AF Chalmers - What is this thing called Science?

Chapter 5 – Introducing falsificationism



Return to "Chalmers - Contents"
























Chapter 5 – Introducing falsificationism

Here Chalmers stops focusing on problems in methods and instead tries to exaggerate advantages in what he describes.

Jump to p.103
As arguments related to lack of “absolute certain knowledge” are not new, and the arguments related to science are logically erroneous and irrelevant, i.e. they lack contact with our experienced reality (see e.g. comments to p.141-142), the reader may pass the content and directly jump to p.103.

Chalmers there agrees with citations by Popper in that only falsification is not a tool during hypothesis testing. This was shown already when Robert Grosseteste (about 1168–1253) tried with falsification [according to John Losee – A Historical Introduction to the Philosophy of Science 4Ed. (2001) p.33].

Hence falsification cannot, according to Chalmers, imply a definition of the term science, even if, like Chalmers do, the illogical arguments of Popper would be accepted.

Deductive test
Popper aimed at discussing theoretical hypothesis testing, i.e. test that only uses deductive arguments and therefore should be able to give “absolutely certain knowledge”, which is commented at p.103.

Falsification of hypothesis
Definition according to
Falsification implies a verification of the negation of a hypothesis.

The definition demonstrates that if verification is denied (as done by Popper) falsification is simultaneously denied.

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

Neither verification nor falsification result in “absolutely certain knowledge” about our world, a term that cannot be shown to have any content within epistemology.

“Logically possible observation”
Popperians use the expression “falsifiable” hypothesis, which implies that it was possible to discover a “logically possible observation” that can falsify the hypothesis (see p.62). A “logically possible observation” raises problems of understanding, as logic is not related to our perceived reality. All types of observations are therefore logically possible.

As example the hypothesis about gravitation should be falsifiable by dropping a stone and see that it do not fall, see p.62. It is also logically possible that a god suddenly should step up from its divine doghouse and using short and long barks give a Morse coded message about that Popper’s theses are not working in our perceived reality.


The clash between Carnap’s supporters and those of Popper was to be a feature of philosophy of science up until the 1960s

The English version of Poppers book was released 1959. The “clash” probably consisted in that no established philosopher of science found Popper’s theories relevant.


Popper became suspicious of the way in which he saw Freudians and Marxists supporting their theories, but accepted Einstein’s theory based on a single verification.

Popper argued for that a number of social sciences should be regarded as non-scientific. Oddly enough it is within those sciences that Popper’s theses are greeted with the deepest admiration.

Please note that Popper appreciated the verification of Einstein’s hypothesis.


Falsificationists freely admit that observation is guided by, and presupposes, theory.

Observations are often, but not always theory dependent. See comment to p.12 and 13.

Popper argued that hypotheses should be created independent of observations.


Falsificationists abandon any claim implying that theories can be established as true or probably true in the light of observational evidence.

Nobody claim that neither hypotheses nor theories can be established as “true”, neither based on observational evidence nor on something else.

Everybody believes in that theories may be probably true, see for a discsussion of probability arguments.

That the falsificationists "abandon claims" is probably because by that no falsificationist ever has succeeded in creation of a hypothesis that had any connection with our perceived reality, see below.


Theories are construed as speculative and tentative conjectures or guesses freely created by the human intellect.

Hypotheses are structured observations.

Chalmers has now forgotten the argument he erroneously claimed was a problem for induction on p.52:

If hypotheses are to be freely created, independent of observations, an infinite number of such hypotheses are possible. A finite number of these hypotheses are meaningful. The probability to create a meaningful hypothesis without referring to observations becomes a finite number divided with an infinite number.

The probability to that a meaningful hypothesis could be created according to Popper’s theses hence becomes equal to zero.


Once proposed, tentative theories are to be rigorously and ruthlessly tested by observation and experiment.

Chalmers does not mention the term “verification” because Popper denied verifications and induction.


Theories that fail to stand up to observational and experimental tests must be eliminated and replaced by further speculative conjectures.

As said above, Popper has an infinite number of hypotheses to test. As each hypothesis should be completely randomly guessed, the observations fall within widely separated areas.


Although it can never be legitimately said of a theory that it is true, It can hopefully be said that it is the best available.

Every hypothesis that have been accepted within science is the “best available”, as it otherwise should have been displaced by a “better”. A similar argument has been used as support of induction:

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. Isaac Newton - Opticks, 4 Ed. (1730) p.380


No problem with induction exists because science, according to falsificationists, does not involve induction.

Popper tried to claim that induction was not important, because it cannot be used in case you want to prove something using logic only. He missed that nothing in our perceived reality can even be discussed by using logic only.

Popper’s logical error was that he without induction cannot obtain any experience whatsoever about the world, neither to verify a hypothesis nor to falsify (verify the negation of) it.


Some theories may, according to falsificationism, be shown to be false by an appeal to observation and experiment.

From singular observation statements universal laws may be falsified deductively.

To demonstrate that a hypothesis is not in accordance with observations is equal to a verification of the negation to the hypothesis. Popper claimed that verifications do not exist.

Every observation, including singular, implies a verification of something. A statement about a white raven is identical with a verification of presence of a white raven.

Only a single verification is rarely sufficient. Did the observer saw it wrong? Is the observer lying? Therefore induction is needed (several similar observations that yield the conclusion that they the observations are representative) also in a trivial example like this.


Falsifiability as a criterion for theories



The falsificationist sees science as a set of hypotheses that are tentatively proposed with the aim of accurately describing the behavior of some aspect of the world.

A popperian does not aim at describe anything in the world accurately.

A description of an object in the world implies, to start with, a verification of that the object actually exists in our perceived world. Only then the properties of the object may be described, which presupposes observation and often induction, e.g. as a method to decrease the risk of erroneous interpretations.

