The results of a survey of university scientists are surprising and odd.
Surprisingly, 87% of scientists think there is a scientific method that describes the way scientists do their work. Most of them believe in the old hypothesis → testing –→ theory view that hasn’t been popular among experts for many decades.
Almost half (49%) of natural scientists and 29% of social scientists thought that science was independent of social and cultural biases.
Almost half (48%) of all scientists believed that a theory becomes a law when it is proven.
You all know that none of those are correct, right? There is no single scientific method, of course science is shaped by culture, and nope, theories and laws are different beasts. I’m used to civilians getting these wrong, but that’s a big surprise that so many scientists don’t understand the nature of their work.
Umm, what’s wrong with the hypothesis → testing –→ theory view?
I would love it if all science majors had to take a Phil Science course. I’m teaching one this spring on the demarcation problem (that is, how we can demarcate “science” from “pseudoscience,” much less “bad science” and “non-science”), and I have some hard sciences majors signed up for it. I’m very excited.
Who’da thunk it? Philosophy ain’t so popular with scientists! Eh?
As near as I can tell, the notion that there are multiple scientific methods rests on the assumption that a different experimental technique, such as a cross-sectional versus longitudinal survey are conceptually distinct. I’m not so sure that those who believe “a” scientific method aren’t simply thinking more abstractly…which is to say, they aren’t in error by most standard thinking, which is precisely why the idea is so widespread. I suggest that the real problem is the notion that “science” is defined by its method(s,) rather than referring to the body of knowledge about the real world, collectively tested so that we are confident it’s true (even if we are nonetheless still going to have to correct errors.)
And holding to the hypothesis–>test–>theory view is much more sensible than Popperian falsificationism in my opinion. And the problems with Kuhnian paradigms ignoring correspondence to fact are kind of notorious. So as far as I can tell, if you’re going to go down the wrong road defining science as a method, they’re taking the least damaging detour.
As for proven theories being dubbed laws, this is a semantic problem. Since most people will insist on using “theory” to mean “opinion,” this kind of effort to forestall confusion by a simplistic definition of law is understandable enough. Plus, since so many people deny any status as knowledge to any empirical evidence, reserving it for a priori mathematical/logical proofs, trying to invest something to with mundane experience with the status of a law is also understandable.
In other words, I’m not so sure that the experts are doing so much better.
As to fancies about the immunity to social and cultural biases? It is quite difficult to prove bias in others. Believing it about yourself is very difficult. Even more to the point I think, it is the academy’s job to tell the world what’s right, not to doubt itself. And the more they get paid, the more important it is to be the expert.
#1 Not all science can rely on this method, this is rather obvious when you think about it. For example , the social sciences have much more elaborate (but necessary) approaches to finding “truths” about, say psychology, here the facts are derived from a combination of data-analysis, quantitative (ie mass-surveys of patients) and qualitative research, such as analysis of in-depth interviews with patients.
@#5 Still sounds like testing a hypothesis to me.
Suddenly, the confusion among the general public about the definitions of “hypothesis”, “theory”, and “law” is much more understandable.
Just wow. I got nothing.
Up next: 37% of mathematicians believe an axiom is used to split wood!
If you’ve never had to defend the theory of evolution against the claim that it hasn’t been observed, i.e. “tested”, by repeated experiment, the obvious truth is that the sciences makes use of both inductive and deductive reasoning.
“The only rules of scientific method are honest observations and accurate logic.”
-Robert MacArthur
#4
No. Theories and Laws are apples and oranges. Foe example, Ohm’s Law E=IR. This is an equation that has been so thoroughly tested that it can be accepted as proved, but it tells you nothing about what electricity is or why the equation is true. For that you need a theory of how electrons interact with each other and the medium they are moving through.
@ stevenjohnson2:
I think it’s more than a semantic problem. Theory and law are two very specific words.
For instance, Thunderf00t’s law tells us that mentioning feminism in public will attract a specific type of commenter, i.e. gives us the what. It’s the Theory of Asshats that tells us why, and can possibly be refuted by the Theory of #NotAllAsshats, given sufficient evidence.
A better example would probably be that you that Newton’s Law of Gravity will describe the vertical position of a thrown object (under certain conditions), but it sure as hell won’t tell you why gravity does that.
Take it back, I like the Theory of Asshats.
Not to me it isn’t. It’s just a fairly straightforward instance of Sturgeon’s Law. But even getting 10% of something good is IMO still quite an achievement for a bunch of domesticated primates clinging to a rock hurtling through space.
Ninja’ed by moarscienceplz. SUMMON THE ASSHATS!
I read once that the Theory/Law distinction is really this:
A law is an observed mathematical relationship between measurable quantities with no underlying explanation of why that is. Think Ohm’s Law, Newton’s Law of Gravitation, Hooke’s Law for springs, etc.
A Theory includes justification. General Relativity explains gravitation through the curvature of space-time by mass/energy, The theory of evolution explains the origin of species through natural selection, etc.
It may just be an accident of history, that earlier scientists like to label things “laws” but it fits the usage better than the idea that a theory is in some sense more provisional than a law, and that with more evidence a theory can “graduate” to a law. That does not match usage at all.
@1
You depend on theory to get rid of a bad (untestable) hypothesis. i.e. there are nargles at the bottom of this lake.
So the method(s) kind of eats itself.
moarscienceplz@10 and Patrick G@11
Let me rephrase…since “theory” is popularly assumed to be “opinion,” then “law” can be co-opted to mean “fact.” And you really cannot blame scientists for wanting to have a word to convey that something other than a measurement is a fact.
David Wilford@8
Strictly speaking, both deductive and inductive propositions can be tested. What you’re dealing with here is the Popperian notion that propositions can only be falsified by controlled experiments…which is another popular but alternative definition of “the” scientific method. Your implied criticism of that is correct, but your target is wrong I think.
#16
The subject was scientists understanding the terms theory and law in a scientific context. Just because some schmo on the street is unfamiliar with that context has no bearing on the question at hand.
I think the traditional scientific method (as emphasized in school science fairs) is a reasonable starting explanation of why we can trust scientific results. Even there, it is probably incomplete, and says nothing about consensus, which is a lot more subjective, and explains why certain theories gain prominence (often with good reason) despite nothing being quite as definitive as the naive view would have it.
The main thing missing in the textbook scientific method is science as a creative process. Hypothesis generation is rarely carried out very methodically, since there is not time to test every possibility. The design of experiments and refinement of equipment are both primarily creative processes as well.
Take something like PCR, for which Kary Mullis was awarded a Nobel Prize. It’s an amazing innovation for carrying out DNA amplification, and it advances science. But it only fits the hypothesis testing framework in a narrow sense of proving that PCR can be used to copy DNA in large quantities. The important implication is not this fact, but all the other things you can discover after building equipment to do this.
I’m not surprised. Going to elementary school in MD in the mid-90s, this is what they taught in science classes.
From my own field of physics I can assure you that one can easily not only get a degree but a PhD, go on with post doc work and become a famous university professor without once having been confronted with these subtleties of philosophy of science. For 99% of the concrete stuff scientists do in their work, it is simply not necessary. So I’m not surprised at all.
In fact, I almost had finished my PhD when I started looking into these things precisely because I needed to defend evolution and modern cosmology against creationists.
I must say though that I find the nomenclature hypothesis, theory, law, simplistic, and to go on about the differences between what a theory is, and a law, and so on is not that interesting except when another creationist tells you that evolution is “just a theory”. In the end, good scientists usually know what they are doing, and these semantic games seem to be mostly a thing for charlatans who want to discredit science. Ohm’s law is just as “wrong” as any theory because it is an approximation. It’s more important to know how Ohm’s law derives from the underlying principles of electromagnetism and/or quantum mechanics, and where its limits are, than knowing the philosophy textbook difference between the terms, which aren’t much used anyways.
All the quoted issues mostly come down to subtelties and variabilities of semantics.
(1) “There is no single scientific method”
Well, if by single scientific method you mean that all scientistst do exactly the same thing, all the time, then yes, this statement is correct. On the other hand, if you interpret ‘single scientific method’ to mean that there is a single set of principles allowing the closeness of some procedure to the scientific ideal to be assessed, then the statement becomes bang on. (That set of principles is called probability theory.)
