The "Stage Theory of Theories" - Cause for Concern?

[foghorn]

To think, you need food (Observation) for thought.
You seem to be saying the scientific method is just a a pile of log books with no ‘thinking’ allowed, I don't think that's the case.

 

[axocanth]

To think, you need food (Observation) for thought.
You seem to be saying the scientific method is just a a pile of log books with no ‘thinking’ allowed, I don't think that's the case.

No, that's not what I'm saying at all. Once again, what I'm saying is there is no such thing. If I were saying what you think I'm saying I would be asserting "There is such a thing as the scientific method and [it is inadequate in one way or another]".

Let me try to be more clear.

You'll have no difficulty whatsoever in finding some Tom, Dick or Harry's idea of what The Scientific Method (TSM) is supposed to be. Everybody and his dog has one! First thing you'll notice very quickly is that they do not agree with each other (see also "The Theory of Evolution"). That's the first thing that should raise suspicion. Second thing you'll notice -- if you make the effort and get the history books out -- is that they do not map onto actual scientific practice.



Let's begin with the kind of thing commonly seen in elementary textbooks and elsewhere (see post #226). This might most aptly be described as a "bad joke", indeed Stuart Firestein does describe it as such, the kind of thing fit for children and kittens. No serious thinking person pays it any mind. But I'll point out just two problems with it for now:

(i) Vagueness: It's likely gardening fit the description given; it's likely astrophysics does not. Astrophysicists, unlike gardeners, can hardly go around banging galaxies into each other to see what happens. Some areas of science do not involve experimentation (e.g. systematics); gardening does.

A problem encountered time and time again with characterizations of TSM is what we might call the "porridge test". Make your criteria too vague or permissive and a great deal of what we do not intuitively think of as science gets included, perhaps even things that you would certainly not want included. Gardening gets in to the science club, police work and plumbing might get in, and perhaps even your worst nightmares (e.g. ID and Creationism). If ID proponents do not conduct experiments now, it would be the easiest thing in the world for them to start doing so tomorrow. It may turn out that virtually the whole world is following TSM except astrophysicists and systematists -- a result not many would be willing to embrace. And if the whole world is following TSM, indeed was following it before Galileo, why are we calling it the scientific method? Why not the goatherd method?

(Einstein does indeed say that scientific reasoning is nothing but everyday reasoning with a bit more refinement.)

On the other hand, if you make your criteria too restrictive, a great deal of what we would intuitively regard as good science gets excluded. You might find -- assuming you took these criteria seriously -- that Einstein, Darwin, and Galileo weren't doing science.


(ii) We are not given a methodical procedure for carrying out the steps, especially the critical Step 2 (form a hypothesis), as Firestein also notes. We have to agree that the word method properly applies to certain procedures, and not to others, otherwise we degenerate into triviality. How would you feel, for example, if you coughed up 50 bucks for "My Secret Method for Winning the Lottery" at a bookstore, only to be told . . . Step 1: Buy a ticket, Step 2: Compare your number against the results; Step 3: Collect the dosh if you win.

You call that a method !!??



Moving on to more serious thinkers. Those who have contemplated and written on TSM include Descartes, Bacon, Newton, Mill, Whewell, Mach, Poincare, Duhem, logical positivists, Einstein, Popper and many many others. As noted at the start, you'll find precious little agreement among them, and on close scrutiny of actual scientific practice you'll find that they do not hook on to what actual scientists are doing.



I'd just close by noting that asserting -- as I do and many other thinkers do (including top notch scientists themselves) -- there is no such thing as TSM often provokes the most bizarre and almost hysterical reactions. It may be said that I'm rejecting science itself, or that I'm implying the whole edifice of science is a "house of cards" (pretty sure I saw that one in another thread yesterday). How these people arrive at these conclusions I have not the foggiest idea.

It is not being asserted that there is no such thing as science, or that science is a load of bollocks. Indeed, it's embarrassing to have to say this at all, but certain people do arrive at these ridiculous conclusions for reasons known only to themselves. What is being asserted -- for the gazillionth time -- is that there is no single, timeless, all-encompassing Method of science, used by all (good) scientists in all times, all places, and all scientific disciplines, and used only by scientists.


Two questions to end:

1. Would you even want there to be such a thing? Suppose it might be possible to cure cancer, but to do so would involve departing from the canons of The Scientific Method. What now, chaps? Forget cancer, let's get back to banging hadrons together! Would you want scientists to be constrained in their attempts to carry out whatever tasks they have assigned themselves?

2. Do you think the Beatles were following a step-by-step Method for generating songs? Would you describe the Beatles as "a successful enterprise"?

 

[axocanth]

Cf. (emphasis added)


"Not only does the fact that the standards of scientific success shift with time make the philosophy of science difficult; it also raises problems for the public understanding of science. We do not have a fixed scientific method to rally around and defend. I remember a conversation I had years ago with a high school teacher, who explained proudly that in her school teachers were trying to get away from teaching just scientific facts, and wanted instead to give their students an idea of what the scientific method was. I replied that I had no idea what the scientific method was, and I thought she ought to teach her students scientific facts. She thought I was just being surly. But it's true; most scientists have very little idea of what the scientific method is, just as most bicyclists have very little idea of how bicycles stay erect. In both cases, if they think about it too much, they're likely to fall off."

-- Steven Weinberg (from "Facing Up", essay 8, "The Methods of Science . . . and Those by Which We Live")




"I know enough about science to know that there is no such thing as a clear and universal "scientific method". All attempts to formulate one since the time of Francis Bacon have failed to capture the way that science and scientists actually work. Still, under the general heading of scientific method, we can understand that there is meant a commitment to reason, often though not necessarily crystalized as mathematics, and a deference to observation and experiment. Above all, it includes a respect for reality as something outside ourselves, that we explore but do not create."

