Talk:Falsifiability/Archive 3

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

→

Archive 9

Recent edits

I have defended recent edits by a revert because they seemed to me a good attempt to improve the article. However I have some concern about them. 1) The paragraph "Purpose of falsifiability and common misconceptions" speaks about Popper's view, while the title sugget we are speaking about the "general" purpose. We should change the name of the paragraph to emphasize that we are talking of a particular POV. 2) Should we assume that the topic "falsifiability" is a subtopic of Popper Falsificationism? Or is it a more general topic? If it is more general (that0s what I think) then the paragraph "Purpose of falsifiability and common misconceptions" should be a subparagraph of "Falsificationism". --Pokipsy76 (talk) 13:33, 24 December 2007 (UTC)

I'd previously moved this "common misperceptions" essay back into a later place in the article, in the section on criticisms--with extensive notes saying hey, guy, look, I'm not deleting your stuff, I'm just moving it later in the article where it belongs-- but this guy who keeps sticking it in seems to have some kind of hobby-horse POV he's pushing, and keeps pasting it back into the beginning of the article where it makes little sense. So this time I just cut it completely; it's mostly irrelevant anyway, and the article doesn't suffer from having it deleted. Geoffrey.landis (talk) 16:11, 30 December 2007 (UTC)

Ok but now I can't understand why did you delete the paragraph.--Pokipsy76 (talk) 17:06, 30 December 2007 (UTC)

See, Popper originally developed falsifiability in the context of an attack on scientism (the view that science is based on evidence or can be based on evidence, or at least strongly supported by evidence). Perversely, it then became almost universally understood as in line with scientism, something that made Popper very sad. And describing the facts now is "hobby-horse POV", because this being the majority opinion legitimizes to describe this myth as the truth and so abuse Popper's name to support the views he criticized? --rtc (talk) 10:42, 6 January 2008 (UTC)

Ok, I understand your concerns, I agree we can mention the misconception. I wask just suggesting to move them in the appropriate subsection.--Pokipsy76 (talk) 10:59, 6 January 2008 (UTC)

I have no problem at all about putting it into the appropriate subjection, but "criticism" is certainly not the appropriate one. Popper's own views on the matter are not criticisms of falsifiability. --rtc (talk) 11:24, 6 January 2008 (UTC)

The appropriate subsection from my point of view is "falsificationism".--Pokipsy76 (talk) 11:43, 6 January 2008 (UTC)

150 years example

"all men die before they reach the age of 150 years" Exactly how is this example falsifiable rather than simply empirical? —Preceding unsigned comment added by CSears (talk • contribs) 17:07, 11 February 2008 (UTC)

What do you mean by "rather than simply empirical"? where do you see a difference between something being empirical and something being falsifiable? According to Popper, both words mean the same and are interchangeable. --rtc (talk) 16:43, 20 February 2008 (UTC)

And the 150 years example can be falsified by producing Mel Brooks' 2000 Year Old Man and thus it is testable.

can we have a reference or quote for this "The Popperian criterion excludes from the domain of science not unfalsifiable statements but only whole theories that contain no falsifiable statements". Why couldn't we just add some true falsifiable statement to a theory, thus making the theory "scientific". I will edit in a couple of days if no feedback 79.70.50.117 (talk) 05:30, 1 March 2008 (UTC)

based on which source will you edit the article? --rtc (talk) 06:53, 1 March 2008 (UTC)

Falsibiability and Physics

One of Popper's major reasons for criticising verificationism is that a universal law cannot be verified by finite evidence, given that it makes infinite predictions. At first sight, no such problem is there for falsificationism: a single anomaly proves a universal law wrong. However, contemporary physics does make certain predictions, such as the statement "Magnetic monopoles exist." This is a claim that can never be falsified, and as such, Popper must denounce it as unscientific. I can't recall reading this in the article, as it makes falsificationism a less probable conception of the scientific method.DDSaeger (talk) 23:57, 5 April 2008 (UTC)

Is "people have souls" nonfalsifiable?

In the intro it is suggested that "people have souls" is a nonfalsifiable statement. But a limited and perhaps impractical falsification may exist, depending on what is meant by "soul". A soul is usually described as a nonmaterial entity that contains the identity of a person, and which animates (but is independent of) the person's material body. If a soul existed, reincarnation and resurrection would be possible. So one could (for example) test whether a possibly reincarnated person remembers things from their past life. If no one ever reported this, it would be a falsification of the theory. Granted it is impossible to check all bodies that ever existed to see whether they contain reincarnated souls, but the same difficulty confronts a person trying to check whether all swans are white.

So what is it about "people have souls" that makes it so much less falsifiable than "all swans are white"?141.211.61.254 (talk) 02:42, 6 May 2008 (UTC)

In fact, not all swans are birds, see for example The Swan (TV series).Kmarinas86 ^(6sin8karma) 17:50, 6 May 2008 (UTC)

“All swans are white” is easily falsifiable because there are well-established conventional definitions of what constitutes “white” and what constitutes a swan. “People have souls” is unfalsifiable because there’s no clear and/or agreed upon definition of what constitutes a “soul”.

Furthermore, claims of having lived past lives are unverifiable because there’s no way to distinguish between someone who would have actually lived a past life, and someone who just studied the life of a dead person very thoroughly. I could claim I was Napoleon in a past life and bring up a multitude of facts about Napoleon’s life, but nothing can prove that I actually was Napoleon. This would be impossible to verify because anybody whose life can be researched by someone verifying the claim, could have been researched by the person making the claim as the basis for their claim. Anybody whose life cannot be researched is a dead end as far as research goes so there's no way to prove having lived a past life there. — NRen2k5^(TALK), 22:36, 25 May 2008 (UTC)

What about the definition I presented in my last comment? It's not as clear and agreed-upon as a swan's definition, but I don't think it's critically lacking in clarity or universality either. You can note a million questions I haven't yet answered about a soul, but I'm sure I could ask a million questions about a swan that wouldn't be in the common definition either. No definition is complete, perfect, or comprehensive.

Further, it is possible to distinguish between someone (call him John) who was Napoleon in a past life and someone who merely studied his life. For example, John might know some verifiable facts about Napoleon which are not in history books because they are presently unknown or known but of little interest. John might claim that he (as Napoleon) used a certain rare perfume and stored it in a secret drawer of his desk shortly before his death in exile at Saint Helena. The perfume could then be found and John's claim very strongly verified. Furthermore, John's relatives testify that he has never left his country, let alone traveled to Saint Helena. This would be very strong testable evidence of John's claim to have lived as Napoleon, no? 141.211.61.254 (talk) 22:31, 26 June 2008 (UTC)

Look up verificationism to see what you are describing.Kmarinas86 ^(6sin8karma) 21:33, 1 July 2008 (UTC)

Paraphrased list of Popper's observations

This article is densely written, or perhaps I'm just too dense to give it justice. This summary of Popper's position from a old version of the philosophy of science article seemed more straightforward to me:

Popper described falsibility using the following observations (paraphrased from Conjectures and Refutations):

1. It is easy to confirm or verify nearly every theory — if we look for confirmations.
2. Confirmations are significant only if they are the result of risky predictions; that is, if, unenlightened by the theory, we should have expected an event which was incompatible with the theory — an event which would have refuted the theory.
3. "Good" scientific theories include prohibitions which forbid certain things to happen. The more a theory forbids, the better it is.
4. A theory which is not refutable by any conceivable event is non-scientific. Irrefutability is not a virtue of a theory.
5. Every genuine test of a theory is an attempt to falsify or refute it. Theories that take greater "risks" are more testable, more exposed to refutation.
6. Confirming or corroborating evidence is only significant when it is the result of a genuine test of the theory; "genuine" in this case means that it comes out of a serious but unsuccessful attempt to falsify the theory.
7. Some genuinely testable theories, when found to be false, are still upheld by their advocates — for example by introducing ad hoc some auxiliary assumption, or by reinterpreting the theory ad hoc in such a way that it escapes refutation. Such a procedure is always possible, but it rescues the theory from refutation only at the price of destroying, or at least lowering, its scientific status.

68.167.254.2 (talk) 08:04, 6 July 2008 (UTC).

Quantum Mechanics

Wikipedia:Reference_desk/Science#Quantum_Mechanics

Question moved to the appropriate location.--OMCV (talk) 23:34, 9 October 2008 (UTC)

Naïve Falsification

I read this entire section, but am still unclear on what makes a falsification naïve; the term isn't clearly defined in its section.

I'm unclear on the whole damn thing. This is what I'm able to figure out...

U -> ~O

If the canadian goose is extinct, no canadian geese will be seen.

O

A canadian goose is seen.

~U

The canadian goose is not extinct.

V -> O

If the media reports a canadian goose is seen, then a canadian goose is seen.

V -> ~U

If the media reports a canadian goose is seen, then the canadian goose is not extinct.

U -> ~V

If the canadian goose is extinct, then the media will not report a sighting of one.

...

...

...SO??

Please someone ground this out with examples or explication so we simple-minded college-educated folk can figure out what you're trying to tell us or take it out! —Preceding unsigned comment added by 68.163.241.201 (talk) 06:22, 8 March 2008 (UTC)

Despite your uncertainty on interpreting this section. You have it completely right! Congratulations!Kmarinas86 ^(6sin8karma) 18:16, 19 June 2009 (UTC)

Conspiracy theories

This sentance:

Conspiracy theories are often essentially unfalsifiable because of their logical structure.

makes no sense.
1) Conspiracy theories (like other theories) can have very different logical structures.
2) What does "often" mean? Often according to whom?
--Pokipsy76 09:44, 2 June 2006 (UTC)

PaineDisciple If the social and political "powers that be" have almost unlimited possibilities of collectively "conspiring," why is it so difficult to understand that ordinary people tend to be interested in them and even to create them? Could it be that, contrary to the "big lie" which is hard to refute because it is beyond the capability of ordinary people to refute it, "conspiracy theories" involving government or corporate and financial elites are so big that they are beyond the capability of the average person (even the "average intellectual") to confirm or deny? Are they simply to much trouble or too dangerous to deal with by elite people concerned for their social status? It does seem that there is a social class basis to the automatic rejection of political or elitist conspiracies. The comforting stereotype for elitists is that only "uneducated" people would believe in large-scale, collective conspiracies by the "powers-that-be." Maybe it would be useful to distinguish between "conspiracy theories" and "scapegoat theories." The latter are what really cause most of the trouble in society. —Preceding unsigned comment added by 66.99.3.7 (talk) 18:20, 26 January 2009 (UTC)

Language too complex

"In science and the philosophy of science, falsifiability, contingency, and defeasibility are properties of empirical statements such that they must admit of logical counterexamples. This stands in contradistinction to formal and mathematical statements that may be tautologies, that is, universally true by dint of definitions, axioms, and proofs. No empirical hypothesis, proposition, or theory can be considered scientific if it does not admit the possibility of a contrary case."

The language here is too complex. It's full of philosophical buzz-words, which are supposed to make the writing sound precise and intelligent... but really we're just going to alienate anyone who hasn't already been schooled in this philosophical language. I'm going to try to dumb it down a bit. (e.g. "stands in contradistinction" = "contrasts")

Amen. Ironic too that more than one of the great philosophers discussed in the article were notable for simple writing styles. This was not b/c they were incapable of complex thought but b/c they chose to try to remain accessible. Specifically, as to the above quote, it's arguably in violation of WP:LEAD guidelines for accessiblity -- one hopes not for long (Popper et. al, RIP). Aim should be to inform & attract, not drive away. --Thomasmeeks

PaineDisciple There is a very strong tendency in modern times, for ALL types of experts to develop the most complex jargon possible. In my view, this parellels the development of organized religion where "priesthoods" try to make the religious ideas as mysterious as possible so that the masses of believers cannot possibly figure it out on their own (or can never be completely sure if they've got it right). This guarantees the privileged status of the "priesthood." If modern science is going to follow this path, it is no wonder that the "masses" will get tired of it. It is also true that the attitude of elitism tends to make scientists think of themselves as a kind of "infallible" priesthood. Carl Sagan, in "Science as a Candle in the Dark," said that scientific truths do not have to be taught as if they were handed down from Moses (or something to that effect). The scientific method and sceptical attitude is what should be taught in schools and to the public. I believe this is happening less and less today and that those scientists who are sincerely concerned with public support of science as a necessity in a democratic republic are "shooting themselves in the foot" if they don't challenge the elitist, "religious" attitude of scientists. —Preceding unsigned comment added by 66.99.3.7 (talk) 18:03, 26 January 2009 (UTC)

Necessary?

In the criticism section, an argument against Falsifiability is that "...falsifiability cannot distinguish between astrology and astronomy, as both make technical predictions which are sometimes incorrect." Would it make sense to respond to this saying that falsifiability might be a necessary condition rather than a sufficient one for something to be considered something?

I apologize for my lack of clarity. Feel free to go ahead and delete this if it doesn't make sense. Pete4512 07:21, 26 March 2007 (UTC)

Well there has been the impression among those who jump to conclusions that "Falsifiability is sufficient". However, that is not what the article says:

Popper uses falsification as a criterion of demarcation

And in the same way, it IS what the article says:

Verifiability came to be replaced by falsifiability as the criterion of demarcation

See, it's easy to get the wrong impression.

So, yes, it would make sense to point that out.Kmarinas86 08:36, 26 March 2007 (UTC)

There are other criteria, but falsifiability is a necessary criteria. Without it one cannot be certain whether one is dealing with objectively verifiable reality or with metaphysical and/or pseudoscientific speculations and beliefs. Belief often downplays experimentation and prefers to use rationalization, circular reasoning and speculation.

The process of creating knowledge involves the use of various rules of logic, used in such a manner as to sift through vast amounts of often confusing information, to produce universally useful bits of knowledge. These bits of knowledge can then be used, like building blocks, to build a reliable foundation upon which to base a repetition of the process. If some of the bits of knowledge in the foundation, with time, prove to be false, they are discarded.