(2) “of course science is shaped by culture”
Yes, the behaviour of scientists, the questions people are interested in pursuing, the projects that get funded, and the approaches that are favoured are all influenced by culture. But the principles that determine the degree of validity of the methodology are independent of culture.
(3) “theories and laws are different beasts”
From the original paper: ‘Laws are descriptive statements of relationships or patterns among observable phenomena in nature. Theories are well-supported explanations for scientific phenomenon.’ I’ll be endebted to anybody that can provide an adequate account of the difference between descriptions of relationships and explanations. I certainly think anybody who is unable to memorize the officially recognized difference between them can be forgiven. If PZ feels that there is a significant difference, I feel that his is (a) one opinion of many possible opinions, and probably (b) an opinion he will struggle to explain coherently. I’ll be happy to be proved wrong!
On the weirdness of the results of the survey: it’s worth noting that they surveyed faculty “of a state university in the northwest of Turkey”. Results could easily vary in the US / Europe.
Let me provide an example of science that doesn’t begin with a hypothesis: phylogenetic analysis of DNA sequences. You start out (or should) with no hypothesis of relationships. (It’s even been suggested that you should do it blind, i.e. have someone encrypt the names of the species before you start aligning them.) And you emerge, if you do it right, with a confirmed hypothesis. Do you emerge with a theory? Maybe; there are arguments about what “theory” means. But the only part of that standard model that seems to fit is the “testing” part, and even there you’re testing not one hypothesis but all possible hypotheses at the same time. To generalize, some science is purely exploratory and/or descriptive.
Meh. If that was fundamentally true, then we’d never have been able to demonstratively prove that acquired characteristics aren’t inherited. It’s one thing to say that cultural influences affect that acceptance of some scientific theory, like the fact that Comrade Stalin is always right, and he agrees with Comrade Lysenko. But science of course does place a higher priority on how well a theory explains the phenomena being studied, irrespective of cultural influences. That’s one of the reasons why the sciences were able to prosper across cultural lines and emerge during the Renaissance in Europe.
P.S. – aggressivePerfector @ 21 beat me to this point, I see. Oh the lost priority of discovery!
Descriptions in and of themselves only account for the relationship among variables, and not what causes those relationships. Saying that E=mc^2 is a description of how energy and matter are related, but it does not explain why that relationship holds. Or, to move out of physics, one can describe how an array of symptoms cluster to produce a disease without having an explanation as to what the aetiology of the disease is. (For example, psychiatric disorders are diagnosed by cluster of symptoms, i.e., “syndromes”, but there is often little understanding of what causes those cluster of symptoms.)
Or, more concisely, descriptions are not causal, whereas explanation are.
I think if one were to come up with a general definition of science, I’d say it is empirical investigation guided by a falsifiable/improvable provisional theory, with strict estimation and control of errors in the process. If it lacks any of these aspects, it probably falls short of scientific method, but that is a pretty broad definition.
Alex:
No, these “semantic games” are the ones we play so we can understand each other and ourselves. No matter what “usually” happens with “good” scientists, in the end, what something means is important to everybody interested in the subject in question, not just charlatans. Indeed, they quite reliably use fuzzy, equivocal, distorted or just-plain-wrong meanings to push their agendas. However, that does not give the rest of us a free pass to pretend as if we automatically know what something is actually about. That’s a process you still have to go through, whether you realize you’re going through it or not.
For fuck’s sake…. To reiterate what others already said above, Ohm’s law isn’t a theory to begin with. To the extent it is an accurate approximation, it is simply a factual description of things which happen in reality, rooted in some way in the best theories we have about those aspects of reality. It is not all by itself an attempt to explain anything, much less does it explain itself somehow. It’s a mistake to think facts are self-explanatory, or to fail to recognize that an explanation is lacking in some cases even if you do have the facts at your disposal (some of the relevant facts at least). If there are fundamentally any “brute” facts, those wouldn’t be explained at all, for instance: they just are. You can have a theory which takes such facts into account, to explain whatever we can (and want to) explain, but just describing the way things are and putting a “theory” label on that simply will not do. Such a label does not get me any closer to understanding how and why things are the way they are. Don’t everyday, regular-guy/gal, number-crunching, lab-running, working scientists need to recognize the difference between a description and an explanation? Why wouldn’t they?
tulse:
Do explanations cause things to follow certain descriptions?
Can you also give an example of a theory that does explain why it is a good account of the phenomena it pertains to?
consciousness razor @27:
Yes, and they do, which you would have noticed if you had read the sentence immediately following the one you quoted.
It’s also very important to recognize context, rather than cherry-picking quotes.
I agree with with not setting rigid boundaries on what is scientific truth or the scientific method. However, too much PoMo and social constructivism can open the doors to woo. I dare anyone to jump out of a tall building and joyously shout “gravity is simply a social construct!!” And I’ll wager no one here agrees with PoMos like Steve Fuller who support Intelligent Design.
Rob, I guess I have a very different impression in context. Are we reading the same comment? There are several bits scattered throughout, which at least seem to make claims to that effect (my emphasis):
To guard against cherry-picking, here’s also the end of the sentence you decided not to quote for some reason:
If Alex wants to clarify any of that or walk it back somewhat, then okay. But the surface reading of it looks like it goes along just fine with my interpretation.
Also, knowing how Ohm’s law derives from underlying physics is not even in the same ballpark as understanding the difference in general between descriptions and explanations. Both are good and important things to do, if you ask me. But I have no idea how the former could be an adequate answer to questions about the latter.
What exactly are you asking? The definition of a good theory is that it has predictive power. Not sure what that has to do with a theory explaining why it’s good. This question really confuses the heck out of me.
Sorry, speaking of semantics: Change “the definition of” to “a characteristic of”. :)
PatrickG:
An example was given by another commenter (to whom my question was addressed): E = mc^2 is supposedly not a theory because it doesn’t explain why E = mc^2. I’m therefore entitled to ask for an example of a theory, which logically must surmount a similar hurdle.
Predictive power is a fine definition of a good theory in my book. But it fails to meet my earlier request for an account of the supposed fundamental difference between a law and a theory.
aggressivePerfector:
Since I’m not an expert in General Relativity, I’ll attempt to explain using Ohm’s Law vs. what, the Theory of Electrons (I’m sure there’s a better name).
Ohm’s Law is a blunt statement that for an electrical circuit, the voltage drop across a circuit element is the current * the resistance of that element. That’s it. It says nothing about why that happens. Ohm’s Law could very well have come from a theory that says little fire-breathing dragons ceaselessly roam the barren wilds of your electrical circuit, and the fire is what makes your voltage drop.
Of course, the Theory of Electrical Dragons doesn’t hold up very well. How do you quantify your dragons? How do you even see or observe them? What is the mechanism by which a dragon roasting your conductive material causes voltage drops. One quickly has to abandon the Theory of Electrical Dragons.
Instead, we’ll go to the Theory of Electrons. In this theory, we posit there are these things call electrons (verification done elsewhere, fortunately). We then hypothesize HOW electrons behave, HOW they move through materials, HOW their behavior changes when moving through different materials, HOW we quantify resistance/conductivity, and so forth. We then go forth and test these hypotheses, and refine our understanding of how electrons work.
In terms of predictive power, the Theory of Fire-Breathing Dragons is likely to burn down your house. The Theory of Electrons gives you semiconductors, electrical grids, and your personal computer. Of course, we may one day discover that electrons are actually the primordial-soup version of internet trolls, in which case we’ll have to revise our theory and test predictions again. And we’d feel really, really bad about enabling trolls.
In summary: A law says “This is so”. When it’s wrong, you chuck it out the window and go with something more apt for your situation. A theory says “We think this is so”. When it’s wrong, you revise the theory, which incidentally will probably lead to new/revised laws.
Does that make more sense?
@ Alex: Not trying to be offensive, but your school should probably work harder on teaching people these distinctions.
Hell, when I was a TA, we covered this as a matter of course in elementary fluid mechanics courses (please, PLEASE don’t take Bernoulli at his word. He’s often wrong!).
cr @31: Do you seriously believe that physicists fuck up because “oh no, I thought that was an explanation, but it’s a description”? Give some credit to people doing their job without constantly referring to an approved (by whom?) nomenclature manual. They seem to be doing just fine.
