AF Chalmers - What is this thing called Science?

Chapter 9 - Theories as structures II: Research programs



Return to "Chalmers - Contents"
























Chaper 9 - Theories as structures II: Research programs

Mathematically formulated hypotheses form a convenient and effective method for creation of structure among observations.

When hypotheses created from various types of observations are possible to connect, this connection dramatically increases the confidence for both the connecting and the connected hypotheses.

Observations are then described by reference to the hypotheses.


… there was no clear guidance concerning which part of a theoretical maze was to be blamed for an apparent falsification.

See comment to p.74.


Some laws or principles are more basic than others … some are … fundamental … they are not to be blamed for any apparent failure.

Lakatos hence argued that the “most securely” hypotheses should not be questioned, i.e. they are to be considered as “certain knowledge”.

Within science there are expressed laws (e.g. energy cannot be created/destroyed, heat cannot flow from colder to warmer) that have been verified many times and have not been falsified, in spite of many attempts. These laws are not regarded as being “true” (as all other hypotheses) but are still considered by very high probability to be in accordance with our perceived reality.


A science can then be seen as the programmatic development of the implications of the fundamental principles.

Every philosopher wants to launch his own term, and “program” seems to be one of them. The term “program” seems to be similar to the concept “group of interlocked hypotheses”.

This is the old dream of the “ultimate cause”. In case you can find such a “true” cause, you can deduce the consequences from it and find “absolutely certain knowledge”. Many have tried.


Scientists can seek to solve problems /not by modifying the fundamental principles, but/ by modifying the more peripheral assumptions as they see fit.

Insofar as their efforts are successful they will be contributing to the development of the same research program

Within the sciences, scientists would very much contribute to modification of the most important principles within each science. If you succeed you get a Nobel prize.


To protect the hard core from falsification, additional hypotheses supplementing the hard core as the protective belt may be formed.



A heuristic is a set of rules or hints to aid discovery or invention.

Lakatos divided guidelines for work within research programs into a negative heuristic and a positive heuristic.

The negative heuristic specifies what the scientist is advised not to do, e.g. tinker with the hard core of the program.

The positive heuristic of a program specifies what scientists should do, e.g. how to modify or sophisticate, the 'refutable' protective belt.

Heuristic is a term for “pragmatic non-strict investigation method” and is, of course, not strictly defined.

Before the scientific method was seriously established, discussions about how science should be performed were active. Today science is defined by how results are reported, see for details.

Lakatos seems to be stuck in a pre-science context.

Positive heuristic obviously consists in that scientists may “sophisticate” “the protective belt”, which implies that verification of a hypothesis is approved.


Various discussions, including the examples that appear to be known by philosophers of science like Copernicus, Galileo, Newton and Einstein.



Popper … denied that comparison with the history of science was a legitimate way of arguing for a philosophy of science.

As said in comment to p.59, Popper was not keen on how our perceived reality fitted his theses.

This is logical: A philosopher that denies verifications should not refer to verifications.


Lakatos came to see the main virtue of his methodology was not  a source of advice for scientists but an aid to the writing of the history of science.



Theories should not be rejected in the face of apparent falsifications.

This is in accordance with comment to p.59:

Verification of a hypothesis implies that the confidence in the hypothesis increases. Falsification of a hypothesis implies that the confidence in the hypothesis decreases.


Chalmers hesitates about that the fundamental principles should not be questioned.

See comment to p.131.


Lakatos could not give criteria for progression and stagnation within a program or for 'elimination' of whole research programs"

This was not the first time a philosopher tries to overrate his hypotheses.

The interesting here was that somebody actually cared about detailed argumentation about that the details were not in accordance with our perceived reality.


Lakatos in effect conceded that his methodology was in no position to diagnose any contemporary theory as non-scientific … /but/ could only make judgments in retrospect, with the benefit of historical hindsight.

Lakatos hypotheses fall into pieces and demonstrate once again that claimed rationalistic philosophy readily is affected by huge problems.




Denna sida på Svenska


ver. 1.0