|
|
Science is an organized, social method for the production of reliable
knowledge
It is social in the sense that scientists build on the contributions
of their predecessors and their peers
The image of the mad, lone wolf scientist is way off!
One trains a long time to absorb prior contributions and the
culture of the scientific community
Communication of results is critical to the scientific enterprise
Progress reports at conferences to get feedback to improve
the work before its publication
Publication entails submitting your work to a panel of
referees called on by the editor, who read your
manuscript and recommend that the editor publish it,
publish it after certain changes are made, ask that it
be published in a journal better suited to it, or flat
out reject it as crap
Having survived the review process, your work once published
is then open to the criticisms of your colleagues,
including their attempts to replicate your findings
It is also social in the sense that projects are often worked on by a
team of collaborating scientists
These are sometimes groups of peers
Sometimes they are one or two principal investigators together
with a team of technicians or students
This is becoming more and more common as the equipment needed to
do research gets more and more refined and expensive and as
host institutions need to spread its usage among many scien-
tists
Not only is science a social affair, it is also a highly organized
affair, and a lot of this lecture will focus on how science
organizes its inquiries (more on that later)
Scientists see themselves as producing knowledge, as much as discover-
ing the truth
You might wonder how one "produces" truth -- isn't truth just out
there, unchanging, and waiting passively for us to "discov-
er" it?
Scientists have a particular view of "The Truth," which varies
substantially from the common concept of it, and this leads
to many misunderstandings between scientists and the larger
society around them
For scientists, truth is the highest and best interpretation of
their observations, as agreed upon by the majority of
qualified scientists working in a given field at a given
time
In other words, for science, truth is both consensual and change-
able or provisional
This drives some elements of the public nuts
Most people conceive of truth as something eternal, unchang-
ing, and, perhaps, revealed a long time ago through a
religious figure
Science accepts change in truth for the following reasons:
Reality itself changes, forcing a change in our concepts
Usually more importantly, our tools and technology change
and this exposes us to information we couldn't get
before and so forces us to change our views ... and
allows us to ask different new questions
Astronomy -- naked eye, Galileo's telescope, modern
huge telescopes (e.g., Palomar), telescopes and
scanners sensitive to other wavelengths than
visible light radiation (e.g., radio, microwave,
infrared, ultraviolet, X ray), satellites (e.g.,
Voyager and the outer solar system, its last image
of the whole solar system, its participation in
the supernova of 1987)
Information processing technology -- computers, statis-
tical methods (e.g., ALS LandBank)
Geography -- field observation and hand-drawn maps,
computerized cartography, computerized statistical
methods and large-scale and multivariate data
sets, geographical information systems, remote
sensing, and computerized qualitative data
reduction
Our interpretations shift, too: more elegant explanations
cover old and new information better
Newtonian physics incorporated into and superseded by
Einsteinian relativity
Simple classification of life and the Lamarckian
interpretation of evolution superseded by the
"Modern Synthesis": Darwinian theory of evolution
coupled with the Mendelian system of genetic
inheritance and now the genome projects
Crustal contraction and elaborate land bridge systems
gave way to the plate tectonic theory, which ele-
gantly explains the distribution and mechanisms of
earthquakes and volcanoes, climate changes over
the millions of years, and the distribution of
closely related species across widely separated
landmasses
Scientists, then, accept that truth is changeable in these
senses and that the majority of scientists will move
toward higher and more sophisticated interpretations of
real world observations
When majority opinion among scientists in a given
field shifts, then the "true" interpretation
shifts
Now one funny thing about this whole process is its
humanness:
Scientists like to think of themselves as logical
individuals who can be moved to a higher and
better truth through simple force of
compelling logic, sort of Mr. Spocks
These major shifts in scientific truth rarely take
place in such a smooth and reasonable manner,
however
Scientists are people and, like people everywhere,
they tend to be fond of their own opinions,
particularly those to which they have devoted
a lifetime of work
The way research proceeds, as we'll see a bit
later, is that everyone contentedly accepts
the transmitted wisdom of the scientific
culture and works out its minor details
Through time, however, new observations pop up,
which don't quite fit into the accepted world
view
They are often not even perceived: one sees
what one expects to see
These anomalies continue to pile up until
they attract the attention of some,
usually younger scientist
One such individual will eventually come up
with an interpretation of the anomalies,
which integrates them with the older,
more routine observations
The old guard is aghast, because the new
model quite possibly invalidates the
premises of an older scientist's entire
corpus of work: no human being easily
accepts the revelation that s/he has
been wrong all this time
The old guard resists the new model, no
matter how elegant it is, and only the
young folks secretly pay attention and
begin to work with it once they've
safely gotten their Ph.D.'s under their
belts
As these young folks move into the powerful
stages of their careers and as the old
guard retires and dies off, the majority
of scientists then shifts to the newer
and higher truth through simple demo-
graphics
One last aspect of the definition of science given above: science is
concerned with the production of reliable information
Reliability depends on replicability
Knowledge is reliable to the extent that one can repeat a given
process and keep getting the same results:
You have to be able to get a phenomenon to repeat itself
consistently
There are some phenomena that can't be observed directly
and, hence, can't be repeated: for these, you have to
get consistent repetitions of enough small scale
aspects of the phenomenon -- testing in bundles, then
With this understanding of science, consider the following issues:
Is creationism scientific?
