues claims and propositions
of fact. The focus on facts is especially consistent with the scientific values of knowledge
M13_RIEK0596_08_SE_C13.indd Page 207 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 207 1/3/12 3:26 PM user-f501 F-401F-401
1. What kinds of things are there [or were there] in the world of nature?
2. How are [or were] these things composed, and how does this makeup affect their behavior
or operation?
3. How did all these things come to be composed as they are [or were]?
4. What are the characteristic functions of each such natural thing and/or its parts? (315)
(1) Is it occurring? (2) What has been causing it? (3) Does it pose dangers or problems to
human society as a whole, as well as to specific segments of human society? (4) How seri-
ous are those dangers? And (5) What can be done to either slow or stop the warming trend?
1. Has earth been warming over the past few centuries?
2. How has the rate of warming changed over time?
3. What is causing the warming?
4. What have been and will be the consequences of global warming?
5. What can be done to remediate the change?
By the early 1980s, most scientists had answered the first question affirmatively and had
begun to explore the details of the others. At the beginning of the twenty-first century,
there is a strong consensus among scientists that the earth has been warming during the
past few centuries; that human activities associated with industrialization have caused a
significant increase in the rate of warming since about 1850 (with another sharp increase
about 1950); that consequences of this change already are being felt by some segments
of society, while other consequences are expected; and that these consequences will have
negative impacts on most segments of society. Scientists also tend to agree that human
Search for Truth over Personal Gain
Scientists are not supposed to act for personal gain, but for the betterment of society.
This is especially consistent with the scientific values of knowledge and usefulness; given
that knowledge refers to true understanding of nature, and usefulness refers to improv-
ing the human condition. Yet public identification of discoveries has been, since the early
CHAPTER 13 Argumentation in Science 209
eighteenth century, the basis on which scientific achievement is credited (Gross 90). Robert
K. Merton identified the “paradox at the heart of the scientific enterprise” years ago:
While the general progress of scientific knowledge depends heavily on the relative subordina-
tion of individual efforts to communal goals, the career progress of scientists depends solely on
the recognition of their individual efforts. (Gross 89)
The paradox has always been there, yet scientists are expected to have their work sub-
jected to peer review in which evidence and argument are tested by the scientific value of
developing new knowledge, not the professional advancement of the scientist. This is one
reason the scientific community is leery of climate research funded by the petroleum in-
dustry, which has lobbied successfully to prevent U.S. energy and environmental policies
from responding to mainstream science on climate change. For example, Dr. Willie Soon
Testable Results
Science exists, according to its own rules, in an atmosphere of the free exchange of ideas,
and to withhold information inhibits scientific progress. This claim is especially congruent
with the scientific values of prediction and order. Probably most important, this argument
calls attention to the fact that science is far more than a collection of observations or theo-
ries. Science is a comprehensive system of empirical knowledge building. So, the theory
revealed, the methods followed, and the evidence used are all part of a comprehensive
Established Theory Changes Slowly
Scientific theory evolves slowly over time, and requires significant argumentation to
achieve change. The resistance to sudden change relates directly to the scientific value
of order, in that it requires overwhelming evidence to change established patterns. Even
Thomas Kuhn, who used the term scientific revolution to characterize major changes in
theory, agrees that there is no sudden overturning of theory (Suppe 135). The replace-
ment of Newtonian physics by the theory of relativity did not come suddenly when Albert
Einstein said “ E = mc
2 .” There was a continual building of the theory, as one anomaly
after another was found in Newtonian physics. Newtonian mechanics were being disman-
tled for many years before Einstein. Likewise, the theory of evolution had been around
for some time before Charles Darwin. He provided evidence and the unifying explanation
M13_RIEK0596_08_SE_C13.indd Page 209 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 209 1/3/12 3:26 PM user-f501 F-401F-401
210 PART 3 Applications
and got credit for it. These were not scientific revolutions; rather, they were theoretical
statements that built on, and made sense of, previous theory and findings.
One reason established theory changes so slowly is that the scientific or hypothetico-
deductive method requires scientists to ground any new research in past research. Unlike
the descriptive method, the scientific or hypothetico-deductive method focuses on hypoth-
esis testing and the use of a methodology (usually statistical) designed to maximize ac-
curacy in the interpretation of research results. This approach requires scientists to move
through a predetermined series of steps, all of which should be quite clearly described for
fellow scientists. This method encourages scientific endeavor to focus on questions of fact
FIGURE 13.1
Hypothetico-deductive Argumentation
1 . Identify the research problem.
