How Science Works

How Science Works


This video is about science.
In particular, about the role culture and community in science, how they influence what scientists do and how they reason. I’d like to start by talking about how scientists are not the only people
who argue that they’re pursuing truth. People of religious faith and philosophers
also talk about truth – but they use different methods. What characterizes science and scientists
above all is their use of scientific methods. And those methods are characterized by
two things. First is evidence: the comparison of theories about the world with evidence
about the world, often taken from experiments. The other is the application of
reason. The way that scientists combine
evidence and reason is what makes them scientists, and what has made science so
effective. The question is, how is this actually done. One use of reasoning and evidence that
is often associated with science is induction. This is the technique of looking at
repeated observations and evidence, and performing on the basis of
them generalizations. The classic example offered by
John Vickers is the idea that all swans that we have
seen are white. Based on this evidence, one might
conjecture therefore, all swans are white. This leap in reasoning is induction. The problem is that the conclusions cannot be
guaranteed. Perhaps there are swans that aren’t
white. All it takes is one observation –
one black swan – to disprove the theory. And in fact, there are black swans in
australia – which is the point of the example. This account of science as inductive
was rejected by the philosopher Karl Popper, who was concerned that people might make
claims that things were scientific when in fact they were false. If all you’re looking to do is to
confirm a theory that you have, it may be easy to find evidence supporting it. But if there’s evidence contradicting it,
the theory may in fact be false. Popper wanted to rule out pseudosciences and bad science. He argued that scientists should be
critics: that they should be constantly looking for flaws with the theories that
they came up with. And he also argued that rather than
being inductive, science in fact is deductive. Deduction is contrasted with induction. Where induction begins with a number of specific examples or pieces of evidence, and from
those formulates a general theory or principle, deduction goes the other way. It begins with some general categories and produces a specific result. Deduction begins with premises, or
assumptions. It applies to them rules of logic and
reasoning to produce conclusions. And these conclusions are irrefutable
unless there’s a problem with one of the premises. The classic example is the syllogism:
All men are mortal. Socrates is a man. (Those of the premises.) Therefore, Socrates is mortal. Socrates must be mortal unless for some reason the assumption that all men are mortal
is false, or Socrates is not actually a man. Popper argued for falsification. Whereas induction begins with a
collection of evidence and proceeds to generate theories, Popper said that scientists do the
opposite. They begin with the theory and they compare it with evidence. But rather than seeking out evidence
that would confirm the theory, he said, they seek out evidence that will
disprove it. As in the example of the whites swans, the
theory that all swans are white can be disproved by single piece of evidence: the single black swan. Popper said, “In so far as a scientific statement
speaks about reality it must be falsifiable; and in so far is it is not falsifiable,
it does not speak about reality.” What distinguishes science from
non-science, in Poppers mind, is that science can be tested – and those tests could potentially turned out to fail, if the theory doesn’t correspond with
reality. There are a number of challenges to Poppers characterization
of falsification. One of them, the Duhem-Quine thesis, holds that any
hypothesis depends on a collection of assumptions and
background. And so that when there’s a single piece
of evidence that may appear to contradict the hypothesis, it’s not clear which part of the
hypothesis is contradicted. Critics also pointed to a problem of
judging which evidence is sufficient to falsify an established theory. Popper himself acknowledged that a single piece of evidence is seldom
enough. Sciences is rife with anomalies, uncertainties, errors,
effects of observations and experiment. But then if multiple pieces of evidence
are required, how much? Falsification begins to look a little
bit like induction. Philosopher Thomas Kuhn examined the history of science, and concluded that this model of
falsification is wrong. It’s not what scientists actually do.
He wrote, “No process yet disclosed by the
historical study of scientific development at all resembles a
methodological stereotype of falsification in direct comparison with nature.” This seems a little bit unfair to Popper. On the one hand, there have been decisive
experiments in science. On the other, while Popper thinks that he is
accurately described what scientists do from day-to-day, he is also making an argument about what
they should be doing in order to do good science. His argument that scientists need to make theories that are testable is important. Seeking to describe what scientists
