Talk:Falsifiability/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Newton

Minor note: Newton stated many of his original laws in terms of calculas in such a way that they held up their validity under Einstein's theories.

logic

A misunderstanding in scientific method prompted me to write a more formal account of falsification and its innate difficulties. I think it is better here than there, or in induction (philosophy), but am open to suggestions. Also it may need some filling out, depending on the reaction. Banno 07:50, Jul 24, 2004 (UTC)

Swan argument

Love the pictures, and the new discussion of the swan argument is concise; but the argument already appeared earlier in the article. Some editing is needed to bring the two together without loosing the connection with ad hoc hypotheses. Banno 21:39, Jul 24, 2004 (UTC)

Merged demarcation sections

Three demarcation sections from different articles, including this one, are now merged to Demarcation problem. The other two articles are Pseudoscience, Falsifiability and the Scientific Method. -- Chris

I’ve replaced the stuff that was deleted. It contained much that is essential to the understanding of falsifiability. It would be better to edit the material than to simply give a glib reference. Banno 11:02, Aug 8, 2004 (UTC)

I understand. I've made a link to the demarcation article for the time being. -- Chris 17:43, 8 Aug 2004 (UTC)

POV in solipsism?

If I'm not mistaken, isn't the opposite of solipsism ("The universe does exist") also non-falsifiable? What evidence could a solipsist offer a non-solipsist that would falsify non-solipsism? The non-solipsist might reply that if he couldn't see or sense anything, then that would be sufficient evidence. However, such an argument assumes non-solipsism, is therefore circular, and thus, not scientific or falsifiable, no? Bottom line: Is it possible to scientifically prove that the universe exists? I don't think so. I'm not a solipsist, but the article seems to have a bias against solipsism.

Differentiate between science and philosophy. I'd posit it's possible to scientifically prove the universe exists, we just begin hitting some variant of an incompleteness theorem if trying to go further (solipsism seems to me uncomparable to Gödel's 2 incompleteness theorems as stated). The section does seem to need editing, as falsifiability is not axiomatic in philosophy. 24.22.227.53 14:52, 11 October 2005 (UTC)

Well, non-solipsism isn't falsifiable, but non-solipsism is a necessary assumption for any argumentation between different individuals to make sense. If a solipsist were to offer a non-solipsist anything, the non-solipsist would have won the argument! The article has a bias against solipsism, because it's actually meaningless to argue for non-solipsism - it being a premise for the existence of anyone else than the individual who asserts non-solipsism. It's equally meaningless to argue for solipsism - because if you're right, there is noone else around to care about your argumentation. --FreezBee 11:23, 13 March 2006 (UTC)

Thank you

I am studying Sociology in school and am doing an essay on the scientific method and it's application to sociology. I have poured over all of my books to find a definition for falsifiability other than that given in the oh-so-trusty Websters to no avail. I came across this web page and can now continue on with the essay. Thank you so for having a wonderful, free information source.--Traci 01:26, 17 Dec 2004 (UTC)

Intro

Someone changed the psychoanalytic example in the intro about homosexuality and denial to alcoholism and denial, thereby making the example miss the point. If an alcoholic could abstain from drinking, he would have convincingly demonstrated he can control his drinking. But the point of the example is that a man who engages in no homosexual acitivity and who has sex with women may be regarded as acting "in denial," not simply as exercising control as in the drinking example. The equivalent might be that the less the alcoholic drinks, the more we accuse him of being an alcoholic, which would make no sense. So I've changed the example back to homosexuality. SlimVirgin 01:52, Feb 16, 2005 (UTC)

Oddly, the alcoholiic example would still be good. Alcoholics Anonymous defines alcoholism such that someone who is an alcoholic but has stopped drinking is still an alcoholic. Ken Arromdee 21:30, 10 August 2005 (UTC)

on conspiracy's by karl popper

I do not wish to imply that conspiracies never happen. On the contrary, they are typical social phenomena. They become important, for example, whenever people who believe in the conspiracy theory get into power. And people who sincerely believe that they know how to make heaven on earth are most likely to adopt the conspiracy teory, and to get involved in a counter-conspiracy theory against non-existing conspirators. For the only explanation of their failure to produce their heaven is the evil intention of the Devil, who has a vested interest in hell.

Conspiracies occur, it must be admitted. But the striking fact which, in spite of their occurrence, disproved the conspiracy theory is that few of these conspiracies are ultimately successful. Conspirators rarely consummate their conspiracy.

Why is this so? Why do achievements differ so widely from aspirations? Because this is usually the case in social life, conspiracy or no conspiracy. Social life is not only a trial of strength between opposing groups: it is action within a more or less resilent or brittle framework of institutions and traditions, and it creates - apart from any conscious counter-action - many unforseen reactions in this framework, some of them perhaps even unforseeable.

Critical Tests

Removed from the article:

In demarcating science from non-science Popper suggests we determine if statements have associated critical tests. If they do, then the statements are potentially falsifiable, and therefore scientific because the test is critical to the continued existence of the theory. For Popper’s theory of demarcation to survive, critical tests must exist.

Critical tests are comprised of basic observational statements. The falsificationists insist that these statements are theory-dependent. That the observation of a rise in electrical current, requires some theory of electricity, or that the observation of Jupiter’s moons with a telescope, is based upon the laws of optics. This means that if the observational statements comprising a critical test appear to falsify a theory X, there is no way of knowing logically if the falsity resides within the theory X or within the theories and assumptions that were required for the test. In science the tendency is to doubt the test rather than to refute long established theory. However, even if great confidence resides in new tests, and a theory is thus overthrown, this move cannot be logically supported. For it will always be conceivable that advances in scientific knowledge might reveal new inadequacies in the tests that overthrew the theory.

This is an oversimplification, which misses some of the points carefully made in the article. What is falsified is the complete hypothesis, which consists of theory X, together with all other theories needed to predict the result of the observation. For example, this is clearly the case in cosmology, where even apparent fundamentals such as the speed of light and other physical constants may be considered to be variable in some theories. -- The Anome 10:00, May 24, 2005 (UTC)

I have only read a little in this area but this was not my understanding of Falsificationism. If you falsify the theory plus all other theories then you may be disregarding theories that were true. It is unlikely that they are all wrong. This total deconstruction via falsification of a theory plus all theories which lie behind it has the qualities of a Kuhnian paradigm change but not I thought of Popper's Falsificationism. Perhaps my problem is that I am only familiar with Popper and not other falsificationists?

I'm new to all this. In the future if I think an article should have a paragraph or sentence added should I put it up for discussion first?--Garsanllean 10:29, 24 May 2005 (UTC)

Theory refinement

As a newcomer to these topics, I'm having trouble wrapping my head around this statement:

"At each stage, experimental observation made a theory untenable (i.e., falsified it) and a new theory was found which had greater 'explanatory power' (i.e., could account for the previously unexplained phenomena), and as a result provided greater opportunity for its own falsification."

How does a refined theory with a more precise universal statement provide greater opportunity for its own falsification? For example, how is 'all swans are white except those in Australia' more likely to be falsified than 'all swans are white'? And is there an implication here that the refined theory is somehow less credible? Clarify in the article, if you can. Thanks. — mjb 23:48, 29 July 2005 (UTC)

Demarcation

OK so something is going wrong here. We have:

Popper noticed that the philosophers of the Vienna Circle had mixed two different problems, and had accordingly given a single solution to both of them: verificationism. In opposition to this view, Popper emphasized that a theory might well be meaningful without being scientific, and that, accordingly, a criterion of meaningfulness may not necessarily coincide with a criterion of demarcation.

followed by

However, much that would be considered meaningful and useful is not falsifiable. Certainly non-falsifiable statements have a role in scientific theories themselves. The Popperian criterion provides a definition of science that excludes much that is of value; it does not provide a way to distinguish meaningful statements from meaningless ones.

The second excerpt seems to be criticising a straw man. Popper differentiated between scientific and meaningful. And then his criterion of demarcation for science is accused of not distinguishing meaningful from meaningless. When the difference is exactly the point he makes.

I would edit this, but I am a big fan of Popper and would probably do him too much justice.

Swans

All swans are white, that black thing you're pointing to is therefore a "Sban" not a "Swan". All it takes is one allele difference to make a new name. ??????????? 16:27, 11 October 2005 (UTC)

:-) Well, Popper's argumentation against logical positivism actually was that there was nothing such an elemental observation - all observations are theory-driven. Therefore there is not such a thing as uninterpreted evidence. As you say, one way to defend the claim of global whiteness of swans is to interpret non-white swans as something else. A theory is never falsified by evidence, only by another theory. --FreezBee 11:29, 13 March 2006 (UTC)

Astrology does not make falsifiable predictions

Astrologers may make the mistake of making falsifiable predictions, but the various kinds of astrology do not. Astrology is very abstract, and it takes a human to translate that into something which could actually be true or false. For example, some forms of astrology predict that around the age of 28 (when Saturn returns to the place it was in the sky when they were born), a person will experience a period of extra trials and tests. There is nothing concrete about this. It's a prediction about the individual's feelings, and only that person can know if the prediction is useful, much less accurate.

Perhaps a less controversial example is in order? Crag 18:12, 13 October 2005 (UTC)

Minor edit: Falsificationism

I split off the last two paragraphs of Naive falsification into "Falsificationism" so people can find that information more easily if they are looking for it.