PatrickG @36: I’d say it’s more important to know the distinction between Special and General Relativity ;)
Interjecting my little “definition” of law vs theory.
Heirarchy: Laws first; i.e. mathematical equations of the relationship of magnitude of cause to the magnitude of the effect. V= I R , for example.
Theory is a set of laws and an explanation of how those laws relate to each other. Quantum Electro Dynamics explain how Ohm’s Law and Maxwell’s Eqns relate to each other, and where they come from.
. Such is my understanding of Theory vs. Law. I.E. “proven Theory => Law” is just the layman’s euphemism for “W.A.G. shown to be correct => Law”.
Heck, it’s even hard to tell the difference between a fact and a theory aside from scope. There’s no such thing as a bare observation after all — every observation has some theory, however implicit or rudimentary, guiding it. Even something as simple as distinguishing objects requires a theory of objects, which has had to be revised many times over when something didn’t quite fit. When something is established as a fact, it is thereby nested in a large web of theory. So although we want to think of “facts” as being “whatever is the case out there,” epistemologically it’s theory all the way down. Theory doesn’t “graduate” to law, but rather laws are subsumed into theories (and I would argue they are themselves theory given a sufficiently small scope). E=mc^2 states more than just a mathematical relationship between quantities, because each term is itself theory laden, and in the right context it may even have explanatory power.
PatrickG:
But this is not the distinction that is predominantly being put forward. The cheap words that keep getting chucked about claim that the difference pertains to exlanatory power. Nobody has managed to explain (!) to me what this means.
From the quotation in the opening post:
Thus, I’m led to infer that according to the holy philosophy texts, a theory does not become a law even when the extent to which we rationally think the theory might be false becomes negligibly small. I’m therefore still searching for what the distinction might be, and why it is so important to the practice of science.
you know what’s also odd? that so many scientists, who even studied probability and statistics, and likely even covered how polling data works, the limits of it, and how hard it is to control for sample bias via question construction, seem to take polling data so seriously they make conclusions about their own profession.
that’s really odd.
LAW is being confused with its legalistic definition. A Bill is proposed, then voted upon to become Law. Let’s see that with the Law of Gravity. There are plenty of congressmen willing to delegislate the Laws of Thermodynamics, & etc. That’s why they always start with “I’m not a scientist, but …” Too many don’t understand that the Laws of Science are not voted into place. They just want to generalize that their job applies to Everything. That preamble there is an attempt to portray themselves as superior to scientists, and scientists are trying to take over the congressman’s job.
johnharshman #23
Actually, you are starting out with a “law” of common descent, which is explained by the Theory of Evolution. Creationists would deny this of course, but I think it’s safe to say that the vast majority of biologists would agree that common descent has been tested so often and so successfully that it can be accepted as a given (for life on Earth). Then, you are using data, either DNA sequence similarities or morphological similarities, or something else, to derive a hypothesis of a particular phylogenetic tree for your subjects. Then you would gather more data and see if your hypothesis is still supported, or you would tweak it to better fit all the data. Hopefully at some point, your hypothetical tree will fit nicely with all available data and a consensus would emerge that you have a successful theory of the development of those species.
The “hypothesis → testing –→ theory” view does oversimplify the process. Some data has to be available before a hypothesis can be proposed, and there should be a feedback loop from the testing phase to the hypothesis generating phase. So maybe that is what the OP meant about that view not being popular anymore.
Sure, we have very good accounts for the aetiology of many diseases. We know, for example, that the disease smallpox was caused by variola viruses. Exposure to these viruses caused the disease, and vaccination against the viruses provided immunity, to the point that the disease has been wiped out and the virus no longer exists in the wild. In other words, we have a causal account for the disease smallpox, one so well-supported empirically that it radically changed public health. (Contrast this with, say, our understanding of schizophrenia, which doesn’t have a single strongly supported theoretical account.)
Is this the kind of thing you wanted?
Another Matt @39:
Just to be clear; this is not a ‘description’ that Einstein pulled out of his arse (just in case anyone’s thinking that). It’s a derived relationship.
tulse:
Absolutely not. You said “Saying that E=mc^2 is a description of how energy and matter are related, but it does not explain why that relationship holds.” If that is the distinguishing difference between a law and a theory, then you haven’t met your own criterion. Saying that variola viruses caused smallpox does not explain why variola viruses caused smallpox.
Why is saying that I fell down because F = G*m1*m2/r^2 not a theory?
it’s not. you seem to be missing his point.
in fact, the observation that gravity follows a consistent formula is what makes it a law, very much like the law of conservation of momentum.
a consistent observation that follows definable rules is pretty much what makes a law. It’s more an extension of an observational fact than anything else.
A theory includes laws, like it includes other facts, and hypotheses to explain these observations, and testable predictions with consistent results based on the outcomes of hypothesis testing.
why is this so confusing?
frankly, it’s all semantics anyway, and likely the reason the poll looks so odd is because really, nobody pays much attention to this in their actual work, mostly because it doesn’t matter.
There’s so many great explanations on this thread (a lot better than mine!) that I just don’t know how to approach it further. My last attempt:
The Law of Hammers says “Swinging a hammer this hard at an object produces a force on that object”. This law says nothing about what you’re swinging it at. It could be a nail, or it could be your head.
The Theory of Tools says “Tools are defined as implements that are used for certain purposes. There are many instances of tools, each of which has a certain defined purpose. A hammer is defined as a tool used to apply force to objects through lever action (define levers), and is characterized by the following parameters (parameters here). Objects a hammer can hit can be evaluated for suitability (define criteria here, probably involving pain somewhere). Hitting nails with a hammer is a suitable use by these criteria (for a certain definition of “nail”). Hitting yourself in the head with a hammer is an unsuitable use by these criteria”.
If that’s not clear, I’ll give up and hit myself in the head with a hammer.
@ Rob Grigjanis:
Fuck offOops. :) I’d meant to fix that before posting, but hey, work and all that. Thanks for the correction.If you gave me a supply of real Quidditch bludger balls and enough time, I could probably create a lot of equations that would tell you how fast you have to travel to outrun one, their turning radii at different speeds, and how hard they can hit you, and if it turns out that all bludgers obey these equations, they might even be called the bludger laws. But that would tell you almost nothing about the magic that makes them move.
Let’s face it, we don’t learn to drive by sitting down and studying theory of driving. We learn by driving. Science is similar–we learn to do science by doing science, by making mistakes and having our teachers and advisers correct us. Most scientists never have to think about what it is the do. Where scientists tend to run into trouble for not having thought this stuff through is when they are exposed to a new scientific field outside their expertise. That is why you may see experienced and respected physicists (my field) making fools of themselves by claiming that climate science or geology or biology, etc. are not science.
@ Rob Grigjanis, 37:
Until they start redefining terms:
Falsifiability police! They’re coming for us all!
Taken from discussion here, in response to an opinion piece in Nature.
Here’s an example of gases from chemistry. Early on, certain researchers able to predict certain properties of gases based on temperature, pressure, and volume. They called their equations Laws. If you combined them, you would get the Ideal Gas Law, PV=nRT. They held for their limits, as they presumed mass points for the gases. The Kinetic Theory of Gases allowed for the derivation of the Ideal Gas Law from basic theory, and why what was seen was seen.
Of course, the presumptions of mass points is wrong, as Molecular Theory says the gases have both size, and that they can interact with each other in various ways. When the Kinetic Theory of Gases is combined with Molecular Theory, the van der Waals equation with corrections for molecular size and attraction/repulsion of the molecules expanded the range of predicting gas behavior, especially at higher pressures.
The equations (laws) predict the behavior, and the two theories say why this is seen.
Ichthyic:
Maybe because it’s not what everybody else is saying.
I agree entirely. People keep mentioning the magical power of explanation that only theories possess. I believe these people are confused.
moarscienceplz:
So explanation is only achieved when we have a description of a process at the most fundamental level possible?
PatrickG:
So it becomes a theory, when? When it includes an account of everything in the universe? The general impression I’m getting is that a theory is distinguished by it complexity – it must contain many laws. How did you discover this fact? Why did nobody show me the shelf in the library that contains this sacred book? What is the minimum number of laws required? Is there a fundamental difference between a theory synthesised from two laws and a theory that contains only one? Is the difference between 1 and 2 more fundamental than the difference between 2 and 3? What if I write my 2 laws on a single line, enclosed in and AND() function? Why the hell is any of this important?
