Monday, January 16, 2017

Hypotheses, Theories, Laws and all that jazz

When I was in high school, I learned about hypotheses, theories, laws and principles.  The problem is that I was taught that these were hierarchical.  It took a long time for me to learn on my own that is not the case. They aren't necessarily stages in the understanding of our Universe.  A single hypothesis does not become a theory.  A theory does not eventually become a law.  A law does not eventually become a principle.  Furthermore, this list is missing the category of models.  Each of these are different things that serve the Scientific Method in different capacities.  Berkeley University of California states,
"Hypotheses, theories, and laws are rather like apples, oranges, and kumquats: one cannot grow into another, no matter how much fertilizer and water are offered. Hypotheses, theories, and laws are all scientific explanations that differ in breadth — not in level of support."[001a]
Hypothesis 
Question mark
A hypothesis is a proposed or suggested explanation for a phenomenon.  The hypothesis is stated in such a way as to allow for scientific testing for specific expectations.  The hypothesis must be testable in a falsifiable manner.  That means, to test the hypothesis, you must be able to conceive of and test methods that can potentially disprove the hypothesis.

The value of the hypothesis is that it allows us to simplify initial observations into a testable statement so that we can determine if the basis for the hypothesis is true or false.  You can test to find supporting evidence for the hypothesis.  You can also test to find refuting evidence which disproves the hypothesis.  

Hypotheses are typically formed by one or a few persons who then conduct tests as experimenters to prove and disprove it in the pursuit to solve a problem.  A hypothesis is often not a single point in research.  Experimenters may test and reject several hypotheses before solving a problem.  Disproving one particular hypothesis is just as important to Science is proving another hypothesis.[001b]

There is a subcategory of hypotheses called working hypothesis, which have some evidence to support them.  As such, they are tentatively accepted as a basis for further study.

Theory
Barbara McClintock in her lab conducting genetic research
A theory is a substantiated and unifying explanation for some aspect of the natural world.  Substantiation is acquired through the Scientific Method, with repeated testing and confirmation using written and predefined protocols for observations and experiments.

Theories are testable and make falsifiable predictions.  They allow for predictions to be made about a phenomenon, and they also explain the causes for the phenomenon.  

Science historian Stephen Jay Gould said, 
“...facts and theories are different things, not rungs in a hierarchy of increasing certainty. Facts are the world′s data. Theories are structures of ideas that explain and interpret facts.”[002]
Theories are typically formed by consensus by many different people over a substantial period of time.  Theories aren't just thought up by one individual and then magically accepted by everyone else.  They are often heavily debated while they are being formed.  This debate drives further hypotheses and experimentation that eventually helps develop the theory.

Theories are not absolute points.  Once you have a theory, that doesn't mean the matter is settled.  It just means that the evidence up to that point allows you to create a structure that provides both reliable predictions and explanations for phenomenon.  Theories are often modified or replaced when better structures allow for better predictions and explanations.

A classic example of the process of debate to come to an eventual consensus is the Big Bang vs. Steady State theories debate, in which experimentation on both sides eventually lead to a much better understanding of our Universe.[003]

Model
Atmosphere composition diagram representing a scientific model
A model is a often overlooked scientific tool to make a particular aspect of the natural world easier to understand, define, quantify, visualize or simulate based on commonly accepted knowledge.  Modelling requires selecting and identifying relevant aspects of a situation in the real world, then applying techniques such as conceptual models to better understand, operational models to operationalize, mathematical models to quantify and graphical models to visualize the subject.

A model seeks to represent empirical objects, phenomena, and physical processes in a logical and objective way. All models are in simulacra, that is, simplified reflections of reality which allow for useful approximations.

A model is evaluated by its consistency to empirical data.  Inconsistency or irreproducibility of observations must force modification or rejection of a model.  A model must be able to explain past observations, predict future observations and have refutability, just like theories from which the models are typically built.

Law
Thermodynamics and negative resistance
A law is a description of a phenomenon in a particular situation without considering the cause.  Peter Coppinger of Rose-Hulman Institute of Technology states,
"Laws are descriptions — often mathematical descriptions — of natural phenomenon; for example, Newton’s Law of Gravity or Mendel’s Law of Independent Assortment. These laws simply describe the observation. Not how or why they work."[004]
Laws are compact generalizations about data.  As with other scientific elements, laws are not immutable.  As more information is learned, laws can be changed.

It is important to note that laws can exist without theories.  Sometimes laws exist for many years before theories explain their causes.[005]

Principle
A principle is really just a law that is true by definition.  The terms law and principle are often used interchangeably in Science.  A principle is not a higher grade above a law.  In fact, if you look up "Scientific Principle", your searches will inevitably lead to information regarding laws.

Some persons have suggested that laws can typically be reduced down to precise math formulae, such as the Laws of Thermodynamics and Ohm's Law.  Conversely, the suggestion is that principles are more general descriptions of the nature world.[006]  Examples of such principles are Principle of Original Horizontality and Pareto Principle.

However, even this comparison is not an absolutely held distinction.  For example, Heisenberg's Uncertainty Principle is highly mathematical in nature. Conversely, the Law of Superposition has no mathematical reduction.  So even math provides no real distinction between the use of the words principle and law in Science.

Confusion
I guess the confusion about the relationships between laws, principles, hypotheses, theories and models is that it is not hierarchically ordered.  It seems counter-intuitive that Science, being the mechanism that has brought so much order to our understanding of the world, is itself not similarly ordered.  But, there's good reason for this.  Science doesn't work in absolutes.  Nothing is absolutely knowable.  As such, everything we know is subject to be revised based on what we later learn.  Having some sort of truth gradient would slow down the progress of learning since managing such grading would be an unnecessary distraction from the search for knowledge.

Response:
reddit-1 and reddit-2

No comments: