The Stage Theory of Theories

[Cenderawasih]

There is a distinction regularly drawn nowadays--in science textbooks, in the popular media, etc.--between the terms hypothesis and theory. Presumably many of our members have seen or heard it. It goes something like this:

"The scientist begins by proposing a hypothesis. The hypothesis is then tested. If the hypothesis fails the test(s) it is rejected. On the other hand, if the hypothesis passes testing, and thus is now supported by evidence, it may be upgraded to the status of a theory. All theories begin life as a hypothesis."

This what I refer to as the "stage theory" of theories, or STOT for short.

Now, if STOT is supposed to be itself an empirical hypothesis/theory, a "law" if you like, based on a survey of how actual scientists use these terms, then I submit it is clearly false. Close attention to the usage of actual scientists in actual real world situations will quickly reveal its falsity; any doubters are encouraged to keep a hawk's eye on the literature from now on. But a few examples to get us started . . .

1. Never once have I heard anyone, including Einstein himself, refer to general relativity as a 'hypothesis'. This includes even the period prior to its completion when clearly, given its unfinished state, there could be no talk of testing or supporting evidence. It was invariably "Mr Einstein (or "I") is working on a new theory". It would appear, then, contra STOT, that not all theories begin life as a hypothesis.

2. Darwin's ideas on evolution circa 1859 were variously referred to using both terms: hypothesis and theory -- apparently depending only the whim and caprice, or perhaps the breakfast, of the writer involved.

3. We are routinely told by scientists themselves that string theory thus far remains untestable. It is nonetheless invariably referred to as 'theory' and not 'hypothesis'. It would appear, then, contra STOT, that not all theories have been tested and are supported by evidence. Moreover, I've no memory of string theory beginning life as "string hypothesis".


Another way to look at STOT, though is to regard it as prescriptive rather than descriptive. In other words, as opposed to an accurate description of how scientists actually use these terms, it is some pedant's prescription of how these terms ought to be used.

For comparison, consider the linguistic legislators of a past age (thank God!) and their views on sentences containing double negatives and split infinitives. They were obviously not trying to describe how people actually speak, because as we all know, people like William Shakespeare and Captain James T. Kirk do routinely use double negatives and split infinitives. Instead, the pedants were prescribing how language, in their opinions, ought to be used.

In the case of a prescription, unlike an empirical description, the question of truth or falsity does not arise. A suggestion (e.g. "You should drink less"), unlike a statement, is not in the truth-falsity business.

I suspect proponents of STOT are simply confused: they think they are describing the actual usage of scientists, while a cursory glance through the scientific literature quickly belies the myth.

What I'm wondering, then, is how this all came about. Who came up with this stage theory of theories? All I know is that it seems only to have gained currency in recent years. Can anyone out there shed some light on this?

Another suspicion of mine is that STOT may have been conjured up by some frustrated evolutionary biologist sick to death of hearing his Creationist bêtes noire smirk, "Evolution is just a theory".

One can almost hear Richard Dawkins groan, "It's not just a theory, you idiot!!! It is overwhelmingly supported by evidence!!"

LOL

Any comments, critiques, insights welcome.


Edit P.S. Lest anyone get the wrong idea, I'm not a Creationist.

 

[exchemist]

There is a distinction regularly drawn nowadays--in science textbooks, in the popular media, etc.--between the terms hypothesis and theory. Presumably many of our members have seen or heard it. It goes something like this:

"The scientist begins by proposing a hypothesis. The hypothesis is then tested. If the hypothesis fails the test(s) it is rejected. On the other hand, if the hypothesis passes testing, and thus is now supported by evidence, it may be upgraded to the status of a theory. All theories begin life as a hypothesis."

This what I refer to as the "stage theory" of theories, or STOT for short.

Now, if STOT is supposed to be itself an empirical hypothesis/theory, a "law" if you like, based on a survey of how actual scientists use these terms, then I submit it is clearly false. Close attention to the usage of actual scientists in actual real world situations will quickly reveal its falsity; any doubters are encouraged to keep a hawk's eye on the literature from now on. But a few examples to get us started . . .