-- Steven Weinberg (from "Facing Up", essay 4, "Confronting O'Brien")




"We fundamentally agree with what [Paul] Feyerabend says about the scientific method, considered in the abstract: "The idea that science can, and should, be run according to fixed and universal rules, is both unrealistic and pernicious." "

-- Alan Sokal, "Cognitive Relativism in the Philosophy of Science" in "Beyond the Hoax" p198




"The scientific method has acquired a certain hold on the popular imagination. Every adult remembers something about the scientific method from high school classes; it figures prominently in textbooks with such titles as Reasoning Together, and it is a polemical bruiser in its weight class, useful under circumstances when members of the scientific community are persuaded they are under attack. It is then that the determination is made that members of the public have failed to understand the scientific method or properly to revere it. No effort need be made actually to exhibit the method or tie it to an argument." [ . . . ]

"I will draw down the curtain of charity over this scene. Golf has no method beyond the trivial. Neither does science."

-- David Berlinski, The Devil's Delusion, p55

 

[TheVat]

On the other hand, in terms of understanding, knowledge, explanation of the unobservable entities and processes responsible for the observed phenomena. I think we do have to write off Newton as a complete loss. On the assumption that the core posits of his theory (absolute space, absolute time, attractive force, etc.) do not refer -- i.e., do not exist (see post 190) -- then it's in precisely the same predicament as the phlogiston or the Santa Claus theory. That is, it explains nothing, and yields no genuine knowledge or understanding at all. Would you agree with this?

Having said all that, it does seem that something more than just instrumental power is carried over from the predecessor to the successor theory. We don't merely have a better tool now; Newton latched onto something real in nature. But what is this "something more" if not the unobservable entities and processes?

I will have to skip a day here, got painted into a temporal corner (not kitten generated, in this case), so will try to wrangle this in a day or two. Keep your lamps phlogisticated!

 

[foghorn]

No, that's not what I'm saying at all. Once again, what I'm saying is there is no such thing. If I were saying what you think I'm saying I would be asserting "There is such a thing as the scientific method and [it is inadequate in one way or another]".

I’m sorry my posts seem short compared to your posts.
What do you think the origin of models in science, especially in physics is?
Does it include observation, thinking, predicting and agreement with others?
Is it common sense, and so shouldn’t have a title such as Scientific method?

What is a ‘galaxy,’ is there an agreement somewhere of what a galaxy is?
Moving galaxies, is there an agreement somewhere that galaxies move?
Can these ideas lead to the prediction that galaxies can collide?
Where are these agreements made and on what evidence?

Models are built on models--- Would you say the stars in those galaxies are made of plasma / atoms? Is it right that the model of red/blue shift be used as a tool?

Would you say Einstein’s fame had nothing to do with his model’s predictions?

 

[axocanth]

What do you think the origin of models in science, especially in physics is?

I'd go along with Einstein, viz., it's a creative rather than a methodical process. If you feel, by contrast, that all scientists in all times and all places are not only constructing their models methodically, but using the same method, the world would very much like to hear what this method is. No one has ever been able to identify such a thing, to my knowledge at least.

You might pause to wonder, say, why it took Einstein a tortuous eight years or so to complete his GR model/theory. Why do you think it took so long if the construction of a model or theory is simply a matter of following the steps of a Method?

Does it include observation, thinking, predicting and agreement with others?
Is it common sense, and so shouldn’t have a title such as Scientific method?

No doubt these things are involved, but you'd have to be far more specific. Observation and thinking, for example, are so vague as to be utterly trivial. Who doesn't observe and think?

re common sense, Einstein says this: “The whole of science is nothing more than a refinement of everyday thinking.”

So yes, if scientific reasoning at the broadest level is not essentially different from everyday modes of reasoning (inductive, deductive, explanatory inferences, etc.) -- as I believe to be the case -- I see no justification for science claiming these modes of inference as their exclusive property. Do you? Science can scarcely claim priority on these modes of inference if they've been used by humankind since time immemorial, and continue to be used by the entire human race.

Would you say Einstein’s fame had nothing to do with his model’s predictions?

Certainly not! Einstein shot to worldwide fame in 1919 precisely as a result of one of general relativity's predictions (light bending) being confirmed in a solar eclipse expedition. How this, and your other remarks, relates to The Scientific Method, however, eludes me.



Perhaps it's still not clear what's being claimed here, so I'll say a bit more. You might describe what we're doing as a search. We're trying to identify a method that -- more or less -- has been used by all scientists, and only scientists, in all times and all places. Now, the only way we can find out whether such a method exists is by examining actual scientific practice, both past and present. If we identify this all-unifying Method then our search is successful. We've found it!

Many scholars have conducted such a search and come back empty handed. There does not seem to be anything both common and unique to all scientific practice in terms of methodology. Actual scientific practice just doesn't match up with what Isaac Newton says, or Karl Popper says, or anyone else says.

It could be that scholars just need more time to identify this elusive Method. This is certainly a logical possibility, and cannot be dismissed entirely. As time goes by though, and scholar after scholar comes back empty handed, it begins to appear more and more as if they're on a wild goose chase -- there's no such thing -- and indeed many scientists and probably most philosophers of science already accept this conclusion.

Either way, the heart will go, and science will go just as it did before. Rest assured!

 

[TheVat]

Many scholars have conducted such a search and come back empty handed. There does not seem to be anything both common and unique to all scientific practice in terms of methodology.