Many of the rules of logic used by scientists are also used by quacks in everyday life. Otherwise they would not be able to communicate or function. So their rejection of scientific logic is somewhat hypocritical. Their immunity to cognitive dissonance allows them to at the same time be logical in one setting and illogical in another. They are simply less concerned than a good scientist is, with discerning the difference between perceived fact and proven fact. The importantce of falsification is not understood or deemed necessary by them. -- Fyslee (collaborate) 09:12, 26 March 2007 (UTC)

The new tightening of discussion and primary-source references to Popper and Watkins in Hypothetico-deductive model might clarify. --Thomasmeeks 00:47, 28 March 2007 (UTC)

PaineDisciple Why not just stick to the scientific routine that first we have an hypothesis (an "educated guess"); then when some substantial PROOF (obviously "falsifiable") has been demonstrated, it is a "theory"; and finally, when it has been thoroughly tested and is capable of predictions, it is called a scientific "law?" —Preceding unsigned comment added by 66.99.3.7 (talk) 18:07, 26 January 2009 (UTC)

'"God doesn't exist" is falsifiable'

On another hand assertion 'God does [not] exist' is falsifiable. [Note that a not seems to be missing here!] This assertion can be falsifiable by demonstrated [sic] the God.

If any evidence surfaced suggesting that God existed, this would show that God is a natural phenomenon, and certainly not a "God" as most people would intend the word to mean - it would probably just be a member of some previously unknown species (such an organism existing by itself is unlikely), and we would not really call it a "God".

The statement is just as unfalsifiable as its negation. But I may be misinterpreting the claim. Is it really justified to claim that this statement is falsifiable?

Colin 00:25, 1 June 2007 (UTC)

'God does not exist' is falsifiable. If he appears and talks to us and does some miracles that would show he exists. Of course 'God exists' is not falsifiable. Xavier cougat 19:15, 1 June 2007 (UTC) Xavier cougat 19:16, 1 June 2007 (UTC)

I know which statements the article is claiming is falsifiable. I have presented an argument above that "God does not exist" is not falsifiable, at least not without a very particular definition of "God" that would seem counterintuitive. I do not believe an entity appearing and claiming to be God would falsify a claim that God does not exist, because this entity would just be something natural, not the mystical stuff people want to call "God".

I would just remove the claim myself, but I was hoping somebody would have a better explanation (or a citation) that this claim is actually falsifiable. Colin 17:19, 2 June 2007 (UTC)

This might be a little amateurish, but I hope the following argument helps. The statement 'God does not exist' is falsifiable IMHO since God cannot be verified to be unobservable. In other words, 'God is observable' is itself not falsifiable. Of course, this meaning of God varies amongst cultures. I think references to God being unobservable should be qualified as assumptions since a lot of people (many of who try to convert others to some form of theism) actually believe they have 'observed God'.

I do agree that 'An unobservable God does not exist' is not a falsifiable statement. This is probably a more accurate statement and makes the article more explicitly clear.

It would be much easier to accept and understand the article, if 'God' were replaced by 'flying spaghetti monster'. 'An FSM exists' is not falsifiable since one cannot positively verify it's absence, but 'An FSM does not exist' is falsifiable since an FSM is observable. Theuse of an FSM instead of, say, a 'dog' is simply to make the article 1337 :) Viggyjiggy 16:29, 5 June 2007 (UTC)

If you saw a pasta on a flying plate [saucer ;)] over your house, does that make it god?◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 01:12, 28 June 2007 (UTC)

If you saw something that looked, felt, behaved like a chair, would that make it a chair? According to you, the answer is NO, it could be something else in disguise. If God was to manifest into physical form and this form looked, felt, behaved (, etc) as it needs to for it to be accurately described as a physical manifestation of God - then it is fair to say it is God. When it comes to God, people are always more critical/picky than they are for everything else... (Just to drive the point home: logically, it is perfectly consistent to hold that "god does not exist" is completely falsifiable ... At least hypothetically - I wouldn't hold your breath.)TranscendTranslation 01:57, 16 November 2007 (UTC)

"If you saw something that looked, felt, behaved like a chair, would that make it a chair? According to you, the answer is NO, it could be something else in disguise." Right. Just because I "saw" something does not mean that observation made it to be that way. That's a kind of sophism. And conversely, just because it appears to be a chair in many aspects does not mean that it cannot in any way be a chair; the maintain so would be untenable. People of all ages, but especially children, make mistakes when learning words. One example is by concluding that they saw a cow, when it was really a bison. Among adults, another type of mistake would be to conclude that what they saw in the sky was swamp gas, when it is really the remnants of fuel dumping from a multi-stage rocket. Given the ambiguity surrounding the terms used in alleged falsifiable (or alleged unfalsifiable) statements, there is no universal prescription to determine which of any such statements are falsifiable. Falsification of statement can only occur in principle if an act of verificationism can unequivocally make tenable the observation of a condition which the statement claims to be impossible. If want to avoid such ambiguities, a common lexicon would better be established, with emphasis on operational definitions. Strict standards with respect to the use of words must be coupled with the precision control of laboratory settings in order to maintain hard sciences as better at advancing our understanding of the world in comparison to soft sciences, historicism, philosophy, and religion. The only way these other studies of human thought would be hard sciences is from the perspective of some entity who relates to all that we know in the same way that scientists relate to their own controlled experiments. The degree of operational intelligence that must be gathered to make that possible is not fathomable except through perspectives most likely regarded as devotedly limited to the paranormal, mystical, or exclusive (i.e. secret knowledge). A abject failure to be taken seriously in this area is recognized by labels such as crackpottery, fraud, quackery, historical revisionism, and science by press conference.Kmarinas86 ^(6sin8karma) 19:15, 19 June 2009 (UTC)

The problem here is that "falsifiable" relies on a surrounding framework. For instance, even if God did appear and shout "I exist" at everyone in the world (while performing miracles to prove it's him), it wouldn't be publishable scientifically as a falsification of "God does not exist" because you can't independently repeat the experiment later. Furthermore, the following day one could argue "it was an illusion/conspiracy ... show me evidence better than my fallible memory and your fakable video tape" and the statement would return to being un-falsified again. What makes "God does not exist" unfalsifiable is the framework around "what would be accepted as falsifying evidence". Indeed, for instance, Christians would assert that Xavier cougat's scenario (earlier in the thread) has already happened -- that God has appeared, spoken to us, and worked miracles -- but that we are now in that "day after" situation claiming it didn't really happen. Effectively telling the (hypothetical) God "but you've got to work miracles on demand and unceasingly or as soon as you stop we'll claim it was just an illusion/conspiracy/didn't really happen". —Preceding unsigned comment added by 124.171.142.223 (talk) 03:32, 18 July 2009 (UTC)

Popper and confirmation

The article states:

"Sokal and Bricmont write, "When a theory successfully withstands an attempt at falsification, a scientist will, quite naturally, consider the theory to be partially confirmed and will accord it a greater likelihood or a higher subjective probability. ... But Popper will have none of this: throughout his life he was a stubborn opponent of any idea of 'confirmation' of a theory, or even of its 'probability'. ... [but] the history of science teaches us that scientific theories come to be accepted above all because of their successes." (Sokal and Bricmont 1997, 62f)"

But, IIRC, a largish section of one of Popper's books (Logic of Scientific Discovery or Conjectures and Refutations) contains quite a bit of dense probability talk aimed at providing a measure of confirmation of a theory. Obviously, someone needs to take a gander at the original sources (I left mine in storage in California) to confirm what I recall. If that does indeed match up, then what we have is a relatively straightforward demonstration that the assertion by S&B above is overwrought and does not belong in this article. It's possible that I got confused and at the end Popper says that all the probability work shows that confirmation can't be considered even in principle, but that isn't my recall offhand. --Wesley R. Elsberry 13:36, 13 August 2007 (UTC)

Popper explicitly denied that the degree of confirmation (a misleading terminology, which he first used, but then rejected) or degree of corroboration of a theory is a logical probability, several times and as explicit as it can be. This degree is merely a report of the tests that have been performed on the theory.

What then of corroboration? "Why do corroborations matter?" [...] It must be remembered that it is part of falsificationist epistemology that the passing of a test does nothing to secure, confirm, or in any way brighten the prospects of a theory. Nonetheless the question can be answered satisfactorily.

The answer is that corroboration doesn't matter, even in the practical realm. It has no epistemological significance at all, as Popper always insisted. But testing matters, and has undeniable methodological significance. We want true theories. Testing is important because it is only be subjecting our theories to tests that we have any opportunity of eliminating those that are false; and the more severe the tests, the more generous the opportunity. We might put things this way: when a theory fails a test, we learning something but end up knowing nothing (since what we knew, our theory, has been eliminated). But when a theory passes a test (when, that is to say, it is corroborated), we learn nothing (since we already knew what the result of the test was going to be) but we continue to know something. Corroboration is doubtless needed if science is to exist, for if no theory were ever corroborated there would be no science; but it makes no contribution to the growth, or to the progress, of science. (David Miller: Critical Rationalism, section 6.3)

In recent years, it turned out that the key point of Popper's philosophy is really his anti-justificationism,

• "We must regard all laws and theories as guesses." (Objective Knowledge, 9)
• "There are no such things as good positive reasons" (The Philosophy of Karl Popper, 1043)
• "In short, positive reasons are neither necessary nor possible" (The Philosophy of Karl Popper, 1974, p. 1041)
• "The truth of any scientific theory is exactly as improbable, both a priori and in relation to any possible evidence, as the truth of a self-contradictory proposition" (ie., 0 or "almost" impossible)
• "Belief, of course, is never rational: it is rational to suspend belief." (The Philosophy of Karl Popper, 69)
• "I never assume that by force of ‘verified’ conclusions, theories can be established as ‘true’, or even as merely ‘probable’." (Logic of Scientific Discovery, 33); by ‘true’ in quotation marks he means justified true belief; not absolute, objective truth, by ‘probable’ in quotation marks, he means logical probability à la bayesianism; not that statistical theories such as "this dice when thrown here will show 1 with probability 1/6" cannot be true.
• "[O]f two hypotheses, the one that is logically stronger, or more informative, or better testable, and thus the one which can be better corroborated, is always less probable—on any given evidence—than the other." (Logic of Scientific Discovery, 363)
• "[I]n an infinite universe [...] the probability of any (non-tautological) universal law will be zero." (Logic of Scientific Discovery, 363)

This is the central element that distinguishes him from all other philosophers. As Bartley puts it, Popper provided "the first non justificational philosophy of criticism in the history of philosophy." (William W. Bartley: Rationality versus the Theory of Rationality, In Mario Bunge: The Critical Approach to Science and Philosophy (The Free Press of Glencoe, 1964), section IX). Popper says that it is never rationally justified to accept a theory. It is always an act of free will and of personal responsibility. Popper's philosophy is concerned only with suspending the belief in a theory if it has been successfully criticized. Popper is misunderstood so often and so badly because basically all the world is justificationist today, and thus basically anyone reads his ideas in a false justificiationist interpretation. --rtc 03:06, 22 August 2007 (UTC)

Rtc, I happen to really, really like you and much of what you've said in our previous exchanges, though I recognize that is irrelevant to this discussion. At the moment, the problem, IMO, is that Popper's criticisms of what he perceived to be misunderstandings of his earlier advocay of falsifiability are not appropriate for the lead section of this article. Therefore, I'm removing the newly inserted material in the lead and will put it below on this page so it can be discussed as to where it might possibly be appropriate. Sincerely yours, ... Kenosis 04:04, 22 August 2007 (UTC)

My point was more narrow. Sokal and Bricmont are certainly correct about general tenor, but what I was pointing out that there was a possibility that Popper had taken the trouble to consider confirmation or corroboration. I think I was careful before to emphasize that my recall of the specifics may be faulty, but I still think that it warrants a few minutes time of someone who may have the two cited books on hand to pop them open and have a look to see. --Wesley R. Elsberry 17:39, 22 August 2007 (UTC)

He does, and he rejects the view that corroboration is a logical probability. (chapter X—not to be confused with the chapter VIII about statistical probability statements—, new appendices *VII, *IX, *XVI, *XVII, *XVIII, *XIX) Admittedly, he still presents corroboration as a "measure of the rationality of our beliefs" (new appendix *IX). However, Bartley has clarified that, to make sense out of Popper's position in a coherent way, "[i]f one is to avoid unprofitable hermeneutics here, one must ignore these passages."[1] So in this aspect, S&B are true to how Popper at least should be understood. If you want to read a rebuttal of S&P's negative judgement on Popper, have a look at [2]. Note: I am using the most recent German edition of the original Logik der Forschung. It contains many additions that seem not to be present in Logic of Scientific Discovery. The book I have ends with new appendix *XX, added shortly before Popper's death. Please pressure the publisher of the English version to bring it up to date if you can! --rtc 23:05, 22 August 2007 (UTC)

PaineDisciple Why not just stick to the scientific routine that first we have an hypothesis (an "educated guess"); then when some substantial PROOF (obviously "falsifiable") has been demonstrated, it is a "theory"; and finally, when it has been thoroughly tested and is capable of predictions, it is called a scientific "law?" —Preceding unsigned comment added by 66.99.3.7 (talk) 18:10, 26 January 2009 (UTC)

moved section

I disagree strongly with the edit [3]. You attempt to protect the mainstream misreading of Popper against the facts. You obviously moved it down to criticism to play down its significance. It is not a criticism, it is a clarification. --rtc (talk) 16:37, 21 December 2007 (UTC)

I'm sorry you disagree with the edit. The material you added seems reasonable to put in the article, but it does not belong in the introduction-- the third sentence-- before the index. I put it in under "criticisms" because that seemed closest to the right place to put it, but if you think a different section works, that's fine with me.

I didn't move it down "to play down its significance," I moved it down because it's doesn't belong in the introduction.

As for your remark that your material is apparently correcting what you call "the mainstream misreading of Popper"-- in that case, if the mainstream reading of Popper is a "misreading" then it this apparently does belong in under criticism.Geoffrey.landis (talk) 00:47, 23 December 2007 (UTC)

So let's be explicit: Do you want to suggest that falsifiability has to be described in the way the mainstream misreads and abuses it, and that Popper's desparate struggle against this should be described only as a criticism? I'd suggest we take mainstream misunderstandings of a position only as an indicator for its relevance, never for its content, and if we describe the position, we describe it as it was actually held, not as it was misunderstood. The mainstream misunderstanding of Popper is "well known and well documended" (M. Artigas, The ethical nature of Karl Popper's theory of knowledge, p. 15). --rtc (talk) 02:37, 23 December 2007 (UTC)

This is better. However, the logical flow of an article is that is should first clearly explain what falsification is, and then discuss the criticisms. This section is apparently a critique of falsification (and the fact that you are quoting Popper makes it no less a critique-- Popper can critique his own theory, and the ways in which other people use his theory).