1. Never once have I heard anyone, including Einstein himself, refer to general relativity as a 'hypothesis'. This includes even the period prior to its completion when clearly, given its unfinished state, there could be no talk of testing or supporting evidence. It was invariably "Mr Einstein (or "I") is working on a new theory".

2. Darwin's ideas on evolution circa 1859 were variously referred to using both terms: hypothesis and theory -- apparently depending only the whim and caprice, or perhaps the breakfast, of the writer involved.

3. We are routinely told by scientists themselves that string theory thus far remains untestable. It is nonetheless invariably referred to as 'theory' and not 'hypothesis'.


Another way to look at STOT, though is to regard it as prescriptive rather than descriptive. In other words, as opposed to an accurate description of how scientists actually use these terms, it is some pedant's prescription of how these terms ought to be used.

For comparison, consider the linguistic legislators of a past age (thank God!) and their views on sentences containing double negatives and split infinitives. They were obviously not trying to describe how people actually speak, because as we all know, people like William Shakespeare and Captain James T. Kirk do routinely use double negatives and split infinitives. Instead, the pedants were prescribing how language, in their opinions, ought to be used.

In the case of a prescription, unlike an empirical description, the question of truth or falsity does not arise. A suggestion (e.g. "You should drink less"), unlike a statement, is not in the truth-falsity business.

I suspect proponents of STOT are simply confused: they think they are describing the actual usage of scientists, while a cursory glance through the scientific literature quickly belies the myth.

What I'm wondering, then, is how this all came about. Who came up with this stage theory of theories? All I know is that it seems only to have gained currency in recent years. Can anyone out there shed some light on this?

Another suspicion of mine is that STOT may have been conjured up by some frustrated evolutionary biologist sick to death of hearing his Creationist bêtes noire smirk, "Evolution is just a theory".

One can almost hear Richard Dawkins groan, "It's not just a theory, you idiot!!! It is overwhelmingly supported by evidence!!"

LOL

Any comments, critiques, insights welcome.


Edit P.S. Lest anyone get the wrong idea, I'm not a Creationist.

I think what you have written makes a lot of sense. In my opinion there is a certain amount of sententious bunk written about something called, grandly, "The Scientific Method". In reality the development of theories often involves an untidy mixture of data, guesses and insights, not in any particular order. What is for sure, though (per Popper and others), is that scientific theories are distinguished from other types of idea about the world by needing to be testable against observation of nature.

I think this is the reason why tidy-minded people use the term hypothesis, to denote a theory that has not yet been put to the test. In reality, Einstein was all the time working with observations of nature that had already been made, trying to fit them into a framework that could account for them. So calling his work merely a hypothesis would seem unfair, especially since it was fully worked out quantitatively, rather than being just a new idea about nature. Darwin's original idea was more of a hypothesis, as it was his idea that perhaps nature could do what he observed animal and plant breeders had been doing for centuries. Collecting data to support the hypothesis and make it into a theory largely came afterwards, though even in his case there seems to have been an overlap between data and insight (Galapagos finches etc.).

You are right, I am sure, that that these distinctions have become emphasised more recently, due to the baleful influence of creationism and the associated disinformation it spreads. Some of the common creationist tropes like "it's only a theory" and the whole "Intelligent Design" scam, have needed to be rebutted by explaining what makes a theory in science a very particular thing, viz. able to make testable predictions about what we should be able to observe.

P.S. It is obvious you are not a creationist.

 

[Cenderawasih]

What a coincidence. Just posted this

http://www.sciforums.com/threads/the-scientific-method-is-useless.106991/page-7#post-3691071

 

[Cenderawasih]

I think what you have written makes a lot of sense. In my opinion there is a certain amount of sententious bunk written about something called, grandly, "The Scientific Method". In reality the development of theories often involves an untidy mixture of data, guesses and insights, not in any particular order. What is for sure, though (per Popper and others), is that scientific theories are distinguished from other types of idea about the world by needing to be testable against observation of nature.