Jmchen 18:46, 7 November 2005 (UTC)

Economics

The subsection on economics needs work. It seems odd that it would focus on one area where the discipline is (arguably) non-falsifiable. Those empiricists in economics do nothing but falsify (in some sense) hypotheses. John McAdams 10:46, 12 November 2005 (UTC)

Popper's swan argument

You realise that all it would take to keep this from being proven false would be to destroy all black swans. If you can't find it, you can't say it exists. --Guthrie 19:13, 12 January 2006 (UTC)

But how would you know that you had killed all the black swans? Banno 19:54, 12 January 2006 (UTC)

You'd have to make a theory. You'd say, "I think I have killed all the black swans." Of course, if someone found a black swan, you'd be disproven. Unless you killed all the black swans that conflicted. But how would you know you had killed all the black swans that conflicted? You'd have to make a theory... :D --Mathwizard1232 05:46, 26 February 2006 (UTC)

Guthrie's first sentence is true, but the last one suffers from a logical fallacy - attempting to prove a negative. This applies here because it has been established that black swans have existed wild in nature. One could only claim to have "thought" that one had succeeded in killing all of them. One cannot be certain. This has occurred with some supposedly extinct species. Strangely enough, a few isolated examples suddenly turn up in some remote location. It would be an entirely different matter if one had proof that all black swans had only ever existed in a certain laboratory room, that none had been stolen or escaped, and that they were all killed. (That would be too bad, since they are rather beautiful birds!) -- Fyslee 12:47, 26 February 2006 (UTC)

Would it not be a good idea to include a falsifiable argument that is not rediculously false - like "there is a black swan". What use would such a blatantly false statements be to a scientist iyswim Would it be worth including that there is no critereon of truth to Popper (so we are aproximating something that we cannot identify), and that observation sentences are acepted by convention - we accept that that was a black swan not because it is true that that is a black swan (as ob statememts are neoither incorrigible nor can we use induction) but becuase of convention. As a consequence some think that the decision is tortally arbitrary. Nor can it be said that we have evidence for a theory, which runs against scientific practice.

Criticism from scientists

I think this section is poorly written. Firstly, the alleged quotes from Sokal and Weinberg are given without references or context. Secondly, the astrology example misses the point. Astrology does not make falsifiable predictions, and this is its major flaw - its easy to interpret horoscopes to make them cohere with almost any set of experiences (the same applies to psychoanalysis and Marxism). The part about "But the same is true of actual science: a physical theory predicts that performing a certain operation will result in a number in a certain range" seems like a vague reference to underdetermination and the supposed impossibility of critical examples. I would propose deleting this entire section and replacing it with a proper exposition of the Duhem-Quine thesis in relation to falsification.

I agree. Unless this argument can be attributed to an authority (and, if so, better explained), the section should go. - John.Conway 02:21, 1 February 2006 (UTC)

Be slow about deleting the section. That Sokal quote is quite profound and makes a good point, at least put effort into verifying it was his statement--and if it cannot be verified at least make this point somewhere else. (By the way, "this is Sokal's statement" is a falsifiable statement assuming he is dead and cannot make any more statements. If he is still alive, he may still write the statement, thus making "this is Sokal's statement" unfalsifiable.) —The preceding unsigned comment was added by 68.97.191.26 (talk • contribs) .

Poor example moved to talk page for discussion

JA: Re:

In scientific inquiry any theory that is not falsifiable is unscientific according to some views. Psychoanalytic theory, for example, is held up by the proponents of Popper's philosophy as an example of an ideology rather than a science. A patient regarded by his psychoanalyst as "in denial" about his sexual orientation may be viewed as confirming he is homosexual simply by denying that he is; and if he has sex with women, he may be accused of trying to buttress his denials. In other words, there is no way the patient could convincingly demonstrate his heterosexuality to the analyst. This is an example of what Popper called a "closed circle". The proposition that the patient is homosexual is not falsifiable. This has found its way into several dystopian novels.

JA: The above text provides a very poor and misleading illustration of falsifiability as it applies to scientific theories, for several reasons. I will mention a couple and return to the issue tomorrow.

JA: Aside from the POV aspects, and the "novel reasoning", or the "reasoning from fiction" at the end of the illustration, the concocted example does not exemplify theory-testing, but it exemplifies only the diagnosis of an individual case. This is a very important distinction in science and has beenn understood to be so from the time of Aristotle at least, who pointed out that there is no "knowledge" of individual cases except in so far as they fall under general descriptions. In a word, if perhaps a bit too briefly, science is not "about" individuals, though it does "apply" to individuals. So a reasoner is justified in applying a theory, in which he or she has developed an overall level of confidence, to an individual case. The criterion of falsifiability applies to the aggregate of data and not to the individual case, and the theory is tested in the long run by its ability to explain phenomena. So a theory about psychological denial is falsifiable as a whole, as it can be falsified by its failing to explain phenomena over the long term. Jon Awbrey 05:16, 20 March 2006 (UTC)

In what way does collecting data which applies to a theory regarding a single person differ from collecting data which applies to a theory regarding the entire world? AFAIK Theories can apply to the psychology of a given individual as well as anything else. Though I see where you're coming from (I think I recognise the coherence theory of justification here), the "falsifiability" problem could trivially be expanded to apply to all cases of people claiming not to be homosexuals, in practice and theory, with nothing but a slight rewording - then re-applied to the particular case of the particular patient dealt with in the example. --AceMyth 05:21, 20 March 2006 (UTC)

JA: I think you may be raising a different question, perhaps having to do with the fact that "individual" is a relative term. Yes, you can do what are called "single case" studies, where you study one individual person over time, but the acquistion of knowledge about this case still involves aggregated longitudinal data. You do not form, or at least you do not sensibly test your "theory of that person" on single data points.

JA: But the question at issue here is only whether a theory that involves a particular type of theoretical construct, namely, "denial", is falsifible or not. Theories involving that sort of construct are falsifiable in the same way that other theories are falsifiable, by failing to explain phemomena and make helpful predictions in the long run. So the example given is not a valid example of an indefeasible theory. Jon Awbrey 05:44, 20 March 2006 (UTC)

Incompleteness of Falsification

Just for the fun of it: Let T be atheory. A metatheory is a theory about theories, e.g. that only falsifiable theories are scientific. If M is a metatheory, then M(T) is M applied to T. In the case mentioned, the result of M(T) is one of the verdicts scientific and not-scientific, depending on whether T is falsifiable or not. Obviously M is a theory itself, so we could state M(M).

Now, let F be falsificationism, a metatheory. So F(F) says that falsificationism is only scientific, if it is falsifiable, that is, if it is at least possible for a theory to be scientific without being falsifiable.

--FreezBee 13:40, 20 March 2006 (UTC)

Popper project

I have removed the following paragraph:

Following from Feyerabend, the whole "Popper project" to define science around one particular methodology—which accepts nothing except itself—is a perverse example of what he supposedly decried: a closed circle argument. The Popperian criterion itself is not falsifiable. Moreover, it makes Popper effectively a philosophical nominalist, which has nothing to do with empirical sciences at all. Although Popper's claim of the singular characteristic of falsifiability does provide a way to replace invalid inductive thinking (empiricism) with deductive, falsifiable reasoning, it appeared to Feyerabend that doing so is neither necessary for, nor conducive to, scientific progress.

The reason should be clear from my recent edits; this is a criticism of naive falsification, not falsificationism. Furthermore, the argument presented is wrongly attributed to Feyerabend; he presented nothing so trivial. It is also clear from the section on demarcation - Kudos to the author - that the criticism that falsificationism is not falsifiable is shallow. Banno 21:37, 1 April 2006 (UTC)

Case Studies

I have also removed:

Multiple universes from the Anthropic Principle and the existence of intelligent life (see SETI) beyond Earth are potentially non-falsifiable ideas. They are "true-ifiable" because they are potentially detectable. Lack of detection does not mean other universes or non-human intelligent life does not exist; it only means they have not been detected. Yet, both of these ideas are generally considered scientific ideas. Some suggest that an idea has to be only one of falsifiable or "true-ifiable", but not both to be considered a scientific idea. That is, it must be at least one of confirmable or deniable.

There may be some merit in this argument, but since it is uncited, and poorly worded (true-ifiable?) and argued, it is better left out for now. Banno 21:50, 1 April 2006 (UTC)

Sokal

The extended quote really needs a citation. Banno 21:59, 1 April 2006 (UTC)

Examples

The examples are a useful part of the article, but are far from encyclopedic. They would all benefit from citations and some weasel extermination. Can any one help in finding citations for the various sections? Banno 21:10, 2 April 2006 (UTC)

Erm, this: "If evidence cannot be presented to support a case, and yet the case cannot be shown to be indeed false, not much credence can be given to such a statement" Surely the first "cannot" should be a can. If not, its unclear. And if SO, then I would have thought that this is contentious, and depends on the truth of falsifiability -- so shouldn't be stated as FACT in a nuetral article.

Empirical & Formal, Secular & Eternal

JA: And here you'd been told that the synthetic-analytic distinction was dead, but it was just so typical of Quine to miss the fact that making a distinction relational is not the same thing as making it disappear. I keep saying that you can't really disentangle the phil'o'sci from the phil'o'math, and this is yet another knot in their texture. We carry out deductively exact reasoning within the math model, but whether that model (logical) is a model (analogical) of anything out there is a risky contingency. The Greeks distinguished Being, the eternal invariant world of math, from Becoming, the physis of coming to be and passing away, and the relation between them is as much a problematique as it ever was, et saecula saeculorum. Amen. Jon Awbrey 18:10, 8 April 2006 (UTC)

Is Epistemological Solipsism Falsifiable?

It seems that the article defines solipsism only as metaphysical solipsism and overlooks epistemological solipsism. The former holds (as reflected in the article and comments above) that the whole universe exists only in "my" brain -- a radical and problematic view. On the other hand, epistemological solipsism is a much more liberal and arguably widely held view which holds that all that any observer really knows is the contents of their own mind. An epistemological solipsist does not hold that there are no other selves, just that for all of them, each knows only the contents of their own minds. If a mind is working well, then its contents present a sufficiently accurate representation of the outside world to facilitate successful interaction with it. Moreover, epistemological solipsism might be falsifiable if it could be shown by brain scans that someone knows something that is outside their brain and their knowing of it has no neurological substrate. Ian Goddard 05:47, 12 April 2006 (UTC)

JA: Maybe it would help to give more explicit statements of the theses in question, as they seem at present to be far too vague to be analyzed and argued, for or against, with any sort of conviction. That said, there seem to be some prevalent misunderstandings about falsifiability. As a condition of being an empirical hypothesis, it is (1) a logical condition, not a practical one, and (2) a necessary condition, not a sufficient one. Thus, there are other conditions of being an empirical statement that take precedence, and which it is economical to examine first, before expending time and wit on the question of falsifiability. Jon Awbrey 13:36, 12 April 2006 (UTC)

The suggestion I'm making is that the article ought to acknowledge that solipsism as the article describes it is only one kind of solipsism, specifically metaphysical solipsism. As it says: "Solipsism claims that the Universe exists entirely in one's own mind." But that is not the view of epistemological solipsism (as described above). So the article's conclusion does not, contrary to its contention, apply to solipsism per se. This then raises the auxiliary question (perhaps suited for other fora) of whether or not the latter form of solipsism is falsifiable. Ian Goddard 15:56, 12 April 2006 (UTC)