2 . Conduct literature review.
3 . Identify broad research objectives.
4 . Collect preliminary data if needed.
5 . Conduct exploratory data analysis.
6 . Formulate research hypotheses.
7 . Formulate testable (usually statistical) hypotheses.
8 . Design methodology.
9 . Prepare research proposal.
10 . Obtain peer review and revise.
11 . Perform experiment, collect data.
12 . Analyze data.
Scientists observe physical, biological, human, or social phenomena for factual claims.
These claims are combined with existing knowledge to form theories that serve as general
laws or rules about the natural condition. The theories develop in a system of peer review
where others can see the claim, the evidence, and the method of argumentation and test
them to confirm or deny. Because theories are built up of many subtheories and empirical
confirmations, they are not, according to scientific tradition, easily overturned. New major
M13_RIEK0596_08_SE_C13.indd Page 210 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 210 1/3/12 3:26 PM user-f501 F-401F-401
CHAPTER 13 Argumentation in Science 211
SCIENTIFIC USE OF ARGUMENT TYPES
All six of the types of arguments discussed in Chapter 4 appear in scientific argument:
logic, generalization, cause, sign, analogy, and authority. Generalization, cause and sign—
particularly cause and sign—are the most important. In addition, scientists apply each of
these types by their own value system based on empiricism, logic, and mathematics.
Argument by Generalization
In one sense, all scientific argument is by generalization . The goal of such argument is to
make observations that will explain a class of phenomena. Those explanations general-
ize about how individual cases behave or about what properties individual cases have in
common. Until well into the nineteenth century, using induction or experiments to form
generalizations that would serve as theories to explain the natural world was the domi-
nant tradition of science known as Baconianism after Francis Bacon, its chief architect
(Campbell, Man Cannot Speak 500). Generalization in modern science functions by what
C. S. Peirce first called abduction , and later referred to as retroduction . Because scientists
more commonly use the second term, we will use it in this chapter. Retroduction refers to
developing a hypothesis that would, if true, best explain a particular set of observations.
Retroductive reasoning begins with a set of observations or facts, and then infers the most
likely or best explanation to account for these facts. Consider the following: All the eggs
in a bird nest disappeared overnight. There were no shell fragments, animal tracks, or dis-
Although both deduction (see Chapters 3 and 4 ) and induction (see the discussion of
examples in Chapter 7 ) are important to scientific argumentation, retroductive reasoning
is, in many ways, the most interesting because it is more likely to result in novel explana-
tions for puzzling phenomena than are induction or deduction. It also is much more likely
to be wrong! Inductive reasoning is an effective way to derive important principles of as-
sociation and is less likely to prove incorrect than retroduction. It has been the workhorse
of science for centuries. Deductively derived conclusions are uninteresting in themselves;
after all, they follow deterministically from the major premise. Instead, the value of de-
ductive reasoning is that it allows scientists to devise ways to critically challenge and
evaluate retroductively developed hypotheses.
So, the scientist begins with a hypothesis formed from limited cases. This hypothesis
may be a valid generalization. But the generalization must be tested for what it might be
M13_RIEK0596_08_SE_C13.indd Page 211 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 211 1/3/12 3:26 PM user-f501 F-401F-401
212 PART 3 Applications
expected to show. To some extent, this is a matter of replication (repetition of an experi-
ment to see whether the results are the same).
Although tests are part of the generalization process, they are not argument by gen-
eralization themselves, but argument from sign, as we will explain shortly. Scientific ar-
gument should build theories that will provide the scientific community with the most
rational explanation of the natural order. This requires extensive experimentation and
evidence, but generalization is only part of the process. Perhaps more important is the
examination of anomalies in the theory. Toulmin, Rieke, and Janik provide an example
from weather forecasting:
Weather forecasting, for instance, presents some serious challenges to science, to find ways of
squaring the observed course of meteorological events with the accepted principles of physical
science. But that does not mean that scientists feel any responsibility for explaining every last
day-to-day or minute-to-minute change in the weather. Presumably, such changes are brought
Argument by generalization is, on the face of it, crucial to science. The generaliza-
tions (theories) require testing to make them more powerful. That testing is not always
by replication. It will more often rely on the next two kinds of argument: cause and sign.