The new version seriously looks to me like you're trying to push a POV. It should go under critiques.Geoffrey.landis (talk) 03:33, 23 December 2007 (UTC)

This is not a critique, it is Popper's own view, and it was his view already when he published The Logic of Scientific Discovery, which says this clearly in the footnotes there. The original German version of the book which contained the mistake without comments had only very limited circulation. If you understand as POV presenting as falsifiability something that contradics the "mainstream interpretation" of it then I will gladly admit that I am trying to push POV. Just because it is an inconvenient truth that the mainstream misrepresents and abuses falsifiability so badly, it does not mean that Popper's actual views on the matter may be made a second-class citizen by grading them down to a mere "criticism". Do you think that this article should describe falsifiability or do you think that it should describe "the ways in which other people [ab]use [it]"? --rtc (talk) 13:52, 23 December 2007 (UTC)

A separate article devoted to Popper's views of falsifiability may be in order and the current one replaced by a disambiguation page that relates the concept of falsifiability to differing contexts.Kmarinas86 ^(6sin8karma) 20:18, 19 June 2009 (UTC)

This section violates neutral point of view quite flagrantly with the sentence beginning "Yet for certain preconceived reasons, hardened creationists...". No citation, no evidence, just a declaration that those creationists are hardened and have preconceived reasons. The previous sentence's strenuous emphasis "not necessary" could also be a non-neutral attempt at an authoritative statement with no citation. —Preceding unsigned comment added by 203.143.164.250 (talk) 05:07, 15 July 2009 (UTC)

Stanford description

There's a reasonably good discussion of falsifiability on the Stanford University Department of Philosophy web site.[4]. Their definition is simple: "A theory is scientific only if it is refutable by a conceivable event. Every genuine test of a scientific theory, then, is logically an attempt to refute or to falsify it, and one genuine counter-instance falsifies the whole theory." That's clearer than what Wikipedia has now.

A theory is an aggregate of hypotheses, some of which are central premises underpinning the entire theory (e.g. In the way the Cosmological principle is to the Big Bang Theory). The majority of hypotheses in a given theory are not critical to the whole set. For example, the understanding of relativity does not disprove every hypothesis of Newton's understanding. Finding a fossil in the wrong place in geological strata need not disprove Evolution so as long as evolution does not need to be the sole determinant of the placement of fossils.Kmarinas86 ^(6sin8karma) 20:26, 19 June 2009 (UTC)

They then go on to a discussion of the hard part of the problem - evaluating whether one theory is "better" than another. It's easy to generate theories which are hard to falsify, but don't predict much. In their words, "for Popper any theory X is better than a ‘rival’ theory Y if X has greater empirical content, and hence greater predictive power, than Y." This is followed by a long discussion of the philosophical history of trying to get hold of the concept of "better" in some formal way. The article currently does not address this well.

There's a practical issue ignored in the article. Engineering is based on theories with predictive power. Engineering is about knowing whether a bridge will hold up or a circuit will work before building it. Only falsifiable theories have reliable predictive power. Hence, engineering is based on falsifiable theories. Philosophers tend to ignore this part, and engineers take it as a given, so it's not mentioned much. I'll look for a cite. --John Nagle (talk) 17:57, 23 December 2007 (UTC)

Whoever claims that "one genuine counter-instance falsifies the whole theory" (which is naive falsification) has never actually read the Logic of Scientific Discovery, which says for example "If accepted basic statements contradict a theory, then we take them as providing sufficient grounds for its falsification only if they corroborate a falsifying hypothesis at the same time." (section 22). That is, it is never sufficient to present a counter-instance; a falsifiable explanation for the occurrence of this counter-instance is needed. The claim that "Engineering is based on theories with predictive power. Engineering is about knowing whether a bridge will hold up or a circuit will work before building it. Only falsifiable theories have reliable predictive power. Hence, engineering is based on falsifiable theories." is also false on closer inspection, and it is wrong that philosophers have ignored this issue. "There is simply no such thing as the scientific knowledge required for the manufacture of a television set or a can of beans; no amount of science will ever tell us how to tackle the job--there may be countless ways that will succeed. All that science can do is to tell us what lines of approach we do well avoid. That is, we do not rely on, or even apply, scientific hypotheses when we act; we exploit them. I should dub this insufficiently appreciated truism the slavery theory of applied science." (D. Miller, Critical Rationalism, 2.2g) It is further false that "Only falsifiable theories have reliable predictive power." No theory, not science, not any other, gives reliability; and predictive power is not exhausted by falsifiable theories. "Science offers no security. Science has no authority." (David Miller: Being an Absolute Skeptic. Science 284 (4 June 1999), pp. 1625–1626) --rtc (talk) 19:18, 23 December 2007 (UTC)

Actually, that quotes Popper out of context. Back up about one page. He's struggling with the problem of experimental error, or, as he puts it, "non-reproduceable single occurrences". He wants the falsifying data to demonstrate some repeatability, which implies some definition of "repeat" for that type of data. He calls this a "low level empirical hypothesis". He's not insisting that a theory cannot be falsified until a "better" theory is available to replace it. --John Nagle (talk) 20:14, 23 December 2007 (UTC)

I am not at all quoting Popper out of context, in fact, I am quoting him in the larger context of the whole book. I did not claim that he is "insisting that a theory cannot be falsified until a 'better' theory is available to replace it", although you have got quite the right idea if you think into this direction, and it is almost what Popper says (he says that one should not give up even a falsified theory as long as there is no better theory to replace it—Doing it would lower the content and hence violate methodological rules). The statement we are discussing is "one genuine counter-instance falsifies the whole theory". Now let's assume that by "falsifiable explanation for the occurrence of this counter-instance is needed" I meant falsification in the methodological sense (as it is described in the article as "Thus, while advocating falsifiability as the criterion of demarcation for science, Popper explicitly allows for the fact that in practice a single conflicting or counter-instance is never sufficient methodologically to falsify a theory, and that scientific theories are often retained even though much of the available evidence conflicts with them, or is anomalous with respect to them.") while you meant it in the logical sense, and let's further assume that you are right that the place I quoted speaks of falsification in your sense and that I was slightly in error there, and let's also assume that the quoted text which I criticized in fact talked about falsification in the logical sense. Then, even if we read "genuine counter-instance" as what you call "falsifying data [that] demonstrates some repeatability", it is then still false. Popper clearly discusses (section 18) the claim that the whole theory is falsified by a counterexample, and rejects it in this generality with several objections, which are primarily that the theory may consist of independent axioms and that the contradiction can then be traced back to a specific one; and further that the theory can be decomposed into theories of smaller generality by analyzing its consequences, of which the problematic one can be identified, corrected and the theory re-assembled into a new one. So in any case, the statement I criticized does not properly describe Popper's views and "The logic of his theory is utterly simple: if a single ferrous metal is unaffected by a magnetic field it cannot be the case that all ferrous metals are affected by magnetic fields" is not quite right; if at all, it holds only for theories of low complexity and generality. --rtc (talk) 22:52, 23 December 2007 (UTC)

Logic

In the article as it currently stands (4-28-09) under the "Logic & mathematics section" contains the statement:

"In this way of looking at things, logic is a science that seeks after knowledge of how we ought to conduct our reasoning if we want to achieve the goals of reasoning. As such, the logical knowledge that we have at any given time can easily fall short of perfection. Thus rules of logical procedure, as normative claims about the fitness of this or that form of inference, are falsifiable according to whether their actual consequences are successful or not."

The most straightforward interpretation of this claim is incorrect. Truth value is defined by reference to the laws of logic. Therefore, any "falsification" of the laws of logic must first make reference to the laws of logic and is therefore self-contradictory. The intended meaning of "the rules of logical procedure" is vague; however, the statement itself must be changed to avoid being misleading at best and completely false at worst.

Dialegomai (talk) 14:12, 28 April 2009 (UTC)

Thanks for starting a discussion about this matter. I have reverted your edit as it is unsourced. Please find a V & RS and format it properly so that it shows up in the references section. The best thing to do before adding such an edit is to try it here and see if it gains acceptance and if the reference is formatted properly. In the end this type of addition will "stick" better, as it is a consensus version and it will be protected by other editors. Please propose your edit here and let's try to make it work properly. We'll help you. -- Brangifer (talk) 00:51, 29 April 2009 (UTC)

__________________________

This edit does not require an external source. This is a simply and directly provable deduction as a 2 premise syllogism:

Premise: The law of non-contradiction (http://en.wikipedia.org/wiki/Law_of_noncontradiction) is false. Premise: The law of non-contradiction states that what is false (non-true) is not true (and vice versa). Conclusion: What is false is non-false (i.e. true).

Thus is it demonstrated that any falsification of the law of non-contradiction is self-contradictory; therefore the law of non-contradiction is non-falsifiable. The other two axioms (law of identity and of excluded middle) can be demonstrated by similar arguments.

My proposed change (which you undid) is as to replace the incorrect sentence with:

"However, the laws of logic themselves (the rules of inference and logical axioms) are not subject to falsifiability per se. That is, since truth values are defined in relation to the laws of logic any "falsification" of these laws would represent a self-contradictory situation though this conclusion has been argued against by philosophers such as W.V._Quine."

I will be the first to admit that the formatting needs some tweaking. I will give you a day or so to review this before correcting it again. Dialegomai (talk) 01:41, 29 April 2009 (UTC)

Logical Possibility

I've heard two different interpretations of the article's opening sentence:

Falsifiability (or refutability) is the logical possibility that an assertion can be shown false by an observation or a physical experiment.

1. Falsifiability is when there is nothing in the logical nature of an assertion that would preclude an observation or physical experiment from showing said assertion to be false.
2. Falsifiability is when there is nothing in the logical nature of an assertion that would preclude a currently practical observation or physical experiment from showing said assertion to be false.

The first says the only criteria is the assertion itself, that what is not falsifiable today will never be falsifiable. The second adds consideration of our technological capabilities, that what is not falsifiable today may be tomorrow. I've always understood the first to be the correct interpretation, but I've heard it argued that Popper was referring only to things that are of current practical value to science. Even if he were not, they say, there is nothing useful in the first interpretation. I understood the issue to be about potential value, not actual value. Some ideas, by their very nature, are incapable of ever being meaningful. Other ideas may or may not be meaningful, but at least the possibility exists. The ability to distinguish between these two classes is what is useful.

I post this seeking the opinion of others. How is the aforementioned opening sentence to be interpreted? Siggimoo (talk) 23:16, 9 December 2009 (UTC)

"Theism" section is just asking for trouble.

The section basically says that the existence of God is non falsifiable but accounts of the activities of God are. I don't see how the latter is falsifiable either. By it's very nature, such an action on the part of God would be defined as a "miracle" and we can't evaluate any particular instance of miracle in the historical record until we have come to a philosophical conclusion on whether or not miracles are possible or not. (C. S. Lewis, "Miracles; A Preliminary Study") This to try to falsify a miracle on the grounds that such an event is contrary to the observed course of nature is circular reasoning.

I suggest we simply delete the section. I don't think the debate over the falsifiability of God is relevant to defining the process of falsification. --BenMcLean (talk) 18:35, 17 December 2009 (UTC)

It's source is self-published. Normally I'd say put in Lewis as a counter claim, but we don't have a third party claim to begin with. Ian.thomson (talk) 18:39, 17 December 2009 (UTC)

Too technical

I've read this page 3 times and I still can't fathom what falsifiability is... I think it needs a clear and simple definition at the beginning of the article. Kat, Queen of Typos 22:45, 9 November 2006 (UTC)

The introduction certainly needs a re-write. It was a gift from our good friend Jon Awbrey (talk · contribs),[5] who never used a five letter word when a half-dozen polysyllabic pronouns could say the same thing. Banno 22:02, 10 November 2006 (UTC)

The best example that I can apply to my interest in scientific modeling is the Arkansas creationism case ruling, where the criteria is that if an idea starts with an intractable conclusion and not an adjustable hypothesis is not falsifiable. So, in the context of modeling, falsifiable means flexible. My criticism of scientific culture is that it is so different than normal society that it makes legal systems appear open-minded. There is something up with this that has to change, because there is nothing open-minded about justice--justice is usually blind.--John Bessa (talk) 14:12, 14 September 2010 (UTC)

Merge with Testability

PRO The testability article does not present sufficient information in order for it to stand on its own. As it is mentioned in the lede of this article as a synonym for falsifiability (or at least part of falsifiability), it should be merged into this article. Neelix (talk) 16:36, 6 April 2008 (UTC)

PRO I agree. It should be merged and a redirect added. lk (talk) 14:19, 10 May 2008 (UTC)

CON I'm not sure there is anything to merge into this article or that there is a clear concept on the Testability article. I'd just make it a redirect. O¹⁸ (talk) 01:07, 12 May 2008 (UTC)

CON I don't think Falsifiability should be merged with Testability. Falsifiability is an important strand of philosophical thought of one of the greatest thinkers of the twentieth century. Its application revolutionized science (e.g. the collapse of the Newtonian paradigm and its replacement with relativity) and is still the harsh "rock of reality" upon which all unsound scientific theories eventually founder. Testability is similar, however it is more of a grab-bag collection of practical problem-solving tools than it is a philosophical bedrock; it is the "applied mathematics" of testing as opposed to the "pure mathematics" of Falsifiability. For Wikipedia to have a page on Popper without a link to his central conceptual idea of Falsifiability would be like having a page on Henry VIII without directly mentioning his six wives, other than as a sub-section under mistresses, concubines, and other female companions. Jackmaturin (talk) 07:52, 12 May 2008 (UTC)

CON I don't think Falsifiability should be merged with Testability either. There is an important strand in the testability article which is missing - I may add it when I get time - which is that testability is also a term used to describe how testable a physical system is (not just a hypothesis). In engineering it is a term used to discuss built in test methodologies and other self check schemes. It may also be a measure of how easy a product is to test by external methods. —Preceding unsigned comment added by 213.120.209.210 (talk) 11:11, 22 May 2008 (UTC)

CON Added highlighting of the positions above. Ours is based on an anticipated build out of the Philosophy of Science here. Testability and Falsifiablity are distinct concepts, in Our English usage, with Testability the empirical projection of Falsifiability plus serendipitous discovery. A theoretical position could be falsified even though there was no known way to empirically test it. 74.78.162.229 (talk) 15:54, 22 July 2008 (UTC)

CON should be kept if only for the fact that it was Popper's technical term. In addition testablitity is only good for experimental science (chemistry & physics) and does nothing for observational sciences (geology, astronomy, and much of ecology/biology).--OMCV (talk) 19:33, 13 August 2008 (UTC)

I have removed the merge tag on the testability article as consensus here seems against the merge. Deamon138 (talk) 23:19, 13 August 2008 (UTC)

CON I know I am late, but let me throw my $0.02 in: Falsification is about whether an observation 'could' be made that would contradict a proposition. Testabiltiy is about collecting affirmative evidence for a proposition. The former is deductive while the latter is inductive. My thought is that these two things could easily be confused, but are quite different in nature (negative and deductive vs positive and inductive). 70.69.189.240 (talk) 04:20, 23 September 2010 (UTC)

Swans

Original caption on "black swan" image:

The classical view of the philosophy of science is that it is the goal of science to "prove" such hypotheses or induce them from observational data. This seems hardly possible, since it would require us to infer a general rule from a number of individual cases, which is logically inadmissible. However, if we find one single black swan, logic allows us to conclude that the statement that all swans are white is false. Falsificationism thus strives for questioning, for falsification, of hypotheses instead of proving them.

