The scientific method

I will describe here the scientific method in a way which is comprehensible for non-scientists. I will neglect all historical references. I will first draw the general lines of the scientific method, so its principles: the ideal way one should always behave when dealing with reality in a “scientific” way (the very way that makes it possible to call it scientific). In the second part I will make an example, which may clarify what it may look to many as too abstract.

In a very rough way, it could be said that Science is about understanding the way Nature works. That is to say,
in a more specific way: the mechanisms through which a single phenomenon occurs in Nature. Scientists try to
describe how something works in the most accurate way, understand the mechanism that makes things work exactly that, observed way, and use the mechanism they have described to predict the behavior of another systems, or of the same system under different conditions from the ones under which it has been observed. All this may sound extremely abstract, so at the end of this paragraph there will be a case-example (a very limited one, as all examples and descriptions, but hopefully enlightening). Before going there, let us try to clarify the way a scientist acts when approaching a “problem”, namely when s(/he) has to get at work developing a theory of something.
First of all, the scientist must clearly identify the natural event (s/)he wants to study. For example, the movement of the Sun around the Earth, or the growth of plants, or the reproduction of lepidotters. It is possible that after embarking in the study the scientist will find that this natural event (or phenomenon) is connected to other phenomena (and this is a beautiful thing of Science, which aims at generalization, the attempt to describe all things under a same mechanism, but we will perhaps discuss this later); but at the beginning of the study the natural event must be as circumscribed and defined as possible. ( Some amendments to this very categoric statement may exist, and they are discussed in Section IIB)

The scientist must therefore:

  • identify the phenomenon (s/)he wants to describe
  • develop a theory, which explains all the observable properties of that phenomenon
  • use the theory to make predictions about the system in different conditions, or systems similar to that where the phenomenon has been observed, or the evolution of the phenomenon
  • check that the predictions of his theory take place
  • see after the next few sentences, to be continued

So far, it is important to stress that when a scientist develops a theory about a phenomenon, such theory must not only describe (i.e. explain) everything the scientist is seeing about that phenomenon, but be able to make a prediction about the phenomenon in conditions not yet observed, or systems similar to the one where the phenomenon has been observed. As we will see this part is absolutely crucial for the scientific method, in what we may call here the validation of a theory.
A theory that simply describes(/explains) what is already seen but makes no predictions, or makes predictions that can not be validated by further observations are not considerered to be valid scientific theories, they do not comply with the scientific method. Once the scientist has fulfilled the previous steps, he is now at the point where he is validating his theory by checking that the predictions it has done. So s(/he) is faced with two possiblities:

a) the predictions of his theory are actually observed in the new experiment
b) the predictions of his theory are not observed in the new experiment

In the first case the scientist can —at least for now— go home happy: his theory has been validated. In the case the predictions are not validated, the theory is said to have been falsified by the experiment, and the scientist has a lot of work to do.
There are different paths the scientist can follow: a theory is in general composed by a mechanism and some working assumptions (e.g. the mechanism can be the Sun going around the Earth, the assumption that the Earth is flat). In general the first step is to keep the mechanism, and modify the assumptions, in order to see if the theory can be only slighlty modified (in the previous example, one may keep the mechanism of Sun going around the Earth, and make the assumption that the Earth is round). At this point, the scientist checks the prediction of this modified theory again, and proceeds to the test procedure as above. If the predictions are confirmed, the scientist is now happy.
If they are disconfirmed (the new experiment does not observe what the theory predicts), the scientist can now either change his assumptions again, or change the mechanism entirely (in the example above, say for instance that is the Earth going around the Sun).

A. The theory of falling apples
The previous description may sound very abstract and theoretical; here we present a practical example that may help to clarify how the “scientific method” can be applied. Let us pretend we forget everything we know about the physics of gravitation, settling ourselves at times before Newton. A team of scientists, fascinated by the problem, decides then to understand “why” apples fall from the trees and go straight down to the ground underneath the tree. They sit together and they devise three possible explanations:

a) the wind makes them fall from the tree
b) a force exerted by the the mass of the Earth, acting proportionally to the mass of the attracted object itself, pull the apples (and anything else) to the ground, wherever they are, with a force depending on the distance from the ground itself.
c) a vegetarian dragon, hiding in the bosom of the Earth, summons all fruits of his liking through a magic force, proportional to the mass of the fruit itself. Within a distance of ten kilometers from the ground surface.

