Start page  
What is
What is
of Science
Verification and
about Science
Paradigm and
paradigm shift
Karl Popper -
Logic, Status
Results of


logo vetenskapsteori

Science according to


Communication between people often deals with observations. About what we have done, heard about or observed.

Unfortunately, we could have been mistaken during the observation or during its interpretation. Or we may even have been lying.

Scientific methodology aims at minimize that such mistakes are freely spread.


This is science (short version)


The term science may be defined as "the content of all scientific reports".

Scientific reports consist of attempts to describe what we call reality.

Every statement in a scientific report should be verified, either by described observations or by other named scientific reports. Therefore the information in a report becomes controllable and possible to criticize.

As a result the probability for that the information in a scientific report corresponds to reality is higher than within most other descriptions.


The result is that the probability for the information in a scientific report to correspond to reality is higher than in many other types of descriptions.

Using terms from philosophy, science is NOT a collection of "proven" "knowledge" and is not representing "absolutely certain knowledge".


Background - Mitigated Skepticism


During the 16th century the opposition towards religious and older philosophical dogmas grew. In addition printing technology permitted distribution of information that were not controlled by religious authorities.

In order to reduce the risk of criticism and punishment, scientists and philosophers began to give detailed descriptions of why they could claim matters that were not in accordance with those of the authorities.

The result was the creation of a "scientific method" and the result from this: Science.


The basic philosophy of "scientific methodology" is a "mitigated Skepticism" that can be summarized with:

Although no conclusion about our perceived reality
can be shown to represent "absolutely certain knowledge",
some conclusions are more probable than other


The "scientific method" aims at differentiating more probable statements from those that are less probable. See the part Induction under Epistemology for details.


Science is non-dogmatic


One of the foundations of scientific methodology is that a statement should be verified by carefully described observations. This important rule results in that science is basically non-dogmatic and non-authoritarian.


In the welter of conflicting fanaticisms, one of the few unifying forces is scientific truthfulness, by which I mean the habit of basing our beliefs upon observations and inferences as impersonal, and as much divested of local and temperamental bias, as is possible for human beings.

Russell (1945) - History of Western Philosophy, final words


Science is not equal to "knowledge"


Science is the sum of all scientific results.

Scientific results are created by activities that follow certain rules. They are called the scientific method.


As science is based on mitigated scepticism, it is well known that scientific results are not the same as the content of the concept "Absolutely certain knowledge".


Science is the result of an activity


When we perform any activity, we acquire experiences and draw conclusions.

In relation to the concept "knowledge", there is no difference between acquiring experience within a scientific area like chemistry, or within any other activity like sewing or sports.


During all activities we learn from observations and from hypotheses that we create from the observations.

The activity that creates scientific results distinguished from many other activities by stressing that experiences should be reported in a manner that enables validation and criticism.


Sport is another activity


Both science and sport are rather vague terms containing several types of activities, e.g. Chemistry, Archaeology and Psychology or Decathlon, Football and Chess, respectively.

Within both activities, observations and theories are used as foundation for development.

Of course differences in what we actually perform in the two activities exist. In one activity the experiences may be used to create a medicine, in the other a football goal.


But from an epistemological view there is no difference in producing scientific results or sport results.

In both activities the performers may have varying degree of professionalism. They improve their results as they systematically learn from their experiences

The results are shown to the public. Scientists demonstrate their results in scientific publications and football players demonstrate them at football arenas.

Paradigm shifts are sometimes clamed to be present both within sport as within science


Open communication

In order for a statement to be accepted, it's background must be accounted for


The most important agreement when scientific results are reported is that every statement should be verified through comparison with what we call reality.

This results in openness and transparency in the communication. Such an openness rise problems in that to keep up a statement that is incompatible with our experiences.


Delusive communication may refer to e.g. "secret material", "deeper insight", that facts "are published" but without referring to where, or "are to be published later"

Such phrases are contrary to scientific methodology with regards to openness.