I deleted most of this, because it was too long for a caption. The existence of black swans disproves the statement, "All swans are white." I think we should leave it at that, and hope that the reader pursues the matter further by reading the intro.

And if the intro is sufficiently interesting, maybe the reader will go on to the rest of the article.

The key point of falsifibility (and this should be in the intro) is that every general rule should be compared with real-world observations, and that we need to make corrections in the rule if too many observations contradict it.

"Things that you drop will fall to the ground." This is an excellent general rule, which can be grasped by children as young as 3 or 4. But then we can show exceptions such party balloons, which can be held up by air currents. So we must then modify the rule to, "Things that feel heavy will fall to the ground." We can then formulate a second rule for light things, like balloons, paper, and dust. "Lights things fall to the ground, unless the wind keeps them aloft." --Uncle Ed (talk) 17:03, 22 February 2010 (UTC)

removing quote "No amount of experimentation can ever prove me right; a single experiment can prove me wrong"

I removed the above quote. Swedish philosopher Sven Ove Hansson has made an analysis regarding this quote, which concluded that this quote is misleading; see article [6]. Ulner (talk) 11:42, 8 August 2010 (UTC)

I have restored the quote. If/when you find a reliable source (preferably in English), you can add a footnote about this philosopher's view on the matter. Note that your source says: "Vetenskap och Folkbildning (VoF) is a Swedish non-profit organization set out to promote popular education...". I don't think that this qualifies as a WP:RS. DVdm (talk) 11:47, 8 August 2010 (UTC)

I removed the quote because it is not a quote by Einstein. As explained by Calprice herself, it is not "an exact quotation", and hence it is grossly misleading to use this quote her. We were quoting Calprice, not Einstein! Ulner (talk) 13:51, 8 August 2010 (UTC)

I have restored the (paraphrased) quotation with two more sources. I have also added the phrase "Albert Einstein is reported to have said". That is a fact, as one can verify in the sources. Note that the first footnote already has a proviso on this. DVdm (talk) 14:41, 8 August 2010 (UTC)

The article by Sven Ove Hansson, a philosopher and chair of the Department of Philosophy and History of Technology at the Royal Institute of Technology (KTH) in Stockholm, Sweden, is a reliable source. According to Hansson, Calprice statements are misleading; quoting from the article in Swedish:

"Calaprices påstående är alltså missvisande; citatet om ”ett enda experiment” kan inte rimligen sägas vara en parafras på det som Einstein skrev i tidningsartikeln om induktion och deduktion. Sammantaget framstår det som föga troligt att citatet har sitt ursprung i något som Einstein själv sagt eller skrivit. Dock är det inte möjligt att bevisa att citatet är falskt. Påstådda citat hör nämligen till det som kan verifieras men i regel inte falsifieras."

In English translation:

"Calprice statement is misleading, the quote about 'a single experiment' can not reasonable be a paraphase of what Einstein wrote in the newspaper article about Induction and Deduction. Altogether, it seems to be unlikely that the quote originates from something written or said by Einstein himself. However, it is not possible to prove that the quote is false. A quote claimed to be true belongs to what can be verified but in general not falsified."

Hansson has carefully read the original text by Einstein, and found that the paraphrase by Calprice is misleading or false. For this reason I think the quote should be removed; we should never repeat dubious quotes. If we want to keep the quote we should find an original source by Einstein which says that he actually used this quote (or something very similar). Ulner (talk) 21:04, 8 August 2010 (UTC)

I don't think this is a reason to remove the (clearly paraphrased) quote. Perhaps it could be a reason to add a note about Hansson's objection, provided we have a consensus that this objection is sufficiently wp:notable, whereas the fact that Einstein (paraphrasedly) said it, is extremely well sourced by wp:secondary sources. I personally don't think Hansson's objection is notable, but let's see what other contributors think about this? DVdm (talk) 21:17, 8 August 2010 (UTC)

Let's see what other contributors say. One problem is that quotes are copied from book to book, so the number of secondary sources can be high although this is a false or misleading quote. Ulner (talk) 21:23, 8 August 2010 (UTC)

Well, you seem to be a somewhat new editor, so perhaps you don't know, but as you can see in this policy, "the threshold for inclusion in Wikipedia is verifiability, not truth." The essence here is that we have an enormous number of secondary sources who say that Einstein said something, so we are entitled to write that Einstein said it, whether (1) what he said was true or false, and (2) whether all these sources are right or wrong about whether he said it or not. Tricky to explain :-) - But yes, let's see what others say. DVdm (talk) 21:43, 8 August 2010 (UTC)

I accept the verifiability criterion, but there may be exceptions in special cases. Paul Renno Heyl seems to have this quote in "The common sense of the theory of relativity" (1924). I concur that this quote seems to be widely used. Ulner (talk) 00:03, 9 August 2010 (UTC)

God's existence irrelevant to Falsification/Falsifiabiltiy

The article currently contains this sentence, "As a result, it would be logically invalid to observe an act directly falsifying the existence of a common ancestor, just as it would be impossible to falsify the existence of an invisible God."

This sentence contains an obvious comparison between the "common ancestor" of Darwinian evolution and the "invisible God" of religion. However religions do not claim to be scientific, and since falsifiability is a principle Popper applied to scientific method, the insertion of religion here is quite unwarranted. Talking about negative evidence may include discussions of religion, but my opinion is that the article is about falsification, which relates solely to scientific theories, of which evolution claims to be (thereby suiting the context), whereas religion makes no claim to a scientific nature and thus does not suit the context. In other words, the arbitrating factor for me is the article title. Religion simply does not fit in this article anywhere, whereas any scientific theory would fit fine.

Please note that this comment is not intended to be prejudiced for or against religion, and I decline to (irrelevantly) share my personal beliefs here. Please forgive the run-on sentence above.70.69.189.240 (talk) 01:05, 17 September 2010 (UTC)

I tried to do some rewriting to keep matters of evolution and theism seperate. If you would like, you might also check out the latest "Theism" section of this page to see if it could use any constructive criticism as well.-Tesseract2 (talk) 06:15, 17 September 2010 (UTC)

"the hypothesis that God created humankind specifically in their modern form, which was falsified by evidence that instead supports evolutionary origins from some common ancestor. Other beliefs have been falsified as scientific understanding has increased, or in some cases as science has gathered evidence of absence."

Well, from my position as a philosopher of science, all this is quite silly -- you see, falsification is relevant only in discussion of scientific theories, and not just for any belief. Additionally there is an epistemic problem when one says, "which was falsified by evidence that instead supports evolutionary origins," since the evidence is not a clear falsification. Falsification goes, "scientific theory/proposition is 'p', we [could] observe '~p', therefore p is falsified [falsifiable]." Obviously the important philosophic point (represented in [ ]) is that a theory or proposition is falsifiable. The foregoing statement was not a clear falsification since evidence that could be interpreted as supporting an evolutionary argument is not of the form '~p' when 'p' is 'the hypothesis that God created humankind specifically in their modern form'. A similar statement would be that one would first have to show an argument form of disjunctive syllogism between evolution and creation, or between whatever two items are at issue, which has clearly not been done.

Furthermore the statement, "Other beliefs have been falsified as scientific understanding. . ." is far too loose as well, since falsification only applies to scientific beliefs anyway. A proper rendition would go, "Scientific beliefs have been falsified as scientific understanding [progressed]. . ." but that would be a little redundant or silly itself, since obviously as scientific understanding progressed the propositions of science have been refined (for quantitative propositions) and corrected (for qualitative propositions). Since this whole page is about falsification, which is a specific principle to philosophy of science, I believe that we as a community would do well to refrain from slack philosophical ramblings... 70.69.189.240 (talk) 04:12, 23 September 2010 (UTC)

Check for copyright infringement

Some of the material appears to infringe copyright: i.e come from this source: http://books.google.ie/books?id=AiZe7fqM41AC&printsec=frontcover#v=onepage&q&f=false . I will remove the copied paragraphs, if someone wants to re-insert the ideas fire ahead but I don't think we can leave up as is. IRWolfie- (talk) 21:23, 3 January 2011 (UTC)

Null of Green Swan

The green swan becomes falsifiable by testing the null hypothesis "no green swan exists". This is falsifiable by finding a single green swan. Symbolically, -(-A)=A, as opposed to +(A)=A which reduces to A=A.108.65.0.169 (talk) 22:33, 13 January 2011 (UTC)

The point of the green swan is that we can't falsify the statement that "a green swan exists", there is no way we could show it does not exist. IRWolfie- (talk) 16:36, 16 April 2011 (UTC)

This is an interesting point raised above in that long discussion "All men are mortal". I will edit in relevant links to evidence of absence.-Tesseract²(talk) 03:12, 17 April 2011 (UTC)

Aren't all positive claims falsifiable?

Technically, I mean. If you say, "A green swan exists", that is falsifiable. Show one. Difficult, as nobody knows where the're supposed to live, but not impossible. If one says "Green swans do not exist", that cannot be proven. You would have to check every swan on the planet, which would only result in the person who made the claim say "Oh, I forgot to mention they live on the moon.". After searching the moon he would tell you to search Jupiter and the rest of the universe. You just can't. If you say "Russell's teapot exists", that is falsifiable. Well, technically. You would have to figure out a way to show it, but future technology may make that possible. However, if you say "Russell's teapot does not exist", Russell makes it clear how that claim is unfalsifiable.

But since the article on Russell's teapot makes it clear his claim is unfalsifiable, it seems to be the case, that falsifiability depends on available resources. Which would mean that saying "Ionizing radiation is lethal" was a non-falsifiable claim 500 years ago? Is this the case? W3ird N3rd (talk) 00:43, 1 June 2011 (UTC)

All men are mortal

At first glance "all men are mortal" may seem non-falsifiable, but if we look and think more closely we find something unexpected: the statement is completely falsifiable. Falsification of the statement "all men are mortal" is impractical, but is nevertheless possible. You simply need to wipe out every other human being in existence till you are the last one standing. To finally drive home the argument, you, as the last human on earth would have to commit suicide - and the human race would end with you. QED.

That is the most flawed argument I have ever heard from someone trying to be intelligent.Kmarinas86 ^(6sin8karma) 18:11, 19 June 2009 (UTC)

That is just useless personal attack. Comment is completely valid. "The set is all men, not all living men."

Article should be fixed.88.114.55.240 (talk) 21:37, 17 September 2011 (UTC)

My bad... mixing up verifiable and falsifiable. Maybe there should be something to clear this up better?

Something like 'Therefore, if "all men are mortal" is true it is verifiable, and if "all men are mortal" is false it is unfalsifiable.' ?

88.114.55.240 (talk) 00:09, 18 September 2011 (UTC)

Again, I agree that the above falsification of the statement that "all men are mortal" is impractical (and a tad destitute of compassion for misery & suffering), but the simple fact is that the statement can hypothetically be falsified. Better and briefer (but more controversial) statements could be:

1 - "God exists" - Non-falsifiable

2 - "God does not exist" - Falsifiable

I hope this helps. Otherwise, the article is almost perfect. TranscendTranslation 01:56, 16 November 2007 (UTC)

The comment I have here is that observation that NOT "all men are moral" by seeing one man die leads to the obvious conclusion that SOME men are mortal, not that ALL men are mortal. this is an important point. §74.13.89.115 (talk) 06:50, 21 November 2007 (UTC)

This comment outclasses the above, despite its tangent on verficationism.Kmarinas86 ^(6sin8karma) 18:11, 19 June 2009 (UTC)

Your comment above isn't perfectly clear to me. Nevertheless, my point is that the statement "all men are mortal" can be falsified by experimentation. Quite obviously, it is NOT falsified just by considering a single dead man, no-one is saying that. However, it IS falsified by killing all men (a theoretically possible but clearly callous/impractical task). Once all men have died, the human race ends with you, so you can be sure there will be new men to complicate the issue. Finally, this whole consideration is semantic, the overall point is that the 'God' examples are certainly true, whereas the mortal/immortal example are not.TranscendTranslation (talk) 18:26, 24 November 2007 (UTC)

Wrong. That is absolutely illogical and incorrect.Kmarinas86 ^(6sin8karma) 18:11, 19 June 2009 (UTC)

When you've reduced the set of "all men" to the empty set, assertions about its members are meaningless... and in fact, not falsifiable! 67.98.226.14 (talk) 00:55, 5 December 2007 (UTC)

Another good point by an anonymous user!Kmarinas86 ^(6sin8karma) 18:11, 19 June 2009 (UTC)

