AF Chalmers - What is this thing called Science?

Chapter 4 - Deriving theories from the facts: Induction



Return to "Chalmers - Contents"
























Chapter 4 - Deriving theories from the facts: Induction

The term induction implies an expectation, based on that several similar cases have given similar results, that an additional similar case will give a similar result as previously.

A very large part of our experience is based on induction, and induction gives us possibilities to make plans in our lives.

It is since the Greek antiquity well known that results from induction not are inferred by absolute security (logically valid), and this is the reason for that arguments based on induction are called probability arguments (were the probability may vary from vey low to very high).


Also in this chapter, Chalmers assumes that appropriate facts can be established in science.

The term “fact”, related to “knowledge” is not defined within serious philosophy and is not used within science. See comments to p.xx. See also comment to p.27.

41-42 Logic is concerned with the deduction of statements from other, given, statements.

The explanation of deduction is correct, and an important term is the word “given”.

A statement about our world that is “given” is based on observation and induction.


If the premises in a valid logic argument are true, the conclusion is also true.

In case this statement was formulated correctly, it should be correct, see also the comment below.

Unfortunately no logically valid (“true”) premises about our perceived reality exist within philosophy.

If furthermore induction as a source for our experience is denied, also the existence of probable premises (based on observations of our world) are denied.


Logical deduction alone cannot establish the truth of factual statements.

In case this statement was formulated correctly, it should be correct, see also the comment above.

But the statement is deceptive. Chalmers expresses that logic “not alone” can establish logically valid (“true”) statements about the perceived reality. He omits that it within philosophy is well known that no reasoning whatsoever have possibilities to establish logically valid statements about our perceived world.


If we can be sure our premises are true then we can be equally sure that everything we logically derive from them will also be true.

A cheap trick in order to fool new students: The term “IF”.

It is well known that no premises from our perceived reality can be shown to be logically valid (“true”).


Can scientific laws be derived from the facts?

“Facts” do not exist, see  comment to p.xx.


Scientific knowledge cannot be derived from the facts if "derive" is interpreted as "logically deduce".

I really feel tired from nagging about sloppy terms!

No logically valid arguments whatsoever can be logically deduced from observations.

We live in a world which we try to understand based on probability arguments. Our only source to experience about our perceived world is induction


A characteristic of inductive arguments that distinguishes them from deductive ones is that, by proceeding as they do from statements about some to statements about all events of a particular kind, they go beyond what is contained in the premises.

A classical skeptic description about induction.

Presumably are we in every moment awake going beyond the premises, e.g. each time we let our experiences decide our choices.


What constitutes a good inductive argument?

What constitutes the term “good”


Chalmers’ definition about induction:

If a number of A’s have been observed under a wide variety of conditions, and if all those A’s without exception possess the property B, then all A’s have the property B.

See comment to p.41.


There are serious problems with this characterization of induction.

Here Chalmers is, for the second time in the book, interested in correct definitions. See also comment to p.xx.


Suppose I reproduced an experiment reported in some recent scientific journal, and sent my results off for publication. Surely the editor of the journal would reject my paper, explaining that the experiment had already been done!

Chalmers incorrectly claims that there is no interest in verifying publications. On the contrary there is, before a publications that contains surprising observations is accepted, often a demand that the observations are verified by a second observer.

If on the other hand Chalmers wants to publish observations from something that is already fully documented, and hence well known, he may suffer the problems that he mentions.


Further problems with inductivism



Chalmers calls the position according to which scientific knowledge is to be derived from the observable facts by some kind of inductive inference inductivism and those who subscribe to that view inductivists.

Chalmers definition of inductivism is given at p.84-85:
“The position according to which scientific knowledge is inductively derived from given facts.”

If the term inductivism should be used at all, a better definition would be:
“The position according to which all experience about our perceived world ultimately is inductively derived from observations.”

Scientific results consist of one part of all experience.


A serious problem inherent in inductivism is that it is not clear what induction amounts to.

See p.41, 45 and 47.


And there are further problems:

Scientific knowledge of the unobservable world like the electron can never be established by inductive reasoning.

Chalmers arguments may be interpreted in two ways:

He is correct: No “knowledge” about anything in our perceived reality can ever be established by any reasoning.

He is totally erroneous in case he should have avoided the term “knowledge”: By induction, indirect observations from electrons were generalized into hypotheses about their presence.

Chalmers wants to claim that only one single observation made only once, can form the basis of a new concept, in this case the electron.


The inductivist therefore has to reject much contemporary science.

Nope (se comment above).


If scientific laws are inductive generalisations from observable facts it is difficult to see how one can escape the inexactness of the measurements that constitute the premises of the inductive arguments.

Chalmers wants to argue that scientific laws are something else than descriptions about what we have observed and insinuates that the basis of scientific laws cannot be our experience.


David Hume claimed that scientific knowledge in all its aspects must be justified either by an appeal to (deductive) logic or by deriving it from experience

Hume analyzed in Enquiry concerning Human Understanding among other things experienced based connection between what we have observed and what we have not observed.

His conclusion was that, in spite of that we cannot logically derive such a connection, both men and animals still have a strong instinct to believe in it, and that our lives would cease if we did not believe in such a connection.