Of course the practice of science by any particular person will be affected by their culture. However, of course the correctness of science is not a social construct. Hammers fall to the ground when released in normal household conditions, no matter what your culture has to say about it. I suspect the scientists of the survey thought that they were being asked a different question, a basic question, and not a question of subtle technicalities.
I’d want to see the exact survey question asked. As we all should know, writing proper survey questions is incredibly hard.
PS: Laurence A. Moran should be the last guy to talk. He’s defending methodological naturalism, which is just NOMA by any other name, which is bullshit. Science is not founded upon methodological naturalism.
moarscienceplz #43 sez:
I presume you’re putting “law” in quotes because you realize it isn’t a law at all. A law need not be mathematical in biology, e.g. Dollo’s law or Haldane’s rule. But it’s a simple statement about a pattern, e.g. “the same species will not evolve twice” or “when in a cross there is hybrid sterility in only one sex, it will be the heterogametic one”. Common descent isn’t any such thing. I’d say, in fact, that common descent is both a theory and a fact, and that facts are just very well supported theories. (Some might object to use of the word “theory” and substitute “hypothesis.) But all that is irrelevant to your point, as far as I can tell, which follows after.
You may say that phylogenetics fits the hypothesis->test->theory idea, but only if it’s as you describe, which in my experience it generally isn’t. First of all, each node of a tree is a separate hypothesis; a tree is merely a set of such hypotheses and can’t be considered as a single hypothesis in its own right. One may tailor a study specifically to test prior hypotheses, but this is very seldom done. And the first study in which a particular node is observed may also be sufficient to confirm the hypothesis that the node exists; no three-part process but a one-part process.
Take a look at the recent paper in Science using 48 genomes to find avian phylogeny. It wasn’t designed as a test of prior hypotheses; it’s merely that all that data produced a tree that confirms some, disconfirms others, and presents some other novel ones. Less “let’s test this hypothesis” than “let’s see where these data lead”. And that’s a perfectly fine model of science.
I don’t know what specifically PZ was thinking of, but that’s certainly one of many alternative styles, all of which can be valid science.
PatrickG @51: Yeah, that’s a whole other ball of string. For some reason I forgot about that when thinking about physics. I blame
societythe multiverse.I’m not surprised; I’m with comments 3, 20, 47 and 50 – most of the time, being aware of science theory is simply irrelevant to doing science, so most scientists never learn it.
That doesn’t mean they run around trying to prove theories so the theories can grow up to become laws! It means they don’t care about such terms at all. They don’t even use them – they write “strongly suggests” where a layperson would write “proves beyond reasonable doubt”.
Well, do they mean the ideal or actual current practice? Maybe they’d say that once you let bias in, it’s not actually science anymore?
There is; but “falsification and parsimony” isn’t terribly precise, so if you want a more detailed description, you won’t end up with a single method. See comment 21.
Theories are bigger than hypotheses (they try to explain more facts, less obviously connected facts etc.), not necessarily better tested.
But you can’t do that when you don’t doubt yourself. Doubt is the… utter… foundation of science. Doubt is good! Doubt everything! Everything must be doubted!
You can tell the world lots of stuff and sound very confident indeed when you don’t doubt yourself. But that way you can’t tell the world what’s right.
PhenomenAAAAAAAAAA!!! The plural of phenomenon is phenomena! It’s exactly like mitochondrion, mitochondria! It’s even the same thing as the Latin -um, -a phenomenon! *flail*
Sorrynotsorry. :-)
…Several people had already done that before you submitted your comment, you know.
…Oh.
Or of anything at all. – Well, you don’t begin with a hypothesis. As you go on to say, you begin, technically, with all mathematically possible hypotheses, and then you throw Ockham’s Razor at them and see what happens. I exaggerate (in that you almost always have too many taxa for an exhaustive search, so various mathematical tricks come in), but not by much.
In short: you begin with all hypotheses, and you test them all.
Well, yes. If it relates too few laws to each other, it’s only a hypothesis. :-þ
what
Terminology is created by convention. It’s not discovered.
David Marjanović:
But half the population are wrong, wrong, WRONG, when they don’t use the terminology correctly.
@ aggressivePerfector:
No. Complexity does not a theory make. You need to step back and re-read what a lot of people are saying. I highly recommend moarscienceplz’s Quidditch example at #49. Probably the simplest and most accessible thing on this thread.
Give me data, and I’ll tell you what’s happening. I’ll even give you some assurance of what will continue to happen, assuming nothing changes. But I know nothing of how those balls are spinning and those wizards are flying, and I certainly can’t explain why so many people enjoy Harry Potter books.
I see what you did there, Rob Grigjanis.
Apparently, scientists AT ONE COLLEGE IN TURKEY need to work on educating themselves.
Is this result representative of all scientists elsewhere in the world? Who knows.
@aggressivePerfector
Laws describe what happens. Theory tells you why.
The Theory of Evolution.
Why is saying that I fell down because F = G*m1*m2/r^2 not a theory?
F(weight), G, m1, m2, and r are all independently measurable quantities. You can plug and chug them into the Law of Gravitation and make all kinds of neat predictions, that’s part of the theory of mechanics. In that sense, the Law of Gravitation is an empirical observation that, taken to be true, is one of the underlying assumptions of the theory of mechanics.
But it is the original “spooky action at a distance”. There’s no underlying mechanism that can be derived from a transmission of force by the usual means known in Newton’s day (physical contact between objects, particles, and waves). People objected to universal gravitation for this reason, and others went nuts try to come up with a particle based model that would satisfy those critics. They failed, probably because Newton’s Law of Gravitation is just an approximation that falls down (a tiny bit) in our solar system wrt the orbit of Mercury. So there probably is no deeper explanation that could work.
Saying that variola viruses caused smallpox does not explain why variola viruses caused smallpox.
It doesn’t have to. The germ theory of disease says infectious diseases are caused by tiny organisms (germs) and not moral character, or bad air, or evil spirits. That is still a useful and testable idea, even if we don’t know how these organisms attack the body, heck, even if we can’t *see* the organisms (e.g., viruses before the electron microscope).
No theory starts from zero. The Theory of General Relativity assumes an interaction between mass/energy and space/time and from there derives a more correct description of gravity than Newton’s Law of Gravitation. There’s still assumptions there, but they are more basic ones that result in a better prediction.
And a theory doesn’t have to explain everything to be useful. An incomplete germ theory of disease can still convince you to wash your hands frequently (and it did). General Relativity can predict the observed expansion of the universe (and it did) but it can’t explain the very beginning, because it doesn’t work at quantum scales.
Personally, I would be happy if the word “Law” was replaced everywhere in the literature with “Principle” or “Equation” or “Relation” as appropriate, but that’s not going to happen. As long as the current terminology persists, there is a meaningful distinction it’s worthwhile to keep straight. And the most important thing is that the common lay explanation of the difference is wildly inaccurate and actually misleads people.
Sorry about the quotation fails, I thought bq would work for blockquote, but it doesn’t. :-(
Everything comment 63 says. :-)
Well… yes. :-| Is that supposed to be a contradiction?
there’s irony in that there quote.
or, rather than sitting there rambling, you could simply google for “scientific theory” and get what various academic institutions define it as for yourself.
they’re surprisingly consistent.
something tells me you don’t really care though.
so then, we’re left with…. what is it exactly that you want to know?
@14: “A law is an observed mathematical relationship between measurable quantities with no underlying explanation of why that is. Think Ohm’s Law, Newton’s Law of Gravitation, Hooke’s Law for springs, etc.”
Or think of Archemedes’ Law, which totally can be derived from other formulas. Your statement is falsified.
As a teacher physics it always has been my impression that the label “Law” is attached rather randomly – its main goal is to honour someone. But yeah, laws of physics always are part of a coherent and consistent theory, so they are not the same.
@68 broad definitions of English words aren’t mathematically precise and “falsifiable” by single counterexamples, IMHO. As a math teacher I get annoyed when people try to do that. :-)
Nevertheless, the rest of my post was in basic agreement that the use of Law is not rigorous or uniform. That doesn’t mean there isn’t a basic pattern though. And certainly a pattern more useful than the lay belief of how law and theory compare.