JA: I am simply saying that neither of the theses so far formulated approach the level of scientific statements on any grounds. For instance: Are they intended as explanations of any specific phenomena? Then what is the surprising feature of appearances that they are supposed to explain? Jon Awbrey 16:06, 12 April 2006 (UTC)

Since 'science' is derived from the Latin term 'scientia' meaning 'knowledge,' it's not clear to me that a set of statements purporting to define the physical scope of knowledge inherently fails to "approach the level of scientific statements." Moreover, you seem to define scientific statements as only those that purport to explain something "surprising." However, science also includes boring catalogs of empirical observations of mundane matters. Surprise and excitement are not criteria for being scientific. It also seems that the direction your input takes us is well off course. The article at hand is falsifiability, not science per se. Ian Goddard 01:27, 13 April 2006 (UTC)

JA: The relevant context for the concept of falsifiability, as indicated in the first few words of the article, is "science and the philosophy of science". The concept per se is a logical property of propositions, but its use in this context is as a criterion of propositions that can qualify as empirical hypotheses. Science is an activity that is carried out with a community of inquiry in mind, however isolated the work may appear to be at times, and it is carried out by persons with sufficient maturity to have an experience of reality. Folks like that do not waste their time with sophomoric solipsisms of any variety, except perhaps as a diversion from the more serious play of inquiry. These solipsisms are pure non-starters in this context, as anybody with a minimal grasp of what science is about immediately recognizes. Jon Awbrey 03:00, 13 April 2006 (UTC)

JA: Let me try to put this point another way, as I see the possibility of a rhetorical regress already in progress along the current lines of discussion. This is really the same formal point as the one about what you have to show and what you do not have to show in the matter of physical theories being empirical. Contingency is a logical property, and this remains true no matter whether the subject matter at hand is physics, chemistry, biology, psychology, epistemology, ontology, or theology. To show that a proposition is falsifiable, all you have to show is that there are cases in the universe of discourse, which cases may themselves be universes of discourse, that are countermodels of the proposition. This has nothing to do with the physical, psychological, or practical difficulties of constructing anti-grav skateboards, cloning black swans from white swan stock, convincing recalcitrant sophomores that they are not immortal, und so weiter. Jon Awbrey 04:10, 13 April 2006 (UTC)

I'm updating the article to denote that the solipsism it describes is only metaphysical solipsism. Therefore, the conclusion it derives stemming from premises of that form of solipsism do not apply to solipsism proper. Justification for the distinction I'm making between epistemological and metaphysical solipsism can be found in a search of jstor.org. And, while most philosophy dictionaries do not make it, The Harper Collins Dictionary of Philosophy [*] does. I'll enter the first line of each entry from that text:

solipsism, epistemological 1. the theory that one's consciousness (self, mind) cannot know anything other than its own content. [...]

solipsism, metaphysical literally, "I myself only exist"; the theory that no reality exists other than one's self. [...]

[*] Angeles, P. A. (1992). The Harper Collins Dictionary of Philosophy. New York: Harper Perennial.

Perhaps epistemological solipsism should also be listed as another nonfalsifiable example, but if so, it needs a separate justification as well as supporting citation. It strikes me as problematic that most wiki articles pertaining to philosophical issues include statements lacking reference to supporting literature such that readers are left to wonder if some statement reflects an established view or that of a random wiki editor. Ian Goddard 16:22, 13 April 2006 (UTC)

Metaphysical solipsism falsifiable?

Jon, to declare a theory falsifiable as you do metaphysical solipsism (here), you have to show that the set of conceivable (ie, logically possible) contrary observations is nonempty. However, as the argument posted by Wikipedia co-founder Larry Sanger that you so blithely abolished noted: "whatever evidence one might adduce that is contrary to solipsism can be, after all, dismissed as something that is 'in one's mind.'" And convincing a believer has no bearing on the matter. It seems none of us can falsify metaphysical solipsism because any observations that I could add to its set of contrary observations (such as seeing or talking to 'others') can be removed from the set by simply defining them as illusions in my dream.

Jon, unless you can show a contrary view in philosophy literature, please leave the metaphysical solipsim entry as it was. Ian Goddard 07:09, 14 April 2006 (UTC)

What is falsifiability anyway?

JA: There now appears to be at least a couple of different views as to what sort of predicate the predicate of falsifiability might be, and perhaps there can be brought forward an equal number of clear and distinct, as opposed to confused and vague, definitions for the word in question. I feel that I have already given the definition that I learned, and will proceed to go hunt up sources. Perhaps it will help matters if Ian provides a formal definition of the notion he uses, along with some more solid sources than "Our Co-Founder". And then we can begin to incorporate the various usages into the text somehow. I will throughout be operating under the declared article topic that we are talking about the concepts of falsifiability that are relevant to science and the philosophy thereof. If that is not the general understanding then there is a different sort of problem to address. Jon Awbrey 13:46, 14 April 2006 (UTC)

JA: One definition of the predicate "falsifiability" has been given. It is a predicate that applies initially to propositions and by extension to theories that contain propositions. This can be made more formal if need be. The definition is: A proposition is falsifiable if and only if it has countermodels in the relevant universe of discourse. In other words, a proposition p is falsifiable if and only if there is an object x in the intended universe X such that p(x) is false. That is all that this definition says, though the context of use will require one to say what is meant by an "intended universe of discourse" and a few other details.

Agreed. So, simply, what x (where x is an observation) shows that propositions of metaphysical solipsism are false? Ian Goddard 16:11, 14 April 2006 (UTC)

JA: By way of seeing whether there is another coherent definition, let us examine the text of Ian's remarks to see if we can extract one:

IG: Jon, to declare a theory falsifiable as you do metaphysical solipsism (here), you have to show that the set of conceivable (ie, logically possible) contrary observations is nonempty. However, as the argument posted by Wikipedia co-founder Larry Sanger that you so blithely abolished noted: "whatever evidence one might adduce that is contrary to solipsism can be, after all, dismissed as something that is 'in one's mind'". And convincing a believer has no bearing on the matter. It seems none of us can falsify metaphysical solipsism because any observations that I could add to its set of contrary observations (such as seeing or talking to 'others') can be removed from the set by simply defining them as illusions in my dream.

JA: But first a second cup of coffee. Jon Awbrey 14:12, 14 April 2006 (UTC)

IG: Agreed. So, simply, what x (where x is an observation) shows that propositions of metaphysical solipsism are false? Ian Goddard 16:11, 14 April 2006 (UTC)

JA: Just so we don't get into a "garden of forking paths" situation, I copied your last comment into a serial frame. The current "definition" of "metaphysical solipsism" (MS) — or maybe "ontological solipsism" (OS) is more specific? — is this:

Metaphysical solipsism claims that the Universe exists entirely in one's own mind.

JA: Defigurizing the anthropomorphism would give this:

Metaphysical solipsism is the proposition that the Universe exists entirely in one's own mind.

JA: I am hard pressed to guess what that sentence means, but maybe this would be one possible paraphrase:

Metaphysical solipsism is the proposition that the predicate "mental occurrence" applies to everything in the Universe.

JA: I will wait to see if that accords with your intention before proceeding. Jon Awbrey 16:28, 14 April 2006 (UTC)

Jon, metaphysical solipsism is not my concept (consult the text I cited above), so my intention has no bearing on the issue. I've entered no argument to the article. Given that the preexisting argument has stood here for years, I think the onus is on you to justify its removal as well as your contrary argument. Since you've claimed that metaphysical solipsism is falsifiable, you should support that claim by identifying the x your claim requires.

Btw, I agree with Noosphere that the previous entry should also cite a source, but I find it odd that given dozens of new claims recently posted with no references, only one posted by Sanger years ago suddenly requires reference. My own contribution to the entry does have supporting reference (see above). Ian Goddard 16:59, 14 April 2006 (UTC)

JA: That is the statement currently in the article, and you have stood for retaining it in the article, so you are the man on the spot to interpret and justify keeping it. Maybe I missed it, but the sources you gave are WP-internal and so do not qualify. I've seen things by LS on the web, and I'm guessing that there are published works by him or others that have a more formal definition, but so far I don't see any here. And it's not my job to find them. I hope it's not your argument that things that stand for years must stand for aye. Jon Awbrey 17:14, 14 April 2006 (UTC)

JA: In case it was not clear, my standing claim is that the so-called "definition" of MS as it currently stands in the article is simply not definite enough to support logical inferences about its implications. I am trying work with it, but if nobody can speak for what it means or answer even the simplest questions required to clarify it, then I must assume that it is meaningless, and it will have to go. I asked if the following is an acceptable paraphrase:

Metaphysical solipsism is the proposition that the predicate "mental occurrence" applies to everything in the Universe.

JA: Please say "yes" or "no" or suggest an alternative. Jon Awbrey 20:06, 14 April 2006 (UTC)

IG: well, just "mental occurrence" does not specify whose mental occurrence. Berkeley would, for example, agree with that definition, and say the mental occurrences are God's. But the meta-solipsist would see the occurrences as their own. The definition I quoted for meta-solipsism is: "'I myself only exist'; the theory that no reality exists other than one's self." So I'd suggest the predicate would be "is in me." To capture everything in the net of the solipsist's universal me, perhaps the proposition of meta-solipsism would be (where "is in me" = M):

(x)(x=x -> Mx)

Ie, for all x, if x=x (as is the case for everything), then x is in me. However, I'm not so sure logic (which seems to be the route you're preparing to take) is going to set up the conditions to test the empirical falsifiability of meta-solipsism, which flows from it's own set of axioms wherein statements like (Ex)(x=x & ~Mx) will be false. Ian Goddard 02:05, 15 April 2006 (UTC)

JA: I'm just trying to get a clear enough definition of the concept, if there is one, to tell whether it's falsifiable or not. If the concept is not amenable to logical analysis then it's a complete non-starter as far as empirical science goes, and that is the only thing of interest as far as the present article goes. I found a more explicit definition in Runes, so I will see if that helps any. Jon Awbrey 02:32, 15 April 2006 (UTC)

IG: we could also define (x)(x=x -> Mx) as the set M such that M = {x|x=x}, which is of course the universal set U, and thus M = U. If you read the cited definition above (which is clear to me), that seems to be an explicit logical definition of meta-solipsism. The question that arises for me is whether it is a valid theory, I mean, it's just a rule, a definition. What predictions would it make?! And this harkens to its being nonfalsifiable, and thus pseudoscientific; which is to say it's on the nonscience side of the line of demarcation. And as you suggest, it may be "a complete non-starter as far as empirical science goes", which also comports with its being nonfalsifiable, which is exactly what I'm arguing. Ian Goddard 03:57, 15 April 2006 (UTC)

JA: It's a bit too late to think with any rigor, and it may be Monday before I can do more than dabble a bit, but right at the moment it seems to me like we are talking about a claim that two ostensibly different predicates on the universe or discourse are really indiscernibles. Yup, that's just what you wrote.