Consensual truth among practicing life scientists
Replicability
Why do scientists avoid the UFO and psychic issues?
Replicability
Their existence is unpalatable, because they seem to violate
a number of scientific truths that scientists as yet
see no reason for discarding (e.g., relativity)
Purpose of science
To explain the world as far as possible without resort to the super-
natural
Science is, then, philosophically ... or at least operationally
materialist (some scientists are philosophically religious but
function operationally in a materialist manner in the scientific
areas in which they do work)
It holds that there is a "real world" out there, that that world
is reflected in our senses and extensions of our senses, so
that we can acquire knowledge about the "out there"
It seeks to explain the material world by reference to material
processes, those involving the interactions of matter and
energy
Another approach to the world is that of the various idealisms
Idealism holds that there may or may not be a real world out
there (some variants say yes, others no)
Some philosophical idealists feel that what we perceive as the
world is actually a reflection of our individual or
collective minds and expectations (e.g., a lot of New Age
thinkers, old time solipsists)
Others hold that there is a real world, but that it was created
by spirit or consciousness (religion would fall in here, the
world being created by a deity)
Still others say that the world out there is real, whether or not
it was created supernaturally, but that the realm of human
thought dominates material processes (this shows up in
cultural geography, cultural anthropology, and sociology
sometimes, in studies of how thoughts or beliefs create,
say, economic change in and of themselves)
Science, then, being philosophically or operationally materialist
often finds itself inadvertantly upsetting some religious people,
because religion very much depends on an idealist world view and
so science seems to be threatening to some people (despite the
philsopher Immanuel Kant's best efforts at keeping science and
spirituality in their own corners, each dominating that aspect of
human life at which they seem best suited, this back in 1781 in
Critique of Pure Reason)
The process of science
The operation of science is historically contingent: it depends on
the maturity of a given field of inquiry
In youthful fields
There is a rather unstructured search for anything that strikes
you as "interesting"
At this point, there is no theory to guide your opinion of
what might be "interesting," so you let your heart be
your guide, so to speak
Such a field is highly "empirical," that is, focussed on
facts, not theory
It could also be characterized as "idiographic," that is,
concerned with descriptions more than underlying laws
Such a field is often focussed on categorizations,
developing classification systems, as was, for example,
biology in the late eighteenth century and much of the
nineteenth ... or Landsat interpretation back in the
mid 1970s
Just because the field has no theoretical background at this
point does not mean that one's work is unstructured in the
sense of sloppy, however
One is very painstaking and careful in collecting informa-
tion and in communicating how one went about one's work
This promotes replication and, hopefully, your work can be
repeated and thereby supported
As time goes on, patterns begin to emerge in the observations
Classification systems seem better able to cover the things
they're intended to
Some of the regular patterns you see in the date can be stated as
law-like statements: if you do this, you will consistently
observe that
At this point, we see the beginnings of theoretical work:
The statement of elementary patterns
These could be considered elementary generalizations
The movement from data to generalization is called "induc-
tion"
An example of a field in this stage was AIDS research in the
early 1980's, when no explanation for the disease had yet
been identified, but there were consistent regularities in
terms of related symptoms, associated micro-organisms, human
groups most likely to experience the complex of symptoms,
etc.
As these generalizations are stated, they are tested back against
"grubby reality"
Some folks test the generalizations against virtually identical
data sets simply to see if a particular study is replicable
Others will say that if the generalization is valid, then a com-
pletely different set of data could be expected to behave in
thus and such a manner as a means of testing the original
generalization: that is, they formulate and test a
hypothesis
If your results square up with expectation, fine and good
If they don't, there are three possible reasons why:
The generalization may be wrong
You made an illogical connection between the original
generalization and the new data it was tested against
Either you just plain screwed up
Or the connection revealed as illogical might itself
turn out to lead in interesting directions
Most common of all, you may have simply made an error in
your data collection and analysis
This movement from generalization back to the world of data is
called "deduction," and takes the form of potentially allow-
ing the generalization to be falsified
Deduction usually proceeds by testing "if-then" statements.
It allows us to test a generalization by seeing just how
general it is: can it cover new data?