As for Archimedes Law, I’d be surprised if it was derivable from physics known in Archimedes’ day. Just was Ohm’s Law is not derivable from the physics of his day, but can be explained by modern physics.
Last I checked, philosophers of science regard trying to ascribe clear and unambiguous demarcations between “theory”, “law”, “hypothesis”, etc. to be a mug’s game. There may be good reasons to prefer one term over the other in certain situations, but there is no One True meaning for each term that has been handed down from on high, and scientists themselves, who are the only real arbiters as to what scientific terms mean, are far too casual and inconsistent in using these terms. Besides, it’s about as important as hamster snot. Which of these labels you apply is at best only slightly useful in conveying information about content, and not at all about correctness.
I wanted to comment on the hypothesis -> testing -> theory comment, which you hear a lot in pop sci contexts and is so very, very incomplete.
This was Popper’s view, and at best it applies to limited scientific activities. A lot of science is descriptive, like anatomy and astronomy. Even large chunks of things like organic chemistry.
But Popper was off in a more subtle way. The idea that you have one hypothesis, which is then abandoned when falsified, falls apart pretty quickly when you look closely. What usually happens is that people try to save the hypothesis by modifying it. Certain situations get defined as “outliers” or “exceptions,” new elements get added, and so on. And this isn’t just scientist being imperfect humans–it means what really falsifies a hypothesis is often having a better hypothesis come along. You aren’t really figuring out if something is empirically false, as Popper hoped; whether it’s false depends on what else the scientific community is thinking about.
An specific example, involving Newton’s laws of gravity, I had forgotten about until recently. As measurements got better classical mechanics didn’t explain the movement of Mercury exactly. People started hypothesizing a new planet. No one could find the planet, but a lot of people were willing to believe it was there. Or maybe something else we couldn’t quite measure. But no one thought the discrepancy “falsified” Newton, which had been confirmed in so many other ways. Until, of course, Einstein did general relativity. Then suddenly the old measurements, which hadn’t changed at all, were enough to falsify Newton.
FWIW I agree with many here that the differences don’t matter much to a working scientist. Few scientists change how they work based on descriptions by philosophers of science.
I think that I am enjoying this discussion but I really don’t know what the fuck y’all are talking about.
—
and my iPad is now offering suggestions for every letter I type.
complete with spaces
fuckinghell fuckwhitterty fuck
Ichthyic:
What I’d like to know is why PZ cites these 3 findings as evidence that scientists need to educate themselves.
Why should I care how academic institutions define terms like ‘theory’ and ‘law’. What difference could it make to my capacity to conduct science?
latveriandiplomat:
I’m very grateful for the lecture, but the point of my question was not to prove that germ theory is not a real theory, but to illustrate how an offered explanation of the difference between law and theory was incoherent.
Lady Mondegreen:
Thank you, but my point throughout has been that nobody can offer a coherent description of what it means for a theory to say why something will happen, as opposed to simply predicting that it will. I understand that a good theory uses a few concise statements to generate predictions about a wide variety of situations, but ‘laws’ also do this. You need to give a better account of your proposed distinction.
No good. The theory of evolution does not explain why the theory of evolution is an accurate theory. It is data that validates the theory, as with ‘laws’.
PatrickG
What is your point? Are you suggesting that the hypothetical law of spinning balls is not a theory because it does not explain the underlying mechanism at the most fundamental level?
I’m happy to explain my position, if people have genuine curiosity and difficulty understanding, but perhaps it is you that needs to step back here, and consider whether you are really the expert you think.
aggressivePerfector
golly
you’re so smart
Chigau @ 72
I am SO with you! I have little real knowledge of the sciences, but I thoroughly enjoy discussions like the above. My self-education has been greatly enhanced by reading the comments on Pharyngula over the past six or seven years. And all hail iPad suggestions/corrections!
AtheistPilgrim #75
I expect that there is grog available in the Lounge.
http://freethoughtblogs.com/pharyngula/2015/01/08/lounge-487/comment-page-2/
A theory is an accumulation of tested and verified hypotheses which together inform us about some aspect of the world.
Facts are what are observed, by multiple independent witnesses. (They can change when we develop new means of observing).
Hypotheses are educated guesses about these observations, that can be tested, and are falsifiable (if there isn’t a way of testing if it’s wrong then it’s not a hypothesis).
Laws come from a verified hypothesis, are usually a mathematical relationship, and also usually have a particular scope within which they are valid (for example not all materials have an ohmic resistance, metals usually do but semiconductors frequently do not, especially in diodes).
@71:
Falsification was Popper’s way of trying to get around the problem of induction. It doesn’t really work. Trying to falsify theories (or laws/hypotheses/whatever) runs into the exact same problem as your example demonstrates. To the extent you can falsify anything, it requires being able to prove some sort of converse statement. Hence, the problem of induction still exists, just in slightly obscured form. Probabilistic induction is almost certainly a better description of what scientists do than falsification is, but that apparently is not without issues either. Also, Popper’s views are more complex than popular treatments typically allow, and I haven’t read all the source material, and…
If you look at the history of science, you actually find that a lot of our “laws” we’re not discovered through hypothesis testing at all. Instead they were determined through
Observation -> figuring out a way to quantify the observations -> fiddling with math until you can curve-fit an equation into your data-plot of your quantified observations.
The equation you produce becomes a law if your peers confirm that it is consistent and reproducible, and they decide to honor you by naming it a law after you. (oft times you are the one who proposes that it be called a law and it officially becomes one when enough of your peers agree to call it so in their own writing.)
Regarding the hypothesis -> test -> theory thing, the majority of routine science actually goes theory -> hypothesis -> test.
Because it is usually the underlying theory that generates the hypotheses, and each hypothesis so generated that is successfully tested increases the level confidence that the underlying theory is a reliable approximation of reality.
As for the theories themselves, they can be initially produced in a variety of ways. Einstein famously did not do any experiments when he derived relativity theory. Darwin did many small experiments, but no single set of them resulted in the derivation of the whole of his theory of evolution. More often than not he was testing individual ideas that were later incorporated into his theory, as he came up with them.
When you boil it down, theories are usually produced through acts of associative reasoning. Individuals have flashes of insight that link disparate phenomena or ideas (ie they recognize a pattern in the data) into a unifying framework. The things that they link together, the data that they do the pattern recognition on, can come from many sources, including previous hypothesis testing, but not necessarily exclusively from that.
Another misconception is in portraying scientific method as a linear process when it actually is a cycle.
Rather than observation -> hypothesis -> test, it is actually observation -> hypothesis -> test -> new observations from the test -> new or modified hypothesis -> test, etc.
Theories spin off from the observations, and feed new hypotheses back into the cycle. Laws ARE the observations, when they can be expressed in formal mathematics.
It is actually kind of like the Krebs Cycle.
Hi,
@consciousness razor #31
I certainly didn’t mean to give the impression that scientists don’t need to be aware of the difference between a description and an explanation. I rather think that this is so self-evident to anyone working in their field that the nomenclature never really comes up. Also, I don’t think that the separation is as clear as you make it sound. You could say that Kepler’s laws are an explanation of why the orbits of planets relate to each other as they do. If we’re talking about why planetary orbits look like they look in the first place, it is more fitting to say Kepler’s laws are merely a description of planetary motion, whereas Newton’s laws are the real explanation. I’m sure in the context of general relativity, one could make the case that Newton’s laws are an approximate description of certain configuration of the metric, and einstein’s quation give the explanation.
So I don’t see how insisting on “A is an explanation, B is a description” is at all helpful. It completely depends on which question you ask, and I’d say working scientists who filled out the questionnaire in the OP wrong can have a better understanding of these subtleties than someone who insists on using this simple terminology.
Damn it, that last sentence was not supposed to be in the block quote
Everything Amphiox said. :-)
The “with no underlying explanation of why that is” part meant that no such explanation is part of the law. It doesn’t mean that an explanation can’t be found or hasn’t already been found. All three of the original examples have been derived from the theories of quantum electrodynamics or general relativity; they’re still laws. Since Einstein, physics has been on a quest to find out if any laws are truly fundamental (impossible to derive from anything) or if all of them can be derived from some overarching Theory Of Everything.
