Dear Marcel

Yes, very much so. It is said that 20th century was the century of information and knowledge; and 21st century will be the century of 'Concepts.' Only those who have ideas that can delight the fastidious customer will survive. Before we build concepts, they are in the raw form of ideas--to be developed in teams or individually. Idea accumulation is taking place because it takes a long time for their development into concepts/innovation. For, it requires planning and action. Creative people are limited, who have the motivation to do that. That is the reason that ideas get piled up.

Yes it is. An example is the book Jules Verne. Why is this happening. Generating a new idea does not imply its immediate implementation. Embodiment of the idea implies the availability of resources, technology, manufacturing, new theory. To create these technologies, and production takes time. So there is a gap between the generation of an idea and its embodiment.

Yes that's true, the rate of the production of new ideas is higher than the rate of studies to test these ideas. The reasons for it can be attributed to several factors depending upon the idea itself, availability of required technology, data sets, expertise, and funding etc. Moreover, it is always easy to say but cumbersome to prove it. To prove, one needs resources as mentioned above.

Dear Marcel

Yes, very much so. It is said that 20th century was the century of information and knowledge; and 21st century will be the century of 'Concepts.' Only those who have ideas that can delight the fastidious customer will survive. Before we build concepts, they are in the raw form of ideas--to be developed in teams or individually. Idea accumulation is taking place because it takes a long time for their development into concepts/innovation. For, it requires planning and action. Creative people are limited, who have the motivation to do that. That is the reason that ideas get piled up.

There is not only accumulation of untested ideas in the published literature. Many novel ideas probably remain hidden in (1) manuscripts that were refused for publication (e.g. because there was no enough empirical evidence to justify publication), (2) unpublished thesis manuscripts (e.g. because many students stop research or leave the laboratory before the information is published), or (3) unpublished activity reports (e.g. because many sponsored research projects are short-term and researchers are not given enough time to publish).

Dear Andrei,

In many cases, new ideas presented in Introductions of manuscripts will not be accepted for publication when the ideas are not empirically supported by data. On the other hand, new ideas might often be hidden in discussion sections to provide an alternative explanation for what has been observed.

Everybody agrees we need replicated empirical research. How many times the same idea has been tested twice, at different places, by different teams? Any statistics available?

Example:

Clint D. Kelly 2006. Replicating empirical research in behavioral ecology: how and why it should be done but rarely ever is. The Quarterly Review of Biology 81, 3: 221-236

I agree entirely. A great idea can occur in a matter of seconds. But to test this idea requires time, patience, perspiration and a pile of trash (as Edison puts it).

For second part of the Q;

'Does this imply that the publication of untested (fantastic?) ideas is accumulating in the published literature, and if so why? I refrain from commenting on what other researchers may do. But for myself, I test my ideas empirically.

It depends on the way and rate of application of the idea, it may be fast or slow.

I will give an example related to my field:

Nobel prize in Physiology/Medicine 2012 was jointly awarded to Dr John B. Gurdon and Dr Shinya Yamanaka for the discovery that mature cells can be reprogrammed

to become pluripotent.

The idea first was by Dr Gurdon in 1960s by the process of nuclear transfer using an oocyte that was successfully applied in frogs and finally in Dolly in 1997. Then he was awarded Nobel prize after 50 years.

On the other hand, Yamanaka did his gene mixture (without using oocytes) to turn the cells to pluripotent in 2006 and awarded Nobel prize after 6 years because of advanced materials to apply his idea.

Nowadays, it is very hard to find a "Virgin" idea !!

Dear Islam,

From a philosophical point of view, perhaps it is not so complicated to find a 'virgin' idea accepting that each spatio-temporal event is unique in physical expression and each event can provoke an idea to explain that event. For instance, new physical events are continuously created because of nature's dynamics, which can be accompanied by a continuous flow of 'virgin' ideas explaining these new physical events....

Example:

- At time X at place X I saw phenomenon X, and my virgin idea is that this phenomenon X at time X at place X can be explained by factor X.

- At time Y at place Y I saw phenomenon Y, and my virgin idea is that this phenomenon Y at time Y at place Y can be explained by factor Y.

ETc....

Dea Marcel, Thank you very ugh for your question. As mentioned, I do think that the answer i: yes, no doubt about it.

I would like, though, to make to specific remarks, thus: on the one hand, regarding Vyacheslav's answer, it is the freedom of fiction to freely play with possible worlds, without having to go for experimental tests. Such, notoriously, is the possibility of science fiction, from Verne to G. H. Wells, f.i. Fiction does not have the constraints science does.

On the other side, technically said, such is exactly on the Millennium Problems, namely the P vs NP problems. It does fit here perfectly. Summarizing: It is always easy (= easier) to find a solution to any problem. That is P. The trouble, however, consists in proving that the solution encountered is adequate. That a NP problem.

The P vs NP problems gets much more complex than this - it goes without saying.

Of course, new ideas rarely can be oublished without solid data load - no high ranked journal will take such paper (however, what about "letters"?).

I fully support this point of view. Then, what do you think about presenting new ideas as presentations at the conferences, where abstracts are oublished? You will not loose authorship - and maybe get supporters of your idea?

Now about productivity. It's simple. Everything what is new, gives a boost of energy. And the routine work/experiments, vice versa,make you tired. Thus, not all good ideas, produced in ligthning speed, will be supported by gathered evidence.

What might be the (evolutionary) benefits for humans to create so many untestable ideas? Is it not more than an artefact/by-product of the brain? 

Nope!

Period!