Some scholars have come back with hands full. Checking scholarly hands does not a good argument make. Also, and we may have chatted on this elsewhere, calling it a A Method seems like a straw man. Why can't science proceed to make its hypotheses, observations and experiments with a wide array of methods that vary from discipline to discipline and are not individually unique to science - these methods fitting loosely under an umbrella we call science? I just don't understand why method has to be a singular thing that wears a tight lexical corset when the actual world out there is so robustly complex and multilayered.

One handle on meaning in the methods of science might be through Wittgenstein's concept of Family Resemblance...

Family resemblance - Wikipedia

en.wikipedia.org

Family resemblance (German: Familienähnlichkeit) is a philosophical idea made popular by Ludwig Wittgenstein, with the best known exposition given in his posthumously published book Philosophical Investigations (1953).[1] It argues that things which could be thought to be connected by one essential common feature may in fact be connected by a series of overlapping similarities, where no one feature is common to all of the things. Games, which Wittgenstein used as an example to explain the notion, have become the paradigmatic example of a group that is related by family resemblances.

 

[axocanth]

Also, and we may have chatted on this elsewhere, calling it a A Method seems like a straw man. Why can't science proceed to make its hypotheses, observations and experiments with a wide array of methods that vary from discipline to discipline and are not individually unique to science - these methods fitting loosely under an umbrella we call science?

Why can't science do this, you ask? It can. And it does.

But if science uses a vast array of methods, varying from discipline to discipline and from time to time, new methods being introduced and old methods discarded constantly, then the definition of "The Scientific Method" -- as I've been characterizing it -- is not satisfied, and we can logically infer that TSM does not exist.

How would you characterize TSM if not as I have done?

I just don't understand why method has to be a singular thing that wears a tight lexical corset when the actual world out there is so robustly complex and multilayered.

If TSM -- as I've been characterizing it -- was real, it would be a precious thing indeed. It would serve the following functions:

* Unification: It unifies all the various scientific disciplines.

* Explanation: It explains the success of science. Science is successful (where astrology, say, is not) because of this Method that is unique to science.

* Demarcation: What distinguishes bona fide science from non-science, pseudoscience, metaphysics (insert other horrors here) is precisely is adherence to The Scientific Method


If there is no such thing as a timeless, unchanging Method of science, none of these functions is served. If you're going to dismiss Intelligent Design, say, as non-science, you'll have to appeal to something other than method.

 

[TheVat]

If TSM -- as I've been characterizing it -- was real, it would be a precious thing indeed. It would serve the following functions:

* Unification: It unifies all the various scientific disciplines.

* Explanation: It explains the success of science. Science is successful (where astrology, say, is not) because of this Method that is unique to science.

* Demarcation: What distinguishes bona fide science from non-science, pseudoscience, metaphysics (insert other horrors here) is precisely is adherence to The Scientific Method

Wittgenstein was key, in my reply. TSM does not have to fit that shortlist you've got there. It is simply some methods and models with a Wittgensteinian family resemblance. It doesn't have to unify any more than "game" unifies chess, hopscotch, skeet shooting, pissing contests, and kittens pouncing on hallucinated butterflies. It doesn't have to explain success - consistent results and accurate predictions explain success. (they ARE success) Nor does a method need to distinguish good science from bad - any jackass can use a method and award himself a seal of approval. The demarcation lies not in some particular method but in, again, that which supplies evidence, solid predictions, and reproducible results. (all of which pseudo science has in short supply, if any at all) Science is about results not enshrined Method. Which is precisely why any semantic family of methods can only hang together in that Wittgensteinian sense. Expecting more is like nailing jello to the wall.

 

[axocanth]

Wittgenstein's point, if I understand correctly, is that certain concepts -- e.g. game -- have no common essence. There is nothing common to all activities to which we apply the word "game".

Now, it may well be the case that the Wittgenstein analysis applies to the concept "science" too.

But isn't this precisely the wrong conclusion from the one that defenders of TSM want? After all, they are claiming that there is a common essence to all (good) scientific practice, and that common essence just is its Method.

What do you think?

There cannot be a common "method" (or anything else) to all games or to all science if, as Wittgenstein argues, they have nothing universally in common.

 

[foghorn]

I'd go along with Einstein, viz., it's a creative rather than a methodical process. If you feel, by contrast, that all scientists in all times and all places are not only constructing their models methodically, but using the same method, the world would very much like to hear what this method is. No one has ever been able to identify such a thing, to my knowledge at least.

You might pause to wonder, say, why it took Einstein a tortuous eight years or so to complete his GR model/theory. Why do you think it took so long if the construction of a model or theory is simply a matter of following the steps of a Method?

Thinking involves the brain, so if someone does ‘creative’ thinking that would come under thinking.
Einstein took sometime to come up with General Relativity because he was stuck instinctively in flat spacetime.
If Einstein wanted to go further and see if his model described Nature, that would mean an observation to see if the amount of ‘bending’ of light near the sun, was that which his model predicted.
So, that is, the first observation, which would be knowing the expected outcome of the then existing ideas.
In the eclipse case of back then, that would be the expected amount of light ‘bending’ in a flat space time.
The rest is history.

So, Observation, thinking, finding a way to mathematically model his idea, predicting an outcome in Nature and checking for that outcome by observation.
Thinking means the things the brain does to come up with answers, you listed them in your post #246.

 

[axocanth]

So, Observation, thinking, finding a way to mathematically model his idea, predicting an outcome in Nature and checking for that outcome by observation.
Thinking means the things the brain does to come up with answers, you listed them in your post #246.