Argument by Cause
The assumption of science is that there is order in nature and that order is held together
by cause-and-effect relationships. High- and low-pressure changes cause changes in the
weather. Changes in the social order cause changes in the way individuals live their lives.
Once climate scientists began to focus on causes of climate change, they discovered that
natural factors dominated climatic fluctuations up to somewhere between the years 1750
and 1850. From about 1850 to 1950, human factors associated with industrialization grew
to sufficient potency to rival natural factors, leading to climate variability derived from a
M13_RIEK0596_08_SE_C13.indd Page 212 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 212 1/3/12 3:26 PM user-f501 F-401F-401
CHAPTER 13 Argumentation in Science 213
This requirement of approaching certainty poses an increasing problem in all sciences,
but particularly in the human sciences. No cases of human behavior meet certainty stand-
ards. Consequently, the human sciences use a more open statistical probability as the basis
for judging cause, called a conditional cause . The claim is the best explanation, but it is
conditional because it is supported at a statistical level that admits of some evidence to the
Self-ratings for argumentative trainees and other participants were compared by t -test, with
results showing that trainees perceived themselves as significantly more self-assured ( t = 2.78;
df 36; p < .01), more goal-oriented ( t = 2.65; df 36; p < .01), more quarrelsome ( t = 2.85;
df 36; p < .01), and in the direction of significance for summarizing ( t = 1.81; df 36; p = <.08).
Untrained participants did not significantly alter their ratings after the second session. (560)
Before the training, there was no difference between the experimental group and the
control group, but after the experimental group was trained, they showed a significant
difference in the perception of their leadership qualities versus the control group. The
only difference between the two was the training that became the necessary and sufficient
If the design of the experiment was perfect, the elimination of chance as an explana-
tion should leave only the “treatment” (training) as the cause. But this experiment does
not prove that the training caused everyone to change as would be expected in a tradi-
tional understanding of cause. As a matter of fact, some participants changed, some did
not, and a few probably regressed. Statistically, however, the total group changed. The
change was dependent on an unknown characteristic, so we can say there is cause but the
cause (training) is conditional.
Scientists look for the necessary and sufficient causes of phenomena. There is always
some question, but certain causal relationships, particularly in the physical and biological
sciences, come closest to producing the ideal relationship between cause and effect. For
much of science, a conditional relationship is the best that can be expected.
Argument by Sign
A major way to test a theory is to look for observable phenomena that the theory predicts
should be there. The theory of global warming relies heavily on sign argument. Because
the theory is based on propositions of past fact, the types of evidence all are signs of
what has happened throughout earth’s history. The Intergovernmental Panel on Climate
M13_RIEK0596_08_SE_C13.indd Page 213 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 213 1/3/12 3:26 PM user-f501 F-401F-401
214 PART 3 Applications
Change noted several signs to support its argument. For example, the global average sea
The argument from sign becomes particularly crucial in the human sciences. All sur-
vey research is a sign argument. The sample of the population is taken as a sign of the
whole population. It is also an important part of experimental research in the social sci-
ences. For instance, Cynthia Hoffner and Joanne Cantor wanted to find out what factors
affect “children’s enjoyment of a frightening film sequence.” They studied 5- to 7-year
olds and 9- to 11-year olds. The children viewed a sequence from Swiss Family Robinson
in which two brothers encounter a snake. The researchers varied the introduction and the
ending to provide either a threat or happy circumstances.
First, note that a video of the Robinson boys encountering a snake is taken to be a
sign of a frightening film sequence. After viewing the sequence, the children were asked if
they felt happy, scared, or just OK. They were also asked a number of questions such as
how worried, scared, and so on they were. Note that what the children said is taken as a
sign of what they actually felt.
Social scientists cannot show you “enjoyment” or “frightening,” and they cannot
show you an “attitude,” “violence,” “communication conflict,” or “deception.” They
must test their theories against things, events, or behaviors that are signs of those ab-
stract concepts. In Chapter 2 , we talked about worldviews that some people believe guide
human mental processes. However, no one has ever seen a worldview, held one up to the
light, or poked it with a finger. So, what makes the concept believable? Worldviews are
accepted because numerous studies using outward signs point to the existence of such
organizing principles.