A field that has begun to create and test law-like statements is
said to be more "nomothetic" (law-seeking) than "idiograph-
ic" (descriptive)
As time goes on, this process of deduction is going to prune the
number of generalizations, by validating some and disproving
others
The falsified generalizations are thrown out
The survivors often show a pattern that itself can be stated as a
higher level generalization
So the field proceeds again by a higher level inductive episode,
stating the laws that seem to link lower level laws or
modelling the relations among them
These higher level generalizations are also tested and weeded
through deduction and themselves linked into even higher
level generalizations, or theories
A theory, then, is a high level generalization, one of great
sophistication
This is another point of conflict between scientific lang-
uage and normal speech, leading to poor communication
In popular parlance, a "theory" is "just" a "specula-
tion"
To a scientist, it is a sophisticated, high level
grouping of regularities
Popular speech has it that evolution is "just" a
theory, so JQ Public thinks it's just a wild guess
and so is creationism, so why not teach them both
in science classes?
High level theories are also subjected to testing in another
deductive round
This time the generalization is at such a high level that
you can't test the whole thing at once in a single test
So what happens at this point is that "hypotheses" are
derived from the theory and tested singly or in bundles
If the whole theory is right, then the following things
should take place in a defined situation
Each of the things is tested singly or all together in
a related study project
As time goes on, a given theory will withstand so many tests that
virtually everyone in the field comes to accept it as the "truth"
Such a theory is called a "paradigm" (pronounced "pair a dime")
A paradigm is the theory which an entire field takes for granted
as the accepted intellectual background for all the work
going on in it
You don't bother describing and defending the paradigm's basic
logic in your articles any more
You simply accept the theory and know that your audience is
familiar with it, and try to extend it into new areas
It is possible that, in the process of so doing, you may
come up with truly anomalous results, which you then
publish almost sheepishly
You can't explain it within the paradigm, but in publishing
it, you expect someone else will see a connection you
didn't
Over time, if consistent patterns appear in the anomalies
and the anomalies are numerous enough, they may force a
change in the paradigm or its overthrow when someone
comes up with a better mousetrap
Examples of paradigms:
Physics: relativity and quantum mechanics
Biology: the modern synthesis of evolutionary theory
Geosciences: plate tectonics
Note: there are no paradigms in the social sciences
Partly because of their relative youth
Partly because we are studying ourselves
Partly because of the economic and cultural interests behind
the possible results
It is quite possible that because of these last two reasons,
there will never be paradigms in the social sciences
I should note that there are often debates in some fields about
whether they should be idiographic or nomothetic at given stages
of their development
There is the possibility of premature generalization
Such premature theorization is just wild speculation with
little connection to the data
Geography and other fields with a social science component
had an embarrassing episode of this around the turn of
the century up till about the 1920's
Geographical determinism: physical environment deter-
mines human social and genetic patterns
Theory stated long before any data collected or lower
order generalizations made
This was inappropriate theorization based on the exten-
sion of Darwinism to culture (a form of Social
Darwinism)
This neatly "justified" racism and colonialism
It became obvious that it didn't fit the data once
people bothered to go get the data
The different definitions of the "perfect" envi-
ronment, depending on which empire you were
from
Sequent occupance
Geography was so burned by this that it retreated into pure
empiricism and an idiographic method until the 1950's
in the US
Some archaeologists are making similar objections to theor-
izing in their field
Even as one can jump to the theoretical level too hastily, it is
also possible to do the opposite: avoid theory and cling to
mindless empiricism and thereby sterilize any inherent
interest in your field
The role of abstraction in science
One of the biggest problems in science is the sheer complexity and
confusion of the real world
Abstraction one way around that: simplify the conditions under which
you collect data or analyze it
Methods of abstraction
Laboratory experiments
Computer simulations
Thought experiments
Principal components analysis and classification systems
Generally involves analysis:
Setting up your cartoon world by isolating it from grubby reality
Stating your expectations in a rigorous manner
Seeing if your expectations hold up under these special, isolated
circumstances, if you in fact understand your system
Once you've done this, you can plug grubby reality factors back into
your cartoon world
Synthesis
If the model resembles the real world after a few rounds of syn-
thesis, you may be justified in claiming you understand the
basic factors underlying that reality
Applications
Science and technological development can eventually produce a body of
tested theory and technological tools, which can then be applied
to the resolution of human problems
An applied and technological investigation, though it involves
scientific concepts and theories, is actually quite distinct from
science and is judged by different standards than pure science
A technology-focussed investigation is justified, not by its situation
in a body of theory, but in terms of the need it is designed to
meet:
So, if you were doing a planning study in a city planning office,
perhaps a revision of the general plan safety element, you
would review the laws that mandate general plans be made for
California cities (social need) and the changes in our
understanding of, say, seismic processes and seismic
engineering (pure science and applied science) that force a
revision of the safety element (social need again)
If you were working on using fine-resolution hyperspectral
imagery to identify evacuation routes, you'd refer to the
social need of getting people safely out of the path of a
fire and firefighters quickly in to the area where the fire
is being fought (social need). You'd review the spatial,
spectral, and temporal resolution of the imaging system
you're working with and the software that allows you to
process it (the science) and how you'd like to test out
whether this is a feasible application (an open-ended
endeavor more than a formal hypothetico-deductive exercise).
|