What.
None of that is purely descriptive. Being part of biology, for example, anatomy doesn’t make any sense “except in the light of evolution”. Without hypothesizing about homology – without the theory of evolution, in other words – you won’t even know which words to use to describe what you see!
Popper was simply stricter with words than you are: in such a case (and yes, it is common), the original hypothesis is falsified, and people switch to a very similar one that is nonetheless not the same.
That’s because a more parsimonious explanation was now available. Parsimony is the one big thing Popper appears to have overlooked; it’s hidden within falsification a lot, too.
…We’re really talking very far past each other.
Where are you even taking “accurate” from? Theories aren’t by definition accurate! Lamarck’s theory of evolution was a theory; it explained the whole of biodiversity as known back then – the explanation just happened to be wrong.
Yep. It has become cultural. It’s the kind of habit you pick up from reading papers and writing your own – just like writing “strongly suggests”. :-)
Hm. I need to think about that. :-) In any case, Kepler’s laws are a special case of Newton’s, for mathematical meanings of “special case”.
Other considerations that might affect scientists’ answers to this survey:
1) Who is conducting this survey?
2) For what purpose is the survey being conducted?
3) Where are the results of this survey going to appear?
4) What’s the intended audience for the survey results?
5) What other questions were asked prior to these that might condition the results?
6) How was the survey conducted? Mail? Internet? Phone? In-person? Under what conditions?
It’s hard to ascribe any value — good or bad — to the results of this survey. A scientist would, I’d hope, understand that data of any kind is pretty meaningless without context.
Assigning
I never said or suggested they were. You might try reading my comments, before responding to them.
Another commenter told me that laws are not theories because they do not explain why things are that way, i.e. why they are accurate descriptions of reality (“Saying that E=mc^2 is a description of how energy and matter are related, but it does not explain why that relationship holds.”). Others echoed this sentiment. I was merely pursuing that line of reasoning, in the attempt to demonstrate that it doesn’t work.
Laws can be determined empirically, with no theoretical framework, and then theory can be developed later that derives the law’s equations. In fact I think the majority of our scientific laws are of this type.
When a theory derives a law already known to be an accurate representation of reality, that is essentially a “postdiction” of the theory, and is good independent evidence in support of the theory, assuming said derivation was not part of the initial formulation of the theory. Note that this sequence does not follow the standard observation -> hypothesis -> test pattern.
Even stronger evidence for theory is if it derives an equation which is clos to, but not identical to the law (or of which the law is one special case), and you go do the measurements on those cases where the theory differs slightly from the law, and you find that the theory is more correct than the law.
Other times the theory can derive an equation describing something for which no empirical law had previously been known, which then can become a law (though usually there is less impetus for people to officially call it a law in that case, since most are content to just refer to the theory itself when talking about it – the naming of official scientific laws is often a shorthand for admitting “we don’t really understand why it works, but this is how it works dagnabit”) layer once the equation is tested and verified. Some of the population genetics equations pertaining to selection coefficients are like this, and the equations describing the Higgs Boson are probably like this.
The second case is actually a minority in the history of human science.
You are misunderstanding precisely what E=mc^2 as a theory explains (the theory is actually the derivation of that equation rather than the equation itself). The “whys” that the theory explains are things like why nuclear fusion can produce energy and why the speed of light cannot be exceeded by anything with mass. The explanation that the theory provides is that mass can change into energy and vice versa. At the same time, the theory provides a derivation of the precise manner in which energy and matter are related, as a testable prediction, which you can go test by finding an energy producing event from which you can get right kinds of measurements from, measure the amount of energy produced and the mass before and after, and check to see if they relate by c^2.
It does not, indeed, explain WHY mass an energy are related (it does explain, through the math of the derivation, why, *given* that mass and energy are related, they are related by that, specific relationship and not, say, c^3), but that was not what the theory attempted to explain in the first place.
Because Relativity isn’t the ultimate theory of everything.
This is how all theories work. It explains X. In the process of explaining X, it also derives prediction Y that is a consequence of X working the way the theory, and sometimes Y generates new questions of its own, which leads to the development of new theories to explain the why of Y.
Once again everything Amphiox has said. :-)
Then why did you write “accurate theory”? (Your comment 75: “The theory of evolution does not explain why the theory of evolution is an accurate theory.”) Providing an explanation for “why that relationship holds” does not entail that this explanation is accurate, even if the relationship is accurate.
The theory of evolution explains a lot of facts. :-| I’m not sure if it explains any particular law; there aren’t really any in evolution.
That’s how neutrinos were discovered. Beta decay appeared to annihilate energy somewhere; some thought this falsified E = mc², others thought there must be a particle that is difficult to detect and carries the missing energy away, and then just such a particle containing just the right amount of energy was found.
David Marjanović:
I’m puzzled. Your comment seems to be directed at me, but I don’t think Amphiox’s was. Certainly there was nothing in that related to anything I said.
The principle of mass energy equivalence (I don’t care if this is a good example or a bad one – it wasn’t my example, and my point is that the distinctions between ‘law’, ‘theory’, ‘hypothesis’, ‘principle’, etc. are arbitrary and matters of degree) is reportedly not a theory because it doesn’t explain why mass energy equivalence is true (or approximately true, or probably true, or whatever variant you wish). By this logic, if there are theories, they must have this property of explaining their own accuracy (or even supposed accuracy, to relax the original proposal). (A theory is accurate if the statements it supports lie close, in some high dimensional hypothesis space, to statements that are true.) By this standard, a theory must include an account of why things work that way. It is my strong impression of the impossibility of meeting this condition of a theory explaining its own accuracy (or supposed accuracy) that leads me to infer that the distinction is bogus. A theory, after all, is synthesized from laws, principles, or hypotheses, or whatever the hell you want to call them. If the basic laws don’t explain why they are true, then how can the theory? I await a counter example to change my mind.
I really don’t see (longish thread) how you jumped to the conclusion that someone was insisting that theories hard to explain their *own* accuracy.
Theories explain. Laws describe.
Theories explain … things. Laws describe…. things. (Things includes processes here.)
Theories don’t have to explain THEMSELVES. ANOTHER higher order theory does that, all the way to the Theory of Everything, assuming such a theory exists and is knowable. Indeed there are some theoretical work that suggests that it may well be IMPOSSIBLE for any theory to explain itself (Godel, I think, perhaps)
Mass energy equivalence may be a bad example. I’m not a physicist but I vaguely recall that the equation is in fact derived through general relativity’s basic assumptions, such as the universal constancy of he speed of light. If this is correct, then the mathematics of the derivation in general relativity do in fact explan why there is mass-energy equivalence, and the explanation is it is because the speed of light is a universal constant (assuming that is the correct assumption fom which the equation was derived). Basically, to translate into simplified English, if the speed of light in a vacuum is a universal constant and maximum possible velocity then it follows that mass and energy MUST be equivalent, and they MUST be related by a factor of c squared.
But the distinction between law and theory is not arbitrary (any more than anything named in a human language is arbitrary) nor a matter of degree. A law is not a mini-theory. No matter how much you expand or extend a law it will never cross some magic hers hold and turn into a theory.
Theories are explanatory. They explain SOMETHING. NOT themselves. Certainly not everything. But they explain SOMETHING.
Laws are descriptive. Laws do not explain ANYTHING.
So, you don’t know what the conversation is about, yet you feel compelled to vomit your opinion on everyone. Interesting.
This is what so many keep saying, yet when challenged to enlighten me as to the actual difference between explaining and describing, nobody has come up with anything credible. Capitalizing words, and embelishing statements with the word ‘things’ is not productive, it is just helpless flailing.
The theory of relativity, to continue your line, is composed of basic hypotheses (‘laws’). Yes, it explains things, but not in a manner that the laws do not. Any explanatory ability of a theory can only be derived, in fact, from the laws it is built from. Any time I see some symmetry in natural phenomena, then that symmetry is explained just as well by some law describing it, as it is by a theory derived from that law (A will do B, because it is a fact that all things like A always do B. Why is that a fact? Don’t know, that’s outside the scope of the law / theory / hypothesis / model / …).
The feeling of a radically different kind of understanding that nominal theories provide is, I’m afraid, an illusion derived from a combination of complexity (compounding of principles) and familiarity.