I think you make a lot of sense too, sir. But this "testability" criterion of theoryness doesn't seem to sit very comfortably with string theory, say, which we are told cannot be tested at this stage.

 

[exchemist]

I think you make a lot of sense too, sir. But this "testability" criterion of theoryness doesn't seem to sit very comfortably with string theory, say, which we are told cannot be tested at this stage.

Indeed. I'm far from convinced that string theory is a proper scientific theory - and I'm not alone. It's a cottage industry in mathematical speculation, really, that seems to be going nowhere: https://www.math.columbia.edu/~woit/wordpress/

 

[Cenderawasih]

I think this is the reason why tidy-minded people use the term hypothesis, to denote a theory that has not yet been put to the test. In reality, Einstein was all the time working with observations of nature that had already been made, trying to fit them into a framework that could account for them. So calling his work merely a hypothesis would seem unfair, especially since it was fully worked out quantitatively, rather than being just a new idea about nature.

But, with all due respect, would the same not apply to any theory/hypothesis (or whatever you wanna call it)?

How can one possibly propose a theory/hypothesis in the absence of "observations of nature that had already been made"?

In other words, Einstein was doing what everyone else who proposes a theory/hypothesis was doing, viz., trying to account for the data.

And yet, even before any testing had taken place, even before any testing could take place (because he hadn't even finished), general relativity, even before a name was settled on, was invariably referred to as a "theory".

 

[Cenderawasih]

But, with all due respect, would the same not apply to any theory/hypothesis (or whatever you wanna call it)?

How can one possibly propose a theory/hypothesis in the absence of "observations of nature that had already been made"?

In other words, Einstein was doing what everyone else who proposes a theory/hypothesis was doing, viz., trying to account for the data.

And yet, even before any testing had taken place, even before any testing could take place (because he hadn't even finished), general relativity, even before a name was settled on, was invariably referred to as a "theory".

Now, the only alternative I see is to say that any wotjamacallit (no matter how absurd) which accounts for the data is already supported by said data. In other words, if STOT is correct then every wotjamacallit is a theory . . . and there is no such thing as a hypothesis. That would include, say, gremlins in the attic being responsible those noises I hear at night.

My gremlins are not just a hypothesis!!!!!

Which kinda wreaks havoc with STOT's insistence that every theory begins life as a hypothesis . . . because hypotheses don't exist.

After all, do these noises I hear at night not count as "observations of nature that had already been made"?

 

[Cenderawasih]

Now, obviously I'm being a little flippant with the gremlins: I don't believe in them. But there is a serious point to be made.

If, as you say, exchemist, Einstein's general relativity circa 1915 ought to be instantly awarded theory status on the grounds he was working with "observations of nature that had already been made" . . .

. . . why shouldn't my gremlins conjecture also be awarded the same status?

 

[Cenderawasih]

Oh, and as an afterthought, I wonder what exponents of the stage theory of theories would say about all those abandoned theories (e.g. phlogiston) on the scientific scrapheap . . .

Stage 1: It's a hypothesis with no supporting evidence

Stage 2: It grows muscles, gets upgraded to theory, and kicks sand in the face of wimpy 90-lb hypotheses. After all a theory is that which is supported by lots of evidence, and we have good reason to believe it now.

Stage 3: Um.


What do we say about phlogiston then? Is it still supported by lots of evidence? Or is STOT a lot of rubbish?

Seems to me names aren't that important. Call it a hypothesis, call it a theory, call it George Burns for all I care.

What does seem important to me is whether it's true or not.

 

[exchemist]

But, with all due respect, would the same not apply to any theory/hypothesis (or whatever you wanna call it)?

How can one possibly propose a theory/hypothesis in the absence of "observations of nature that had already been made"?

In other words, Einstein was doing what everyone else who proposes a theory/hypothesis was doing, viz., trying to account for the data.

And yet, even before any testing had taken place, even before any testing could take place (because he hadn't even finished), general relativity, even before a name was settled on, was invariably referred to as a "theory".