The attempt to justify induction by an appeal to experience involves justifying induction by appealing to induction, and so is totally unsatisfactory.

This is true and has been stated by e.g. Hume. Induction can, as every other generalization concerning our perceived reality, only be justified using induction.
If it is unsatisfactory or not is up to each person to decide as induction is the only method we have when discussing our perceived reality.


An attempt to avoid the problem of induction involves weakening the demand that scientific knowledge be proven true, and resting content with the claim that scientific claims can be shown to be probably true in the light of the evidence.


Chalmers has, at last, succeeded in expressing something that is philosophically correct, basic and well known within science! This is not an attempt, but a well established basis within science. This is the reason that nobody claims “scientific knowledge” but uses the term “scientific results”.


The probability of any general law is zero whatever the observational evidence. There will always be an infinite number of general statements that are compatible with a finite number of observation statements.

The general law will make claims about an infinity of possible cases whereas observational evidence consists of a finite number of observation statements.

The probability of the law is thus a finite number divided by infinity, which remains zero.

In our technological world eyebrows are raised upon reading that the probabilities of our general laws should be zero.

Chalmers erroneous logic is created by that he reasons from a starting point that hypotheses are created independent of experience.

Without experience, an infinite number of hypotheses are possible in order to explain a single observation. If  pebble is released it is, without any experience, equally logical to assume that it will fall upwards as that it will fall downwards. This has been discsussed by Hume in Enquiry as a support that logical reasoning alone is not suitable during interpretation of our perceived reality. We have to trust experience.

This reasoning becomes logically correct in case Chalmers instead wanted to criticize Popper’s theses, see comment to p.60.


What constitutes a valid deductive argument can be specified with a high degree of precision, whereas what constitutes a good inductive argument has not been made at all clear.

This type of argument is well known since the Greek antiquity.

It is equally known that no argument at all about our perceived reality is possible to claim using deduction alone. The premises in such logical statements are always created from observations and are therefore only probable, which results in that such deductive arguments are also only probable.

That no valid deductive arguments concerning our perceived reality exist whatsoever could have been expressed by Chalmers with high precision.


The appeal of inductivism



The inductivist view that scientific knowledge is derived from the facts by inductive inference.

The empiricist view is that scientific results are founded on observations that eventually have been generalized to hypotheses using e.g. induction. See also comment to p.49.


Chalmers gives a citation from an article by an economist (!): AB Wolfe – ”Functional Economics” from the book RG Tugwell (1924) ”Trend of Economics” that he had read about in CG Hempel (1966) ”Philosophy of Natural Science” p. 11).

Chalmers gives an indirect inductivist-citation, originally stated 1924 by an economist, which implies that it is virtually impossible for an author of a book about philosophy of science to find serious information about induction (please note the irony).

Chalmers could instead have cited virtually any successful scientist during the eighteenth century, when the scientific method still was less obvious. E.g. Isaac Newton wrote like this:

And although the arguing from Experiments and Observations by Induction be no Demonstration of general Conclusions; yet it is the best way of arguing which the Nature of Things admits of, and may be looked upon as so much the stronger, by how much the Induction is more general. / Opticks, 4Ed (1730) p.380


We have seen that collection of facts prior to the acquisition of any knowledge does not bear analysis.

Chalmers pushes his erroneous armchair philosophical thesis that theory comes before observation. David Hume in his book Enquiry discussed how a human that suddenly was brought to our world, may be thought to gain experiences. See also comment to lower part of p.12.


Parts of what the economist said Chalmers rejects. What remains has a certain appeal



Chalmers draws a diagram

The diagram shows the inductive-deductive method according to Aristotle (342-322 b.C.), that actually is still used (but not because it was documented by Aristotle) within science during investigations of new phenomena.


The laws and theories that make up scientific knowledge are derived by induction from a factual basis supplied by observation and experiment. Once such general knowledge is available, it can be drawn on to make predictions and offer explanations.

It would be appreciated if Chalmers used terms that are philosophically correct, accepted and well known.

Observations are gathered and generalized to a hypothesis using induction. Consequences from the hypothesis are deduced in order to test the hypothesis further and to suggest other observations or other hypotheses.


Chalmers gives an example of how an inductivist may explain the rainbow:

General laws are derived from experience by induction and in this case the laws of optics have been formed. Together with experimental parameters at a certain time it is possible from these to explain haw the rainbow is created.

Chalmers gives a reasonable description, but appears to believe that the “laws of optics” represent some kind of “truth”, i.e. that they are something else than a summary of observations, where the summary receives increased credibility because it is expressed as a mathematical relation between different observations that, through this, verifies each other.


Chalmers summarizes scientific explanations:

1: Laws and theories
2: Observation parameters
3: Result: Predictions and explanations

Oops! Now Chalmers has forgotten the initial observations and claims that it is laws and theories that are starting points. Se the previous description by Chalmers.


Even if the inductivist position appears attractive it is, at best in need of severe qualification, and at worst, thoroughly inadequate.

Chalmers again excludes that reasoning based on observations is our only source for experience regarding our world, including such experiences that are summarized in scientific hypotheses.


Denna sida på Svenska


ver. 1.31