If you encounter a really difficult problem and have used every inductive and deductive technique at your disposal to discover the answer to the problem but to no avail, the only recourse is to use the scientific method. The purpose of the scientific method is to make certain that Nature hasn’t made you think you know something that you don’t know. The scientific method starts with keeping a notebook. In the notebook you must write down everything you do to solve the problem. If you don’t write everything down you are more than likely to get confused and forget what you know and what you don’t know, what you have done and what you have yet to do. If you do not write everything down in your notebook you are almost guaranteed to become baffled.
The scientific method involves six steps that must be taken in order:
- State the problem
- Formulate hypotheses about the cause of the problem
- Formulate experiments that test the hypotheses
- Predict the results of the experiments
- Observe the results of the experiments
- Formulate conclusions based on the results of the experiments
Perhaps the most critical skill necessary to use the scientific method is to state the problem using no more than you absolutely know about it. For example; Why does the heart stop beating? This may sound stupid but the question is logical and correct. It presumes you know only that the heart was beating and then it stopped. An incorrect stating of the problem might be; Why does too much fat in the diet make the heart stop beating? This statement of the problem implies that you know that too much fat in the diet will make the heart stop beating.
If the problem statement is limited to only what you know, that the heart was beating then it stopped, you can formulate a number of different hypotheses that could be tested and many more that you probably cannot design experiments to test. One of the testable hypotheses might be that feeding too much fat to pigs for two years will cause the heart to stop before the pigs reach the end of their normal life span. However, this would be a poor hypothesis because it is likely that the type of fat or the ratio of different fats is what is important. The ART of the scientific method is stating a hypothesis that can be tested with the proper experiment or series of experiments. Since in this example we presumably can’t use humans to test the hypotheses we have to start by identifying a suitable animal model and do the experiments necessary to provide evidence that the model is suitable or cite the work of others who have done those experiments. Another potential problem is the actual cause of the lack of a heartbeat. Can too much of a specific type of fat, for example cholesterol, cause a blockage in one of the coronary arteries? How much cholesterol in the diet does it take to cause the blockage? Is there a threshold level for cholesterol circulating in the blood that will result in an infarction, a piece of the blockage that breaks away, is carried downstream and completely obstructs the artery? Does a coronary artery infarction always result in death? Is the location of the infarct in a particular coronary artery important?
What is called for is a very specific hypothesis that can be tested experimentally. For example: Will feeding a diet containing 25% animal origin cholesterol in an otherwise balanced diet for 16 weeks result in higher than normal blood cholesterol levels and the accumulation of fat deposits in the left anterior coronary artery of year old pigs but not in year old pigs fed exactly the same diet without the added animal origin cholesterol?
A major advantage of formulating a testable hypothesis is that the actual experiment can never be a failure. If feeding some appropriate number of pigs, the 25% animal origin cholesterol diet for 16 weeks does not result in the accumulation of fat deposits in that specific artery you have added to the body of knowledge, even with a negative result. You can still draw a valid conclusion from the experiment. Of course you might be able to repeat the same identical experiment in a different breed of pig and get a different result.
Now you are poised to do what all good scientists do. Formulate other hypotheses and design experiments to test them and those results will lead to other hypotheses. Most remarkably the end result of this exercise is that there is no end. As each hypothesis is tested more hypotheses come to mind and more experiments must be devised to test them. As hypotheses are tested and confirmed or eliminated their number increases exponentially. Parkinson’s law was an adage applied to the untrammeled growth of bureaucracies; “work expands so as to fill the time available for its completion”. Robert M. Pirsig in his bestselling book, Zen and The Art of Motorcycle Maintenance, has suggested a Parkinsonian-like law that says; “The number of rational hypotheses that can explain any given phenomenon is infinite.” Pirsig suggests that, if true, this law is “… a catastrophic logical disproof of the general validity of all scientific method”.
My own experience was that successfully crafted hypotheses and experiments resulted in ideas for more grant proposals, to secure more funds, to conduct more experiments, to learn more and more about less and less. I concluded there was no absolutely final answer to the ultimate problem because new questions to be answered kept arising. But the journey was so much fun!