Amphiox @92:
At least one more postulate is required; that physics looks the same in any inertial frame. Specifically, Einstein demanded that electromagnetism look the same, and that led to the mass-energy equivalence. See the link I provided in #45, and the paper that preceded it. They are beautiful, and will be 110 years old this year!
Every explanation given to you has been credible. That YOU, individually, refuse to accept that credibility through obstinate obtuseness in the most egregious JAQing manner is not our problem.
I know EXACTLY what this conversation is about. What I do not comprehend is whether you are being stupid or deliberately intellectually dishonest in your continued refusal to understand, or even evince the slightest hint that you actually paid attention to the details of the many excellent replies given to you or did the expected intellectual work of trying to understand them, as would be expected in any honest conversation.
And this provides strong evidence towards the second hypothesis, in which case further attention paid to you will not be productive.
And since intellectual dishonesty is the one thing that I will not abide nor forgive in a discussion such as this, I shall now exercise my freedom of association and excise you from the sphere of individuals with whom I am willing to engage, and in the tonal spirit that you yourself, through willing action and conscious choice, introduced to this conversation, fuck off and goodbye.
latveriandiplomat:
Well, no, it isn’t like that, in a lot of important ways. Gravity isn’t an instantaneous phenomenon, and it does propagate through the intervening the spacetime and depends on the stuff between the two distant points in question. Also, even though it presumably has an infinite or nearly-infinite range, the magnitude of it drops off with distance, meaning you fairly quickly get no appreciable action at a distance to even talk about. It’s also not “spooky” in the sense that it supposedly only happens whenever you measure it, as if it were a matter of you telepathically controlling what’s real and what isn’t. Even Newton himself would have apparently agreed with all of that; he simply didn’t put forward his ideas about gravitational particles doing all this, I guess because he couldn’t get it to work or didn’t care enough to try.
What you go on to say about the Newtonian theory’s lack of a well-defined and well-understood mechanism is okay, though. That’s just not the kind of idea Einstein was objecting to when he used that phrase. It’s just a nitpick, really, and it’s not important in this thread unless we start talking about quantum theories. (Let’s hope it won’t come to that). / SIWOTI
Alex:
Fair enough, but sometimes it does come up and becomes an issue. Think of Krauss’ universe-from-nothing bullshit, for instance. Or consider people arguing over whether or not certain features of the world demand explanation because they’re supposedly unlikely (e.g., low entropy in the past, various sorts of fine-tuning or various other unresolved “problems” people talk about). It sometimes isn’t obvious/self-evident that something needs explanation or can be explained or exactly what form the explanation ought to take if one could be given. Or for a less-drastic example, think about how non-scientists tend to interpret results and theories which aren’t being articulated very clearly by scientists — that really matters, and it doesn’t all reduce to cranks/charlatans trying to justify magic or confuse all of the “good” people or whatever. Even the “good” or “rational” people exhibit these kinds of mistakes or carelessness or whatever you call it — they at the very least don’t all follow the party line about what is supposedly “self-evident” or trivial according to you (or to science itself, or whatever the argument is supposed to be here).
I think a lot of commenters are being unfair to aggressivePerfector. After all, the distinction between description and explanation is still debated amongst philosophers of science and theorists. Hell, they can’t even agree on what causality is (which is pretty central to explanation in science).
I think we all have an intuitive sense of a distinction between description and explanation. All the definitions I’ve seen in the comments tap into this (“an explanation explains things!”), without presenting a tangible distinction. In fact, most of them are quite recursive, highlighting aggressivePerfector’s point (as I understand it).
So, AP is doing is mental wanking about a point they see, while the real scientists responding have no problem seeing a distinction? And one can wonder why philosophy has such a bad rap around here, compared to science.
but it isn’t incoherent.
and you JUST PREVIOUSLY said you didn’t even care if it was.
troll much, asshole?
run along.
then you have really poor reading skills.
maudell @97: I’m not sure what you think the debate is. ‘explanation’ and ‘description’ have always been, and may always be, relative terms. The Standard Model is a theory which explains a lot of what we observe. It also contains 19 parameters which have to be measured (“why is the Higgs mass 125 GeV?”). I’d call that description. These things are understood by the practitioners, and not debated, because they are understood.
you know, shoving your head up your ass whenever someone meets your idiotic “challenge” doesn’t say much for your reasoning skills.
What Imre Lakatos was on about, then? (I’m a mathematician, very much not a “scientist” in any sense that appears to be described any of the various competing accounts you all are giving here; but I enjoy reading them.)
maudell:
Yes, that’s exactly my point. I’m glad somebody appreciates it!
they misread you.
you clearly DO NOT HAVE a point, and have been deliberately wasting everyone’s time.
just fucking admit it and go.
Nope, you missed the point that was explained to you many times. You lost. Your inability to admit you are wrong is telling.
When he dismisses answers altruistically provided for him in good faith as “vomit” then frankly the commenters have been nicer to him than he ever deserved.
He has proven himself the type of self-entitled leech who thinks the universe owes him answers, on terms of his own choosing. None of us here are paid to educate him, or have sworn some professional oath of conduct that requires us to patiently answer internet trolls, or even to answer at all.
Every reply given to him was freely given AS A FAVOR. Information provided as a GIFT. An act of altruism.
He took the gift horse and spat in its face.
It is my understanding that per general relativity, gravity moves at the speed of light. Thus, if somehow the sun were to suddenly vanish into an alternate dimension, earth would continue moving in an orbital arc for 8 or so more minutes, the time it takes for light to travel from the sun to the earth. We on earth would notice the change in the gravitational effect at the same time we see the sun’s light disappear, not before.
So at last an admission from the troll that he deliberately wasted everyone’s time for a pointless discussion about mere semantics.
Disgusting.
AP, there is a very easy way to get your point across, and it is much less aggravating.
Simply say “this is what I think, and this is a couple of examples of what I mean”. Examples since this was philosophy, not evidence which is required for science. The discussion could have continued from there without the rancor.
Philosophers need to keep in mind, that science doesn’t try to be perfect knowledge, just good enough enough knowledge, and better than before. A little sloppiness in early theory development is expected.
Part of the claim in the original post was that there is an important and fundamental difference in science between laws and theories, and that scientists who don’t know this are somehow inadequate. I stated my opinion that this supposed distinction is actually only a silly matter of semantics, and asked if anybody can prove me wrong by explaining what the difference actually is. I got a lot of contradictory answers. Many of the replies suggested that the differnce lies in the ability of theories to explain things. I asked how that is different to what laws do. Most of the the replies to this simply restated the hypothesis that theories explain, laws only describe. One productive person gave partial steps towards an answer, by stating that laws are different because they lack a certain property, so I asked for the next step: an demonstration that theories possess this property. A few attempts were made, and I explained why they failed, and why indeed they must fail.
Now a lot of people are cursing at me, and accusing me of wasting their time with ‘pointless discussion of mere semantics,’ while, on the contrary, it was the original poster who used a pointless matter of semantics as evidence to unfairly characterize the scientific community as ignorant of their own discipline. Just as baffling, I’m accused of bringing philosophy into disrepute by pointing out that a pointless distinction dreamed up by poor philosophers of science is being used to undermine faith in the abilities of scientists.
Finally, a commenter has admitted that he/she was unable to keep track of what my comments were about, but then proceeded to try to tell me why my comments were wrong anyway, and then apparently got very grumpy indeed when I complained.
I will now utilize my magical power of forcing you all to reply again, thereby wasting your time some more. Don’t forget to use plenty of CAPITALS.
fuk off, troll.
*eyeroll* If it were, I’d have tried to send you an e-mail or something. It isn’t “directed”. It’s a public announcement on, you know, a blog. Because I don’t want to post 5 comments in a row, I pack all my responses into a single comment every time I visit a thread. I respond to quotes, not to people.
Not that I can entirely escape it, but I’m not actually trying to have social interaction here. I’m trying to have a discussion, the same sort that I have in a much slower way when I write a scientific paper that shows something was wrong with some earlier scientific paper (including my own ones).
Finally you’re getting it: supposed accuracy is what everyone but you was talking about. Being actually accurate is not a necessary criterion for being a theory.