actually do, Kuhn’s answer to the problem of evidence
is not an objective rule or simple comparison with nature, but the judgment of scientists within
their communities according to the standards of those communities. In fact, debates around evidence, and error, and what counts and what doesn’t, are central to the pursuit of science. Kuhn further says the scientists are
seldom critical. Most of the time they pursue their work
on the basis of what has gone before without questioning it, without trying to
falsify it, but rather trying to fill in gaps. Occasionally, however, there arises a situation when there’s
actually a revolution within science, and this is the topic of his famous book, The Structure of Scientific Revolutions. Central to all of this is Kuhn’s
idea of the paradigm. A paradigm as he describes it is the collection of background assumptions questions and what kinds of questions
count within the discipline, the methods that are used by scientists, and the kinds of answers that count. It can be difficult to recognize the taken-for-granted
assumptions but we have about dominant paradigms. It is perhaps easier to highlight them with respect to past paradigms. For example, the old idea that the sun
orbits the earth, which of course was replaced in and the
Copernican revolution by the understanding that the earth
orbits the sun. This is one of Kuhn’s prime examples. Geological catastrophe used to be what
geologists felt produced mountain ranges and other large
physical features. It wasn’t until the understanding of
plate tectonics came in the twentieth century that geologist recognize that many of these
things happen gradually over time. The Newtonian vision of a clockwork
universe that dominated physics up until the twentieth century
is another example. Kuhn described two kinds of science. The first of these is normal science. This is what scientists do almost all of the time. They operate within the bounds of the
existing paradigm. Every scientific paradigm has
anomalies and puzzles the need to be solved. In fact, for scientists this is much of the value of the
paradigm: they find it worthwhile not only because
it explains things, but because it leaves things for them to
do. If there were no problems to be solved
or addressed, scientists would have nothing to do. However, sometimes there can be a
challenge to the dominant paradigm. Perhaps anomalies buildup to such an
extent – there’s evidence that doesnt fit – and scientists, or a few scientists, come
up with an alternative. Kuhn says that a new paradigm will be
incommensurable with the old. That is, that the one cannot be
understood in terms of the other. The idea that the earth orbits the sun
is not a progressive approximation of the idea that this on orbits the earth. It does not fill gap in that theory: rather, it replaces a foundation of our understanding of the movement of
celestial bodies. Kuhn calls this kind of radical
discontinuity a paradigm shift – his book is the origin of that term. Now for scientists, their main value for a paradigm is not
so much that he answers questions or accurately describes reality, as that
it offers questions for them to answer. So the interests differ between new
scientists and established scientists. For established scientists, the new
paradigm can be a danger. It can put at risk the research of a
lifetime. Whereas for someone new or someone outside the
field that risk doesn’t exist: a paradigm that
is new may offer new avenues for research and questions to ask. Therefore, Kuhn argues, that in most cases new paradigms are
introduced by the young and outsiders in the field, whereas the old may simply choose to stick to the
paradigm that they have always applied. It may be generational change that is
required to finally accomplished a paradigm shift. It may be that Kuhn overstates the case for revolution and
paradigm shifts. The changes in science may actually be more gradual than he describes. But what particularly interests me
here, in relation to culture, is the role that consensus plays. There is the problem, again, of evidence. If Kuhn is right, and he’s not the only
person who has made this argument, then ultimately scientific method, scientific reason, does not offer an objective solution to the
problem of evidence. In the end, it’s something that stands outside of
science that must be addressed by the community of scientists. This is an argument made by Juergen
Habermas, another philosopher, who says that ultimately scientific reason stands on the back of the reason and
communication of the community, and is not itself scientific. This is not to discard the tremendous
value of science or the scientific community: it is to emphasize that science is not
simply a method but also the community and the culture that carries it out. And in that case, one of the most fundamental questions of
science isn’t simply how it is done, but by whom. Who are the scientists who judge what
counts of science, and what isn’t. And this brings me back to the quote by
Kuhn with which i began. To understand science, “we shall need
to know the special characteristics of the groups that create and use it.”

Leave a Reply

Your email address will not be published. Required fields are marked *