### Chapter 2: The Scientific Method

Up to this point, we have covered rudiments of deductive logic, particularly forms of logic that involve conditional statements. We have covered two valid argument forms that involve conditional statements — modus ponens (MP) and modus tollens (MT) and one invalid argument form (affirming the consequent).

Next, I will discuss the Scientific Method, and how that looks in the context of logic. But before I go on with that …

How did you do on the homework assignment I gave you in Chapter 1?

There is no need to respond to that question. I already know the answer: you got it wrong.

I have shown this photograph to many scientific experts, including forensic engineers and other forensic scientists. “Tell me what happened to result in this damage,” I asked them. All of them gave wrong answers. Several speculated about some form of traffic accident, involving collisions with another car or another object on or beside the road. Others said something heavy, like a tree, fell on the hood of the car. Some speculated about crashing against a low-lying abutment. All were wrong. If they did not do well with this question, I do not believe you did any better.

When a friend of mine sent me this photograph, saying that her husband was in a horrible accident, I did not try to determine what happened by looking at the photograph. I simply asked her a two-word question, “What happened?”

After years of affirming the consequent — of trying to determine antecedent past events by looking at consequent physical evidence — I have come to realize that it is an exercise in futility. It does not matter how much of an expert someone is, it is impossible to look at physical evidence and determine from it the complex sequence of events that occurred in the past that led to the physical evidence. Doing so would be like pulling a tiny needle from an infinitely large haystack!

She responded to my question by saying that her husband was driving south on an interstate highway when a northbound tractor-trailer lost two wheels from lug nut failure. One of wheels flew rapidly northward to the southbound lanes, like a bullet from a gun. The rapid motion of this wheel of course would be predicted by the Newtonian concept of conservation of momentum — the wheel would fly northward from the truck at roughly the same speed as the truck once it separated from the truck. The heavy and rapidly flying spinning wheel collided with the hood of my friend’s car with great energy, and then it bounced over the occupant compartment. My friend’s life was spared, but his engine was destroyed.

Look at the photograph. Does the story fit?

Now, on to the Scientific Method. The Scientific Method is defined as “a method of procedure that has characterized natural science since the 17th century, consisting in systematic observation, measurement, and experiment, and the formulation, testing, and modification of hypotheses” 4. Essentially, the Scientific Method is about developing truthful conditional statements. Such statements of cause and effect are what science is made of. These statements are the way we understand phenomena in our universe.

The Scientific Method is essentially the hypothetico-deductivo-inductive method. It involves 1) observation, 2) hypothesis, 3) prediction, 4) experimentation or controlled observation, and 5) more of the same.

At first, a scientist observes some things that are startling. Let us call the items he observes “P” and “Q.” The scientist then asks, “I wonder if ‘P’ causes ‘Q’?” He or she then develops a hypothesis to explain “P” causing “Q” (P → Q).

Now a hypothesis cannot stand by itself without testing. It is simply a “hunch” that needs to be tested. The scientist then makes a prediction. “If P does lead to Q, then an experiment designed in a certain way will so demonstrate.” Then the scientist goes about designing this experiment. In such a design, the scientist manipulates P. One way of manipulating P is with experimental and control groups (items with P and the same items without P) in order to isolate out P as the explanation for Q. In such an experiment, the experimental group with P will lead to Q while the control group without P will not. This experimental method is the method of difference, one of several experimental methods described by logician, John Stuart Mills (“Mill’s Methods”).

If Q does not result as predicted, then P → Q is falsified (shown to be not true). This allows an elimination of P as an explanation for Q in a fashion that is guaranteed because of MT:

1. If P, then Q.
2. Not Q.
3. Therefore, not P.

The hypothesis is eliminated so that other hypotheses can be considered.

If Q is the result, it does not mean that P is automatically the cause. To come to that conclusion would be logically invalid. It would be affirming the consequent:

1. If P, then Q.
2. Q.
3. Therefore, P

A scientific statement (P → Q) cannot be proven deductively but it can be falsified deductively.