Both verification and induction is denied by the popperian.


If a hypothesis is to form a part of science, it must be falsifiable.

Within science, a hypothesis should be verified in order to be considered interesting. This decreases the proportion of loose statements and fantasies in scientific reports.

Popper claimed that a hypothesis instead have to be falsifiable to fit as science. One of his logical errors is that the number of such hypotheses is infinite, and includes all verifiable erroneous hypotheses.

A problem with the usability of falsification as a definition of science can be seen by that the falsifiable phrase “Harry Potter was able to cast charms with his magic wand” according to Popper is scientific. See also comment to p.62, below.


It is Important to be clear about the falsificationist's usage of the term "falsifiable".

A strict definition of the term “falsifiable” should here be clarifying, (see comment to this chapter’s title, p.59), but reveals merciless deficiencies with Popper’s theses.

It is actually not interesting at all how the popperian uses the term falsifiable, because Popper’s dogma, due to erroneous logic in his inferences, create results that are not relevant.

Something like a definition is given by Chalmers at p.102, where falsifiable is said to be equal to “testable”, i.e. both verifiable and falsifiable.


“A dropped stone falls downwards” is falsifiable because it is logically possible that it may fall upwards, although it may be that no such statement is ever supported by observation.

Everything in our perceived reality is, without references to observations, logically possible. See comment to this chapter’s title, p.59. Everything is hence falsifiable according to Chalmers.

A possibly more correct treatment of this statement proceeds from the definition of the term “falsify” (see comment to this chapter’s title, p.59):

“Falsification implies a verification of the negation of a hypothesis”.

Falsifying in the form “If a hypothesis results in a certain consequence, and this consequence is shown to be erroneous, the hypotheses must be erroneous” is called modus tollens. This form of logical reasoning is known since the Greek antiquity, and was also discussed during medieval times e.g. by Robert Grosseteste (~1168–1253).

Please note that “show a consequence to be erroneous” implies to verify a negation to this consequence.

The fruitful negation to “A dropped stone falls downwards” is “A dropped stone do not fall downwards”. This phrase is not verifiable through observations and is hence, according to Popper’s theses, not scientific.


A hypothesis is falsifiable if there exists a logically possible observation statement or set of observation statements that are inconsistent with it, that is, which, if established as true, would falsify the hypothesis.

A “wonderful” sentence because:

Every observation statement concerning our perceived reality is logically possible. See comment to this chapter’s title, p.59.

On the other hand can nothing in our perceived reality be shown to be “true”, which is well known within epistemology.

See instead the definition of “falsify” in the comment to this chapters title, p.59.


Degree of falsifiability, clarity and precision

A reasonable definition of “more falsifiable” is that a “more falsifiable hypothesis” is a combination of at least two falsifiable hypotheses.


The logical situation that renders the derivation of universal laws and theories from observation statements impossible, but the deduction of their falsity possible (p.67).

Therefore falsifications become the important landmarks, the striking achievements, the major growing-points in science (p.67).

The more precisely a theory is formulated the more falsifiable it becomes.

See comment to p.61 for the first statement. Chalmers himself claims that the middle statement was wrong on p.79. See comment to p.74 for the lower statement.


Falsificationism and progress

Without observation, verification and induction, we cannot reach any experience nor any progress. Therefore Chalmers here tries to argue for that progress can be reached by falsification.

Progress, and not even hypothesis testing (see comment to this chapter’s title, p.59), cannot be reached using only falsification.


Science starts with problems associated with the explanation of the behavior of some aspects of the world

The word “problem” that is less often used within science than e.g. “question” or “possibility”. When you repair a television set, you fix a “problem”. But when you investigate various types of birds or the properties of a semiconductor, there is no “problem”.

If a “problem”, associated with an explanation of some aspect of the world, exists, then this identification of the “problem” was not the “start of science”.

The "start" was when somebody observed this aspect and created hypotheses about its cause or its properties. The description came before the “problem” of it.

Sometimes a scientific publication is based on questioning earlier publications. And sometimes during research work, something that had not earlier been described is detected (serendipity).


Falsifiable hypotheses are proposed by scientists as solutions to a problem. The conjectured hypotheses are then criticized and tested.

Chalmers deliberately omits certain details:

The method actually implies that infinitely many falsifiable hypotheses, without use of experience (as experience is founded on induction), should be suggested as solution of the problem.

There is no advantage in that the infinite number hypotheses should be suggested by scientists as a child, a monkey or a random word generator could do as well.


Chalmers gives his own version of four examples of what he calls problems that have confronted scientists in the past.

The problems arise from observations.

Problem 1: How can bats fly so dexterously at night?

Problem 2: Why is the height of a simple barometer lower at high altitudes than at low altitudes?

and so on.

Does this not mean that science start with observations? The answer is a firm “No”.

As usual, Chalmers historic examples appear to be invented as attempts to confirm his hypotheses.

“Problem 1” is raised after it was observed a: that bats existed and b: that they fly dexterously at night. Science started with these observations. Then questions like why came.

One cannot ask “why” before it is known “that”.

“Problem 2” was answered a considerable time span after the barometer and its properties was observed and without what is claimed by Chalmers was questioned. Blaise Pascal wanted to transport a barometer to the mountain in order to measure the weight of air. Science also here started with observations.

History of the barometer:

Does this not mean that science start with observations? A better answer than offered by Chalmers is: “Of course, but isn’t that obvious?”


The claim that science starts with problems is perfectly compatible with the priority of theories over observation.

Aha! This was the background to the arguments above. See comments on p.12 and 13 about claims that theory comes before theory.

For comments on armchair philosophy see p.12 and 54.



Denna sida på Svenska


ver. 1.2