I don't think so. There is the example of Darwin, for instance. Or Kekulé. In both cases, sure there was some data that put them on the track, but the insight was a leap in the dark, justified later by evidence. With Einstein's SR, he took existing data (invariance of the speed of light), threw away all preconceptions about nature and then just followed to its conclusion what the data implied.

But all these things are on a kind. of continuum. It is mistake to force the essentially creative process of theory development into a flow diagram.

 

[exchemist]

Now, obviously I'm being a little flippant with the gremlins: I don't believe in them. But there is a serious point to be made.

If, as you say, exchemist, Einstein's general relativity circa 1915 ought to be instantly awarded theory status on the grounds he was working with "observations of nature that had already been made" . . .

. . . why shouldn't my gremlins conjecture also be awarded the same status?

Because you have no observations of nature, unless I'm very much mistaken.

 

[exchemist]

Oh, and as an afterthought, I wonder what exponents of the stage theory of theories would say about all those abandoned theories (e.g. phlogiston) on the scientific scrapheap . . .

Stage 1: It's a hypothesis with no supporting evidence

Stage 2: It grows muscles, gets upgraded to theory, and kicks sand in the face of wimpy 90-lb hypotheses. After all a theory is that which is supported by lots of evidence, and we have good reason to believe it now.

Stage 3: Um.


What do we say about phlogiston then? Is it still supported by lots of evidence? Or is STOT a lot of rubbish?

Seems to me names aren't that important. Call it a hypothesis, call it a theory, call it George Burns for all I care.

What does seem important to me is whether it's true or not.

That's where you are wrong. In science, people strenuously avoid claiming anything is "true" - beyond the observations themselves (suitably confirmed).

You will find that where theories are concerned, they say careful things like "consistent with the evidence".

 

[Cenderawasih]

That's where you are wrong. In science, people strenuously avoid claiming anything is "true" - beyond the observations themselves (suitably confirmed).

You will find that where theories are concerned, they say careful things like "consistent with the evidence".

First of all, thank you for you the stimulating discussion.

Second of all, this is manifest nonsense lol. Are you seriously telling me no scientist has ever claimed anything to be true?

Would you like quotes?

 

[Cenderawasih]

First of all, thank you for you the stimulating discussion.

Second of all, this is manifest nonsense lol. Are you seriously telling me no scientist has ever claimed anything to be true?

Would you like quotes?

The problem is: you think you speak for all scientists.

Your epistemic humility is comforting, but totally at odds with the historical facts.

 

[Cenderawasih]

That's where you are wrong. In science, people strenuously avoid claiming anything is "true" - beyond the observations themselves (suitably confirmed).

Let's begin with Max Planck, shall we?

"A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it." -- Max Planck

 

[Cenderawasih]

"Experiment is the sole judge of scientific truth" - Richard Feynman

Now what was that again about "In science, people strenuously avoid claiming anything is "true" "?

 

[Cenderawasih]

You must listen more closely to your own scientific brehtren, friend, with all due respect.

To claim that scientists never say anything is true is just plain silly.

 

[Cenderawasih]

With Einstein's SR, he took existing data (invariance of the speed of light), threw away all preconceptions about nature and then just followed to its conclusion what the data implied..

Of course he didn't. He simply stipulated that the speed of light is constant . . . and let's see what happens, kinda thing.

"There is, of course, no logical way leading to the establishment of a theory but only groping constructive attempts controlled by careful consideration of factual knowledge." - Einstein

 

[Cenderawasih]

I'm not saying there's anything wrong with what Einstein did. What's wrong is your understanding of what the great man did.

 

[Cenderawasih]

With Einstein's SR, he took existing data (invariance of the speed of light), threw away all preconceptions about nature and then just followed to its conclusion what the data implied.
.

(my emphasis)

Er, more nonsense. What you're telling us here is that data leads inescapably to one conclusion, right? Data implies one and only one conclusion? You just said it.

Ever heard of the "underdetermination of theory by data"?

Ever heard of scientists who arrive at mutually inconsistent conclusions given exactly the same data?