That's not Science, Lead: Willem deVries

2/13/2006

Summary by Ashlee Cieslak

That's not Science

Readings:
Ben-Ari “Just a Theory”, Ch. 4, 5
We were suggested to browse the following websites

The discussion started with Moti Ben-Ari’s definition of theory from his book Just a Theory. The definition stated that “a scientific theory is a concise and coherent set of concepts, claims, and laws (frequently expressed mathematically) that can be used to precisely and accurately explain and predict natural phenomena. A theory should include a mechanism that explains how its concepts, claims, and laws arise from lower-level theories.” Ben-Ari proposed the “Tee-shirt test” claiming that any real scientific theory can fit on a tee shirt.
Using the theory of gravity as an example, a comment was brought up pointing out that the theory is made up of all equations, which would fit on the tee shirt, but do the terms in the equation have to be defined to explain the theory?

This discussion brought about the next question asking what counts as a theory? Dr. deVries explained that Newton’s theory of gravity includes much of the mechanics of the theory (the equations) and describes how gravity depends on masses and their distances, but does that count as a theory? Also brought up was the idea of considering physics and biology as a theory or a set of theories. For example, is the whole study of physics a theory or is it made up of individual theories, such as the theory of relativity and the theory of gravity.

The next question Dr. deVries asked was what needs to be the level of understanding when explaining a theory? If a theory were to be put on a tee shirt, should someone be able to use the shirt to prove the theory or if someone would read the shirt and say that they “understand” what would be the adequate level of understanding?
A comment that was brought up was the determination between a claim and a theory. According to Ben-Ari’s definition 1+1=2 could be considered as a theory.

Another question brought up by Dr. deVries was how mathematical does the theory have to be since that characteristic was only brought up in parentheses. For example physics is very mathematical, but what about biology, plate tectonics, psychology, or sociology? There is a difference between using math in a study and using math to describe, express, and explain a theory.
Some students talked about their area of study and sociology was explained to be both quantitative and qualitative, not just one or the other, and economics had the idea that math may not be the best explanation for what it’s trying to show in society.
A side note that was brought up was that God is metaphysical so it can’t be explained by a theory by definition.

The idea of Universality was also brought up and how universal a theory needs to be. For example physics is pretty universal because it is the same everywhere, but in biology it is dependant on certain variables.
Another question brought up was what did it mean to explain something. Multiple points were brought up by Dr. deVries including:

           • Explanation=deduction
               Statement of initial conditions
           + Statements of general laws that apply to conditions
                  => the explanation
Some problems of explaining something are:
           • The symmetry of explanation and prediction. Ben-Ari claims that there is a direct relation between explanation and prediction.
           • Statistical explanation means you can only predict the probability for something to happen, not when, how, or which one. A prediction is really just why something happens at a specific point. This applies, for example, to quantum mechanics.
           • Statistical Relevance Model. Most comments made about anything are very general in content, but if they become more specific can they still be true? For example can questions such as why, how, and what be used? Dr. deVries brought up the situation where bacteria will evolve to become more resistant which is true, but which one, why, and how are hard to determine. Prediction can be accurate up to a percentage, but is that good enough?

A side conversation about explanation arose and some comments brought up were that an explanation is the predictive part of a theory and the math is used to describe the theory. Also that there are levels of explanation, for example to a PhD student as compared to a 3rd grader, and the level of explanation deals with the background of the subject. An example used in class was the explanation of where rain came from. You could explain to a 3rd grader that it was “Zeus pissing through a sieve,” or to the PhD student about the complexities of the moisture in the clouds, condensation, and water particles. Each example explains where rain comes from, but is shaped differently according to the background the subject has on rain. In essence only one of them provides a physical explanation.

One final question that Dr. deVries brought up was is prediction essential to science. A debate arose concerning whether a law is a prediction. The laws themselves don’t change but the explanations we give to them do. It’s not a theory’s job to explain why, but a theory is put to the specific phenomenon that it describes. The definition of both a law and prediction came up and a prediction was identified as a vision of a future event, while a law identifies a specific variable without reference to time.
A couple people argued that there has to be a time element when referring to a law.

The final take home message was that laws explain phenomena up to the knowledge that we have accumulated thus far. In the future we might find evidence that conflicts with the law, but for now the law explains what we know. This does not in all cases overthrow the previous law altogether. For example going from Ptolemy to Copernicus it did, but going from Newton to Einstein it didn’t. Newton’s laws are still perfectly valid approximations in our everyday world.