You're example "God does not exist" is also non-falsifiable because of the very nature of what people call god. God is so that "I don't understand" becomes "I understand", for that reason the proof which generates the faulsensee of the statment also generates a "we don't understand yet" response, and the choice taken is irrelevant because the 2 competing ideas "god"(proves the statement false) or "a lack of understanding" (doesn't prove it false) cannot be compared for the choice is based on personal beliefs and facts. For example any observation made to prove it false, "oh look, frogs from the sky", can be explained through a scientific method because the very existance of miracles are explained through probabilities and if one isn't happy with the probability explanation, argument of manipulation with human (or even non-human) technology can be use. If you witness god it could be drugs, if you and your friend witness god how can you argue that the 25foot tall guy in robes wasnt actually a giant robot. —Preceding unsigned comment added by 99.226.149.107 (talk) 22:11, 22 April 2008 (UTC)

Win!Kmarinas86 ^(6sin8karma) 18:11, 19 June 2009 (UTC)

There is a lot of confusion here between logic and falsifiablelity. The statement “All people are mortal” is falsifiable. I need only find one person who is not mortal to show that the statement is false. The fact that there has never been an observation of a non-mortal person just goes to supporting the statement. Killing every one on the planet would not show the statement is falsifiable; it would just be another piece of evidence supporting the argument.206.195.19.43 (talk) 10:24, 24 April 2008 (UTC)

Observing an alleged immortal person for any finite duration of time is not enough to verify if that person is indeed immortal. The statement "all men are mortal" is unfalsifiable because the claim that "there is a man that is immortal" is unverifiable, given the absence of observations of unlimited duration in time. Applying the same principle of falsifiability to the claim "all observations are of finite duration in time" leads to the same problem. Valid falsification of the claim "all men are mortal" is not knowledge accessible to mortals. But for immortals, even if they believe (or know) they are immortal, there is never any present act of falsification, for the act of falsification is delayed to an ever distant future.Kmarinas86 ^(6sin8karma) 18:11, 19 June 2009 (UTC)

The concept of an immortal man, required to falsify the proposition 'all men are mortal', is entirely non-cognitive by virtue of immortality having no ostensive examples in reality; we can't after all point out an immortal man for the reason explained above, an observation made over a finite period is not logically coherent to claiming immortality. Thus an immortal man cannot be observed, and hence cannot provide the basis for an empirical falsification method. So in this respect at least the original statement can't be falsified. Apathy92 (talk) 09:54, 2 December 2009 (UTC)

"The concept of an immortal man...can't be falsified." = TruthKmarinas86 ^(6sin8karma) 09:00, 3 December 2009 (UTC)

Quote: "an immortal man cannot be observed". Response: It is more accurate to say, "a man cannot be observed to be immortal". Just observing someone does not prove they are mortal, although it is very, very, very, very, likely that they are mortal anyway.Kmarinas86 ^(6sin8karma) 09:03, 3 December 2009 (UTC)

Thanks for the pointer :) Apathy92 (talk) 11:28, 3 December 2009 (UTC)

Also, I'm grappling a little with this- Quote: "You simply need to wipe out every other human being in existence till you are the last one standing"- it occurs to me that if you reduce the human race to being non-existent any assertions about the set become meaningless and unfalsifiable as a result. Am I along the right lines here? Apathy92 (talk) 11:56, 3 December 2009 (UTC)

Why not think on terms of "All men die before the age of 200"? —Preceding unsigned comment added by 200.27.146.116 (talk) 01:09, 30 January 2010 (UTC)

'When you've reduced the set of "all men" to the empty set...' This logic is even more flawed. The set is all men, not all living men. Do men cease to exist when they die? Nope, they are just dead men. So the original commenter is absolutely correct: if all men died, that would be absolute proof and the statement would be falsified. 66.54.122.127 (talk) 16:43, 25 March 2010 (UTC)

Yes, but eventually the men will break down somehow; meteors, decomposers, e.t.c. - and it is unfalsifiable because you cannot prove that that will not eventually happen, because it takes infinite time to prove. And if a dead man is destroyed he is not a man anymore, no? 96.232.25.169 (talk) 01:07, 26 March 2010 (UTC)

Men breaking down has nothing to do with mortality. When a man dies it proves he is mortal, therefore if all men died it would be proof that all men are mortal. I don't understand why we have to make this so complicated... 66.54.122.127 (talk) 05:44, 27 March 2010 (UTC)

TranscendTranslation, you've just reversed those statements. "God exists" is a perfectly falsifiable claim. All you have to do is show him, and there's your proof. "God does not exist" however, is non-falsifiable. You can say he's not here. So somebody will suggest God is on the moon. You search the moon: no God. Somebody suggests he may be on Mars. You search Mars. Nothing. Somebody suggests another place. You search the entire universe (impossible, but let's go for it): nothing. Somebody will suggest he's in another dimension that we cannot see in any way. Not falsifiable. As for "all men are mortal", that is also a reasonably falsifiable claim. All you have to do is find somebody who is immortal, shoot him/her, throw this person off buildings, use poison, hanging, give him/her a lethal injection, etc. All this would be accepted as reasonable proof this person is immortal. Think of Jack Harkness from Torchwood. As always in science, if at some point something is proven wrong (Jack Harkness dies), that's ok. You've made a mistake. Similarly we could at some point figure out all the black swans were actually white swans painted black as a prank by the Australians. If that would happen you would also have to find another swan somewhere that's not white. Similarly you would have to find another immortal human if the one you thought to be immortal dies. On another note, killing all humans doesn't make the claim falsifiable. There may still be some humans on another planet (transported by spaceships a long time ago), underground, in the jungle etc. You could never verify everyone is dead. We don't have to watch somebody immortal for an infinite amount of time, if we can reasonably argue he is (shot, crushed, 500 years old etc and still standing, unaged) we can consider that proof. The same way we say comic books and atoms exist, even though we could still be proven wrong about that and figure out we're all living in The Matrix. We talk about history and consider it truth, despite having no absolute proof for everything we know (some bones or finding rusty tools in the ground is no absolute proof, it can be proven wrong). Good proof doesn't have to be impossible to be proven incorrect, in fact, virtually every proof of anything can be proven incorrect by the "Matrix" argument.W3ird N3rd (talk) 00:18, 1 June 2011 (UTC)

Falsifiable-statements and Godel-statements

This does not appear to drectly concern the wikipedia article- collapsed per WP:NOTFORUM

Treating pure-mathematics as purely formal hypothetical knowledge; it is hard to see how 'falsifiable-statement' in a strict scientific sense is applicable. Never-the-less falsifiability in a mathematical as opposed to ' objectively observable' sense; does seem applicable.eg. Pythagoras's Theorem establishes that: for 'right angled triangles' in Euclidean space; H^2 = A^2 + B^2, for triangles with A and B in any ratio and hypotenuse H. The set of such triangles is necessarily infinite, but the set which will ever be calculated is necessarily finite.The theorem (a far from obvious tautology) establishes that the latter will always be a subset of the former. But surely this in no way contradicts the statement that "Pythagoras's Theorem is 'mathematically' falsifiable". If by mathematically falsifiable statement; we intend a statement which asserts explicitly or implicitly one or more mathematical 'facts' in some axiom system.Then we can catagorise such statements as: A/ Tautological falsifiable-statements (theorems); the asserted 'mathematical facts' of which; we are confident will always be a super-set of the set of calculated 'mathematical facts' and B/ Contingent falsifiable-statements the asserted 'mathematical facts' of which ;have thus far always been a super-set of the set of calculated 'mathematical facts'; but may not remain so. Now consider 'Godel-statements'; by which is meant the unprovable theorems of Kurt Godel’s famous Incompleteness Theorem, interpreted as:“ Every axiomatic system is either complete or consistent, but not both”.Which implies that every consistent axiomatic system is incomplete; which implies that Godel-statements must exist in every consistent axiomatic system.Where by 'complete axiomatic system' is intended ,one in which: 'all statements constructable within such a system, are provably true or false'; and whereby 'consistent axiomatic system ' is intended one, in which : 'no two provable statements, can be found to conflict' Now another theorem of Godel, his Consistency Theorem established that consistency of an axiomatic system is itself not provable; which implies the statements:“This axiom set is consistent” along with it's corollary ,“This axiom set is incomplete “ ; are necessarily Godel-statements in any axiom system, which is not formally inconsistent; ie. containing conflicting axioms. Now a difficulty arises when trying to decide what exactly a Godel-statement, 'IS'..! Statements which seem true, but can't be proved , superficially resemble the axioms themselves. In that axioms are statements, which are necessarily true, since they define the particular axiom system, but necessarily can't be proved using the other axioms; as they would then be unnecessary ; ie. they would be theorems. But if Godel-statements are considered to be additional axioms, and formally added to the set, one starts an escalating vicious cycle; because this new enhanced axiom set, will have it's own additional Godel-statements. etc. Returning to Godel's Incompleteness Theorem; we have in any axiomatic system: 1/ 'theorems' which are true and proved and 2/'theorems' which are true, but unprovable It occurred to me that: if 1/ can be identified with A/ above then perhaps 2/ can be identified with B/ above. The contention is that Godel-statements be identified as non-tautological 'mathematically' falsifiable-statements, Then we have a place for Godel-statements, without having to add them as axioms, thus avoiding the above vicious cycle. Queries (1) If Godel-statements can be considered to be 'mathematical falsifiable-statements' in this extended sense,of 'non-tautological statements, thus far found consistent with the facts of the axiom system, but possibly false by comparison with facts yet to be considered.'(1)Are Godel-statements , which are consistent with only a sub-set of the facts of the axiom system (a) possible ?; (b) necessary ? (2)Given the very general interpretations of Godel's Incompleteness Theorem which now exist, outside number theory; Can we conclude that in any consistent axiomatic system, non-tautological 'falsifiable-statements'; must exist ? Rhnmcl (talk) 08:29, 8 October 2011 (UTC)

Confusion between logical and belief falsification

Popper's argument was that theories are falsified/falsifiable in logic, not belief. For example, while Galileo's observations logically instantly falsified the Ptolomeic theory, the beliefs of other scientists did not change instantly, nor did such beliefs have to change at all in order for the logical falsification to exist. 108.65.0.169 (talk) 02:24, 15 July 2011 (UTC)

The trouble is that these other scientists may have disagreed that Galileo had done his measurements properly - or that the logic of saying "The observation of the phases of Venus proves that Ptolomy was wrong" is an incorrect statement. Clearly, if Ptolomy was right, aliens living on Venus could have painted the entire planet black and covered it with tiny lights that they could turn on and off at will in order to fake the phases. This would have fooled Galileo. One can only ever assign a probability to any experimental 'fact' that we know - almost any result could turn out to be incorrect at some time in the future. So the other scientists didn't have to believe what Galileo said. However, that doesn't change the falsifiability of Ptolomeic theory - only whether the actual falsification was widely accepted or not. SteveBaker (talk) 15:07, 21 November 2011 (UTC)

Why the "atoms do exist" example?

The example used in the first paragraph in the abstract, "atoms do exist" seems confusing. The word atom is used in the metaphysical meaning, something indivisible, but I think that a lot of lay people might think it is referring to the scientific notion of the atom, meaning electons orbiting a nucleus. The former is not falsifiable, but the latter is. How hard would it be to use a different example here? Oktal (talk) 13:17, 4 September 2011 (UTC)

I agree, it's not a clear example and appears to be original research. --George100 (talk) 23:03, 5 October 2011 (UTC)

It's been over a month since this was pointed out and it's obvious to anyone with even a lay person's understanding of the scientific concept of an atom that this is absurd. Deleting the entry. --Bart simpson rules (talk) 20:22, 8 October 2011 (UTC)

Also if anyone is thinking about restoring it, examples do not belong in the summary. There is already an examples section and it should be placed there if not removed entirely. --Bart simpson rules (talk) 20:34, 8 October 2011 (UTC)

I have restored the intro. Examples are important for understanding. And the example of atoms is the standard example, and one of the examples used by Popper himself. The example is not absurd. Oktal says "The word atom is used in the metaphysical meaning" -- This gets the point completely wrong. Words do not have "metaphysical meaning"; words can merely be used in the context of a statement, and it is the statement as such that may be metaphysical or falsifiable. It is completely irrelevant whether the statement "atoms do exist" refers to atoms in the old greek sense or in the modern scientific sense. The statement remains metaphysical in both cases. It is very easy: The "scientific [theory] of the atom, meaning electons orbiting a nucleus" says that atoms have certain structural properties and behave in certain ways. But If we have found an atom that does not have these structural properties predicted by modern theory, it does not contradict the claim that atoms (in this modern sense) exist. For there may always be a different atom that has these properties and that we have not looked at yet. What is falsifiable is the theory that *all* atoms obey the laws of the modern theory. But I have changed the statement slightly anyway to help understanding. --rtc (talk) 14:58, 6 November 2011 (UTC)

You have no idea what you're talking about. Furthermore, no citations exist for any of the claims in the summary. If you restore it again you will be reported for vandalism. Bart simpson rules (talk) 22:05, 18 November 2011 (UTC)

Please be aware of WP:NPA. IRWolfie- (talk) 01:09, 20 November 2011 (UTC)

I have restored the intro once more and I am considering your reverting vandalism. If you have any objections more specific than "You have no idea what you're talking about", please raise them here. --rtc (talk) 19:33, 19 November 2011 (UTC)

While your edits where not vandalism and were good faith edits they should still be discussed here. IRWolfie- (talk) 00:59, 20 November 2011 (UTC)

The point Oktal appears to be making is that the use of the word atoms in the metaphysical sense is confusing in the lede. The swan example seems to be much more clear and less terse. IRWolfie- (talk) 01:09, 20 November 2011 (UTC)

As I already said, the atom example is one of the key examples used by Popper himself. Because of its special logical form making it metaphysical even as a "for-all-statement", it goes beyond the swan example. You can read Popper's discussion at here --rtc (talk) 01:18, 20 November 2011 (UTC)

The purpose of the lede is to summarise the article. Add it to a relevant or new section in the article with citations etc with it but do not add it to the lede. IRWolfie- (talk) 01:21, 20 November 2011 (UTC)

There is no urgent reason to remove this stuff from the intro. Oktal's argument that the example is confusing was correct, but I fixed that already before User:Bart simpson rules reverted again, accusing me of vandalism. It may be true that this content can be moved to a section one day, but there is nothing that makes this so important that it would be necessary to keep it out of the intro at all costs until then. --rtc (talk) 01:31, 20 November 2011 (UTC)

The lede should summarise the article. Your additions do not refer to the article. They are not sourced. IRWolfie- (talk) 14:42, 20 November 2011 (UTC)

You are repeating yourself. Please undo your revert. --rtc (talk) 00:36, 21 November 2011 (UTC)

Sorry, no. IRWolfie is correct. This doesn't belong in the lede. According to WP:LEDE the lede should be a summary and not introduce any information not contained in the main article...and without a reliable source (per WP:RS), they shouldn't be in the body of the article either. So this information should not be here because it's not in the main article and it shouldn't be in the main article. QED. SteveBaker (talk) 14:08, 21 November 2011 (UTC)

Sorry, but you are repeating IRWolfie-'s arguments that I have clearly refuted above. The atom example is one of the key examples used by Popper himself. The lede can and should contain this information until a better place is found. There is no urgent reason to delete it. I will restore it myself now. Please try to be more constructive. --rtc (talk) 17:12, 21 November 2011 (UTC)

While your intentions are well, please don't further decrease the quality of the aritcle by adding text without references which support your statement. In particular, I looked up Poller's quote you cited, and it says not what you wrote here. Loew Galitz (talk) 22:43, 21 November 2011 (UTC)

Of course it says that. Maybe not in the exact same wording, but the meaning is the clearly the same. Please let's not be literalists. If you still think the reference says something different than what I use it for, please argue here in what specific way you see a conflict. I cannot do much with a statement like "it says not what you wrote here". And not every example is original research if it has not been published in the exact same wording before. Else we could delete most examples from math articles. Falsifiability is a concept as unambiguous as most mathematical concepts, at least for those who have read the respective books. And BTW, it's Popper, not Poller. --rtc (talk) 23:50, 21 November 2011 (UTC)

A theory not scientific until we know where it fails?