Again, this all strikes me as trivial and vacuous -- look, think, and test ideas -- hauling in the entire human race, and perhaps cats and dogs too, in its net. We all do that! It's far too permissive.

Not sure whether I told TheVat before, but sometimes I would mischievously leave my (dearly departed!) cat's food bowl in a place other than its usual location, then watch him as he apparently put on his feline thinking cap, entertained various hypotheses ("Perhaps that bastard put it in the bathroom") and proceeded to test them. Are we to say then that Mimee joins the illustrious ranks of Galileo, Newton, and Einstein in doing science?

But two things to note: the way you describe ("finding a way") Einstein's tortuous path to GR sounds quite accurate to me, indeed the way Einstein describes it himself on countless occasions, i.e., fumbling around in the dark, trying out different things, guided only by a few heuristics that his completed theory would have to satisfy, until voila!

But surely fumbling around in the dark, etc. is the kind of thing people do in the absence of a set of a clear set of instructions -- a method -- to guide them. Budding chefs following a recipe in a cookbook do not need to grope around in the dark. Einstein could not be more clear on this point: the construction of a scientific theory is not a logical or methodical process, at least in his view. (Once you have a theory, that's another matter: propositions can be logically deduced and tested, Einstein goes on.)

The only part of your characterization which is not trivial, in my opinion, is your inclusion of "mathematical models". Clearly, the whole world does not go around constructing mathematical models. But now the "porridge test" rears its ugly head again. Equally clearly, not all scientists go around constructing mathematical models either. Much of what we intuitively regard as good science is qualitative as opposed to quantitative, and unless I'm mistaken, you will not find a single mathematical equation in The Origin of Species.

If mathematical armature is part of The Scientific Method, and to do science just is to follow TSM, then we arrive at the conclusion that Charles Darwin and many others beside are/were not doing science -- a bullet not many are willing to bite.

 

[TheVat]

There cannot be a common "method" (or anything else) to all games or to all science if, as Wittgenstein argues, they have nothing universally in common.

Quite. There are similarities that somewhat span the range of fields we call science, but there is no one universal binding method. I guess the exception, by virtue of its great vagueness, is observation. At that point it's worth asking if we can call observation a method, or is it just a broad verb that is loosely descriptive? In any case, observing is hardly confined to the disciplines of science, and so it can't really fulfill anyone's dream of A Method. We are left, it seems to me, with Wittgenstein family resemblances, all the different fields like cousins who seem familiar, or games that all involve a certain playfulness with the faculty of perception and a penchant to tinker with innards.

 

[axocanth]

Quite. There are similarities that somewhat span the range of fields we call science, but there is no one universal binding method. I guess the exception, by virtue of its great vagueness, is observation. At that point it's worth asking if we can call observation a method, or is it just a broad verb that is loosely descriptive? In any case, observing is hardly confined to the disciplines of science, and so it can't really fulfill anyone's dream of A Method. We are left, it seems to me, with Wittgenstein family resemblances, all the different fields like cousins who seem familiar, or games that all involve a certain playfulness with the faculty of perception and a penchant to tinker with innards.

Notice also, Wittgenstein's analysis, assuming it holds water and applies to science as well as games, implies not only that you will find no feature common to all games and to all good scientific practice -- no internal essence of science, as it were -- but also any feature of some instances of good scientific practice (e.g. experimentation) is likely to be found externally, in extra-scientific activities, gardening perhaps.

By analogy, you will not find that all members of the Jones family have blue eyes, say, and whatever color of eyes any particular Jones family member has will be found in other families too.

Wittgenstein may have been ahead of his time. I'm not sure if you've kept abreast of research in the area of concepts in recent decades. Whereas for centuries or more it was assumed that our concepts (e.g. BIRD) were encoded in our brains as lexical definitions (e.g. feathered biped), this now seems very unlikely to be the case.

It seems, instead, that concepts such as BIRD are encoded as something like an exemplar or a prototype, e.g. a sparrow-ish beastie.

If the concept BIRD was encoded in definitional form (e.g. feathered biped) all birds would be equal, as it were. But they're not. Subjects in experiments take longer to determine that a cassowary, say, falls under the concept BIRD than something more common-and-garden like a thrush.

All birds are equal, but some are more equal than others!


Perhaps the mistake all along was to suppose that science could be captured adequately in definitional form at all. How do we decide whether or not homeopathy, say, falls under the concept SCIENCE? By comparing it against a definition? Or by comparing it against an exemplar (e.g. some dude with unkempt hair who looks a lot like Einstein mixing chemicals together in the lab)?

Perhaps we've been on a wild goose chase - which ranks somewhere between a wild sparrow chase and a wild cassowary chase.

 

[James R]

You might pause to wonder, say, why it took Einstein a tortuous eight years or so to complete his GR model/theory. Why do you think it took so long if the construction of a model or theory is simply a matter of following the steps of a Method?

A general method is never a complete description of all the steps in a process.

For example, if you follow a recipe to cook a cake, the recipe won't tell you where to get the cup you need to measure the sugar (or where to get the sugar). It won't tell you what an oven is or how it works. It won't tell you about how sugar cane is grown and harvested so that you can use it to bake a cake. It won't tell you what it means to "mix" ingredients. It won't teach you how to stir something. etc.

In practice, grand Methods tends to skip over the nitty gritty detail. They make a lot of assumptions about prior knowledge, prior attitudes, aims and such.

The Scientific Method no more taught Einstein how to do tensor calculus than a cooking recipe teaches you how to process sugar cane into refined sugar.