Argument by Analogy
Scientists usually argue from generalization, cause, and sign. They also use analogy. In
one sense, all argument is by analogy because claims are not the same as the incidents,
things, or beings on which they are based. For example, a sign can be seen as an analogy
All I need to do is decide between which two marks on the meter stick the right end of the table
lies. Deciding whether two points are coincident boils down to making judgments about the
distance between two points. Point A on the table and point B on the meter stick cannot liter-
ally be coincident, for two objects cannot occupy the same place at the same time. (21)
M13_RIEK0596_08_SE_C13.indd Page 214 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 214 1/3/12 3:26 PM user-f501 F-401F-401
CHAPTER 13 Argumentation in Science 215
In that sense, even measurement is an analogy. Analogy serves in this system as an expla-
nation. There are quarks in physics, DNA in genetics, and attitudes in communication.
They help explain phenomena that cannot literally be seen.
In the scientific tradition we are examining here, analogy, although useful as expla-
nation, is not as useful in making an empirical argument. Because scientists try to under-
stand the natural world, there is a problem with the comparison of two things that are
not the same. This problem applies particularly to figurative analogy, as in the example
(in Chapter 4 ) in which undocumented immigrants are compared with snowflakes. Even
a literal analogy, such as the one drawn between the oil industry’s treatment of poor
African-Americans in Norco, Louisiana, and its treatment of aboriginal inhabitants of
Ogoniland, Nigeria, has its problems. There are at least as many differences as similarities
between the treatments accorded the two populations. It is very unlikely that residents
of Norco would be hanged for organizing a peaceful protest, even if it caused a work
Although analogy is not as forceful a scientific argument as the three we discussed
before, analogy is argued, particularly in the biological sciences. The term used for an
analogous relationship in biological sciences is homologous , which means there are ex-
John Lyne and Henry F. Howe use the example of E. O. Wilson, who moved far from
his own area of expertise as an entomologist (“his publications prior to 1971 concern
such topics as chemical communication among ants, and castes within insect societies”).
certain baboon behavior” provides a basis for an analogy in public argument but not the
homology necessary to a geneticist (142). It is probably most significant that sociobiology
has had considerable popularity outside biology in the public sphere. Decision makers in
the public sphere hold analogy to less rigid standards than scientists do.
However, scientists interested in the effect of carcinogens on rats believe there is a ho-
mology between rats and humans. To develop statistical analyses of the effects of smoking
and cancer on humans when large amounts of tars and nicotines are ingested in a short pe-
M13_RIEK0596_08_SE_C13.indd Page 215 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 215 1/3/12 3:26 PM user-f501 F-401F-401
216 PART 3 Applications
Argument from Authority
An argument warranted by the authority of scientific principle is considerably different
from argument from authority in public argument. In public argument, you might use
authority to prove a claim (“Lower interest rates stimulate the economy, according to the
chair of the Federal Reserve Board”). A scientist, however, is likely to use something more
like a refutational argument. For instance,
2. The countertheory has little evidence for its position.
3. Therefore, the established theory is still valid.
Whether as a constructive claim or a refutational one, such an argument does not
look like the typical argument from authority found in public argument (Albert Einstein is
a credible scientist, so we can trust his theory E = mc
2 ). For scientists making arguments
in the presentation of scientific arguments in scholarly papers and journal articles, you
will not find the argument “ X is true because Y said it.”
Of course, each scientific paper includes a review of the literature in which the scientist
identifies the significant findings to date and shows how the current research fits with it.
The review of the literature gives the appearance of argument from authority because the
cause, and sign. Its use of generalization is somewhat different from the usual understanding of
that term because it is oriented to testing theory rather than simply replicating it. Scientific
arguments depend particularly on cause and, to a lesser extent, sign. Argument by anal-
ogy is used mostly as a method of explanation. In empirical argument it is rare, found in
a form more like generalization. Argument by authority is considered insufficient, although
M13_RIEK0596_08_SE_C13.indd Page 216 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 216 1/3/12 3:26 PM user-f501 F-401F-401
CHAPTER 13 Argumentation in Science 217
SCIENTIFIC USE OF EVIDENCE
There are three forms of evidence, as we noted in Chapter 7 : examples, statistics, and tes-
timony. All can be found in scientific argumentation. We noted earlier that science is not
simply a collection of observations or theories. It is a comprehensive system of empirical
theory building. As we begin to examine the nature of scientific evidence we must look at
that term empirical more carefully.
