Center for Analysis and Design of Intelligent Agents

Philosophy of Science II

Concepts

Induction (ísl. aðleiðsla)	A generalization from a set of observations. Generalization can be about a class of observed phenomena or about a particular unobserved phenomenon that is part of the class. Considered a key to the advancement of scientific knowledge.
Scientific theory (ísl. vísindakenning)	“A set of statements or principles devised to explain a group of facts or phenomena, especially one that has been repeatedly tested or is widely accepted and can be used to make predictions about natural phenomena.”
Prediction, predictive powers (ísl. forspá, forsagnargildi)	A good theory can explain and predict
The value and power of a scientific theory	Measured by (a) its ability to explain, (b) its ability to to be withstand attempts to falsify it and ( c) its ability to predict.
Scientific theories in the natural sciences	Build on comparative experiments as primary method of investigation. Comparative experiments are attempts to falsification; they test the predicitive power of theories. Results are used to tune the theory.
Prediction and falsification: opposite ends of the same dimension	A theory's strength can be measured at least in two ways, (a) by its power to predict and (b) by its ability to withstand attempts to falsification.
The role of “storytelling”	The ability of individuals and groups to create “coherent stories” of how phenomena in the world are connected, and produce rigorous models that support the stories, is a necessary condition for scientific progress.

Induction is not Enough: Refutability of Scientific Theories

Karl Popper	Claims that induction as a process for producing scientific theories, does not exist in the form most commonly thought (a series of observations → creation of a theory).
We can agree with Popper	IFF: we look at how individual scientists work, and we ignore the fact that they have not read up on a host of scientific results that they themselves did not produce, and which have accumulated since the last theory was proposed. In other words, if we look at the scientific field as a whole, induction as a process for creating theories is readily apparent.
Popperian extremism	While we can say that Popper was an extremist when it comes to his discussion of theories, we can agree with him that the role of science is to come up with theories that are falsifiable. That is, they be formulated in a way that makes the theory theoretically falsifiable. This is done by tying it to facts, both known and unknown.
Popper's favorite examples of pseudotheories	Sigmund Freud's and Karl Marx's
Karl Marx's economic theory	The rich will be richer and the poor poorer ( - the path of the West?)
Popper's idealism	We can of course not simply take the last 100 or 200 years of natural science as the ideal of science and assume that anything that doesn't look like it is psudo-science. That would be ignoring at least 2000 years of what Popper would call psudo-science: Philosophical beginnings and proto-scientific stages of all our current dearly-beloved scientific fields.
Scientific methods (and fields) always evolve from philosophical considerations	All sciences have begun as “mere speculations”. Math was developed to help with managing complexities in the natural sciences; we need other, more powerful tools for some of the remaining problems.
Darwin's theory of evolution	According to a large proportion of modern scientists in all fields, Darwin is the greatest scientist of the last millennium, even of all time. Because his theory is not falsifiable (as a whole) it would rank high on Popper's list of bad scientific theories. However, we must remember, the theory of evolution is enormous, and it has many smaller components each of which is falsifiable (and even on its own is larger than prevailing “grand” theories in other fields).
Freud	Popper didn't like Freud's theories for obvious reasons (lack of falsifiability). An important thing to keep in mind is not to let egos and personal admiration get in the way of critical thought (a lot of the interest in Freud has to do with that). But this cuts both ways: Perhaps it may be justified that Freudian methods are still being used in clinical treatment of mental patients, but perhaps not. We should let the numbers, as produced by comparative experiments and comparative studies, tell us whether these methods are worth holding on to. (If such comparative studies are considered too expensive to do properly, perhaps we're selling mental sanity too cheaply!)
Freud and Marx: twilight zone	These teachings sit somewhere between philosophy and a mature science. (Freud was one of the first major names to put forth the idea that the mind has parts - a central tenet of modern brain science, AI and cognitive science.) As with all philosophy and science one must look at what they said when they said it and put aside idealism and personal admiration.
Falsification: bottom line	The bottom line is this: If you are trying to rate theories on a scale of power, pseudo-theories are very low on the ladder, just above myths. Be aware of this. However, remember that inspiration comes from many sources - these proto-theories might be useful (but they might also be extremely harmful). And inspiration is the fuel for the motor of the progress of all human endeavor: Creativity

The Importance of Creative Thought in the Progress of Science

Creativity in Science	No good theory exists without some creative activity of a human! Creativity is the source of new ideas and insights. It is one of the main driving forces of science, because without it little or no progress could be made.
The role of induction	In an effort to make the creative process more “objective” or “rule-like”, philosophers and scientist have often argued that the main ingredient is “logical induction”, generalization that flows “logically” from a number of observations. This view ignores the creative component of scientific work and may be responsible for why science education is in many cases taught in a boring fashion.
Pure “Popperism”	Pure Popperism suppresses the “storytelling” part of scientific enterprise By shifting empahsis onto the falsification as the most important part of science the creative element is suppressed
The result?	We end up with a much lower number of creative people* in science than otherwise. It is possible that this slows down the progress of science.

_{*NB: This does not refer to the common phrase “creative people” as in those who play instruments, create clay sculptures or work at advertisement agencies. This refers to the common notion of the concept of creativity – the ability to produce (good) ideas.}

EOF