|
NATS 1700 6.0 COMPUTERS, INFORMATION AND SOCIETY
Lecture 2: The Method(s) of Science II : Inductivism
| Previous | Next | Syllabus | Selected References | Home |
| 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
8 | 9 | 10 | 11 | 12 | 13 |
14 | 15 | 16 | 17 | 18 | 19 |
20 | 21 | 22 |
Introduction
Topics
- Facts and observations. When we speak here of observations, we are not referring of course to a few casual
glances at what may come our way. We mean instead a careful process of data gathering. In particular, we agree that
we must try to observe particular phenomena in the world as objectively as possible. The number of observations
must be sufficiently large as to minimize random errors and occasional mistakes. We must also repeat these
observations under a wide variety of circumstances. For example, the scientists who investigate global
climate changes must record temperatures, pressures, winds, precipitations, etc. at various times of the year, for
many years, in different locations, etc. If they did not do so, a perfectly normal hurricane could skew the data.
- The question of course is what do we mean by "a sufficiently large number of observations"? How do we
decide we have made observations "under a sufficiently wide set of conditions"? How can we ensure that
"we are objective observers"? The simple definition of the scientific method we examined in the previous
lecture does not give us any clues about such questions. Practicing scientists will tell us, among other things, that
somehow their own experience as scientists provides the clues. But how did they accumulate such experience? Well, for
example, they inherited the accumulated experience of their teachers. But where did the teachers get their experience?
You realize this is an endless recursion. We do not solve the problem by passing the buck.
- And what about the hypothesis? Where does it come from? In the previous lecture we said that we don't care where
the hypothesis comes from, that the idea may well come "from a dream, from a genie, from the depths of your
unconscious, from wherever." But what is the stuff our dreams are made of? It would seem we are caught in a
vicious cycle. In the end, any hypothesis owes much to experience, to observations!
- In addition, we know how easy we find it to see familiar shapes in the clouds, in rock formations, and so on.
Human beings (and perhaps other species) have an uncanny ability to project regularities and patterns even on
randomly assembled objects. In this sense we should be rather suspicious of any claim about our objectivity.
- Proving or disproving the hypothesis. These terms are commonly used, even by scientists,
when they check their hypothesis against new observations for consistency. Unfortunately 'prove' and 'disprove'
derive from a mis-translation or mis-interpretation of the word 'prove' itself. 'To prove' derives from
the Latin 'probare,' which primarily means 'to test.' Hypotheses can not be 'proven'--certainly not
conclusively--because it is always possible to gather new data, to make new observations, and before these are made,
there is no way we can be assured they will confirm our hypothesis. Thus, instead of saying that 'we prove
a hypothesis,' we should rather say that 'we test a hypothesis.' If the test is positive, then we have
extended the validity of the hypothesis, and we are more confident (but not at all assured) that new observations
may continue to be consistent with it. If the test is negative, then we must revise our hypothesis or discard it.
It can not be used 'as is.'
- Let's probe a bit further the notion of observation. In his book What is This Thing Called
Science? Chalmers states: "Observation statements [...] are always made in the language of some
theory and will be as precise as the theoretical or conceptual framework that they utilize is precise. The concept
of 'force' as used in physics is precise because it acquires its meaning from the role it plays in a precise,
relatively autonomous theory, Newtonian mechanics. The use of the same word in everyday language (the force of
circumstance, gale-force winds, the force of an argument, etc.) is imprecise just because the corresponding theories
are multifarious and imprecise. Precise, clearly formulated theories are a prerequisite for precise observation
statements. In this sense theories precede observation." Before I comment on this statement, please notice
that the word 'theory' essentially mean a hypothesis that has been tested repeatedly and successfully,
and refined to be consistent with a whole body of observations. Relativity, Darwinian evolution, etc. are examples
of theories. Now, what Chalmers says is that when we make observations, we do so wearing certain goggles, using
certain pre-conceptions, certain other hypotheses. We say that observations are theory-laden. When a biologist
studies the behavior of chromosomes in a cell, he assumes a lot about such an object. For example, that chromosomes
do not exist freely, outside of cells; that cells have a membrane that separates their interior from the environment;
that the intense light of the microscope does not alter the inner structure of the cell; that certain internal
structures of the cells are in fact stable organelles called chromosomes; that most cellular structures need to be
suitably 'stained' in order to be visible in the field of the microscope; and so on... If our biologist didn't
have these 'theories,' his observations would not differ much from those of a layman who peers for the first time
through a microscope. Indeed, if you have the opportunity, do look through a microscope at anything you like: you
will hardly see and recognize anything.
- The problems of induction. Let's return finally to the notion of hypothesis and its relationship to observation.
Clearly our hypothesis is also inspired by the observations at hand. In fact this is a common icon of science:
the scientist, having made and noted a number of observations, starts then to...meditate on them. He hopes he has
a sufficient number of observations from which he may be able to generalize--to intuit a pattern. What do they have
in common? Is there a pattern, a trend, some regularity? He tries to let...the facts speak for themselves. What do
they suggest? If he is lucky, his observations and his experience as a scientist in his particular
field will lead to a tentative a hypothesis. Here is an example. A German chemist, August Kekulé (1829-1896),
was studying the chemical structure of certain compounds such as benzene. He wanted to figure out the actual
arrangements of the component atoms in the molecule. He knew, for instance, that in water (H2O) the atoms
are arranged thus: H-O-H. How are the atoms arranged in benzene (C6H6)? He had tried various
straight chain-like arrangements (the only ones known at the time). But these hypotheses did not match the properties
of the compound. One day he had a dream, where he saw a snake biting his own tail. When he woke up he experienced
Archimedes' eureka: I got it! Instead of straight, open chains, the benzene atoms may be arranged in closed,
loop-like chains. He tested the new hypothesis. It was indeed consistent with the observed properties of benzene.
It is clear that you and I could have had the same dream (the image in the dream has a long history in the mythologies
of several cultures). Yet, even though we may know some chemistry, including perhaps something about benzene, it is
unlikely we would have been ready to jump on this analogy. Kekulé had the experience, the knowledge and the
dream, and his intuition did the rest. He had found the molecular structure of benzene:
(The black balls denote carbon atoms, and the light-gray ones hydrogen atoms. The single and double lines denote
the various types of bonds between the atoms.)
- What can we learn from this example? In the previous bullet we said the scientist hopes he has a sufficient
number of observations from which to generalize--to intuit a pattern. This is a controversial point. When we
formulate a viable hypothesis (i.e. one that fits the data at hand) we do imply (or more simply, we do hope)
that it will work for observations yet to be made. How can we justify such hope? This is the problem of
induction: to jump from a finite number of observations already made, to a potentially infinite
number of observations not yet made. The history of philosophy shows that this is a problematic jump, one
that perhaps may not be rationally justifiable.
- It would seem therefore that we don't really say much when we claim that science is the application of
the scientific method. This method is riddled with unanswered questions, with hidden assumptions, with
impossible implications. Yet science exists, and it appears to be successful in providing us with some
understanding of the world. Perhaps we need to replace the notion of scientific method with something
else. That's what we will explore in the next lectures.
Questions and Exercises
- Find an example (meaningful to you) of the use of the scientific method, and summarize it, identifying
the various stages of the method.
- Apply the critique developed in this and the previous lecture to your example.
Last Modification Date: 07 July 2008
|
Chris Clarke, Nature by Design:
How Disney's treatment of animals has altered our sense of the wild and cleared the way for
environmental decline