Testing if an hypothesis is falsifiable is a basic filter in skeptic reasoning, yet appears to be widely misunderstood. The concept that something has to be falsifiable in order to be validated may indeed come across as paradoxical at first.
Personally I do not like the term “falsify” in this context, because it is too easily misunderstood as to mean “fraudulently alter” and then gets confused with data or evidence tampering. I therefore refer to “potentially refutable” here. Karl Popper (see below), who developed the concept, also often used the term refutable, so I regard it as appropriate.
Even if observations confirm an hypothesis a thousand times, it will still not prove such an hypothesis to always hold true. Conversely, it would require only one contradictory observation to discredit the hypothesis. The test for universal validity does therefore not rely on positive confirmation, but on potential refutation. If an hypothesis is phrased such that there is no means by which it can be refuted, it becomes impossible to test its validity. Another phrasing used by Popper specifies that it must be “capable of conflicting with possible, or conceivable, observations”. Failing this, an hypothesis is unscientific.
Salient questions to ask would be:
• How would one go about refuting the hypothesis?
• What evidence would be required to refute the hypothesis?
If the answer to these questions is “impossible”, the hypothesis is unscientific.
Popper argues that unscientific theories are not necessarily worthless in that some of them have been shown to evolve into valuable theories in the past. However, it is clear that they are of little worth.
An extract from Conjectures & Refutations by Karl Popper on this topic is available here. I recommend reading the whole extract, but here’s a bit:
It is easy to obtain confirmations, or verifications, for nearly every theory — if we look for confirmations.
Confirmations should count only if they are the result of risky predictions; that is to say, if, unenlightened by the theory in question, we should have expected an event which was incompatible with the theory — an event which would have refuted the theory.
Every “good” scientific theory is a prohibition: it forbids certain things to happen. The more a theory forbids, the better it is.
A theory which is not refutable by any conceivable event is nonscientific. Irrefutability is not a virtue of a theory (as people often think) but a vice.
Every genuine test of a theory is an attempt to falsify it, or to refute it. Testability is falsifiability; but there are degrees of testability: some theories are more testable, more exposed to refutation, than others; they take, as it were, greater risks.
Confirming evidence should not count except when it is the result of a genuine test of the theory; and this means that it can be presented as a serious but unsuccessful attempt to falsify the theory. (I now speak in such cases of “corroborating evidence.”)
Some genuinely testable theories, when found to be false, are still upheld by their admirers — for example by introducing ad hoc some auxiliary assumption, or by reinterpreting the theory ad hoc in such a way that it escapes refutation. Such a procedure is always possible, but it rescues the theory from refutation only at the price of destroying, or at least lowering, its scientific status. (I later described such a rescuing operation as a “conventionalist twist” or a “conventionalist stratagem.”)
One can sum up all this by saying that the criterion of the scientific status of a theory is its falsifiability, or refutability, or testability.
This is interesting, not least because by Popper’s definition of ‘scientific’ quite a few scientists are indulging in non-scientific work. I refer to the likes of Stephen Hawking and Roger Penrose who devote a lot of their time to hypothesising about what happens inside black holes, which hypotheses are impossible to either verify or refute.
And what of Russel’s famous teapot, which while theoretically possible to refute, is unlikely ever to be so refuted because of technological restraints?
Surely there is nothing wrong with generating untested hypotheses, as long as they are portrayed as such. As a matter of fact, generating such hypotheses is a core element of the scientific process.
And what of Russel's famous teapot, which while theoretically possible to refute, is unlikely ever to be so refuted because of technological restraints?
Being potentially refutable is one prerequisite for validity, not proof of validity.
I agree–nothing wrong at all, but still unscientific according to Popper.
I wouldn’t dream of suggesting that Russel’s teapot hypothesis was actually true. Validity is neither here nor there: it applies to arguments, not statements or hypotheses.
OK, being potentially refutable is one prerequisite for a scientific hypothesis, not proof that an hypothesis/claim/statement is necessarily scientific in nature.
One needs to be very careful here. Many real-world phenomena that are the business of science are not directly observable and can only be assessed as a result of certain consequences they entail. So, for example, subatomic particles can presently only be “viewed” by means of the trails they make in cloud or bubble chambers, or the scintillations they prompt in a detector. What Hawking et al. are doing is to originate and develop mathematical models and techniques in the hope of constructing an accurate model that reveals or facilitates just such observable consequences – as Hermes indicated, they’re generating untested hypotheses. But it’s more than just that. The models they build can also potentially indicate what else to look for. Hawking Radiation is one of them; being able to distinguish rotating from non-rotating black holes is another, as are the Penrose-Hawking Singularity Theorems. The last of these, when incorporated into a suitable mathematical model of the “interior” of a black hole (i.e. the dark side of the hole’s event horizon), could conceivably produce physical effects via gravity perturbations or gravitons or some other as-yet undiscovered means, which physical effects may be properly observable and so potentially support or refute the model. The principle to bear in mind here is that the cutting edge of science is always speculative. Testability only comes later. General Relativity, for example, went through much the same process, and it wasn’t even Einstein himself who suggested the experiments that were undertaken to test it.