JA: Which cited definition is clear to you? The problem with the formulation "I myself only exist" is that it's not even a single proposition, as the pronoun "I" is an indexical sign, that is, a variable on a par with things like x. Thus it makes a different proposition for every one who utters it.

JA: As I've said, there are many more ways to scratch at the starting blocks than being indefeasible. Some strings of words don't even get that far, and I personaly suspect that this might be one of them.

JA: Other than this, the time has come to say more about the universe of discourse, one or many, and so on. But later. Jon Awbrey 04:40, 15 April 2006 (UTC)

Someone please tell me this discussion is not a figment of my metaphysically solipsist mind. The possibility that I think at all (at a minimum in what I imagine to be "perceptably" interactive, symbolic, perhaps arguably rational terms within myself) is, it seems to the metaphysically solipsist "me," only because I have been taught to think as a direct interactive result of relationships with my fellow human beings — this concept, while not fully explainable to either committed schizophrenics or fully committed metaphysical solipsists, appears to me (after the figments of my metaphysically solopsist mind known to me as C.S. Peirce and K. Popper, of course) to have more of a basis in probabiltity assessments and relationships with putative fellow human beings than it does in strict falsifiability. It is (within my own metaphysically solipsist mind of course) a very difficult argument to make to someone whose dentist has given less than adequate anesthesia while fully believing one is undergoing dental surgery, that one's dentist is a constuction of one's mind. To the extent there is arguable falsifiability in such a situation, it appears that the probabilities seem to shift fairly dramatically under the pain of a [presumed] intrusion by someone without [perceived] mutual consent. If there is falsifiability here, I'm afraid it's limited to modus tollens when one attempts to convince one's self that the [perceived] negligent dentist is the cause of one's pain. Suddenly, and somewhat magically, one all-at-once seems to prefer an "objective" third party to argue that it was the dentist's fault.. Of course, none of this is strictly falsifiable in the most stringent sense of the term "falsifiable"-- indeed why bother arguing the case, merely based upon one's perception that one thinks there are others to argue to via Wikipedia...Kenosis 07:43, 15 April 2006 (UTC)

Then again, perhaps MS is falsifiable by what is forbidden by MS. For instance, forbidden by MS might be that if one runs out into the middle of traffic on the nearest [putative] interstate that one has no actual traffic to which to be accountable. Of course, that may not be persuasive to the [possibly no longer existent] person who takes MS as a valid position...Kenosis 07:57, 15 April 2006 (UTC)

Right, the meta-solipsism (MS) believer could just say it's part of their dream that they got hit by a car. Others watching it could say it's part of their dream that they saw someone else get hit. Basically, it seems that under MS any observation can be interpreted as part of one's own dream, or universal mental program. One can, of course, have dreams where one does not have control of events, or where they go against one. And of course we usually assume, during a dream, that what's happening is out there. While there are differences between dreaming and waking states from which we could argue that the waking state corresponds to interaction with an external world, the solipsist could simply counter that there are two levels of dreaming (or of being in my universal mind), one I call "dreaming" and one I call "waking." I think, along the lines of the preexisting entry and the citation I just added for it, that solipsism could absorb any counter example, and as such is persistently nonfalsifiable and thus pseudoscientific.

Btw, Jon, your edits to the solipsism are really nice. Good job! Although I wonder if we need to list other varieties that are not being identified as nonfalsifiable. Certainly such a listing of its varieties ought to be included on the solipsism page, where I didn't see such articulation. Most (but not all) dictionary entries fail to make such distinctions, so wiki readers will be getting a better outline then they'll find in most resources. Ian Goddard 17:38, 15 April 2006 (UTC)

The reference to Descartes is quite apt I think, although back when Descartes wrote there weren't so many categories of the DSM that could readily be applied to those taking as gospel his similar line of reasoning in the Meditations. Falsifiable or not, MS, when taken as gospel, seems to me rather inflationary as to the influence of the self in cognative and practical matters, and radically depreciative of contributions by other [putative] human beings-- but I digress from the point of the article. Incidentally, this article has progressed immensely of late. Nice work everybody...Kenosis 20:09, 15 April 2006 (UTC)

Well, the reference to Descartes isn't mine. The Solipsism entry was written and posted in 2001 (here). The other day Noosphere hit it with a "citation needed" request. Maybe someone can find a cite for the Cartesian skepticism claim in the last sentence, otherwise, with its even lacking a supporting argument, in agreement with Noosphere, I'd say delete it. Ian Goddard 03:48, 16 April 2006 (UTC)

JA: For my part, I thought that we were still in discussion mode about these issues, but I have easter eggs to seek and miles to go before I sleep, so maybe not today, maybe not tomorrow, … Jon Awbrey 04:00, 16 April 2006 (UTC)

IG: Jon, you said: "the pronoun 'I' is an indexical sign, that is, a variable on a par with things like x." True, but we can define a predicate M, as I did, that describes the property of "is in me," or of "being in me." We can then define a set of elements with the property M that includes all my organs, each being a subset of me. I see no problems there. Now, the solipsist just has a far wider set-inclusion criterion for their own M such that: M = {x|x=x}. Then M contains everything. We can also define a set for the predicate U, "is in the universe," such that: U = {x|x=x}. So by the definitions, everything that's in M is in U and visa versa, thus (x)(Mx <-> Ux) and the two sets are indiscernible. That seems to be an accurate logical model of meta-solipsism, whatever the flaws of that philosophy may be. Btw, that supplementary page you made and inputs on others re solipsism are good work! Ian Goddard 16:40, 17 April 2006 (UTC)

Maths

I started to edit this section, with the aim of reducing its length. Also, the prosaic style sets it apart from the rest of the article. Then I decided that I could not quite follow the argument. On thinking about it, I find it curious that it discusses Peirce in detail, since Peirce predates Popper (even given his preemption of falsifiability). I think that what we have here is more a discussion of whether maths is empirical, rather than whether it is falsifiable. What do others think? Banno 09:38, 15 April 2006 (UTC)

JA: (1) I'm not the one who put mathematics in question with regard to falsifiability, but merely tried to accord the benefit of the doubt to whomever did, and tried to clarify the issues as best I could. Since it's the working assumption of this article that falsifiability is a necessary condition of empirical propositions, a discussion of maths as empirical implies a discussion of maths as falsifiable. Jon Awbrey 03:32, 16 April 2006 (UTC)

JA: (2) I realize that Popper has achieved some Popperlarity in some circles of late, but I hope that you have not bought his self-promotions to have "invented" falsifiability or "solved" the riddle of induction. Popper deserves credit for reviving the questions that Peirce treated under the "Logic of Science" after the Logical Atomizers reduced logic to a fine misterium of its former good scents, but the basic ideas here are as old as Aristotle, at least. Jon Awbrey 03:32, 16 April 2006 (UTC)

I have removed the following, pending further analysis of it by the editors...Kenosis 13:16, 15 April 2006 (UTC)

What is the conclusion then? Are mathematical theorems falsifiable or not? The most that can be said of them is that they are true of what they are true of, but what they are true of may not be the object of a given experience, and thus there can be things of which they are false.13:16, 15 April 2006 (UTC)

Thanks, Kenosis, for your edits to this section. The rhetorical questions are certainly not "encyclopedic" in style. Banno 17:20, 15 April 2006 (UTC)

I just removed the following, and am placing it here for further analysis and reconsideration as to possible partial re-use and placement somewhere in the Math section...Kenosis 18:24, 15 April 2006 (UTC)

As a conditional or hypothetical science, it is in the popular idiom "contingent all the way down". Its apparent certainty is but a relative certainty, relative to the axioms and definitions that are taken as the basic descriptions of one or another hypothetical universe. One can say that its theorems are tautologies, so long as one remembers the original meaning of tautology, which is a repetition of something previously asserted. Mathematical theorems merely say more acutely what the axioms more obtusely have already said.18:24, 15 April 2006 (UTC)

Statements and theories

The distinction between naive falsification and falsifiability is based on the distinction between falsifying a statement and falsifying a theory. This was not clear in the wording of the modus tollens section; hence my re-wording. Banno 18:17, 15 April 2006 (UTC)

Mathematics — Deleted Material

JA: I don't know where some of this stuff came from, but it's completely off-base. Jon Awbrey 03:12, 16 April 2006 (UTC)

Although they may appear on first encounter to be unexceptionally true, much consideration has been given to whether the theorems of logic and mathematics are falsifiable. Recall that it is not necessary to actually falsify or show a that an actual exception exists, only for it to be logically possible for a mathematical construction to be shown false. The multiple classes of use of mathematical propositions complicate the analysis of falsifiability. Because mathematics uses axiomatic methods derived deductively, it must be logically possible to falsify their structure deductively. And because mathematics participates in making assertions about the nature of the physical world (through what are termed non-logical axioms), it must also be accountable to the standards used for falsifiability of empirical propositions when used in such a way. In the case of a mathematical proposition, then: (1) it must be logically possible to show whether the proof is fallacious or invalid, and (2) when a mathematical proposition makes an assertion about some aspect of the natural world (for instance, E = mc2), it must be capable of being falsified by empirical observation.