Several important parts of science


Documented observations


The main part of all natural scientific literature (maybe 90%) contains a background to the publication, how observations were obtained, descriptions of the results from observations. They often also include a more speculative part, discussing the results.

The description of how observation results were obtained is very important. Natural science students are taught that it should be so detailed that the observations can be reproduced.

The description makes it possible to estimate the probability for that the observation was correctly performed, or if errors could have influenced the observation results.


The description also makes it possible to use the result from the observation in the future, even it a later scientist does not agree with the implementation made of the first observer.

A third advantage with the descriptions is the pedagogic value. Later scientists may use the methods described in the initial report, and do not need to learn them from the ground.


Reproducible observations


A criterion demonstrating that an observation is well documented is that it is possible to be reproduced.

This means that the documentation should be described in such a detail that the observation should be possible to repeat.


Reproducing the observation should also give the same result as in the original report. One effect of this requirement is that it increases the probability that the first observer makes efforts to eliminate random factors in the report.


Verified results


An additional factor that has contributed to the reputation of science in society is that scientists review their colleague's results. A scientist verifies or falsifies other's observations regularly. New and important results even require verification by independent scientists to be regarded as reliable.


In case a reproduction of the observation does not give the same result as in the original report, the new result may be published. This implies that the original author may be criticized in public, which of course is a negative experience. Openness hence leads to a personal motivation to deliver results of high quality.


Controlled reports


A scientific article may of course contain obvious errors. To decrease the risk of such errors, the article is reviewed prior to publishing by one or two referees, persons skilled within the area in question. The procedure is called "peer review".

The system with "peer reviews", may unfortunately give rise to possibilities of abuse in areas where a few people dominate their research area and often are chosen as referees.


My own experience of referees is that their suggestions for completions have been well motivated.

In one case, the referee maybe was more seeking to demonstrate his own splendidness than to provide an objective review. In that case, it was sufficient to defend the original results for the magazine editor to get the article published.


Building further


The high requirements for documentation and verification have led to an incredible development within scientific disciplines. Instead of "starting from the beginning", the documents provide a possibility to "build further" on previous results. Hypotheses based on the results may rapidly be judged as plausible or be rejected.


A similar type of building on previous experience can be found within almost any human activity. Experience is transferred e.g. through journals, handbooks or through personal contacts.


What is significantly separating science from other activities are the higher requirements of reproducibility. High standards of documentation are also found within e.g. house building. Within this area, however, it appears that the normative information (regulations) is dominating over the describing.

The importance of "building further" has been mentioned within philosophic circles:

Both philosophy and science seek not mere opinion but knowledge. The sciences, however, have by now won a vast body of knowledge, and daily make positive additions to it, not-withstanding their theoretical controversies.

In philosophy, on the contrary, the same great problems are discussed by generation after generation with rather meager results other than a multiplication of theories and schools of opinion.

Ducasse (1935), Philosophy of Science 2, p.121


Interpretation of results


Results from observations are interpreted, meaning they are placed in a context. Interpretation may i.e. mean that visual observations are claimed to be reliable.


Interpretation may be subjective but, according to my opinion, is a part of scientific work, provided that the process is so well documented that it can be reproduced or criticized by an independent colleague.


Hypotheses, theories and laws


A hypothesis is a statement, an idea, a fantasy or just some simple though of no specific type. It may, or may not, show a connection to what we call reality.


In a scientifically documented publication, a hypothesis is often an attempt to explain or generalize observations. Hypotheses are interesting and important components within scientific publications.


Verification, falsification, theory, law


As observations results in that a hypothesis is verified, or that its consequences are verified, the belief in the hypothesis increases.

A falsification, i.e. a verification of a negation to the hypothesis or to its consequences, is of course disturbing; it decreases the belief in the hypothesis and may result in that the hypothesis must be discarded or modified.

When observations that verify a hypothesis are generally accepted, the belief in the hypothesis is increased and it may become termed as a theory.

The border between hypothesis and theory is very diffuse. Hypotheses that have been widely discussed and are generally believed to be in accordance with our perceived reality may be called "laws".