On the other hand, when you buy a decent-sized cookbook, often there will be a few pages at the start that set out some general principles and tips for how to cook. The Scientific Method is sort of like those handy hints and good advice.

Also, obviously there's more than one way to bake a cake, which means there's no singular prescribed Baking Method. Similarly, there's more than just one way to do science, so why would we demand that there be a single, prescribed Scientific Method?

Perhaps it's still not clear what's being claimed here, so I'll say a bit more. You might describe what we're doing as a search. We're trying to identify a method that -- more or less -- has been used by all scientists, and only scientists, in all times and all places. Now, the only way we can find out whether such a method exists is by examining actual scientific practice, both past and present. If we identify this all-unifying Method then our search is successful. We've found it!

Many scholars have conducted such a search and come back empty handed. There does not seem to be anything both common and unique to all scientific practice in terms of methodology. Actual scientific practice just doesn't match up with what Isaac Newton says, or Karl Popper says, or anyone else says.

Similarly, if you went looking for an all-unifying method for making cakes, probably your search would be unsuccessful. It is very likely that you will find that the method will vary depending on what kind of cake you want to make, along with many other factors.

This is not to say that you couldn't identify any commonalities or themes in how people typically bake cakes. And the same can be said for doing science.

But two things to note: the way you describe ("finding a way") Einstein's tortuous path to GR sounds quite accurate to me, indeed the way Einstein describes it himself on countless occasions, i.e., fumbling around in the dark, trying out different things, guided only by a few heuristics that his completed theory would have to satisfy, until voila!

To inject a bit of historical fact in here... Einstein only gradually came to the realisation that what he needed mathematically for his General Theory was differential geometry, which was most conveniently formulated using tensors. So he spent quite a lot of time between 1905 and 1915 getting up to speed on what was, at that time (and probably remains today, for most people), a rather obscure area of mathematics that had seen little in the way of practical applications. Einstein got a little help from his friends with the maths.

Einstein tended to downplay the intellectual effort involved in his public interviews about his work. (He didn't need to play it up, since the press were very willing to do that for him.) While he might have described "fumbling around in the dark" or something similar, that shouldn't be taken as an accurate description of what he was actually doing at any time during the development of his Theory of Relativity. "Guided by a few heuristics" is also something of an understatement of his actual work flow.

But surely fumbling around in the dark, etc. is the kind of thing people do in the absence of a set of a clear set of instructions -- a method -- to guide them. Budding chefs following a recipe in a cookbook do not need to grope around in the dark.

Chefs come to a recipe already knowing what an oven is, what sugar is, how to stir stuff in a bowl, and such. Similarly, Einstein came to General Relativity with the preliminary work of the Special Theory behind him, not to mention all of his prior training in physics and mathematics, including his PhD.

Einstein could not be more clear on this point: the construction of a scientific theory is not a logical or methodical process, at least in his view. (Once you have a theory, that's another matter: propositions can be logically deduced and tested, Einstein goes on.)

The scientific method is often called a hypothetico-deductive method. To call this "not logical or methodical" is bizarre, if you ask me. That's not to say that it is not also a creative process that often involves imagination (or at the very least mental visualisation).

Wittgenstein may have been ahead of his time. I'm not sure if you've kept abreast of research in the area of concepts in recent decades. Whereas for centuries or more it was assumed that our concepts (e.g. BIRD) were encoded in our brains as lexical definitions (e.g. feathered biped), this now seems very unlikely to be the case.

It seems, instead, that concepts such as BIRD are encoded as something like an exemplar or a prototype, e.g. a sparrow-ish beastie.

If the concept BIRD was encoded in definitional form (e.g. feathered biped) all birds would be equal, as it were. But they're not. Subjects in experiments take longer to determine that a cassowary, say, falls under the concept BIRD than something more common-and-garden like a thrush.

All birds are equal, but some are more equal than others!

Interestingly, our mental categories do not always correspond meaningfully to more natural ways of classifying things. For instance, biologists and geneticists will sometimes say, with justification, that there's no such thing as a fish. Why? Because lots of animals that superficially look very similar turn out to be wildly divergent in their ancestries, genetics, internals, and such.

Most of us have a mental template of FISH that involves fins, a tail and eyes, at a minimum. Historically, whales were referred to as great fish. But these days it is well documented and widely publicised that a whale is fundamentally different from a trout in many important ways, as far as the evolutionary sciences are concerned - to the extent that even the general public is aware that whales aren't 'really' fish. But, as in the case of axocanth's BIRD example, when we drill down into the anatomy, or the genetics, of lots of types "fish", we find that certain "fish" are much more closely related to other groups of animals than they are to one another. As a biological classification, the label 'fish' is almost useless, because there's no way to list the anatomical characteristics of 'fish' in such a way that includes all of the animals we would ordinarily want to include in the category while simultaneously excluding all of the kinds of animals that we would ordinarily want to exclude from the category.

Perhaps the mistake all along was to suppose that science could be captured adequately in definitional form at all.

It really comes down to what your criteria for 'adequacy' are.

How do we decide whether or not homeopathy, say, falls under the concept SCIENCE?

A good start might be to consider what it has in common with accepted sciences and what differences it has from those.

For instance, most of us accept that medicine is a kind of science. Homeopathy is often referred to, by its practitioners, as alternative medicine. But what does that mean?

The great Tim Minchin suggested that "alternative medicine" is 'medicine' that has either not been proven to work, or which has been proven not to work. That seems like a reasonable starting point to me.

 

[axocanth]

James, I just want to focus, for now, on one thing in your post above:

Einstein could not be more clear on this point: the construction of a scientific theory is not a logical or methodical process, at least in his view. (Once you have a theory, that's another matter: propositions can be logically deduced and tested, Einstein goes on.)