They set up to apply the scientific method in order to test these three theories, to see which is the correct one. In order to test the first one they set up the following simple experiment: observe the fall of apples from a tree on a windy day, and on a day without wind. It is important to notice that conditions are all the same but the wind, which is the only thing we are therefore testing, shielding the experiment from contaminations (other possible explanations). So the scientists perform the experiment and find out that apples fall from the tree in exactly the same way. They have therefore to conclude that wind has nothing to do with the fall of the apples, and are left with two alternative theories, which they now want to test.
Notice that both the “Vegetarian Dragon” theory and the “Gravity” theory predict that apples fall the same way, with or without wind, so now the scientists have to look for something that the two theories predict differently, so to make a distinction between the two. They therefore decide to check a pear tree, and find out that pears fall from trees exactly the same way as apples do, with the same dependence on the mass of the pear. They are very satisfied of this, but still this is not relevant, because both the Vegetarian Dragon theory and the Gravitational theory predict pears to fall exactly the same way. So nothing can be said again. A very clever scientist does therefore suggest the following experiment: to tie sausages (of the same mass of the apples) to trees with a very very thin rope. This way: if the Gravitational theory is correct, a force should be exerted on the sausage equal to that of the apple and the rope should break (it has been calibrated in such a way). If instead the “Vegetarian Dragon” theory is correct, the Dragon force should not apply to sausages, in which the vegetarian Dragon is not interested, and the sausages should not fall. They perform the clever experiment and find out that the rope beaks, and sausages fall. Great success for the Gravitation theory! But a very clever scientist supporting the Vegetarian Dragon theory suggests that perhaps the latter is not to throw away, they just have to apply some change in assumptions: the mechanism of the Dragon stays valid, but he cleverly changes the assumption that the Dragon is vegetarian, and suggests that the Dragon eats anything. Under this assumption the “Omnivor Dragon” theory can explain the sausage observation, and it is again difficult to tell which one, the Gravitation of the Dragon theory is the correct one.
For a couple of years scientists are incapable of thinking of an experiment that can help them tell the truth, but in the meantime a young brilliant student is working at some details of the Gravitation theory, and realizes that the same Gravitational force that in this theory pulls the apples to the ground should make the planets, and the Earth, revolve around the Sun. He is so good that he manages to calculate the position of the planets in the future ten years, according to the Gravitation theory. This can absolutely not be done with the Dragon theory because that applies only to objects within 10 kilometers from the surface of the Earth, and planets are farther away than that. He proposes this computation to the community and they decide to set up an observational campaign of the planets in the following years, to check if the Gravitation theory is indeed correct.

B. Few comments on a scientific theory
So far we have discussed the “scientific method”, that is to say a general procedure needed to test some requirements that a theory needs to have to be called a scientific one. However, beside complying with all of the strict requirements (and once they do), theories can also have different qualities, that sometimes scientists use to weight the “appeal” of one theory with respect to another (again, and never enough to stress, when we use the word “theory”, we mean something that has passed the scientific method test).
So, it happens also in the scientific world tat a theory can be defined as “more efficient” than another one, and in some cases even “beautiful” or “elegant”. How is it possible that some very human, subjective, and loosely defined qualities can be associated to objects as seemingly dull and rigid such as scientific theories? It is first worth stressing that scientific theories which pass the test of the scientific method are all valid, that is to say are validated through the described procedure and are able to describe phenomena and make predictions.
Nonetheless, some theories may have more appeal, or show more “strength” than others. To make a practical example, let us take again the previously discussed case of the Dragon vs Gravity theory, at the moment when scientists just find out that the motion of planets can be described by the Gravity theory, and not at all by the Dragon one. At this point some theorist supporting the Dragon theory, says that some special Dragons are not only Omnivorians, but extend they powers well beyond 10km, and devises a very good fitting description, under the new dragon assumptions, of the motion of the planets as attracted by the special Dragon force. He now makes all the tests and finds it out to be working, so the new Dragon theory is accepted. At this point there are two theories, the Dragon and the Gravity one, both working on a strictly practical point of view, and both valid according to the procedure of the scientific method. It is to be remarked however, that the Gravity theory has never, since it was created, needed a modification in order to explain the observations, and it has always made predictions, even farther beyond the original scope it was created for. It is clear that, whereas the Gravity and the new Dragon theory both can describe the data and make predictions, that the Gravity theory has an appeal of elegance and “naturality”, something which is difficult to measure or evaluate, but that nonetheless the scientists are sensible to, as human beings.

It is also to be noticed that -in our example- the Gravitation theory has been formulated in order to describe the fall of apples from the three, but has eventually proven itself perfectly fit to explain the motion of planets in the sky, which anyone would agree (and certainly the scientists of our example) that at first glance are a very different problem from that of apples falling from tree. (And the Gravitation theory is proving not to be). So the Gravitation theory is a good example of another quality that scientists usually regard as a very good one in scientific theories: that of “generality”. The fact that apparently very different phenomena belong can be described by very same theory (and therefore have a common nature) is a very powerful outcome of “elegant” and “very efficient” theories.

– Fabio Iocco

Leave a Reply

Your email address will not be published. Required fields are marked *

Cosmology, astrophysics, astroparticle physics