Verified hypotheses, theories and laws are, together with observations, important parts of science. But serious scientists never express the opinion that something within science should be "true" or should represent "absolutely certain knowledge". Every scientist is, sometimes painfully, aware of that an article may be published that will claim other hypotheses than their own belief.

An example of this skepticism about hypotheses is that "energy cannot be created", a fully tested and accepted law, is called for "postulate" or "law".


Theoretical publications


An additional and common method to create hypotheses that gain confidence is to connect previously accepted hypotheses. Often mathematical relations are used for connection, which imply that the result may be quantified and be compared to observations.

When such a connection has been verified by observations, or by traceable logic like mathematics, a very strong belief in the starting hypotheses and in the result hypothesis is created


If we believed strongly in the starting hypotheses, an emotion that we call understanding may then be experienced.

Science is largely structured using such connections.

This may contribute to that philosophers, unaware of that the original hypotheses described results from observations, claim that theories are the basis for what we believe about our world.


"Scientific" and "non-scientific" hypotheses


A hypothesis is hence a more or less credible statement. The credibility is based on observations and how these are reported. In cases they are reported according to scientific methodology the hypothesis may be accepted in a scientific magazine.

Discussions about a hypothesis being "scientific" or "non-scientific" is hence basically illogical and unnecessary - a hypothesis is not in itself neither of these two.


The question is whether the observations, forming the basis for the hypothesis, are reported using something reminding of scientific methodology.

The interesting feature of a hypothesis is if it is entertaining, stimulating or if it, through its origin or consequences, can be shown to conform to what we call our reality.


Hypotheses not based on observations


Quite often, hypotheses are expressed that are neither based on carefully described observations nor can be verified through its consequences. The source may as examples be observations that are not possible to reproduce during testing, analogies to other areas, assumptions that are not generally accepted, or a wish of the hypothesis' creator to receive a favour of some kind, without having to care too much about our perceived reality.

Such hypotheses may later be verified by scientifically described observations, or may be shown not to be in accordance with such observations.

In order to avoid mistakes, the community of (natural) scientists views hypotheses that are not supported by observations, as unreliable.


This skeptic view is sometimes criticized, e.g. by people advocating alternative reality.

A vast majority of not verified hypotheses have been forgotten or have been shown to be erroneous.

But also a few examples of hypotheses exist that were not accepted by the scientific community, but later were shown to be in accordance with observations and were accepted.

Some examples of creators of hypotheses, that unfortunately only after their death were acknowledged, include Ludwig Boltzmann (thermodynamics), Alfred Wegener (tectonics) and Milutin Milanković (climate).


Logic and mathematics


Logic and mathematics are tools that help us to organize observation results. As tools they contribute to what we call understanding. We may for instance understand that different observations have the same cause, that experiences are possible to pass from one area to another, or that a certain result always follows from a certain cause.

Use and development of logical tools, e.g. mathematics, are very important within scientific work.

But is mathematics science?

It is sometimes discussed if mathematics is a part of science, as mathematics is not considered with empiric methodology. In my opinion it is similar to a discussion if letters are a part of science. Both are tools used in description of what we call reality.

My opinion is that a well-documented development or discovery of a mathematical relationship is a scientific result, similar to e.g. the discovery of a new species of fish. The observation consists of the relationship and the documentation determines if the development or discovery should be accepted within the scientific society.


Use of mathematics and logic to connect different observations, and by this to create increased acceptance and feeling of understanding, is according to my view also a scientific contribution.

Freely adjustable parameters:

Use of known mathematical relations does not necessarily imply that a publication becomes "scientific". It is unfortunately common that "models", or mathematical constructions that contain freely adjustable parameters are used in order to increase the credit of some reasoning.

When the values of adjustable parameters are not coupled to observations, this type of reasoning becomes as scientifically uninteresting as the phrase "I guess that...", it may only appear as being more scientific.

With four parameters I can fit an elephant and with five I can make him wiggle his trunk.

John von Neuman (according to Enrico Fermi)