To which you respond . . .

The scientific method is often called a hypothetico-deductive method. To call this "not logical or methodical" is bizarre, if you ask me. That's not to say that it is not also a creative process that often involves imagination (or at the very least mental visualisation).

The scientific method is often called this, that, and the other. Hypothetico-deductivism is just one of many things that it is called, and like all the other suggestions (e.g. inductivism), it comes with problems galore. Newton, for example, did not see himself as following the H-D method, indeed he and other inductivists (common at the time) explicitly expresses contempt for hypotheses: they have no place in good science!

But since you brought it up . . .


The hypothetico-deductive method, as commonly understood, at least as I understand it, is a method for testing scientific theories, not for their construction. Again, as typically formulated, it offers no advice whatsoever on how to construct a theory. Rather, it tells you what to do once you have a theory (see top of this post again).

Einstein tells us explicitly time and time again that, in general, there is no logic or method for the construction or "setting up" of a physical theory and, in particular, that he himself did not follow any logic or method in the construction of general relativity. E.g.

"I have learned something else from the theory of gravitation. No ever so inclusive collection of empirical facts can ever lead to the setting up of complicated equations. A theory can be tested by experience, but there is no way from the experience to the setting up of a theory"


Some questions then:

Q1: You are evidently telling us -- his own repeated denials to the contrary notwithstanding -- that Einstein was following a method in the construction of GR. Is this your view? Proceeding on that assumption (ignore the following questions otherwise) . . .

Q2: Clearly, then, if you're right, Einstein was mistaken. Why do you think Einstein is mistaken about this? Why would he deny that he was following a method while you affirm that he was?

Q3: Can you please state this method for us? What method do scientists in general, and Einstein in particular, use to construct their theories?

 

[James R]

axocanth:

The scientific method is often called this, that, and the other. Hypothetico-deductivism is just one of many things that it is called, and like all the other suggestions (e.g. inductivism), it comes with problems galore.

All labels and descriptions come with problems galore if you want to go looking for them. This is something that certain philosophers love to spend lots of time on. (Not judging, just saying. Good luck to them.)

Newton, for example, did not see himself as following the H-D method, indeed he and other inductivists (common at the time) explicitly expresses contempt for hypotheses: they have no place in good science!

I'm not really aware of how well developed these kinds of categories and descriptions were in the 1600s. At that time, "science" itself was only starting to be clearly distinguished from other kinds of intellectual pursuits. "Natural philosophy" was a thing - it's in the title of Newton's most famous book - but "physics" wasn't, exactly.

Newton spent a lot of his time trying to "decode" the bible and trying to turn lead into gold, neither of which is well regarded in scientific circles today.

In other words, I'm not sure we should be over-reliant on Newton's opinion when it comes to what makes for good scientific method.

The hypothetico-deductive method, as commonly understood, at least as I understand it, is a method for testing scientific theories, not for their construction. Again, as typically formulated, it offers no advice whatsoever on how to construct a theory. Rather, it tells you what to do once you have a theory (see top of this post again).

Einstein tells us explicitly time and time again that, in general, there is no logic or method for the construction of a physical theory and, in particular, that he himself did not follow any logic or method in the construction of general relativity. E.g.

"I have learned something else from the theory of gravitation. No ever so inclusive collection of empirical facts can ever lead to the setting up of complicated equations. A theory can be tested by experience, but there is no way from the experience to the setting up of a theory"

The H-D "method" can 'begin' at different points in the 'cycle', which is really more of feedback loop. You can start with experience/observation, then formulate a hypothesis/theory, then test the theory, then tweak the theory based on the results. Or you can start with a hypothesis/theory, then test the theory, then adjust, then observe some more, and so on. This is a continually iterative loop, no matter where it starts.

Einstein is correct that to formulate a hypothesis/theory, one needs to start with an idea about what might be true (or false). That's a creative process.

Seeing an apple fall off a tree does not inevitably lead to a theory of gravity. It takes an Isaac Newton to have an idea and to ask the right kinds of questions before anything further can happen. As for "complicated equations", mathematics is considered by many scientists to be a tool. While some scientific theories start with ideas that come from mathematical insights, I think that more often the mathematical insights come after the initial spark of an idea or the realisation that there is a question that seems to need an answer.

BTW, it's very popular these days for people who don't know much science - especially on the internet - to quote Einstein as saying "Imagination is more important than knowledge". The unexpressed hope of the lazy internet pundit is that they will be able to get away without needing knowledge - that they will get by on imagination alone. They hope in vain. Einstein didn't just have imagination; he also had lots of knowledge, experience and hard-earned expertise.

Some questions then:

Q1: You are evidently telling us -- his own repeated denials to the contrary notwithstanding -- that Einstein was following a method in the construction of GR. Is this your view? Proceeding on that assumption (ignore the following questions otherwise) . . .

In broad brushstrokes, yes, Einstein was following the "scientific method". He studied certain aspects of nature in intimate detail. He had some ideas, which led him to formulate some "what if?" questions. He then drew on his expertise (or, in some case, went out and learned new stuff until he was an expert) to express his ideas and questions in the "language" of science - i.e. to formulate testable hypotheses and to make quantitative predictions about certain observations. Later - in some cases, much later - other scientists tested Einstein's hypotheses and found that his predictions were a better match to observations/experimental results than the predictions of various competing theories (e.g. Newtonian physics).