Empirically Grounded Claims
Traditionally, to be empirically grounded means that a claim must be based on sensory
experience. Scientific explanations are empirical arguments when the evidence can be
seen, heard, touched, smelled, or tasted. That understanding seemed reasonable in earlier
centuries when our scientific theories were limited by our immediate senses, augmented
Specific Instances
Specific instances provide the empirical grounding for a scientific claim. That should be
obvious. Colonies of bees, strata of rocks, actions of individuals all provide the empiri-
cal bases for forming generalizations about those phenomena that lead to hypotheses.
Further examination of other instances serves to replicate, modify, or reject a theory.
A New York Times article about climate change research illustrates how several spe-
slightly declines after 2050; and a shift in energy choices away from fossil fuels. Another
set of scenarios assumes slower diffusion of new technologies, continuation of existing
fertility patterns, and energy usage remaining dependent on fossil fuels. As you can im-
agine, each of these combinations of economic development, population trajectory, and
statistical probability and are always open to further questioning.
Statistics in Science
We noted in Chapter 7 that statistics are essentially a numerical compacting of specific
instances. They provide a means of talking about many specific instances without citing
every one. That approach to explanation is common for most public, even legal, argu-
1. To quantify a set of observations into a set of numbers . This is the descriptive use
of statistics. Statistics reveal the central tendency of the numbers or the averages (mode,
2. To determine whether the sample is representative (a sign) of the population from
which it was drawn . A poll on your campus says that 58% of the students favor national
health insurance. Were the students questioned representative of your student body?
3. To determine by a decision rule whether the characteristics found can be attributed to
an error in sampling (the null hypothesis) or, if not, whether an alternative explanation,
usually the hypothesis, is confirmed (Anderson 175–176).
Recall the study of children and the film sequence from Swiss Family Robinson . The
researchers used a statistical test to determine how significant the results were.
The first task is to quantify observations into a set of numbers and relationships that
will tell the researcher how to describe a population. The first task concerns measurement
much like the general use of statistics discussed in Chapter 7 . The second and third tasks are
concerned with meaning. These statistics are used to develop knowledge about a population.
M13_RIEK0596_08_SE_C13.indd Page 218 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 218 1/3/12 3:26 PM user-f501 F-401F-401
CHAPTER 13 Argumentation in Science 219
For these tasks, a decision rule is necessary to decide what is and what is not worth know-
ing. For the scientist, tasks two and three are at least as important as task one.
A look at these three tasks illustrates that there are assumptions that cannot be
proven but must be taken as givens. For example, in measurement a primary assumption
is that the numbers represent the phenomena. If not, then one’s statistical descriptions
have no real meaning. Another is that there is such a thing as a representative sample. If
not, studies of public opinion, the behavior of chimpanzees, the effect of carcinogens on
Testimony
Testimony is a form of evidence that can stand alone as grounds for an argument from
authority. For instance, you could argue that because E. O. Wilson is a respected scien-
tist, his theory that there are genes for moral principles should be accepted. But Lyne and
Howe criticize that view because his theory is outside his expertise and, more important,
he does not present sufficient evidence for the theory. Is there a place for testimony in
scientific argument? The answer in the human sciences is yes. In such human sciences as
psychology, communication, sociology, anthropology, and political science, testimony is
at least a significant basis of evidence. It is used, however, in a special way.
Sometimes the evidence is taken from a set of categories that people judge. For in-
stance, in Chapter 8 , we introduce a series of value terms by Milton Rokeach to repre-
sent the 18 major terminal and the 18 major instrumental values. People are given one
or both lists and asked to rank-order them, and the group’s composite response is taken
as a hierarchy of that group’s values. Like Benoit and Benoit, Rokeach used open-ended
testimony to discover the values. But both studies rely on testimony in defining categories
and in forming generalizations about their use by a group. The evidence is testimony (and
hangs on the scientist’s assumptions that the testimony is both true and real). It reflects
the opinion and knowledge of the persons engaged in the experiment.
