Touching The Limits Of Knowledge

Cosmology and our View of the World


The Nature of Science
Ben Brown & Connor Dieter-Leeds


Summary by James Kilpeck

The Nature of Science

The discussion began with the presenters introducing their presentation subject; “What is the nature of science?” and “Are there different types of Science?”  They made an important note that what they mean by “types of science” is not the differences in the subject, but the ways we can achieve goals within each science.  The goals of science are to describe, explain and predict, while a law is what can bring these ideas together.  Laws can describe and help us to predict, but they lack the ability to explain things.  One student argued that the theory of evolution can describe a specific field of science (biology), but cannot be used to make predictions.  It was argued that the theory of evolution can predict that something will happen, although we do not know exactly what.  This was argued about and a conclusion was formed that evolution itself is not a law, but it is a process and that the word ‘law’ has no formal definition that will fit all sciences.  Another student added his thoughts on the comparison of sciences.  He argued that not all fields of study have the same importance to us.  One can make the argument that creating a vaccine is more important than chemical interactions within the atmosphere.  The former would be influential to many people while the latter could be researched solely for the sake of human knowledge.  As a response to this, an argument was made that any kind of knowledge is power in some way or form.

The next slide in the presentation was titled “What do we mean by the Natural World?”  Ben opened the discussion by asking the class if the natural world is the entire world.  The discussion soon led to a student asking if it were possible to create a definition of ‘unnatural’ without the use of ‘natural’.  This question was put aside for a few minutes, but the conclusion made was that we cannot create an absolute definition of unnatural, only a comparative definition.   One student gave the example that since a star has a density that varies from the average density of the universe, it could be deemed ‘unnatural’ in a sense that it differs from what we see as natural within our universe.  After the question was asked, another student was curious as to whether human consciousness allowed us, as a race, to opt out of evolution altogether.  The general consensus was that human evolution is different, because we are able to mold our evolutionary pathway in a sense partly by social interactions, while other living beings without self-consciousness are generally only affected by nature itself and not conscious thoughts and interactions.  After this the conversation changed subject for a little while into religious matters and how religion can fit within the definition of natural or unnatural.   Ben ended the slide by talking about how some things can fall into the category of the natural quite nicely with few arguments, while other subjects can easily be argued either way.

Next was the question whether there is a single process to science.  Someone mentioned that the reasoning behind the creation of scientific methods was precisely to create processes that are common for all sciences.  One of the professors said that since the scientific method is so abstractly described by different people, it can be true that all sciences follow this method, but it does not mean anything besides the fact that all science requires an idea to be formed and then tested.  Another person mentioned that the process of doing science itself is defined as inducing knowledge in some way.  Therefore, the scientific method is just a way of organization for the scientific community.  It is the manmade definition of science and hence all science must necessarily follow it.

The next section discussed induction within science.  It was defined as a way to arrive at laws via observations over some period of time.  Connor made a good point in that induction itself can never prove anything, but it can provide us with testable hypotheses.  Having Proof implies that one can know something will happen without a doubt.  Just because we observe something 1000 times, it does not necessarily mean it will happen again under the same conditions. Professor Moebius talked about the fact that not having a causal relation between two events means that no proof can be formed, only a theory.  One major point was brought up within the discussion, which was very insightful: it takes only one event that does not follow a law to disprove a law that has been shown to work many times.  The law must then be adjusted in order to fit the new event or data.

The discussion then flowed into the next slide, which focused on falsification.  Ben started out with a quote (not verbatim) from Karl Popper which said “Falsifiability, not verifiability is the criterion which separates science from not science”.  Basically, this means that to create a scientific theory it must have the ability to be proven wrong.  This in turn implies that laws can never be proven to be true as they only follow our observations and inductions, but they can be proven wrong.  It was mentioned that not all theories can be tested; many are merely hypothesized because current science does not have the ability to test them. A good example is the Theory of Relativity, which was not tested for a few years after it was hypothesized.  For some laws it is not even known whether the future will allow testing, such as String Theory.

Next was the discussion of the Duhem-Quine thesis, which says the view that any single hypothesis or theoretical sentence is not conclusively refuted when predictions derived from it turn out to be false, that is, when it is apparently incompatible with observation. Predictions may also rest upon other hypotheses, which serve as background knowledge. We can always revise this background knowledge to save the hypothesis in question. Empirical tests can only be applied to the whole system of hypotheses, not to single theoretical sentences.  Popper was quoted as saying that it is not possible to have theories; we can only have modified hypotheses.  A discussion was started with the question as to whether or not theories are testable.  According to the Duhem-Quine thesis theories can only be refined down to fit newer data.

In the current model of the scientific method, there must be some point at which an abstract model has been tested enough to be generalized into one coherent theory.  The point at which this happens is not a set value, a theory can be created with some amount of data to back it up, but all that means is that more testing could still disprove or change it easily.  Professor Davis provided a counterargument to the thesis and made the distinction that some theories are only conglomerations of multiple hypotheses, but cannot be proven, while hypotheses themselves can be.  A hypothesis on whether or not life on Mars exists can easily be proven if we were to find life on Mars; it could then become a fact.  An important distinction needs to be made for these cases though.  We can easily say a planet has life, but we can only theorize that all planets have life.  Someone brought up the point that all these arguments seem to say that we cannot believe anything, since no theories can be proven true.  The argument to this was that we should not think of these ideas as false, but as things that are not necessarily true.  However, theories within science tend to be our best ‘estimate’ as to what is really happening and therefore can usually be treated as truth. We can still use them as if they were true as long as they work for the experiment we are trying to do.  A good example is Newtonian physics, which can be used as very good approximations for most cases, but do not necessarily work with calculations for objects travelling near the speed of light.

The presentation concluded with a general discussion of what the goals of science are.  Professor Davis talked for a little while about how reasoning behind people’s personal beliefs is very subjective to their own thoughts and experiences.  One person might have a compelling reason to believe in a Deity of some sort, but another person may believe that there are no good reasons.  Theism and science are unrelated in that neither will ever be able to prove the other.  Science is based on observations and evidence and can be used to answer questions, while theism is based on personal thoughts and judgments to answer questions.