On the other hand, you seem to be looking for a "nitty gritty" method, so that we can say "Step 1 in The Scientific Method is to ... Then we move on to Step 2, in which we ..." and so on, like a recipe. I'm not saying Einstein followed a step-by-step recipe that anybody could have followed that would have inevitably led to the General Theory of Relativity. Science doesn't work like that. However, a lot of scientists will tell you that they believe that if Einstein hadn't "discovered" his general theory, somebody else would have, sooner or later. That's partly because the General Theory is not a purely creative act of imagination by one man. Einstein stood on the shoulders of giants, as was said of Newton. Science is cumulative. Science is communal. Science is social. It's not a mindless process of following a set of instructions.

I believe I have answered your Q2 and half of Q3 above.

Q3: ... What method do scientists in general, and Einstein in particular, use to construct their theories?

If you're looking for a no-brainer recipe with a set of steps that any idiot would only have to follow to arrive at General Relativity, I guess that, in principle, a set of steps could be written down for that specific example. Obviously Einstein himself went through a series of steps and got to the end. If he could do it, other people can too. (People can be taught about specific scientific theories and how they were developed.)

That's a very low, workmanlike level of specificity for what one might want in terms of a blueprint for constructing a theory. What you are asking for, I think, is a general method for constructing theories about anything (or about anything scientific). I guess that I could start to write down a recipe for such a thing for you, too, but I don't know how helpful it would be in terms of giving you specific advice. It might go something like this (not necessarily in this order):

1. Go outside and observe a natural phenomenon.
2. Come up with a question like "Why does it do that?" or "How did that thing come to be the way it is?" or "What would be a good description of how thing X does action Y, which might be more generally applicable to things other than X?"
3. Take some science courses. Learn what our best current scientific theories have to say regarding the phenomenon of interest.
4. Using your acquired expertise, imagine a plausible answer to your question from step 2.
5. Using your acquired scientific skills, formulate a testable hypothesis - preferably quantitative - that can be used to check whether your answer in step 4 is wrong or right.
6. Flesh out any necessary details required to make meaningful predictions from your "theory" or hypothesis.
7. Test the hypothesis experimentally/observationally, or get somebody else to do it for you.
8. Analyse the results.
9. Compare the results to what your hypothesis predicted. Note any discrepancies between theory and observation carefully.
10. Repeat from step 2.

You will notice that there's nothing explicit here about how to learn tensor calculus, or about why curved spacetime might be a good basis for a theory of gravity. Nothing in this list would point Einstein specifically in the "right" direction to get to General Relativity, specifically. But that's not a problem. You asked for a general method, so I've given you a general method.

Just for comparison, I think this is a bit like asking "Can you give me a general method for writing a symphony? If you can't, then I say there's no such thing as a so-called Musical Method! Game Over! You lose!" But is that true? Are symphonies a product of the purely creative human mind, or is How to Write a Symphony something that somebody can be taught? Can we meaningfully provide some guidelines about better and worse ways to go about writing a symphony? I say we can. I also say this isn't so different from the position with science.

There are better and worse ways to go about doing science that can be taught and learned. It's not random. It's not a "methodless" enterprise. If there was no "method" to it, then taking a science course would be a waste of time in terms of preparing a person for the job of "scientist". At best, it would just be an exercise in rote memorisation of a bunch of "facts" that were arrived at by other people more or less by accident or at random, I suppose due to freakish accidents of imagination.

 

[axocanth]

This to begin with . . .

There are better and worse ways to go about doing science that can be taught and learned. It's not random. It's not a "methodless" enterprise.

No one I know of ever said it was. I certainly didn't. Obviously scientists in different disciplines and different times have used all kinds of specific methods. That's not the question we're addressing, which is: Is there a universal method unique to science and used by all (good) scientists in all times and all places?

Most recently, you and I have been examining the more specific question: Is there a method for constructing hypotheses/theories unique to science and used by all (good) scientists in all times and all places?

In response to my questions, the answer in your post above is . . .

5. Using your acquired scientific skills, formulate a testable hypothesis - preferably quantitative - that can be used to check whether your answer in step 4 is wrong or right.

Ignoring all the peripherals that were not requested, your answer to my question about the method by which scientists formulate a hypothesis/theory is "formulate a testable hypothesis", which obviously sheds no light whatsoever on how this is done methodically. You've simply repeated the question.

Do you consider the instructions "formulate a testable hypothesis" to be a methodical prescription for formulating a hypothesis? I'm not trying to be impolite, but is this not as vacuous as offering the advice "write a good book" as a methodological prescription for writing a book?

No doubt there are certain preconditions that must be satisfied for a book to be written; the budding Dostoevsky would have to be literate for one thing. Likewise, you've effectively told us that to formulate a hypothesis about something, you'd have to acquire some background knowledge. Such preconditions, however, hardly constitute a method for writing a book or forming a hypothesis. Don't you agree?

It seems to me you are trivializing the whole idea of a method.

Since symphonies were mentioned, do you feel, James, that all the great composers were following the very same The Symphonic Method? Once again, you're not going to write a symphony without some training in music -- it's a necessary precondition -- but presumably a grounding in basic music theory won't methodically lead you to Beethoven's 5th.

Do you feel all great (and not so great) authors follow The Novelist Method? - exactly the same method in every case?

Are these not creative processes -- as you've already hinted at yourself -- that defy any kind of formal codification?

We certainly never hear composers or novelists speak of such a thing, just as Einstein tells us there is no method for the construction of scientific theories/hypotheses.

If you don't feel there is such a thing as The Symphonic Method or The Novelist Method, why do you think it is appropriate to apply the word "method" to the construction (note that word!) of scientific theories/hypotheses?