So how does one establish a scientific concept if it cannot be proven deductively? Scientists have to take hypotheses that have resisted falsification and establish them inductively (show that they are probably true). One way is by induction by enumeration. P → Q is tested numerous times under varying conditions. Other scientists are given the opportunity to falsify the hypothesis by experimentation and controlled observation. If they are unsuccessful in falsifying it under a wide variety of conditions and circumstances, the odds are increased that the hypothesis is true. In other words, the higher the number of confirming instances, the more likely the hypothesis is true.

Furthermore, one can make an argument by authority. If one’s scientific peers who are all experts agree that the hypothesis is not falsified, then the study is published in a peer-reviewed journal, giving the world indication that some experts (not all experts) are in agreement. This can become a strong argument (a highly probable argument) for truthfulness if many, many experts agree.

Another inductive method is the argument of explanatory power. How well does the hypothesis explain other phenomena? For example, no one has ever seen atoms or molecules. No one has ever seen protons, electrons, or neutrons. But the concepts embodied in these “models” that have never been seen possess great explanatory power to explain a great many things. Similarly, no one has ever seen gravity or force, but these items — these concepts — can be measured and characterized in the physical world as phenomena. This is because over time, these concepts possess great explanatory power for many phenomena. Still, in spite of such explanatory power, none of these items can be proven in the deductive sense but only deemed probable in the inductive sense — even highly probable.

Now, ladies and gentlemen, here is the great news flash…wait for it…wait for it…wait for it…

The Scientific Method does not work for past events!

First of all, I need to make a few comments about the past.

Past events no longer exist in a real and tangible sense. Past events are abstract concepts. The only items that exist are what is currently present; however, if the past can be characterized to “exist” in any form, it is in the form of 1) memory or 2) record. We have the capacity to remember what happened. We also have the capacity to record what happened by writing it down (or typing it on paper or in a computer), by audio and video recordings, and by images. If there was no one to remember or record certain past events, those past events cannot be known.

Furthermore, the past is complex. Even the past concerning one person or item is complex. Over time, unique phenomena occur to that person or item that are never repeated in exactly the same way. Then consider the past involving multiple people, multiple objects, multiple events throughout the entire universe — events that were both observable and not observable. The past is exceedingly complex and beyond the reach of the human imagination in its exceeding complexity. We do not come close to knowing all past events.

Here is why the Scientific Method does not work for past events.

The past is not observable by scientists living in the present. Only people who were present in a certain place at a certain time and who were sufficiently sentient to observe, remember and record are the only ones who observed the past event. The scientific method relies on scientists to do the observing.

The past cannot be predicted. Prediction is an activity occurring in the present that looks to the future. Determining what is past through any form of reasoning is simply relying on Q to surmise P — in other words, affirming the consequent. Affirming the consequent does not make it past the hypothesis stage because it is deductively invalid and unsound. The photograph of my friend’s car demonstrated just how poorly we do at affirming the consequent for past events.

Furthermore, one cannot perform experiments on the past. Experimentation involves manipulating P, and scientists cannot manipulate the past, although some may try in vain to “rewrite history” — to revise the past in the minds of others!

The Scientific Method even falls down in the inductive sense. An argument by enumeration cannot be used because past events are unique — occurring only one time. An argument by authority also holds no water. The only somewhat-real authorities on past events are historians, not scientists, and the historians have to rely on the memories and preserved records of others who were alive and present when the past event occurred. Scientists who all agree on past events they have never seen offer an ad verecundiam argument (appeal to false authority) which is a fallacy. One cannot claim to be an expert on a past event as a scientist if he or she was not there to observe it, measure it or manipulate it for purposes of proof.

Past events surmised by scientists most of the time have poor explanatory power. In my forensic practice, I have found that witnesses provide past-event explanations for physical evidence that have greater explanatory power than the scenarios invented by scientists. That is why these days many like the explanatory power of what I have to say and why people seek out my services. It is not because I am so smart; it is because I listen to witnesses with an open mind rather than try to invent something out of physical evidence.

So much for the Scientific Method. In the next chapter, I will tell you about the Forensic Scientific Method — a deductively valid approach to learning the truth about the past — and the Inferential Test — a statement that is always true. Both can be used effectively to answer The Big Question.