What you can see from this brief summary is that all three of the evidence forms are
found in scientific argument. Statistics are clearly the most critical. Specific instances and
testimony (in the human sciences) are the raw material from which statistical inferences
are made. Statistical inferences are essentially linked to argument by generalization, cause,
sign, and sometimes analogy.
M13_RIEK0596_08_SE_C13.indd Page 219 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 219 1/3/12 3:26 PM user-f501 F-401F-401
220 PART 3 Applications
SCIENTIFIC METHOD AS ARGUMENT
The steps of the hypothetico-deductive method are intended to guide researchers in produc-
ing the most credible argument possible. Identifying the research problem requires scientists
to use generalization, cause, sign, and sometimes analogical forms of argument. For example,
when scientists argue that increased levels of CO
2 lead to global warming, they are claiming
that CO
2 is a cause , and global warming is its effect. When they provide evidence that glaciers
in South America are retreating, that the ice sheet in arctic regions is thinning, and that sea
levels are rising, they are claiming that these events are signs of global warming. When they
explain that global climate change is a complicated system response to multiple individual
Identifying research objectives and formulating hypotheses are fundamental to the
values espoused by science. In Chapters 8 and 9 , we discussed values and credibility.
Science values both order and the discovery of new knowledge. You may recall we em-
phasized that values function systematically, rather than individually. Without the ne-
cessity of identifying specific objectives and formulating testable hypotheses, the discov-
ery function of science could easily destroy its orderliness. Further, without these steps,
scientific replication would be impossible. Formulating hypotheses in a way that allows
someone to test them also is referred to as operationalizing them. When you operational-
ize something, you provide a way to measure, or evaluate it. Most commonly, scientists
operationalize hypotheses by identifying the statistical tests they will use, as well as the
rejection range. The rejection range operates as a decision rule.
Suppose you wanted to know whether class attendance was correlated with the grades
students earn in a class. You might hypothesize that students who attended more classes
would earn higher grades. You could design a methodology that enabled you to compare
daily attendance with final grades for all students. To make your hypothesis testable, you
would need to decide what statistical test you would use to make this comparison. Further,
CONCLUSION
As examined in this chapter, science refers to the “objective account of the material world
based upon measurement and quantification so that structure, process, movement, and
transformation can be described mathematically in terms of fundamental laws” (Peat 208).
Scientists often use the physical sciences as a model and mathematics as a foundation for
explanatory theories. Scientists play a significant political role, but here we are interested
M13_RIEK0596_08_SE_C13.indd Page 220 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 220 1/3/12 3:26 PM user-f501 F-401F-401
CHAPTER 13 Argumentation in Science 221
in how they argue to one another as scientists. Scientific study begins with the value that
there is order in nature that can be discovered. Order is explained through observation,
and observations are characterized in claims of fact. They are combined with other already
acknowledged claims of fact to develop a proposition that is a theory about fact. That
theory can then be used to predict what will happen at another time or in another place.
Argument by cause is basic to scientific argumentation. When study reveals some pre-
viously unrecognized condition, the scientist wants to know its cause. To be established
in theory, it must be both necessary and sufficient to produce the effect. Argument by sign
is important in establishing a scientific theory. Signs in biology indicate that an organism
belongs to a particular species. All survey research is sign argument, as is most social sci-
ence argumentation.
Argument by analogy is used primarily to explain a phenomenon. To be the basis for
an empirical argument, the analogy must have extensive similarities. In biology, it must
be homologous. Argument from authority is the least used type of argument in science.
The review of literature sections of research papers and articles and the authority of es-
tablished theories function as a kind of argument from authority. Outward statements
of authority as a basis for argument are not made, although they may be used in private.
EXERCISES/PROJECTS
1. Interview a faculty member at your college or university who would be considered a scientist as
we have defined a scientist in this chapter. Ask questions about the kinds of arguments and evi-
dence he or she uses with peers (not those that might be used to convince non-scientists). Write
a short paper (no more than five double-spaced pages) about what kinds of argument and
evidence apply. Does the person you interviewed agree with what has been said in the chapter?
How different is the interviewee’s position from that in this chapter? Why do you suppose the
difference exists?
M13_RIEK0596_08_SE_C13.indd Page 221 1/3/12 3:26 PM user-f501M13_RIEK0596_08_SE_C13.indd Page 221 1/3/12 3:26 PM user-f501 F-401F-401