And for the record, the distinction of “hypothesis” and “theory” is indeed a matter of degree where no clear definition exists or is reasonably possible; it’s arbitrary. A theory is just “larger” than a hypothesis, it explains more facts and more laws that seem even less connected at first glance.
…No. Not even close.
When was that actually discovered? And when was it found to propagate at specifically the speed of light?
Some of us, however, have SIWOTI syndrome.
I’m not really capable of just letting a misunderstanding be.
A reply from me isn’t a favor, and neither is it a sign of respect or anything. It’s just a sign that I think something needs to be commented on and nobody else has done it yet.
“of course science is shaped by culture”
Any history of science course would make that pretty clear. Of course the opposite is also true: what we discover in the sciences affects our culture. Are graduate students required to take one? I know some medical schools make a history of medicine credit mandatory but I’m not sure about the other sciences.
You demonstrated nothing but your own ignorance and determination to show PZ wrong. Which you didn’t do. We scientists are not perfect thinkers by philosophy standards, as evidence, not a precise way we think, drives what we do. There doesn’t need to be perfect methodologies as what works in the past for any given discipline, has been superseded by newer methods and better theories. And what works in chemistry, may not work in biology, and needs to altered for type of evidence seen. Which is why there is not a good philosophical definition of science and scientific thinking. Scientific thinking is always adapting, as our knowledge and tools to look at the world change.
Speaking of the influence of culture on science.
The naming of scientific laws is a cultural process. Laws are named laws because the cultural community of scientists collectively decide that they should be so named, usually as a way to honor some individual’s contribution to the work. (Honoring individuals of course being a wholly cultural phenomenon).
Or sometimes the originator of the law will propose that the relationship he or she describes, in the original publication be named X’s Law after themselves (this being an act of ego, which is also a wholly cultural phenomenon) and enough of their peers agree, over time, to use that designation in their own communications/papers/descriptions of the phenomenon, then the name sticks.
I suppose one could call it historical rather than cultural if one wanted to, but then history is a manifestation of a cultural phenomenon as well.
Not everything that functions as a law in science is actually called a law. E=mc^2, for example, is not to my knowledge commonly called “Einstein’s Law” or anything like that, though it functions no differently than stuff like Newton’s Laws of Motion, or Boyle’s Law. The statistical equations used in evolutionary science to calculate selection coefficients, time to fixation, and so forth are also not to my knowledge generally called laws, but they too function as laws.
My understanding is that gravity moving at the speed of light, through the propagation of gravity waves, is a prediction of general relativity.
I don’t know if this particular prediction has been experimentally verified or not, but I have heard of a lot of excitement in the physics community at various points regarding big experiments to look for gravity waves.
More dishonest lying from the troll, I see.
The OP states that “theories and laws” are different beasts. And they are. Theories and hypotheses may exist on a gradient, but laws are separate.
There is no issue of semantics there. Laws and theories are not the same. Period. Full stop. Theories are not bigger laws, or a compilation of a bunch of laws. Theories may *contain* multiple laws (and good theories DERIVE multiple laws) but the meat of any theory is the explanatory framework that links the laws together, explaining all of them as a subset of or a consequence arising from, a greater underlying phenomenon or process. Laws on the other hand, NEVER contain such an explanatory framework (the framework may exist and everyone may associate it with the law, but it is not part of the law itself), whereas theories MUST have that framework. Semantically quibbling over the terminology used to describe the difference between laws and theories doesn’t change the reality that they are different, even if the language used to describe the difference in any particular instance may not be perfectly ideal or appropriate.
David Marjanović:
I never said it was. As I explained several times, I was pursuing somebody else’s line of thought, to try to highlight its problems.
This requires some explanation. I don’t know how to respond to this, because I can’t imagine how you can think this way.
Nerd of Redhead:
I believe in this case PZ is wrong. Is this some kind of crime?
Yet you go on to say
which looks a lot to me like an endorsement of my criticism of PZ’s original post. You think the terms describing the components of the scientific process are not well defined, yet you are offended by the idea that it might be wrong to question somebody’s understanding of science when their account of various scientific terms doesn’t match some supposedly well-established standard! And I’m the one repeatedly being accused of intellectual dishonesty!
Amphiox:
The manner in which a small number have banded together to try to intimidate and bully their way to dominance, with unsupported language like this, in what ought to be an intellectual discussion, is quite disappointing to me, particularly in a community like this. I can see that you love to revel in the self-appointed role of PZ’s bulldogs, but if that is how you derive fulfilment, you might try to reflect on whether this is productive or healthy.
Can you give an example of a theory whose explanatory framework consists of something other than laws / basic propositions? If you can do that, you’ll win this argument instantly.
Laws are the explanatory framework.
@Amphiox 117
No direct observation of Gravity waves, but the Hulse Taylor pulsar measurement is usually accepted as evidence that gravity waves exist and work as advertised. It was good enough for the Nobel committee anyways.
Nope, in chemistry laws are the mathematical descriptions based in empirical studies. They don’t explain, just allow you to calculate. The theories, like the kinetic theory of gases, molecular theory, and the theory of conservation of energy and mass explain. Why are you being deliberately obtuse? You should be smart enough to grasp the concept. Unless, your ego is in the way.
My point is that nobody in this thread has claimed that being accurate is a necessary criterion for being a theory. I’m fully aware that you’ve tried to pursue somebody else’s line of thought; but you’ve misunderstood it and are therefore misrepresenting it.
At that point I was too exhausted to explain everything all over again.
First of all, you equate laws and hypotheses. That’s already wrong. Hypotheses, like theories (only on a smaller scale), explain laws; they aren’t laws and don’t grow up to become laws.
I don’t know enough about special or general relativity to tell if they could be decomposed into distinguishable hypotheses; I doubt that’s the case, though.
That the definitions aren’t as precise as they could theoretically be doesn’t mean there are no definitions at all, let alone that the terms can mean everything and its opposite.
Banded together? Do you think Amphiox and Nerd of Redhead coordinate their “attacks”???
Then explain to me how the theory of evolution by mutation, selection and drift consists of laws. :-| It’s the connection between the laws that makes a theory; a pile of unconnected laws is not a theory.
And no, not every basic proposition is a law.
…Wait, I found two more examples! Theories of combustion:
1) Phlogiston exists. Substances that can burn contain phlogiston, and above some temperature that depends on the substance the phlogiston escapes, leaving a combustion product that weighs less than the original substance – if you could weigh the escaped phlogiston, the weights would add up to that of the original substance.
2) Oxygen exists; phlogiston does not. To burn means to react with oxygen, which happens above some temperature that depends on the substance. That leaves combustion products which together weigh as much as the original substance plus the oxygen; some common combustion products are gaseous, don’t forget to weigh them, too.
3) Phlogiston exists. Oxygen exists in a sense, but it’s just air without phlogiston. Substances that can burn contain phlogiston, and above some temperature that depends on the substance the phlogiston escapes, leaving a combustion product that weighs more than the original substance, because phlogiston has negative mass – if you could weigh the escaped phlogiston, the weights would add up (well, subtract down) to that of the original substance.
I see one law in here, the same in all three: the conservation of mass. And yet, there are three different theories that explain how this law applies to combustion.
LOL, that’s what I get for writing in real time as I think! Three more examples, not two. :-þ
David Marjanović
Perhaps you should come in again.
PS:
@David Marjanović
I’m ready to discuss this issue if you want (parsimony vs induction). I forget about it entirely.
http://freethoughtblogs.com/pharyngula/2014/12/25/thunderdome-57/comment-page-1/#comment-902679
If you want.
Oh! Certainly! The trouble is just that I’m very busy today (yep, on a Sunday) and likely for a few days more; it might be till next weekend till I actually get to it.
I don’t understand…
David Marjanović, your recent article about phlogiston reminded me of the electronics industry claim that all electronics works because of SMOKE inside. We know this is so, because electronics don’t work anymore when you let the smoke out. Any piece of electronics releasing smoke don’t work anymore. No working piece of electronics releases smoke, so smoke must be what makes it work, donchanoe.
David M:
It’s a Monty Python reference (Nobody expects the Spanish Inquisition). How long have you been posting to the web?
David Marjanović
https://m.youtube.com/watch?v=vt0Y39eMvpI
come in again
—-
johnharshman
bless your heart