JA: The hacking away of all logical transitions between points has left the section introduction incoherent, and the subsequent additions completely muff the points about the distinctions between hypothetical, normative, and positive sciences. For those who are worried about the magentic fields generated by pieces of text, the introduction of a pop sci icon like E = mc² not only confounds the proper relationship between mathematics and physics, but it guarantees a perpetual flux of 2¢ contributions from every pop sci tyro who ever accidentally watched an episode of Nova. Jon Awbrey 14:16, 16 April 2006 (UTC)

Charles Sanders Peirce defines a positive science as "an inquiry which seeks for positive knowledge", that is, for knowledge that can be expressed in a categorical proposition (Peirce, EP 2, 144). He goes on to say the following of the normative sciences, namely, logic, ethics, and aesthetics: Logic and the other normative sciences, although they ask, not what is but what ought to be, nevertheless are positive sciences since it is by asserting positive, categorical truth that they are able show that what they call good really is so; and the right reason, right effort, and right being of which they treat derive that character from positive categorical fact. (Peirce, EP 2, 144). However Peirce distinguishes mathematics proper from all positive sciences, and reckons it more fundamental than any of them, saying that any positive science "must, if it is to be properly grounded, be made to depend upon the Conditional or Hypothetical Science of Pure Mathematics, whose only aim is to discover not how things actually are, but how they might be supposed to be, if not in our universe, then in some other" (Peirce, EP 2, 144). For Peirce, logic is a science that seeks after knowledge of how we ought to conduct our reasoning in order to achieve its aims. As such, logical knowledge can easily fall short of perfection. Thus logical theorems, as normative claims, are falsifiable.

JA: Once again, the hacking away of all connective tissue most likely leaves the reader wondering what th'issue's all about. I clarified the tricky point about the difference between logical theorems and normative rules of procedure. Jon Awbrey 14:28, 16 April 2006 (UTC)

Pure mathematics contains no propositions that are not contingent on prior assumptions. Applied mathematics, in particular mathematics as applied in empirical science, is another thing. The application of mathematical abstractions to a domain of experiential phenomena involves a critical comparison of many different mathematical models, not all of them consistent with each other, and it normally leads to a judgment that some of the hypothetical models are better analogues or more likely icons than others of the empirical domain in question. How well a mathematical formula applies to the physical world is a physical question, and thus testable and falsifiable if structured in a way that allows a logical possibility of being shown false. For a very basic example, the proposition that all objects follow a parabolic path when thrown into the air is falsifiable. Indeed, it is readily falsified by the behavior of a feather. A more precise, also falsifiable, proposition is that all objects follow a parabolic path when thrown in a vacuum and acted upon by gravity.

JA: The mushed-together paragraph now begins with a topic sentence about pure mathematics. Then it takes up applied mathematics, which might be okay if the point is to make a contrast, but then it switches levels to a concrete example, which it's normally best for expository purposes to treat in a separate paragraph from the abstract considerations being illustrated. 14:44, 16 April 2006 (UTC)

Imre Lakatos discusses falsifiability in the context of the discipline of mathematics in his Proofs and Refutations. The question whether mathematical practice is a quasi-experimental science depends, according to Lakatos' line of reasoning, in part on whether proofs are fundamentally different from experiments. Lakatos argued that axioms, definitions, and proofs evolve through criticism and counterexample in a manner not unlike the way a scientific theory evolves in response to experiments. Thus, by Lakatos' assessment, mathematical constructions are falsifiable by those persons adequately familiar with the relevant methods of mathematical proof and disproof.

JA: I'm not Lakatos-intolerant, but do take it all with a grain of salt. I simply tried to relate these ideas to some precursors and postcursors of a like stripe, and place them at the proper level, which is cultural and pragmatic. Jon Awbrey 15:00, 16 April 2006 (UTC)

Mathematics Section

JA: Here is the mathematics section as last I remember it. I am not the one who placed logic and mathematics under the Damocletion sword of indefeasibility, but I merely tried to rescue the discussion from the near total misapprehension of logic and mathematics that I found when I read it. Please let me know if you have something of substance to criticize here, hopefully something more substantive than the usual drumbeat to dumb it down and dull it up. Jon Awbrey 03:02, 16 April 2006 (UTC)

JA: Updating to current version. Jon Awbrey 14:02, 16 April 2006 (UTC)

The question may be raised as to whether the theorems of logic and mathematics are falsifiable or not. After all, they appear on first encounter to be unexceptionally true. In considering this question it is helpful to introduce a classical distinction that is frequently emphasized in this connection by Charles Sanders Peirce. On the one hand, he defines a positive science as "an inquiry which seeks for positive knowledge", that is, for knowledge that can be expressed in a categorical proposition (Peirce, EP 2, 144). He goes on to say the following of the normative sciences, namely, logic, ethics, and aesthetics: Logic and the other normative sciences, although they ask, not what is but what ought to be, nevertheless are positive sciences since it is by asserting positive, categorical truth that they are able show that what they call good really is so; and the right reason, right effort, and right being of which they treat derive that character from positive categorical fact. (Peirce, EP 2, 144). On the other hand, Peirce distinguishes mathematics proper from all positive sciences, and reckons it more fundamental than any of them, saying that any positive science "must, if it is to be properly grounded, be made to depend upon the Conditional or Hypothetical Science of Pure Mathematics, whose only aim is to discover not how things actually are, but how they might be supposed to be, if not in our universe, then in some other" (Peirce, EP 2, 144). 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 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. Pure mathematics, on the contrary, contains no propositions that are not contingent on prior assumptions. Its apparent certainty is but a relative certainty, relative to the axioms and definitions that are taken as the basic descriptions of one or another hypothetical universe. One can say that its theorems are tautologies, so long as one remembers the original meaning of tautology, which is a repetition of something previously asserted. Mathematical theorems merely say more acutely what the axioms more obtusely already say. Applied mathematics, in particular, mathematics as applied in empirical science, is still another thing. The application of mathematical abstractions to a domain of experiential phenomena involves a critical comparison of many different mathematical models, not all of them consistent with each other, and it normally leads to a judgment that some of the hypothetical models are better analogues or more likely icons than others of the empirical domain in question. This is, of course, an extremely fallible business, and each judgment call is subject to revision as more empirical data comes in. How well a mathematical formula applies to the physical world is a physical question, and thus testable, within certain limits. For example, the proposition that all objects follow a parabolic path when thrown into the air is falsifiable, indeed, it is false. To see this, one has but to think of a feather. A slightly better proposition is that all objects follow a parabolic path when thrown in a vacuum and acted upon by gravity, which is itself falsified in regard to paths whose lengths are not negligible in proportion to a given planet's radius. What is the conclusion then? Are mathematical theorems falsifiable or not? The most that can be said of them is that they are true of what they are true of, but what they are true of may not be the object of a given experience, and thus there can be things of which they are false. The above discussion addressed the nature of mathematical theorems in and of themselves, and then took up their application to empirical phenomena. But the actual practice of mathematics involves yet another level of consideration, and it may yet involve activities that are very similar to empirical science. Many working mathematicians, from Peirce in his day to Stephen Wolfram in ours, have remarked on the active, observational, and even experimental character of mathematical work. Imre Lakatos brings the concept of falsifiability to bear on the discipline of mathematics in his Proofs and Refutations. The question of whether mathematical practice is a quasi-experimental science depends in part on whether proofs are fundamentally different from experiments. Lakatos argues that axioms, definitions, and proofs evolve through criticism and counterexample in a manner not unlike the way that a scientific theory evolves in response to experiments.

JA: I am curious, and not just rhetorically, as to why anyone would object to a few question marks in an article about an aspect of scientific inquiry. Those questions marks do not mark rhetorical questions, since their answers are not foregone conclusions. Some of the questions do serve a rhetorical purpose, in the way that every bit of text ought to do, in this case, recapitulating the points of the question in question before drawing a provisional conclusion, but that is just the best that can be expected with regard to many questions. Jon Awbrey 15:15, 16 April 2006 (UTC)

JA: Comment on the following paragraph:

What is the conclusion then? Are mathematical theorems falsifiable or not? The most that can be said of them is that they are true of what they are true of, but what they are true of may not be the object of a given experience, and thus there can be things of which they are false.

JA: Yes, this is a tough one. I will work on it. I was trying to gistize the preceding discussion in a single gisture. To discuss it adequately would require introducing a bit about the role of hypostatic abstraction in mathematics, and I was trying to put off that potentially off-putting business. Jon Awbrey 15:32, 16 April 2006 (UTC)

I'm unhappy about the assertion that "objects follow parabolic paths when thrown" is a theory - and a mathmatical one at that. It seems to me simply a hypothesis with specific scope conditions that has been extracted from the greater theory of gravity. Under this model, _Gravity_ is the theory that seeks falsification. Given the theory of gravity I might say that "objects under the following conditions should behave thusly". THis is then a specifically designed experiement with scope conditions to test the hypothesis (derived from my assumptions about gravity) and, indirectly, the theory of gravity. Plus its physics anyway. Andy - Somewhere, at some point.

Your Thematics & My Thematics

JA: Unless I missed something, all you did was replace the nominative pronoun "I" with the accusative pronoun "me". When you convert that to the letter "M", it remains indexical, and inasmuch as anybody can do the same thing, we have the predicates or properties M₁, M₂, M₃, …

JA: Here, you can be 1 and I'll be 2. Your thesis of metaphysical solipsism is expresed by the equation M₁ = U. My thesis of metaphysical solipsism is expressed by the equation M₂ = U. But none of this really matters. The real question was always about U. The question is whether the property U can serve any purpose when taken as an explanatory hypothesis to explain some other fact.