Does falsifiability mean a theory is not accepted as scientific---no matter how many experiments/observations support it---until we find out where it fails?

Was Newton's gravity falsifiable before physics at subatomic levels appeared?

--Roland 20:18, 14 September 2011 (UTC)

You are confusing the term "falsifiable" with "falsified". If we found an example where gravity failed to operate according to that law - then the law would be falsified (proven to be false) - and everyone would agree that it's untrue. A hypothesis is falsifiable if it could in principle be found to be false. Gravity could be found to be false if (for example) we could find an object of sizeable mass with no gravitational field...or a gravitational field ten times too big. The laws of gravitation are certainly falsifiable.

On the other hand, if you take something like "An omnipotent being called 'God' exists", then we can't (even in principle) falsify that claim. That is because this god is claimed to be omnipotent, so he can easily 'rig' any experiment we try to do to prove his existence. So any possible experiment that we could do that would prove conslusively that god doesn't exist may be dismissed by his followers as invalid because God influenced the results to test the faith of his followers (or some such thing). That means that the existence of God is an "unfalsifiable" hypothesis. Quite unlike the laws of gravitation - which, although they have not yet been falsified, could quite easily be falsified in the future (for example, when we figure out what all of this dark matter/dark energy stuff is all about).

Science is generally happy to work with a hypothesis that is falsifiable because there is reasonable hope that if it actually is false, we'll one day know that - and every day that we don't find it to be false strengthens our belief in it. But things that are unfalsifiable have to be taken on faith - and history shows us that this never works out too well. SteveBaker (talk) 18:20, 20 November 2011 (UTC)

All swans are white is _not_ falsifiable

Assume that swans wouldn’t actually exist (for instance, because we replace swan by dragon)? We see that the sentence is actually not falsifiable, because if it is impossible to find a single swan, we sure cannot find a swan that is non white. So why do you claim, that "all swans are white" is falsifiable?

"All swans are white" is equivalent to "there are no swans which are not white". To falsify it, you need only find one swan which is not white, say, a black swan.

Likewise "all dragons are red" is actually true if there really are no dragons, because it means the same thing as "there are no dragons which are not red" -- which is, of course, true, if there are no dragons at all. It is still falsifiable because we could prove it false by finding, say, a green dragon -- which would show that there are some dragons (at least one) and that some of them (at least one) are not red. If it will not, in fact, be falsified, because we will never find a non-red dragon, then it is true: either because there are no dragons at all, or because any dragons there might be (which we haven't found yet) are red.

Positive existential claims ("some are...") are not falsifiable, but negative existential claims ("none are...") are; negative universal claims ("not all are...") are equivalent to positive existential claims ("not all are..." something if and only if "some are..." not that thing), and so are not falsifiable, but positive universal claims ("all are...") are equivalent to negative existential claims ("all are..." something if and only if "none are..." not that thing) and thus are falsifiable.

I once knew a virgin woman all of whose children could walk on water. That is to say, she didn't have any children which couldn't walk on water... didn't have any which could either, but hey. --Pfhorrest (talk) 08:27, 17 November 2011 (UTC)

Forgive me, I am not sure I agree that "all dragons are red" is "true" if there are no dragons. That would mean, if there are no dragons, it is true that "All dragons are completely red" but also that "All dragons are completely blue". That is incoherent, no?

It seems to me that "All dragons are red" translates to "Dragons exist. All those dragons are red. Or in other words, none of those dragons are not red". There is a tacit existence claim, to my eyes. Is it really not contentious to take a sentence like "All dragons that exist are entirely red, and also entirely blue" and call that true? I am tempted to say it is neither true nor false, because no dragons exist..?

Just wondering

-Tesseract²(talk) 17:33, 19 November 2011 (UTC)

I don't know that I would go so far as to say that it's not contentious at all, because the colloquial understanding of universally quantified ("for all...") statements does have existential import, and Aristotelian logic formally included that feature as well, and for precisely that reason there are still a few supporters of Aristotelian logic today who boycott modern predicate calculus because it has this "absurd" result. However virtually all mathematicians (and scientists using modern math) and the vast majority of analytic philosophers use modern predicate calculus, not Aristotelian logic, and in predicate calculus universal claims unambiguously have no existential import.

It's not really so absurd when you think about it though, it's just a surprising ambiguity of natural language which becomes clear when you formalize it. If you'll concede that "all dragons are red" means "every member of the set of dragons has the property of being red", then when the set of dragons has zero members, that obviously has the same truth value as the statement "no members of the set of dragons has the property of being red". Every single one of the zero dragons which exist is red, so all zero dragons are red, and there are zero red dragons; so there are no red dragons, even though all dragons are red. The surprise arises when we find ourselves talking about the (zero) members of an empty set, which we are not usually prone to do, so it surprises us to find that the predication of both some property and its negation to every member of that set can both be true, as we don't usually aim to predicate anything of the members of an empty set, so our intuitions about that are bad.

For any nonzero cardinality of a set of things, to say that all of those things are completely red and all of them are completely blue is incoherent, yes. And we are almost always meaning, in natural language, to talk about members of sets with nonzero cardinality -- sets that have some members. So it is very surprising, but technically correct, when we find that when talking about a set with zero members, it is not incoherent to say both that all of those members are completely red and all of them are completely blue. --Pfhorrest (talk) 06:09, 20 November 2011 (UTC)

"All dragons are red" is certainly a true statement if there are no dragons in the universe - and that's more obvious if you write it out in formal logic because it translates to "For each dragon in the universe, this statement asserts that this dragon is red" - and if there are no dragons then you have failed to find a non-red dragon to falsify that statement - so (weirdly) it's true. It's definitely falsifiable: if you ever find a green dragon - and the statement is clearly false. It's difficult to "truthify" though - to conclusively prove that it's true you'd have to search the entire universe and show conclusively that there are no no-red dragons hiding anywhere within it! "All dragons are not red" is also true. So is "All dragons are yellow with purple polkadots".

Contrast that to "God exists" (a classic unfalsifiable statement) - you could possibly prove it's true (eg by finding an actual God - which if it is true, might not be that tough) - but you can't prove it's false because god is claimed to be omnipotent. If I were to somehow come up with a cunning way to design a fully scientific, 100% reliable, 100% accurate "godometer" (with a scale that shows how many gods there are in the universe) - and we plugged it in and it read "zero" then it can simply be claimed that God chose to hide his existence (eg in order to test the faith of his followers) by creating a miracle that causes the instrument to mis-read so it shows a 'zero' when it should actually read a solid 'one'. So if God is claimed to be omnipotent (and can therefore change the readings on godometers), then we can't conclusively prove that he doesn't exist. Hence this is an unfalsifiable statement.

You could make "God exists" falsifiable if you placed some limits on his powers - for example, if you were to say "God exists, but can't cause logical contradictions to appear in the universe" - then we could conclusively disprove his existence by merely asking if he can create a rock that is too large for him to lift. That is a logical contradiction that should disprove the existence of omnipotency of all kinds - but religious people simply claim that their god can produce logical inconsistency - which means he's still unfalsifiable.

"All swans are white" is undoubtedly falsifiable - because we could find a yellow-polkadot swan someday. In this case, we also know that it's falsifiable because it is actually 'false' - we've found actual, living, black swans. But it was falsifiable before black swans were discovered...and before swans were discovered too. SteveBaker (talk) 14:53, 21 November 2011 (UTC)

SteveBaker (talk) 14:53, 21 November 2011 (UTC)

Deletions

I am not sure all of Loew Galitz recent deletions are necessary. I have two worries. The first worry is that a quick glance at the deleted content does seem to contain some references. It was a quick glance though.

My second worry is a smaller comment. More of a hope. That is, I of course believe that Wikipedia should have sources for all of its information. I just hope that editors attempt literature searches to add sources, especially when this could be done with relative ease. Wikipedia has collected information that its public cares about, and I am not entirely sure that the easier route - of deleting entire passages - is always better than leaving it alone. I am much more convinced that it would be ideal to perform a more challenging "fix" like finding a reputable source (unless the information is altogether incorrect).-Tesseract²(talk) 23:30, 21 November 2011 (UTC)

Global warming

Are there any aspects of the anthropogenic global warming theory (AGW) which proponents and opponents both agree are falsifiable? That is, do scientists in general agree that AGW is a falsifiable theory?

• If so, what sort of observed phenomena could possible disprove AGW? --Uncle Ed (talk) 02:44, 22 November 2011 (UTC)

Certainly it'a falsifiable - we could conceivably find an unforseen natural source of CO2 that mimicked the human input so closely as to fool us. But falsifiability is a matter for judging hypotheses. Once something like AGW has advanced to the theory/law stage, it's already been proven true. It doesn't really need to be falsifiable. A better thing to consider is to think back before we knew that AGW was true...when it was just an unproven hypothesis. At that point, we hadn't measured all of the CO2 sources and sinks accurately - and it was very possible for the hypothesis to be shown to be false. SteveBaker (talk) 04:18, 22 November 2011 (UTC)

Careful there - the distinction between a hypothesis and a theory is one of degree, not of kind. There's a reason only historical theories get called "laws" anymore: we understand now that we are never completely certain about anything, only sufficiently confident for the time being. Even the best-supported theory can still be falsified by new data; just look at what happened to Newton's theory of gravity. So falsifiability continues to be important to theories no matter how well-established they are, and no theory is ever "proven true". --Pfhorrest (talk) 06:28, 22 November 2011 (UTC)

SteveBaker, the term "falsifiable" is formally defined in terms of deductive logic and according to this definition, theories are falsifiable in the same way as hypotheses. In fact, Popper holds in his Logic of Scientific Discovery that from an epistemological point of view, theories and hypotheses are exactly the same, and can at most be distinguished by the fact that hypotheses are usually less general than theories. Also note that Popper disagrees with the whole philosophical approach you and Pfhorrest are implying in what you say: Popper thinks that no theory can ever be proven or even suppported, made probable, given a reason for etc. ("justificationism"). Falsifiability may usually be held together with exactly the popular philosophy of science that supposes the opposite to be true (and we see such use of falsifiability as a criterion for a theory that can achieve genuine certainty or probability-beyond-doubt now a lot in political controversies like creation vs evolution, global warming etc.), but the article should not mix these popular views and Popper's original view. It should make a clear difference between them. About the original question: No, global warming theory is not falsifiable, because it is a probabilistic theory and such theories are never falsifiable. In fact, that was the main point of Logic of Scientific Discovery and the topic on which the book spends most of its pages. However, the book comes to the conclusion that the lack of falsifiability of probabilistic theories is not a problem in practice. If we would keep living the way we do now and would still observe the climate to stay essentially constant for the next 100 years, we could consider global warming as falsified for all practical purposes. Of course that would be a risky experiment. However, be careful. Note that Popper stresses again and again that political decisions (ie., what we should do) cannot be reduced to theories (ie., what is and what will be the case) (and not the question of their falsifiability either, I may add). But I think such discussion does not belong here, but to the articles on global warming. --rtc (talk) 10:45, 22 November 2011 (UTC)

I agree with much of what you say (although I disagree that Popper is still the ruling authority on such matters - so I don't really care what he thinks about the nature of scientific theory/law). But the important point here is that if (hypothetically) something were ever proved conclusively to be true, then there wouldn't (by definition) be any possible way to show that it's false. Any avenue of potential falsifiability would make that proof incomplete.

Consider the statement "Not all swans are white". This is proven beyond all doubt by the finding of a black swan. Before you found the black swan, this statement was only a hypothesis - and it was falsifiable by doing a survey of all swans and discovering that they are indeed all white. But given solid evidence of black swans running around out there in the wild, we have moved from a hypothesis about non-white swans to a proper scientific theory. How can "Not all swans are white" now be falsified?

Admittedly, out here in the real world, there is a possibility of experimental error - perhaps we were mistaken and genetic testing shows that the "black swan" that we found is really a kind of goose or something. Also, maybe next week, all of the black swans die and now the hypothesis is no longer true. No decision as to truth or falsehood is immune to that kind of re-examination (Newtons laws of motion are a good example of that) - and no claim of falsifiability or non-falsifiability is immune to that either. But to the same degree of confidence that we can say "X is true", we can also say "X is not falsifiable" - and "We don't care that X isn't falsifiable, because it's true". If we subsequently discover that X is no longer known to be true - then we also have to decide whether it's now become falsifiable.