As an aside, up until the most recent godbot fracas, I’ve never had an issue with the term “falsifiable.” It never occurred to me that one could easily confuse or conflate “falsify” with “forge” or “fabricate.”
Exactly. the problem with black holes is that–even in theory–there can be [em]no[/em]observable consequences of anything that takes place [em]inside[/em]the event horizon of the black hole; all the hypothesising is conjencture based on physics [em]this[/em] side of the event horizon. Therefore, according to Popper, not science.
st0nes, I’m not sure you properly followed what I wrote. There’s the potential that processes inside a black hole have observable repercussions outside it. In the absence of compensating mass-energy being sucked into a black hole, the emission of Hawking Radiation means that the hole’s event horizon is shrinking and the spacetime curvature inside is decreasing. That is knowledge about events inside the black hole. Conversely, the event horizon and spacetime curvature are increasing in the case of a black hole that is sucking in mass-energy at a greater rate than it is emitting Hawking Radiation, again knowledge of internal occurrences. As I indicated, more inferences about internal action are conceivable from subtler interactions or effects such as gravity or thermodynamics.
No, and I apologise. I was busy cooking dinner and de-worming the wife, so I didn’t pay your post the attention it deserved. That said, first let me point out that I am neither a physicist nor a philosopher, and will gladly bow to superior knowledge. I am confused by the distinction between knowledge about a black hole as a whole (mass, spin, etc) and knowledge of events that may occur inside it. In my amateurism I may have misunderstood, but I thought it was more or less accepted that it was impossible in theory for information about any event occurring within a black hole to emerge therefrom. We may gain knowledge of the attributes of a black hole through the mechanisms you postulate, but not of events within it.
Yes, that’s our best current understanding in terms of models derived from GR. As you will note, I have underlined the key issue. Hawking et al. are in a sense testing the existing body of theory with their efforts towards improved models, and thus they are doing science. I’ve mentioned Hawking Radiation before. Few cosmologists doubt it, and it is thought to have been observed in a few instances, hard as it is to identify. Before Hawking combined a known QM effect with GR, theory forbade that anything at all should escape from a black hole. If his model of Hawking Radiation is correct, it means that the energy for the radiation came from inside the black hole, and so we have good reason to suppose that there’s at least one mechanism that permits something to escape from a black hole. It is still the case that in terms of present theory, information is irrecoverably destroyed when it crosses a black hole’s event horizon. The point, though, is that as compelling as current theory might be, this feature is not cast in stone. Analogous to Hawking Radiation, there may be an as-yet undiscovered combination of QM and GR effects that allow information to leak from a black hole in certain circumstances, and this forms part of what Hawking et al. are probing. Unless an overarching Grand Unified Theory that subsumes both QM and GR is arrived at and explicitly forbids the sought-for effects, the speculative work Hawking et al. are doing towards a deeper understanding of nature cannot be ruled out or dismissed as unscientific for the straightforward reason that it may point towards hitherto unsuspected observable consequences after all. In short, it’s not entirely certain that the hypotheses and models are not testable.
Thank you for the explanation. It seems I have allowed my knowledge to become out of date. Could you recommend an accessible text that incorporates the latest thinking on the subject?
I’m not aware of any up-to-date non-specialist reading as such, but Amir Aczel’s God’s Equation, Michio Kaku’s Beyond Einstein and Kip Thorne’s Black Holes and Time Warps, although about a decade old, may be helpful in giving some essential grounding and locating other suitable sources. Roger Penrose’s The Road to Reality is probably unsuitable for being highly technical and making extensive use of mathematics.
Thanks. I have The Road to Reality (I’m not an arithmophobe); I’ll keep a lookout for the others. It’s infuriating, though, that the science sections in our bookshops are invariable so poorly populated, and even then they contain volumes on astrology and rubbish like ‘quantum healing’. I’ll probably have to order from overseas, pay a fortune and wait 6 weeks…
Yes, the dark side of give-the-people-what-they-want-never-mind-what-they-need free market enterprise rears its hydrocephalic head. Who needs reality when a comforting fantasy can make you feel so-o-o-o much more special!? It’s just like religion – with a less formal clergy and less regimented funds collection strategy.