JA: Now U is unfalsifiable by definition. On your thesis of metaphysical solipsism, M₁ = U, in which case M₁ is also unfalsifiable. Does that make your thesis of metaphysical solipsism unfalsifiable? No, that thesis is the equation M₁ = U, not the property M₁. Two very different things. Jon Awbrey 17:12, 17 April 2006 (UTC)

But your list of properties M₁, M₂, M₃, … flows from our shared assumption that each M is a proper subset of U. So in our discourse of meta-solipsism we're speaking from one set of assumptions (where we are each proper subsets of U) about another set of assumptions where 'I' (the observer) contains everything. The meta-solipsist operates from that assumption and thus from a unique set of premises. We have to evaluate meta-solipsism by assuming its set of premises. So long as no logical contradiction can be derived from its set of premises (which define some of our premises as false), it can be considered a valid (if also unfalsifiable and thus pseudoscientific) theory. Ian Goddard 18:49, 17 April 2006 (UTC)

JA: No, I don't think that I, in the purity of my solipsism, have made any such assumptions. Maybe you've gotten sloppy about your solipsism, and slipped in some posits that you should have bracketed away, phenomenologically speaking, but my solipsism is shorn of any speculative presuppositions about bodies, minds, set theories, and highly suspect usuals like that. All we have is a bunch of symbols, whose intensional connotations are occasionally entertaining but wholly uncontaining of any more content than our respective solipsisms speculatively afford them. Jon Awbrey 19:48, 17 April 2006 (UTC)

Under the definition of meta-solipsism (x)[x=x ® (x Î M)] -- which simply says "Everything is in me (M)" -- speaking of your meta-solipsism (M₁) and my meta-solipsism (M₂) as distinguishable properties leads to contradiction. Proof:

Premises of differential meta-solipsisms

1. (x)[x=x ® (x Î M₁)]

2. (x)[x=x ® (x Î M₂)]

3. $x(x Î M₁) & $x(x Î M₂)

4. M₁ ¹ M₂

Now, by premises 3 and 4, there exists some x such that either x Î M₁ and x Ï M₂ OR x Ï M₁ and x Î M₂. But then we contradict either premise 1 or 2. Therefore, there cannot be different "meta-solipsisms." QED

So a listing of different meta-solipsisms (as suggested "M₁, M₂, M₃, ...") is inconsistent with meta-solipsism, premises of which we could express as (where U = the universe, and the capital X in premise 4 is a predicate variable):

Premises of meta-solipsism

1. (x)[x=x ® (x Î M)]

2. (x)[x=x ® (x Î U)]

3. M = U

4. (X)(x)[(x Î X) ® (x Î M)]

While that set of statements may be a valid logical theory (if no contradiction can be derived from it), it's just an empty definition routine, a name game -- solipsist: "I call the universe 'me.'" The solipsist's name game classifies all observations as "me" a priori, and thus by definition excludes the possibility of counter example, and so is nonfalsifiable (ie, is nonscientific). Ian Goddard 05:21, 23 April 2006 (UTC)

JA: Just off hand, as I've been away from this for a while, I think that you are confusing the predicates $M_{1},\ M_{2},\ M_{3},\ \ldots$ with the corresponding "Theses of Ontological Solipsism" $T_{1},\ T_{2},\ T_{3},\ \ldots$, where $T_{i}\iff M_{i}=U$.

JA: I'm not exactly sure, as I think that the introduction of elements is only clouding the issue of what predcates are discernible on phenomenological grounds, but you seem to be saying that if somebody were to adopt two different theses $T_{i},\ T_{j}$ for $i\neq j$, then of course it would follow that $T_{i}=U=T_{j}\!$. But a solipsist is not required to adopt some other solipsist's thesis of solipsism, so there is no common ground for the equation to go through. Jon Awbrey 12:40, 23 April 2006 (UTC)

Nice articulation Jon! My point is that when we speak of a set of "Theses of Ontological [ie, meta] Solipsism," we do so from a POV outside of them. In that POV the two theses are conjoined, just as I have conjoined them (see 1 and 2) in the first set of premises above. The contradiction simply shows that that POV is not the POV of meta-solipsism. Now I think we agree on that. So my remaining point is that when we evaluate meta-solipsism from within its POV -- which can list no other meta-solipsist POV (as in the second set of premises above) -- the problems you suggest of indexicality and 'I' being on a par with things like x do not arise just as no contradiction should derive from the second set of premises above. To the meta-solipsist, 'I' is a set containing everything; 'I' is necessarily not some limited member x of an set that also has other members outside 'I.' Meta-solipsism is not falsifiable when evaluated from within its POV, or set of premises as above. Ian Goddard 03:42, 24 April 2006 (UTC)

JA: Now that I don't see. I always felt that this whole order of discussion was beside the point of falsifiability. A proposition is falsifiable if it has a logical counterexample. Is it logically possible that U = M has a counterexample? Yes, it is. The end. This has never had anything to do with proof. It is a model-theoretic thing. Much less does it have to do with proofs that are grokkable by specific type of agent. Maybe I can't convince my parakeet of certain theorems in geometry. Their logical properties remain the same either way. Some solipsists may indeed see the logical possibility of a counterexample, but continue to go with their favored hypothesis. What they choose to believe has nothing to do with the logical property of falsifiability. Jon Awbrey 04:24, 24 April 2006 (UTC)

Not to interfere with your interesting discussion here, but since Jon anyway mentioned model-theory, I would say that the difference between epistemological solipsism (ES) and metaphysical solipsism (MS) is that any model for MS is a minimal model for ES: according to ES we can only know what is in our own mind, and that goes for all of us, according to MS "all of us" is "me". Or put differently: ES allows for minds $M_{1},M_{2},...$, whereas MS only allows for $M=M_{1}=M_{2}=...$, that is {$M$} = {$M_{1},M_{2},...$}. As for the falsifiability of MS, Jon says that a proposition is falsifiable if it has a logical counterexample. But is that sufficient? Note that with a "normal" worldview you can say that "all swans (that I know about) are white" - the observation of a black swan shouldn't tremble your entire world-view, only increase your knowledge. With MS a counterexample does not simply increase your knowledge, it shakes the whole foundation of your very self. How can there be something that you do not know, if all there is to know is you? MS is not simply the proposition that U = M, it is the proposition that U = M, that is that a counterexample is not possible ;-)

--FreezBee 15:54, 2 May 2006 (UTC)

Stephen Wolfram

To say Wolfram is a "working mathematician" is a bit of a joke. He is certainly not a crackpot, but his views on mathematics and physics are extraodinarily idiosyncratic and widely dismissed (or worse — ridiculed, held in contempt?) by the math and physics communities. I think we need a better example here.

Coming up with a good reference may not be easy. In my opinion, falsifiability criteria would seem to mostly applicable to mathematicians that describe themselves as "platonists" or "constructivists". The activity of mathematics, for these is one dealing with real objects, such as sets or constructible objects, which are just as real as electrons. In the case of constructivists, (a definite minority, though a very active and fertile one) I suppose one could come up with falsifiability criteria. Aside from these, however, mathematicians are generally not pure platonists or pure formalists or pure logicists They are some convex combination of these, so the whole "falsifiability" debate is very hard to formulate.

It would be nice if Edward Witten had some written opinions on the matter, although I seriously doubt it.--CSTAR 15:30, 27 April 2006 (UTC)

JA: Wolfram is mentioned simply as an example of someone who thinks that there is a non-trivial experimental factor involved in the actual doing of mathematics. He's hardly a voice in the wilderness on that score, and we already cited Peirce as one who devoted considerable thought to analyzing the observational aspects of mathematical reasoning. Jon Awbrey 15:38, 27 April 2006 (UTC)

{fact} = {citation needed}

JA: Cydebot is apparently changing {fact} tags to {citation needed} tags to bypass some kind of redirect. But since these tags are equivalent, and only non-robotic human laziness prevents us from typing out the long form, there is no need to change them back, so far as I can see, anyway. Jon Awbrey 03:56, 13 May 2006 (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)

"Contingent" is different from "Falsifiable"

In this article, we're acting like "contingent" and "falsifiable" mean the same thing. But they don't. A contingent statement is anything that's not a tautology (or a conclusion whose truth depends on whether the premises are true or false). So "grass is green or not green" is not a contingent statement (nor is it falsifiable).

However, "young children go through an Oedipal complex" *is* a contingent statement, even if it's not falsifiable. This statement is not true if young children don't go through an Oedipal complex, and therefore is contingent, even if there's no way to prove that young children don't go through an Oedipal complex. Thus it's possible for a statement to be contingent but not falsifiable.

I say these should have two separate articles. —Preceding unsigned comment added by 24.211.246.113 (talk) 19:08, 10 June 2006‎ (UTC)

Actually, it turns out "contingency" already is its own article, so I'm just going to delete it from this one. —Preceding unsigned comment added by 24.211.246.113 (talk) 20:01, 10 June 2006‎ (UTC)

The intro asserts the Popper POV

The intro says:

No empirical hypothesis, proposition, or theory can be considered scientific if it does not admit the possibility of a contrary case.

But this is the Karl Popper's POV, we can't assert the Popper's POV in the intro without specifying that it is controversial between scientist and philosophers.--Pokipsy76 18:16, 10 June 2006 (UTC)

Agreed. That needs changin'.

Why "many"? Explanation needed.

Someone have replaced

Some philosophers and scientists, most notably Karl Popper, have asserted that no empirical hypothesis, proposition, or theory can be considered scientific if it does not admit the possibility of a contrary case.

to

Many philosophers and scientists, ...

on what grounds should "many" be preferred to "some"?
"Some" don't give any numerical information. "Many" meakes people think that they are "a lot", but a lot compared to what? An explanation si needed, otherwise I think we should return to the more neutral "some". --Pokipsy76 07:15, 23 June 2006 (UTC)

Maybe the words "A number of..." works here. It sounds to me like we're close to an objective summary sentence there. ... Kenosis 13:33, 23 June 2006 (UTC)

Apart from Popper what other scientists and/or philosophers had this point of view? Why does the article speaks just of Popper and the followers?--Pokipsy76 13:45, 23 June 2006 (UTC)

There are at least a few ways of saying this accurately. "Karl Popper, and others who followed in his line of reasoning, advocated ..." seems like a reasonably accurate one too. ... Kenosis 17:48, 23 June 2006 (UTC)

That would be OK, I think.--Pokipsy76 18:16, 23 June 2006 (UTC)

Agreed. Somewhere in that approximate range of language is a reasonably accurate presentation for the reader. Thanks, ... Kenosis 18:54, 23 June 2006 (UTC)

Macroevolution/Darwinism

Some have taken this principle to an extreme to cast doubt on the scientific validity of many disciplines (such as macroevolution and physical cosmology).

One of those "some", it might be worth pointing out, was Popper himself. From Unended Quest:

"From this point of view the question of the scientific status of Darwinian theory—in the widest sense, the theory of trial and error-elimination—becomes an interesting one. I have come to the conclusion that Darwinism is not a testable scientific theory, but a metaphysical research programme—a possible framework for testable scientific theories."