 

[axocanth]

Q1: You are evidently telling us -- his own repeated denials to the contrary notwithstanding -- that Einstein was following a method in the construction of GR. Is this your view? Proceeding on that assumption (ignore the following questions otherwise) . . .

(emphasis added)

and your reply . . .

In broad brushstrokes, yes, Einstein was following the "scientific method". He studied certain aspects of nature in intimate detail. He had some ideas, which led him to formulate some "what if?" questions. He then drew on his expertise (or, in some case, went out and learned new stuff until he was an expert) to express his ideas and questions in the "language" of science - i.e. to formulate testable hypotheses and to make quantitative predictions about certain observations. Later - in some cases, much later - other scientists tested Einstein's hypotheses and found that his predictions were a better match to observations/experimental results than the predictions of various competing theories (e.g. Newtonian physics).

Notice, you are largely addressing issues that were not asked. I didn't ask about subsequent testing or anything else. I asked according to what method he came up with the theory.

For your reference, if you're not already aware, philosophers and scientists too sometimes draw a distinction between the so-called "context of discovery" of scientific hypotheses/theories, and the so-called "context of justification" (subsequent testing, etc.). Right now we're (supposed to be!) addressing the former.

And your answer essentially (ignoring the irrelevancies again): "He had some ideas".

Sounds about right to me, and doesn't anything like a method to me. What about you?

By the way, by far the most valuable source on all this, that I'm aware of anyway, is the magnificent 4-volume "The Genesis of General Relativity".

 

[axocanth]

Just one more thing, somewhat tangential to our current discussion . . .

Science is cumulative. Science is communal. Science is social. It's not a mindless process of following a set of instructions.

Copied from post #1554 (page 78) of the "Intelligent Design Redux" thread . . .


"The two theories of relativity are among humankind's most precious achievements, and with them Einstein toppled Newton's conception of reality. Even though Newtonian physics seemed to capture mathematically much of what we experience physically, the reality it describes turns out not to be the reality of our world. Ours is a relativistic reality. Yet, because the deviation between classical and relativistic reality is manifest only under extreme conditions (such as extremes of speed and gravity), Newtonian physics still provides an approximation that proves extremely accurate and useful in many circumstances [cf. it works - axo]. But utility and reality are very different standards. As we will see, features of space and time that for many of us are second nature have turned out to be figments of a false Newtonian perspective."

- "The Fabric of the Cosmos", Brian Greene, p10




"It was this task to which Einstein boldly dedicated himself, and with the dazzling framework he developed after close to a decade of searching in the dark, Einstein overthrew Newton's revered theory of gravity."

- "The Fabric of the Cosmos", Brian Greene, p64




"To present such new ideas without relating them properly to previously held ideas gives the wrong impression that the theory of relativity is merely at a culminating point of earlier developments and does not properly bring out the fact that this theory is on a radically new line that contradicts Newtonian concepts in the very same step in which it extends physical law in new directions, and into hitherto unexpected new domains"

- David Bohm, "The Special Theory of Relativity", preface, page xvi




"It is now nearly a full century since Einstein destroyed Newton's concept of space and time as absolute, and began laying the foundations for his own legacy."

- Kip S. Thorne, "Black Holes and Time Warps: Einstein's Outrageous Legacy", p523




"A theory that has real predictive content must then, as it were, "stick its neck out". But if it does this it is likely in time to "have its neck chopped off". Indeed, this is what did happen eventually to a great many theories such as Newtonian mechanics, which were confirmed up to a point but then shown to be false."

- David Bohm, "The Special Theory of Relativity", pp 149-150



"We can indeed see from Newton's formulation of it that the concept of absolute space, which comprised that of absolute rest, made him feel uncomfortable; he realized that there seemed to be nothing in experience corresponding to this last concept. He was also not quite comfortable about the introduction of forces operating at a distance. But the tremendous practical success of his doctrines may well have prevented him and the physicists of the eighteenth and nineteenth centuries from recognizing the fictitious character of the foundations of his system."

- Albert Einstein, essay, "On the Methods of Theoretical Physics"




Sorry for the lengthy copying. I thought it might save readers a trip to the other thread.


Now, assuming these writers can be trusted (can they?), it does not sound at all like the switch -- or the revolution, if you will-- from Newtonian physics to relativistic physics was cumulative. We are told not, for example, that Einstein "built on" Newton's work but, quite to the contrary, that he toppled it, he overthrew it, and destroyed it. We are told not that Einstein's theory complements or adds to Newton's theory, but it contradicts Newton's theory. We are told that Newton's theory was shown to be false; it turned out to be a "figment of a false Newtonian perspective".

You have to admit surely that, at least to the uninitiated, cumulative hardly seems consistent with what we've just been told? It sounds a lot more like a process of destruction followed by a complete restructuring, rather than additional construction upon a foundation that itself remains unchanged. Indeed, if you'll forgive me, it sounds more like Einstein took a wrecking ball to the Newtonian edifice!

In what sense, then, do you see this episode as cumulative (assuming you see it that way at all)?

As I said in the other thread, I've no objection to the claim that Einstein's theory represents an improvement of Newton's in terms of its instrumental efficacy -- it works better! I know of no one else who objects to this claim either.

But it's hard to see how this transition represents cumulation of knowledge of how our universe really is.

Do you see this transition -- typical of scientific revolutions -- as an accumulation of knowledge, James? Pure accumulation? Is nothing "lost" in such transitions? Do we understand the unobservable processes (supposedly) at work better than we did before? Or, did we previously have no understanding of them at all? What possible understanding or knowledge of unobservable reality can a fictitious theory (see Einstein above) give us?

I'd be interested to hear your thoughts.