This all comes down to when (or indeed, whether) you consider something to be "true". Saying "Nothing is ever true" might be a valid philosophical point of view - but it's not useful. You can't make useful decisions based on science if you continually doubt well-established theory. The whole point of saying "This hypothesis is now a theory" is to say "To the best of our knowledge we can rely on this being true". SteveBaker (talk) 16:17, 22 November 2011 (UTC)

SteveBaker, I did not say that Popper is a "ruling authority" on such matters, and Popper would certainly disagree with such a statement himself. You say the statement "Not all swans are white" is "proven beyond all doubt by the finding of a black swan". This completely disregards Popper's whole investigation of these matters. Perhaps because you "don't really care what he thinks about the nature of scientific theory/law". But if you hold such views, perhaps you should care. I won't repeat Popper's discussion here. But in fact, the finding of a black swan proves nothing. Even the simplest statements about observation may be wrong. Maybe you are confusing things here. It is a logical consequence of the (highly falsifiable) observation "a black swan lives in Australia at this or that place" that "Not all swans are white". The statement is true, after all. And this is what might trick you to believe that it is "proven beyond doubt." It is easy for true (or highly truthlike) statements to look "proven beyond doubt". But they are not proven. They are merely true. It is too easy to interpret observations that are corresponding to what such statements say as a proof of these statements. But logic tells us they cannot be a proof. "Also, maybe next week, all of the black swans die and now the hypothesis is no longer true" -- see, you say it by yourself: true. Not proven. However, the "next week example" has nothing to do with falsifiability. It has to do with context dependence of the statement. The statement refers to the context (the current time) and so it is acctually changing if the context (the time) changes. If we remove this context dependence, saying for example "Not all swans are white in November 2011" the problem completely disappears. "no claim of falsifiability or non-falsifiability is immune to that either" Let me repeat: Falsifiability, at least in the Popperian context, is defined in terms of deductive logic and there cannot be a question whether something is falsifiable or not. Obviously you are confusing falsifiablity and falsity here. A true statement may be "potentially" falsifiable, but still no correct experiment will produce a result that falsifies it. Ie., it is not "actually falsifiable", so to say; it is not false. Please make a difference between something being false and something being falsifiable. "But given solid evidence of black swans running around out there in the wild, we have moved from a hypothesis about non-white swans to a proper scientific theory." That is the naive philosophy of science we commonly see in the wild. It is positivism -- which Popper criticized -- but reformulated using Popper's terminology (espeically the notion of falsifiablity). According to Popper's actual view, however, there is no such thing as "solid evidence". As you acknowledge, "there is a possibility of experimental error". You do not take this fact serious enough. You use the notion of "degree of confidence". This is just the kind of justificationist philosophy I told you above. Popper criticized and rejected this whole framework of thinking.

"This all comes down to when (or indeed, whether) you consider something to be 'true'." Indeed, and this is a question of the theory of rationality (not falsifiability). Popper says: Everything can be considered true, but we need to question it and try to get rid of it in case we turn out to be in error about our assumption. This is the opposite position of justificationism, which says that something must first be justified and only once this has been done (for example by acknowledging that it is falsifiable and has passed some tests) can then be accepted as true. "Saying 'Nothing is ever true' might be a valid philosophical point of view - but it's not useful." Again, you don't completely get the full consequences if the criticism of justificationism. Only assuming justificationism, you could draw the conclusion "nothing is ever true" from what I say. But I merely say "nothing is ever justified". It's definitely not the same as "nothing is ever true." You claim "You can't make useful decisions based on science if you continually doubt well-established theory." is another popular topic and has been discussed a lot in the literature. Nobody says you should doubt theory (whether well-established or not). Again, only from a justificationist view one could assume that doubting the justification of a theory means doubting the theory itself. But one should question theory. For if one does not, one might overlook errors, and I think we agree that this is a problem for practical purposes. You say: "The whole point of saying 'This hypothesis is now a theory' is to say 'To the best of our knowledge we can rely on this being true'". But there is another problem here. We cannot rely on scientific knowledge for action. Science does not tell us what we should do. It may be able to tell us what we can and cannot do (and exlude some practice as impossible), but that's it. There is no such thing as "relying on science" to find out what we should do. We cannot base decisions on science.

Again, let me stress that your arguments have been discussed a lot in the literature. Perhaps you should read it if you are interested in this. A good contemporary book is David Miller's "Critical Rationalism: A restatement and defense" which answers your questions from a Popperian perspective in a very detailed and clear way. --rtc (talk) 17:41, 22 November 2011 (UTC)

This is getting more into WP:FORUM territory, but I have to make a quick point of clarification on the above discussion. There are (at least) two issues being confused here: the logic of the falsification and verification of existential and universal claims, and the reliability of observations to do so.

It is simple logic (given classical logic with bivalence, at least) that if you falsify a claim, you have verified its negation. When talking about scientific theories, we are talking about universal claims: "all phenomena of this sort are correlated in this way with phenomena of that sort", ∀xF(x)→G(x). Lets say for our example F(x) = "x is a swan" and G(x) = "x is white", so "for all x, if x is a swan then x is white", or "all swans are white". To falsify this is precisely to show that ¬∀xF(x)→G(x), which is logically equivalent to ∃x¬(F(x)→G(x)) which is in turn logically equivalent to ∃xF(x)∧¬G(x)) -- that is to say, "there exists some x such that x is a swan and x is not white", or "some non-white swan exists". We falsify "all swans are white" precisely by showing that there is some non-white swan somewhere; so if "all swans are white" is falsifiable, then "there is some non-white swan somewhere" logically must be verifiable.

But that's OK, because "there is some non-white swan somewhere" is not the kind of thing that would be a scientific theory. It is a particular existential claim, not a universal claim. Existential claims are not falsifiable: "There is some non-white swan somewhere" can never be known to be false; at most we can know that "we haven't found any non-white swans yet", because we can never know if we have checked all the swans or not. ("We have checked all the swans" is another universal claim, and is thus falsifiable but not verifiable). This is what I was going on about above w.r.t. dragons and such: universal claims are falsifiable, existential claims are not, and the negation of a universal claims is always an existential claim and vice versa. But conversely, that means that existential claims are verifiable, and universal claims are not. This does not go against Popper because scientific theories are not existential claims, they are universal ones, and are therefore falsifiable and not verifiable.

Completely aside from all of that is the question of the reliability of observation in verifying existential claims and consequently falsifying universal claims. So what if it appears to me that I see a black swan -- can I really take that as evidence that there exist some non-white swans, and thus take "all swans are white" to be falsified? What if I'm hallucinating, or I'm a dreaming brain in a vat in the Matrix being deceived by an Evil Genius, who knows. That's a wholly separate question from the logical issue above. (Although I suppose there are Quinean issues about the theory-ladenness of observation that complicate the matter further, e.g. if my observation of a black swan is dependent upon interpretation of my senses in light of other, non-verifiable, theories, how can my observation constitute verification that there exists a black swan and falsification of "all swans are white"?). --Pfhorrest (talk) 04:50, 23 November 2011 (UTC)

"It is simple logic (given classical logic with bivalence, at least) that if you falsify a claim, you have verified its negation" Popper stressed from the very beginning that there is an asymmetry in logic (the "asymmetry of verification and falsification" as he sometimes calls it) that causes the opposite to be true. The falsification of a claim is thus emphatically not the verification of its opposite. According to Poppers view, verification is not possible. Logic does not admit verification. It only admits criticism, ie., contradictions. "We falsify 'all swans are white' precisely by showing that there is some non-white swan somewhere; so if 'all swans are white' is falsifiable, then 'there is some non-white swan somewhere' logically must be verifiable." No; this is the classical misunderstanding of Popper. We falsify "all swans are white" not by verifying its opposite, but by making a falsifiable claim about a swan being somewhere such that the opposite of "all swans are white" logically follows from this "Completely aside from all of that is the question of the reliability of observation in verifying existential claims and consequently falsifying universal claims" No, that already is a problem of logic; a problem of justifying the correctness of observation statements. Popper discusses this at length. Please note that you have gotten it wrong that Poppers methodology is about showing something, especially about showing something to be false. It is about finding counterexamples. You have to grasp this crucial point that criticism does not imply justification of the opposite. --rtc (talk) 11:07, 23 November 2011 (UTC)

In response to Rfc above saying "Also note that Popper disagrees with the whole philosophical approach you and Pfhorrest are implying in what you say: Popper thinks that no theory can ever be proven or even suppported, made probable, given a reason for etc" (my emphasis added), I just want to point out in case it wasn't clear that I was cautioning SteveBaker against just assuming the negation of Popper's position. I agree pretty strongly with Popper. Note my words: "we are never completely certain about anything, only sufficiently confident for the time being", "Even the best-supported theory can still be falsified by new data", and "no theory is ever 'proven true'". --Pfhorrest (talk) 04:50, 23 November 2011 (UTC)

Pfhorrest, you seem not to have grasped Popper's views to their full extent. Popper goes much beyond what you say. He disagrees with the possibility to be "sufficiently confident for the time being" -- confidence and sufficiency do not exist, according to Popper -- and with the possibility of support for theories. You are still applying the justificationist framework, accepting only some pieces of fallibilism (that knowledge may be wrong), while Popper's view also include Negativism (that you can never give any argument whatsoever in favor of a claim, only against the claim) and Skepticism (that no "epistemic" reasons whatsoever exist and that reason has nothing to do with reasons). --rtc (talk) 11:07, 23 November 2011 (UTC)

That's a lot of verbiage to wade through. Could someone just tell me what sort of observations AGW theory proponents agree would disprove the theory?

• Would they accept, for example, CO2 changes lagging a few centuries BEHIND temperature changes, in ice cores? In other words, does their theory imply that CO2 changes should PRECEDE CO2 changes? And if so, would they accept FOLLOWING CO2 changes as disproof of their theory? --Uncle Ed (talk) 14:30, 24 November 2011 (UTC)

Please see WP:NOTAFORUM and read the verbiage forest above. What does your question have to do with article improvement for this article? Answer: nothing. Vsmith (talk) 15:43, 24 November 2011 (UTC)

Are you an admin, telling me I've violated a rule? I just finished telling you that the verbiage forest didn't answer my question, so what would be the point in reading it? Are you trying to stop me from improving the article, for some reason? (If so, what is the reason?)

Once again (assuming with good faith that somehow you have really missed the editorial point of my question), what I want to know is whether there is any relationship between the concept of falsifiability and the the last 20 years of AGW advocacy. If so, I'd like to improve the article by using the AGW theory as an example of a theory which is considered falsifiable.

AGW would be a great example - assuming its advocates do have in mind a way it could be disproven - because it's more often in the public eye than any other theory I know of (at least in the USA). Also, our readers would love to know what sort of evidence to look for that could, conceivably, put an end to the controversy which (unless you've been living under a rock) you know has been raging these last two decades.

(And please be civil: telling me in advance of my answer that I don't have an answer is not constructive. You have no authority to do that, whether you're an admin or not.) --Uncle Ed (talk) 20:05, 24 November 2011 (UTC)

Sorry if I came across a bit gruff. You question way up top "what sort of observed phenomena could possible disprove AGW?" was the clue that the whole section wasn't really about improving this article from my viewpoint. Now, with your clarification - seems you were fishing for support or to add an example to this page. What you need is simple: reliable sources saying it is unfalsifiable or raising the question. Without solid sources in hand, your question implies a fishing expedition. If you have sources, then by all means let's see them and your proposed addition to the article. Without sources and a concrete proposal the commentary is pure foruming. Vsmith (talk) 00:20, 25 November 2011 (UTC)

A belated thanks for your courteous answer, but I think I still haven't made clear what I'm trying to do. I'm not fishing for a quote saying that AGW "is unfalsifiable". If it were unfalsifiable, then it wouldn't be a good example of falsifiability at all.

Rather, I'm trying to find something well known to the general public (or at least which is discussed a lot in the media), which we can use as an example for our readers to understand what it takes for a hypothesis or theory to be "falsifiable". Specifically, what predictions does AGW make which can be compared against (discovered) facts or observations which could conceivable disprove it?

Or is AGW not a good example, and should we find something else like the moon is made of green cheese? I read somewhere that an astronomer compared the refractive index of the moon's surface to that of green cheese to falsify this hypothesis.

Got me now? I'm looking for a really good popular example, not something controversial which will just make it harder for people to understand falsifiability. --

Original research by Rtc

Rtc keeps reinserting statements he likes but which are either unreferenced or false, supplied with falase references. For example his claim that "White swans do exist" not falsifiable is false (They may become extinct). Likewise, the statement about indivisivility and atoms. What is worse these are supplied with footnotes supposedly to confirm these statements, which are not. For example Popper speaks about indivisibility for a completely different argument.

In summary, sorry, Rtc, your contributions have lost credibility. If you want to add something, please do this an small, well-referenced pieces which can be discussed. Loew Galitz (talk) 16:45, 22 November 2011 (UTC)

Loew, my statements are neither unreferenced nor false. Please stop fighting and be more constructive. Have you read (and understood) the book Logic of Scientific Discovery? I suppose no. Because otherwise you would not make such claims as you do above. The claim that "White swans do exist" is not falsifiable. Of course you can criticize it with the falsifiable theory that swans have become extinct. But criticizability is not falsifiability. You are confusing falsifiability with a completely different topic, the theory of rationality. Many people naively do that, and it's part of the popular philosophy of science to mix all this. But this article needs to be exact. Please read and understand the works Popper has written on the subject before you try to argue on the subject. The statement about indivisibility and atoms is no different. Popper may not use the exact same words, but the meaning is clear to anyone who has read the book. I already said above: please let's not be literalists. Let's read that text carefully: "I now come to my second argument against the primacy of repetitions. It is this. There are laws and theories of a character altogether different from 'All swans are white', even though they may be formulated in a similar way. Take ancient atomic theory. Admittedly, it may be expressed (in one of its simplest forms) as 'All material bodies are composed of corpuscles'. Yet clearly, the 'all'-form is comparatively unimportant in the case of this law. What I mean is this. The problem of showing that one single physical body--say, a piece of iron--is composed of atoms or 'corpuscles' is at least as difficult as that of showing that all swans are white. Our assertions transcend, in both cases, all observational experience." What Popper is saying here is that even a single instance of the statement "All material bodies are composed of corpuscles" already has the same logical character as the statement "All swans are white". I think we agree that "All swans are white" is nothing that can be observed; it transcends all possible observation (as Popper says). But from this we can conclude that the single instance of the statement about material bodies is nothing that can be observed eithter. But this is just the same as saying that the statement itself is not falsifiable. As already mentioned, it may not be written there literally, but it is completely obvious to anyone who has read and understood the book.