I know that this probably just giving another stick to intellectually dishonest Creationists, but as an honest intellectual I thought this might be worth pointing out. Popper later elaborated that he thought many of the individual theories developed within the framework of evolutionary theory are testable, but the overall framework itself he did not think was. It is not a rejection of Darwinism. As he later elaborated in "Natural selection and the emergence of mind":

"The fact that the theory of natural selection is difficult to test has led some people, anti-Darwnists and even some great Darwinists, to claim that it is a tautology. ... Since the explanatory power of a tautology is obviously zero, something must be wrong here ... I mention this problem because I too belong among the culprits, influenced by what these authorities say. I have in the past described the theory as 'almost tautological' and I have tried to explain how the theory of natural selection could be untenable (as is a tautology) and yet of great scientific interest. My solution was that the doctrine of natural selection is a most successful metaphysical research programme. It raises detailed problems in may fields, and it tells us what we would accept of an acceptable solution of these problems. I still believe that natural selection works in this way as a research programme. Nevertheless, I have changed my mind about the testability and the logical status of the theory of natural selection; and I am glad to have an opportunity to make a recantation."

Just thought I'd post that here, in case it motivated anyone to make any small adjustments. --Fastfission 13:46, 5 September 2006 (UTC)

Doesn't this selection of the article violate the wikipedia:words to avoid page under the extremist section? It implies that people take this method of thinking to an extreme, compared to someone else's definition of "not extreme." This term should be removed. Rockymountains 02:24, 6 October 2006 (UTC)Rockymountains

Memetics

Memetics is falsifiable because one could prove that it does not follow deductively from its premises, which are pretty straightforward (humans evolve via natural selection, ideas transferring from human to human that impact natural selection with themselves be selected accordingly)

Furthermore, it could be falsified using specific, though quite intensive, experiments and data mining. Examples of such experiments can be found here: —Preceding unsigned comment added by 74.39.212.101 (talk) 12:20, 17 December 2007 (UTC)

http://cfpm.org/jom-emit/1998/vol2/lynch_a.html#HEADING18 —Preceding unsigned comment added by 74.39.212.101 (talk) 12:14, 17 December 2007 (UTC)

Cause and Effect

Is it true that the notion of cause and effect is not falsifiable? If one were to observe an effect with no apparent cause would one be able to conclude that cause and effect is false? Or would the general consensus be that the cause of the observed effect has simply not been discovered?

Is that ok to add this:

First known mention of falsification test in history comes from Koran in Surah Nisa chapter 4 verse 82 which any one in the world can try. “Do they not reflect upon the Qur'an? If it had been from (any) other than Allah, they would have found within it much contradiction." Basically it clamis that today if any one wants to prove that the Qur’an is not the word of God, he just has to point out a single contradiction in the Qur’an.

I reverted you edits, because they did not fit the tone of the article, and framed falsificationism in an inappropriate way. Also, I find it very difficult to believe the first time someone admitted they could be proven wrong was in the Qur'an. --John.Conway 09:47, 2 May 2007 (UTC)

Not all swans are birds

That a swan must be a bird (i.e. swan implies bird) implies that if it is not a bird, then it is not a swan. However, in some uses of this word "swan", it does not refer to birds at all, but rather:

“ 2. (noun) Slang term for a female who was once unattractive but somehow became very attractive (referencing the "Ugly Duckling" story). ”

Of course, this does not change the fact that science works, regardless of variable meanings of the word "swan".Kmarinas86 03:55, 24 November 2006 (UTC)

"Even so, the statement all swans are white is testable by being falsifiable" -- In the opening lines, it is implied that testable and falsifiable are interchangable terms. —Preceding unsigned comment added by 61.9.212.224 (talk) 02:21, 11 December 2007 (UTC)

Examples section is mostly original research

I've restored the "needs sources" header to the Examples section. It all sounds quite nice, but the various analyses and the applications to various fields lack references from reliable sources. "So-and-so assesses falsifiability in such-and-such field thusly.(cite)" would be good. Multiple, concurring so-and-sos and thuslys would be better. What this section has, though, is almost entirely original analysis - which really does need to go. - Kelly Ramsey 08:22, 11 February 2007 (UTC)

Economics

Many of the comments in the examples section are highly questionable. For example the comment on economics was clearly not written by an economist. Utility maximisation is not necessarily an assumption in economics or even in "rational expectations theory". A better example might be the new versions of the efficient market hypothesis that allow time varying expected returns and therefore prevent tests of the predictability of the difference between realised and equilibrium returns. —Preceding unsigned comment added by Derryp (talk • contribs) 21:30, 22 October 2007 (UTC)

Nitpick on physics section

The physics section of the article states that F=ma is sometimes taken to be a fundamental (yet falsifiable) law. As it turns out, it's already been falsified. Special relativity states that force is not actually equal to the mass of the object times its acceleration, but rather the derivative of its momentum with respect to time. Consider an object moving at 0.99999 times the speed of light. A force, no matter how large, will not accelerate it by much, but it will significantly alter its momentum. The only reason I didn't change the article is because I think it's valid, given the expected depth of the subject matter. Nevertheless, if there is someone who can elegantly incorporate special relativity's perspective into the article, I think it could be fruitful.--134.173.200.44 08:15, 15 February 2007 (UTC)

I believe this article considers F=ma as a definition. In that sense, "Force" is rather a naming of the quantity "Mass X Acceleration". And well, big mass times small acceleration could still be big, and that's the change in momentum.DyC Ei 23:14, 1 July 2007 (UTC)

Deleting the Subsection I'm extremely puzzled by the whole physics section here, A section about falsifiability in physics would be good, but after I read this section, I don't think I've learned anything. If F=ma is indeed a definition, as the article claims, then it's not really relevant to the falsifiability discussion at all; you can't falsify a definition, since it's always true by definition; at most you can show that it's not useful, and that some other definition is more useful. This has in fact happened with F=ma, if you consider it not to be a law of physics but a definition of a "force"; it was useful in Newtonian physics, but in relativistic physics F=dp/dt is more useful. F=ma is more generally, though, a statement of conservation of momentum, which very explicitly is testable. Except for this statement (which is cited only as "occasionally it is suggested", an example of WP:Weasel words), the rest of the section can be summarized as "in physics, falsifiability is good".

As a net result, I don't think this subsection says anything at all. So I'm deleting it as content-free. Geoffrey.landis (talk) 15:09, 7 December 2007 (UTC)

Forbid?

from the first bullets: "Falsifiable propositions must forbid an observation." How do falsifiable propositions forbid observations? I thought the very nature of such propositions was to admit of and rely upon observations, and conceivable counterinstances? Somaticvibe 19:04, 24 March 2007 (UTC)

Wikirot. Another article all gone to hell. Banno 19:35, 24 March 2007 (UTC)

An observation that would falsify the universal statement would have to be "forbidden" by it. That is to say, that if some observational requirement be met, the universal statement cannot be true and would therefore be rejected or falsified. What makes a universal statement falsifable, such as the statement that "All muslims must exist in Israel" would be the fact that they cannot be true if an observation is made of a situation that would be impossible if the statement were true (e.g. muslims are observed in Iraq). If a statement does not "forbid" an observation, it is not testable by observation simply because in that case there could not be an observation that would show it to be false. "Forbid" does not mean "prove impossible".Kmarinas86 05:00, 26 March 2007 (UTC)

To elaborate further, one of the reasons why scientists make any attempts to observe is that certain procedural methods of observation are believed to have the capacity to generate a result that is for or against a hypothesis. The observation is for the hypothesis (A: "Jumping off a cliff must cause me to fall even further below the level of that cliff") if it matches its expected (or deduced) result (e.g. I still fell as my head was below the altitude of the platform I jumped from). Falsification requires that an observation be forbidden (B: "It is impossible for one to observe me to jump off from one cliff and land on another one that was 5 feet away and 1 foot higher. One cannot see me do it."). Contradicting B would in turn contradict A. Modus tollens.Kmarinas86 05:19, 26 March 2007 (UTC)

Removed content

Some have used this principle to cast doubt on the scientific validity of many disciplines (such as macroevolution and physical cosmology), although some counter that macroevolution is falsifiable, claiming that it could not have occured in the past if tommorow, an animal gave birth to a mythological chimera.

What is in here that is implying that macroevolution isn't falsifiable? It could very well be, provided that it forbid an observation (such as an animal giving birth to a mythological chimera), and given if the observation occured, the nature of macroevolution as it pertains to the origin of higher order taxa would be beyond doubted.Kmarinas86 01:00, 17 April 2007 (UTC)

It's just too weasel-wordish, and the chimera example is pure original research. it gives the impression also that the argument that macroevolution is not falsifiable is even heard in the sciences. that's undue weight. several issues here. that's why I removed it.--Urthogie 01:41, 17 April 2007 (UTC)

Featured Article Candidate?

Ummm? How about no?Kmarinas86 08:11, 29 April 2007 (UTC)

"Little Green Men"

I am a confused of the example given in the article. It is obvious that if "the Earth has not been visited by little green men" is a scientific theory, so is the opposite "...has been visited by...". So how could one be falsifiable and the other couldn't?

DyC Ei 22:52, 19 June 2007 (UTC)

To falsify that something has not visited the earth requires you to make a single observation in a small portion of the earth. To falsify that something has visited the earth requires you to make an infinite number of observations in all portions of the earth. The latter is an impossibility for us.◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 02:07, 28 June 2007 (UTC)

I do not attempt to reject arguments on their falsifiability in that sense. However, I also do not think that they are not both scientific. To clarify my point: I can not trust Karl Popper & co. Or has it got to do with the word "empirical"? I would be pleased to know more (well, for the article) BTW, strictly "the earth has been visited by LGM" can still be falsifiable, depending how it is considered whether an observation could be scientifically interpreted. I think. For example, if the planet spawning little green men is seen not to contain a spaceship, *and if we can say it implies there shouldn't have been any before, possibly learned from other stuff*, the theory is falsified. How this is completely rejected would be interesting... I am afraid to assert myself, after all. But sadly, it is off-topic.DyC Ei 23:34, 1 July 2007 (UTC)

Okay, I've now read "Necessary?". Ignore the above to your own convenience. Sorry. DyC Ei

The statement "the Earth has not been visited by little green men from planet X" may be falsifiable if you can show that the LGM from Planet X are not capable of space flight. But all that means is that the LGM who did visit the Earth did not come from planet X. They must have come from another planet. Banno 23:05, 11 July 2007 (UTC)

The text is an introductory one. Can you think of another example? Banno 23:05, 11 July 2007 (UTC)

Is the statement "Human settlements on Earth have not been visited by little green men between 1900 and 1999" falsifiable?

No. The reason is that anything that happens after the 1990's has no bearing on the falsity of that statement. Implication always occur from causes to effects, not the other way around. It's no longer the 1990's. Therefore, the only way this could be known is by faith, regardless of any proofs that existed and were demonstrated at the time.◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 18:46, 20 July 2007 (UTC)

So is it generally the case that history is not falsifiable? --Damian Yerrick (talk | stalk) 22:28, 22 August 2007 (UTC)

Yes.◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 22:50, 22 August 2007 (UTC)

Please actually read Popper's works before you present his ideas in such a severely distorted manner. I suggest Popper Selections, which should be a minimum if you want to edit such articles. Forget the nonsense that you find on popular web pages. It's simple as that: To state that some idea is not falsifiable, even if correct, is not in any way an argument to reject a claim; it is not a valid criticism, and can be understood so only from a justificationist perspective. Falsifiability has no epistemological significance at all. --rtc 23:45, 22 August 2007 (UTC)

Rational scepticism project

I'm interested that it has been given a "start" rating. Why? What does it need? Banno 23:19, 11 July 2007 (UTC)

Theism

The 1st paragraph is very badly argued

If God is conceived of as an unobservable transcendental being, then one could not disprove his existence by observation. The assertion 'God exists' would be unfalsifiable because of the nature of God. On the other hand, the assertion 'God does not exist' is falsifiable. This assertion can be falsifiable by demonstrating the existence of God.

If God is not directly observable, but influences the course of the history, "God exists" would be falsifiable. A classic attempt at falsification is the (negative) theodician argument -- that if God existed the world would be a better place.

The last two sentences simply contradict the first. 1Z 14:51, 17 July 2007 (UTC)

Can research ethics preclude falsifiability?

If an experiment to falsify a hypothesis might violate the Nuremberg Code or other recognized research ethics, does this make the hypothesis unscientific on account of unfalsifiability? --Damian Yerrick (talk | stalk) 16:36, 20 July 2007 (UTC)

No, it just makes it rather difficult to do so. 130.216.191.182 10:09, 26 July 2007 (UTC)

Removed material from the lead

I've removed Rtc's just-added material from the lead section of the article. Only problem I have with it is that Popper's response to what he percieved as misunderstandings of what he originally meant do not belong in the article lead. I think another appropriate place can be found in this article for Popper's counter-counterarguments. The just-removed material I'm referring to is as follows:

Popper warned that his demarcation criterion had been misunderstood completely. Falsifiability is not a criterion of rational acceptability, nor of scientific recognition, scientific authority or meaningfulness. (Such a view was held for a short time by Hans Albert, but abandoned again.) It must not be confused with the criterion of ‚reinforced dogmatism‘ that Popper uses to demarcate Pseudoscience and Pseudorationality. With William W. Bartley's addition of pancritical rationalism, the criterion of falsifiability has become rather unimportant in critical rationalism and is significant only from a historical perspective.

To avoid the confusion that occurs when falsifiability is understood as a demarcation criterion for science, David Miller, the closest student and former assistant of Popper, now holds that "[i]f falsifiability is to provide [...] a criterion of demarcation in the strict sense, it should rather be between the empirical and the non-empirical." ... 04:19, 22 August 2007 (UTC)

In my opinion, this material is well sourced and has excellent potential somewhere else in the article, after the basics of "falsifiability" have been presented and after the response of other notable sources has been introduced in the article. ... Kenosis 04:19, 22 August 2007 (UTC)

sourcing and OR

For example, "all men are mortal" is unfalsifiable, since no amount of observation could ever demonstrate its falsehood. "All men are immortal," by contrast, is falsifiable, by the presentation of just one dead man. However, the unfalsifiable "all men are mortal" can be the logical consequence of a falsifiable theory, such as "all men die before they reach the age of 150 years". Thus, unfalsifiable statements can almost always be put into a falsifiable framework. The falsifiable does not exclude the unfalsifiable, it embraces and exceeds it.

The thing that bothers me about this is that in science the hypothesis 'all men are mortal' seems to be falsifiable. At least the way experiments are done. We do not have to prove this to an absolute certainty. Now we have not boiled every ounce of water on the earth to prove that it boils at 100 C. So if I say 'all water boils at 100C' that is not falsifiable? What we do is boil enough, enough times to conclude that that all water has the same characteristics.

http://en.wikipedia.org/w/index.php?title=Falsifiability&diff=171235969&oldid=170176711

and this was a good explanation. it really made the point about physical laws. My point is are we being to strict about the OR rule. A good explanation that agrees with sources but stated in a different way should be OK. Massachew 17:44, 13 November 2007 (UTC)

If you find a drop of water that doesn't boil at 373 K (100 C) and 101 kPa, then you have falsified "all water boils at 100 C and standard pressure". To falsify "all men will eventually die of aging-related illness", you'd need to find an immortal man and demonstrate that he is immortal. How would you go about that? --Damian Yerrick (serious | business) 17:57, 13 November 2007 (UTC)

My point is that if we have evidence that millions of men have died it is reasonable to conclude that all men die. Most of our science is done this way. I guess I am saying falsifying is overrated. Like saying no humans have 18 chromozones would be an unscientific statement according to Popper. Massachew 01:21, 14 November 2007 (UTC)

A falsifiable statement is of the form "There exists no X that has property P, and here is how to measure property P." For example, "there exists no hominid that has 18 chromosome pairs" could be falsified by finding a hominid, taking a karyotype, and finding 18 chromosome pairs. The statement "all men are mortal", on the other hand, is of the form "there exist no man who will not die of old age at some point in the future". It is impossible to see a man's future. --Damian Yerrick (serious | business) 04:29, 14 November 2007 (UTC)

Falsifiability of Evolution

I think the specific falsifiability examples relating to evolution are directly contradicted by the article itself, as they are (what the article calls) "uncircumscribed existential statement, such as there exists a green swan".

E.g.; Dawkins says "If there were a single hippo or rabbit in the Precambrian, that would completely blow evolution out of the water. None have ever been found." True, but the trouble with that is to test that would require we search the entirety of the Precambrian (just as we would need to search the world to disprove the green swan statement). The lack of hippos or rabbits in the Precambrian is supporting evidence, but not an adequately falsifiable prediction of evolution. The key issue here is that evolution must make a specific prediction that "you will see something if you look here". Negative assertion of validity "can't find evidence to the contrary so it must be true" is a classic folly (and it's kind of mistake that Dawkins would make, which Sagan would not have)

I would suggest dumping these arguments for some actual falsifiable predictions of evolution. There are plenty of such strong testable predictions; here's a few:

Evolution predicts a (geologically) recent common ancestor of apes and humans. Molecular biology identifies DNA as the mechanism for inherited traits. Therefore if common decent is true, our DNA should be more similar to great apes than other mammals. If it this is not the case, then common decent is falsified. DNA analysis shows this to be the case however (and remarkably so), and evolution passes a falsifiable test.

A more specific example of the above human-ape DNA ancestry is evolution's predictions regarding the differences in human and other great ape chromosomes. Humans have 23 pairs and the other apes have 24. Evolution makes a prediction that the missing chromosome should exist somewhere in the human genome, if it is not there (a reasonable sized search space compared to all the Precambrian), then common decent must be modified or abandoned. This was found on Chromosome 2 and is widely considered strong evidence of common decent.

Other examples include the prediction of transitional fossils, evolution predicts that if you look at rocks of the right age in the right places you should find Transitional fossils, the lack of transitional forms would serve to disprove evolution (note that this is the opposite of Dawkins' statements which is that you should need to *find* hippos to prove evolution false, here you need to fail to find something to disprove).

I suggest we change the evolution examples to at least one, if not two of these above (or other similar) examples. Does anyone object or have comments?

--Robbins 05:36, 4 December 2007 (UTC)

I like the idea of some additional examples, but I'm not sure I like the examples you propose. Your argument against the "Rabbit in the Precambrian" example is faulty. Finding a Precambrian fossil of a modern rabbit most certainly would falsify evolution, your assertions nonwithstanding. That would be a classic example of falsification. The DNA similarity between great apes and humans is a good example (and I think it should be added), but I find the chromosome-24/chromosome-2 argument extremely unpersuasive; it's not at all obvious how the prediction is derived from the Darwinian theory of evolution. Finally, a lack of transitional fossils would unfortunately say nothing about falsifiability. The theory of evolution predicts that transitional forms exist, but it would only predict that transitional fossils existed if there were some reason to make the statement that all animals are fossilized on death. But of course that's not true-- fossilization is actually very rare, and even so, strata are abraded and eroded away. Finding a fossil definitely tells you something was there, but not finding a fossil doesn't definitively tell you something wasn't there. Geoffrey.landis (talk) 23:46, 13 December 2007 (UTC)

What if an alien moves the rabbit fossil from one geologic strata to another? Naive falsification isn't so great. You can ALWAYS add an ad hoc hypothesis so that a hypothesis is still supported. History of nature or of anything else is INHERENTLY unfalsifiable, and it will always be so, without the addition Kmarinas86 (talk) 23:57, 17 December 2007 (UTC)

What an unholy synthesis of a lack of knowledge about Popper's work with fashionable positivistic ideas and the prevalent misreading (rather nonreading) of Popper. "The lack of hippos or rabbits in the Precambrian is supporting evidence". Are you aware that Popper explicitly denied the very existence of evidence and scientific data, and the possibility of support? Are you aware of the degree that Popper rejected the fashionable philosophy of science? Are you aware that Popper wrote more than a few articles about evolution himself, that he rejected naturalistic evolution and accepted much of the criticism of creationists? Did you know that he developed his own theory of evolution, the such-called spearhead theory of evolution, and that it includes an attempt to solve the problem of macroevolution? --rtc (talk) 15:33, 21 December 2007 (UTC)

Naïve falsification

While naïve falsification is a view attributed mistakenly to Popper in his early years by some authors, he never actually held this view. In appendix *XIV of the Logic of Scientific Discovery, which is still missing in the English version, he literally calls claims that this was once his view complete nonsense. The article is reluctant to say this. I further suggest a rename of the article to Falsificationism, which is the Poppperian philosophy applied to science, which includes not only Falsifiability (the criterion) but also Falsification (the methodology). These two belong together. --rtc (talk) 02:45, 23 December 2007 (UTC)