Having said that, I certainly won't insist on this specific example. But examples for these kind of counter-intuitive things are definitely needed. Thus, please find better ones; ones that are more literally contained in the sources, if you think that my example is not good enough for you. But it's not a solution to simply delete the example.

If someone has lost credibility, it's you. I have added references, yet you keep reverting. I have asked for a more constructive attitude, yet you keep reverting. I have given arguments based on literature, yet you keep responding with naive original research counter-claims. --rtc (talk) 17:00, 22 November 2011 (UTC)

PS: I now removed the paragraph to at least stop the discussion from heating up further:

However, not all cases are that intuitive. For example, the ancient Greek metaphysical claim "all bodies consist of atoms, ie., indivisible corpuscles" is not falsifiable, despite its "for-all" form: Even if all divisions of some body made so far did not produce an atom (ie., something that cannot be divided any further), it is still possible that the next divsion does. Note that Popper rejected the orthodox positivist view that sich statements, even if unfalsifiable, can still be verified: In this case, even if something was not divisible any further in one or even all past experiments, it may still turn out to be divisible in the next one.

I hope you will show good faith too and help to look for some replacement for which a source can be found that is more explicit about it. --rtc (talk) 19:00, 22 November 2011 (UTC)

PPS: This kind of counter-intuitive case is called "some-and-all"-statement; you can find this term in the index of The logic of scientific discovery. They are mostly discussed on p. 185, especially in footnote *2, which refers to section *24f of Popper's Postscript. --rtc (talk) 19:18, 22 November 2011 (UTC)

Loew, you seem to misunderstand falsifiability in your assertion that the extinction of white swans would falsify "White swans do exist". A claim that white swans have gone extinct would be in itself a claim that white swans do not exist, so if you could conclusively verify that white swans are extinct then in doing so you would falsify the claim that white swans exist, sure. But that is not the issue at hand: the issue at hand is whether you could conclusively verify that white swans are extinct, or never existed, or in any case do not exist at the present. A falsificationist like Popper argues that you cannot ever properly, conclusively verify that: at best you can say "we have made an exhaustive search for white swans and haven't found any, so barring these improbable (but possible) far-fetched ways in which they might exist and yet evade our search, they probably don't exist". But that is not proof that they don't exist; that doesn't positively verify their non-existence; it's just perhaps a reasonable tentative conclusion to make in light of the burden of evidence; but if someone wanted to cling to the last glimmer of hope that there still were white swans somewhere, they are within their epistemic rights (so to speak) to do so, as we're not completely sure that there aren't any. On the other hand, if we found one white swan, then we could say for certain that "there are no white swans" was false, and anyone proclaiming that was irrationally denying sound evidence. --Pfhorrest (talk) 05:11, 23 November 2011 (UTC)

"A falsificationist like Popper argues that you cannot ever properly, conclusively verify that: at best you can say 'we have made an exhaustive search for white swans and haven't found any, so barring these improbable (but possible) far-fetched ways in which they might exist and yet evade our search, they probably don't exist'." No. Popper would argue that the extinction of white swans is not a statement about observations, but a scientific theory. Of course we can say. And we can of course use this theory to criticize the claim that white swans do exist, because rational criticism is not restricted to falsification; falsification is merely a special case of rational criticism. To say "we have made an exhaustive search for white swans and haven't found any, so barring these improbable (but possible) far-fetched ways in which they might exist and yet evade our search, they probably don't exist'" is positivistic and justificationist thinking, which is opposed to the philosophy of Popper. If you argue like that you can say as well (at least before black swans were discovered) "We cannot say 'all swans are white', at best we can say 'we have made an exhaustive search for white swans and have found this or that many, so barring these improbable (but possible) far-fetched ways in which they might exist and yet evade our search, they probably are all white'." It's the same problem about "But that is not proof that they don't exist; that doesn't positively verify their non-existence; it's just perhaps a reasonable tentative conclusion to make in light of the burden of evidence" and everything else that you say. What you say is still justificationist thinking, you replace verification in the strong sense by some kind of evidence-justificationism, which is still diametrally opposed to Popper's position. Admittedly, all of your claims are part of the very popular philosophy of science, which is basically positivism in an unholy marriage with Popperian terminology. But it has nothing whatsoever to do with Popper's actual views. --rtc (talk) 11:39, 23 November 2011 (UTC)

On the subject of the atoms example in particular, and back to actually talking about the article: I affirm the technical validity of the example as used by Popper and quoted by Rfc, but I think it is a bad example to use in the article for the sake of clarity and the confusion that readers will make between the old metaphysical sense of "atoms" (fundamentally indivisible particles) vs the modern scientific sense of "atoms" (the things that join together into molecules and are composed of protons, neutrons, and electrons). --Pfhorrest (talk) 05:11, 23 November 2011 (UTC)

I agree, the example is valid but is confusing; I think it is grand in the article but just not in the lede. IRWolfie- (talk) 14:48, 27 November 2011 (UTC)

"falsifiable in practice"

Rtc added:

Not all statements that are falsifiable in principle are falsifiable in practice. For example, "it will be raining here in one million years" is theoretically falsifiable, but not practically so.

This is a matter of opinion. I say million years is quite practical. Bite me. Once again, REFERENCES, please. Loew Galitz (talk) 16:50, 22 November 2011 (UTC)

Are you able to test this statement? No. It's not practically possible. Let's not be ridiculous. I am happy to supply references, but you should do the same. Read the standard literature (especially the Logic of Scientific Discovery) and you will find them. And use google book search. --rtc (talk) 17:00, 22 November 2011 (UTC)

This seems to be a case of asking for references that the sky is blue. A related essay: WP:BLUE (not a guideline). IRWolfie- (talk) 14:53, 27 November 2011 (UTC)

I am afraid you are confusing the applicabiliy of WP:BLUE. "Sky is blue" is the statement for which it is exteremely easy to find a reference. I challenge you to find a reference about the rain. Regardless it being true or false, it is Original Research. I can think of hundred similar difficult-to-falsify examples of this and opposite kind. I don't even want to start arguing about the meaning of "practicality" and how this concept is covered within the topic of the article. So, reference, please, or the statement is gone. Loew Galitz (talk) 02:17, 15 December 2011 (UTC)

Confusing usage of "public" and "public school"

In the Article we find this fragment:

"...US law says that only science may be taught in public school science classes (see Science education)."

Following the wiki link to Science education, we find an article in which the first reference to "public school" is in the following fragment:

"The first person credited with being employed as a Science teacher in a British public school was William Sharp...."

Following that wiki link to public school we find an article about schools that are not dependent on government financing. The reader may now conclude, simply by following two wiki links, that US law requires only science to be taught in schools that are not dependent on government financing.

This type of confusion arises because the term "public school" happens to have exactly opposite meanings in the US and UK respectively. In the US, a public school is a school run by the government. In the UK a public school is one that is run independently of the government.

I propose the solution that we always clarify by writing "government-run school" instead of "public school" when the US meaning is intended. And likewise, we should say "non-government school" instead of "public school" when the UK meaning is intended. This usage should ideally apply everywhere in the English Wikipedia.

In the current article, then, we could write:

"US law says that only science may be taught in science classes in government-run schools."

(Citation still needed for this assertion.)

(Edit: added signature below.)

Rahul (talk) 23:27, 2 January 2012 (UTC)

Criticism section has bad citations

The first two citations in the criticism section do not actually support what they're supposed to be citing. I'll give it a day, and then remove it. — Preceding unsigned comment added by 76.19.102.88 (talk) 00:25, 27 July 2012 (UTC)

Re: Examples of "unfalsifiable" statements in Overview section

I may need enlightening here.

Consider that we do have evidence of absence sometimes. The statement "Coins exist in my pocket" is falsifiable, but I can also tell you it is falsified - by checking quick.

My quandary starts when the article suggests that "White Swans exist" is a claim with no logical counter-example. Is that right? Is that the example that Popper gives in "The logic of scientific discovery" (because the reference does not mention a page number to help me find it, and yet it appears after the period - suggesting the whole paragraph IS the example that Popper gives)?

Presumably that claim translates to "White Swans exist anywhere in reality right now". We can logically imagine exploring everywhere and find there are no white swans. Like coins missing from a pocket. The statement does not seem unfalsifiable. It seems (with great practical difficulty) logically falsifiable.

If there are Swans somewhere (on earth) and coins in my pocket, those statements do not seem "unfalsifiable" in theory just because they were not falsified. Or is the idea: in that reality, practically, the claims are 'now unfalsifiable because nothing can counter the prior discovery of Swans on earth and coins in my pocket.

Thoughts? Maybe the article just needs to be more explicit about something. I may have similar issues with "All men are mortal" being called unfalsifiable. -Tesseract²(talk) 17:21, 19 February 2012 (UTC)

The problem is precisely with the falsification requiring the completion of an infinite task. We hypothesize "white swans exists", and then go about testing whether or not that is true. By default, at the start, white swans existing is merely possible, neither verified nor falsified. Then over any given period of time, either we find white swans, and know it's true (though some have issue with the possibility of doing this, but that's a long tangent); or we don't, and it's still merely possible. No matter how long a period of time, no matter how far and wide you search, there is always the possibility that there are still white swans out there which you haven't found yet. Yes, if we could search all reality we could falsify it, if there were no white swans at all; but we can never finish searching all reality, which is not merely a practical or technological problem but a fundamental epistemological one. It's like a mathematical limit: we can approach "all of reality searched" arbitrarily closely, but never actually get there. Likewise with all men being mortal: all we can say is that all men so far have died... any given one still alive might live forever. But until we reach forever, we can't know that. And we will never, by definition, reach forever. So we can never know if a given man is immortal; only that he hasn't died yet. --Pfhorrest (talk) 22:08, 19 February 2012 (UTC)

The statement "White swans exist somewhere within my apartment" is obviously falsifiable. "White swans exist somewhere on Earth" is also falsifiable - albeit with a lot of difficulty. "White swans exist anywhere within the universe" is also falsifiable if the universe is both finite and accessible by some means. So if the universe is infinite (something we're not sure of yet) - or if it's larger than the 'visible universe' (which it appears to be) - or if it's further away than we could travel (which it certainly is) then this is indeed an unfalsifiable statement. But it's only unfalsifiable for annoying practical reasons. IMHO, that makes it a poor example because it's open to more debate than one would like.

The non-existence of God is a vastly better example because "God" is generally defined as being omnipotent/omniscient - and a being with truly infinite/unlimited powers and abilities cannot be proven to exist because he/she/it can simply cause whatever test we might come up with to fail - or reconfigure our brains such that we can't think of the test in the first place.

Something less controversial might be "there is a sequence of 1000 consecutive 7's somewhere in the decimal expansion of pi". This is unfalsifiable because pi is an infinitely long number and a search for this sequence of a thousand consecutive 7's could never terminate.

That said - we can't just come up with our own examples - that would be WP:OR. We need something that's backed by reliable sources - which is why the "white swan" example is our best choice right now.

SteveBaker (talk) 14:51, 20 February 2012 (UTC)

@Tesseract2 "Presumably that claim translates to 'White Swans exist anywhere in reality right now'. We can logically imagine exploring everywhere and find there are no white swans. Like coins missing from a pocket. The statement does not seem unfalsifiable. It seems (with great practical difficulty) logically falsifiable." What you mean here by "logically falsifiable" is really that you can criticize it. It is a common misunderstanding that people confuse the notion of falsifiability with the notion of criticizability. You can criticize the statement "White Swans exist" with the claim "White swans do not exist", which is a falsifiable theory. If try to falsify this theory, by searching for white swans, you will find out pretty quickly that the theory is false. You also observed correctly that the unfalsifiable statement "White Swans exist" can be be the consequence of falsifiable statements like "there are Swans somewhere on earth". Falsifiability is not the absence of a infectious disease. That something is unfalsifiable doesn't mean that anything that produces it as a logical consequence will be infected by its unfalsifiability.

@Pfhorrest "The problem is precisely with the falsification requiring the completion of an infinite task. We hypothesize "white swans exists", and then go about testing whether or not that is true. By default, at the start, white swans existing is merely possible, neither verified nor falsified." Wrong, that would be possibilism (criticzed at length in Miller's Critical Rationalism), not falsificationism. In falsificationism, you start with the assumption that white swans do exist. And then you try to criticize that claim, using all the knowledge you have. Falsificationism incorporates fallibilism. Fallibilism disagrees with the popular view that you may assume something to be true only once you are absolutely certain you are correct. Fallibilism says that it is impossible to ever be certain to correct about anything. There are only two possibilities: either "white swans exists" is right (then your assumption was correct, you made tried to criticize as hard as you could -- not an infinite task! -- and kept your eyes and your mind open after all criticism you could imagine turned out to be unsubstantiated) or it is wrong, then the criticism will find out after a finite number of steps -- not an infinite task either! -- and you have to give up your assumption.

@SteveBaker Popper says that a statement is falsifiable if it logically contradicts at least one basic statement (section 26 LScD). This is a question of pure logic and not one that implies more "difficult" or less "difficult" falsifications. And neither can the question of the existence or non-existence of god be decided using the notion of falsifiability, nor does it give you an answer to the related question whether claims about the existence of god can be rationally accepted or rationally rejected. And what you are discussing BTW is not the question of the falsifiability of the claim "god exists", but the question of whether this claim contains such-called "re-inforced dogmatisms", that is, logical devices that can be used to deflect criticism. And, like the other users, you seem to be confusing the question of falsification with the question of criticism, which are not the same. --rtc (talk) 22:16, 8 March 2012 (UTC)

Pfhorrest, I was going to mention that it can be difficult to give page numbers in citations that are used multiple times, because it would create entries for every one of, say, 7 citations, and can cause significant lengthening of an already long and challenged article. But then I saw that the citation given is to google books. A searchable google book. So, yes, that burden on the reader, to enter a search line, is probably a reasonable one, considering that MLA says it is not preferable to provide a link to an online item in the first place... Grye (talk) 06:54, 15 December 2012 (UTC)