Dear Tobias,

Why not? Does it depend on how you define 'testing ideas'?

When you have suddenly an idea do you always verify it, even with a simple observation? Is this the simplest form of an idea test?

People tested ideas mentally, which resulted in E=MC.... ????

Is the biology-based mental feeling 'Euraka' enough as a succesful mental test?

My information is contrary to the post. Scientists in natural and social sciences have much more data from field and experiments than they could publish.

Just an addition to Vyacheslav's post. Ideas from Verne's books were realized in majority. Now we are in phase of realization of ideas from sci-fi novels from 50' of previous century (Martian Chronicles, smarthouses). The time between idea and its realization might be also interesting.

I agree with  Mohammad Firoz Khan. It depends of scientific fields.

So some people lack inspiration to analyse data? How much ideas can be tested with a single data set, e.g. each time placed in another theoretical framework?

Will people from different fields analyse exactly the same data set in different ways, and therefore also produce manuscripts with distinct contents?

Dear Marcel. I can follow Debi's line of reasoning (explanation). Ideas are instant thoughts (like inspirations) on new problems to be solved. However, In order to get the desired outcome we need concept which is an idea coming out with thorough planning and action (strategy).

Dear Mahmoud,

What is the desired outcome? If we know the outcome in advance, i.e. the desired outcome, why starting/doing the research?

Dear Marcel: said "why starting/doing the research?"

Actually I answered it already! "... new problems to be solved" Research is continued to get more desired outcome (say some kind of optimization, short-cuts, ...) or in your familiar language "evolution!"

One of the truly big differences between growing economies and economies that stagnate is the acceptance of failure. If you don't let forests burn, if you don't let the old trees die out and the new trees grow, you don't get a healthy forest. The ability to manage failure so that enterprises fail but people can still succeed becomes one of the tricks of how you build a society that can reinvent itself as the world changes. YES, Therefore the rate of the production of new ideas should be higher than the rate of testing these ideas.

http://hbswk.hbs.edu/item/6591.html

But then the question is: How different should the rates of production and testing of ideas differ? One tested idea out of 100 produced, or what else? And which idea should then be selected, also accepting that we cannot predict their outcome/success of an idea? 

@ Marcel

Such probabilistic models are tough to be identified and unique with projects and issues.

@Krishnan,

Yes, indeed! But some ideas are financially supported, whereas other ideas are ignored. Why, when we also accept that success cannot be predicted? Is science communication (e.g. the capacity to convince people) more important that science (progress of knowledge)?

Maybe yes, depending on the scientific field.

Some disciplines are hard to be tested, other are easier.

The production of ideas new or updated is a cognitive process on a specific problem or a specific curiosity born from observation, testing,or data trial, or only an imaginative cognitive process built on the assumption "if" (as in science fiction). May be several  ideas are produced each day, but to be accepted or published, is a key question. Publications need scientific evidence through experiments. Many times scientific evidence from cognitive and logical process could be stated, but experimentally not sufficiently supported, cause of lack of tools,  of materials, of financial support, of scientific support, or lack in theory itself...etc  Generally challenging ideas could be accepted and put in practice sooner or latter

Dear Fairouz,

You state: Publications need scientific evidence through experiments. However, we may assume that in very active (scientific) fields experimentation is impossible. I was thinking about History or Cosmology or Astrophysics or Meteorology or ..... human social sciences where experiments are forbidden because of ethical reasons or..  Many try to understand the dynamics of systems based on non-experimental data, or not?

Agree with you dear @Marcel, that many fields of science couldn't support some experiments. However scientific evidence from logical statement and knowledge compilation should be accepted as form of probabilities with minimum error expectation, or new theoretical approaches proved with accepted theorems, axioms and building block knowledge for example

Does RG contribute to this? The production of more and more ideas without the ability to test these ideas?

Dear Marcel :-:

Depends upon what you mean with "testing" and whether you accept other types of testing or not.

But first of all we shouldn't forget that progress in science - formal science like mathematics as well as empirical or experimental science like research psychology - also lives from and by the formulation of plausible conjectures. Here, plausible means: accepted by a majority of peers, equally struggling with the same conjecture.

Most often people think of testing as "putting a theoretical prediction or explanation to empirical test" by adequate sort of methods, e.g. experimentation.

However, in my mind, there is another kind of testing which occurs in more theoretical sience or research, i.e. when someone wants to prove the consistency of a statement (claim) with a given theory. In other words: to test whether a certain claim follows from, i.e. can be derived from a given theory by logical (not empirical) methods. This may be extremely important in order not to delve into expensive and elaborate experimental research about a claim which is after all not compatible with de facto standard theory endorsed by yourself.

yes, including mental experimentation (Gedankenexperiment), but not exclusively.

Rate of Production should naturally be more

Though 'Idea generation' is the initial process; followed by "testing", the flow is constrained by the Researcher's choice either to carry on the succeeding process or not.  There can be "time lag" between ideation and testing; at times testing may not happen too.

Perhaps a 'new' idea is the outcome of an observation (e.g. anecdote) and therefore already tested somehow?

Traditionally, one distinguishes between a creative context of discovery and a reflective context of justification. In practice, these research phases may well overlap. Perhaps we shouldn't even think of those research contexts as phases in time, as if discovery always and entirely precedes justification. Part of discovery may indeed be interpreted and used for justification. And if you have a clear idea about how you would go about justifying the expected result of a long-awaited discovery, this very idea may also be used to your advantage to support any activities for the context of discovery. Remember also, that discovery isn't (always) a sudden event: all important discoveries show a long preparatory stage, in which the discoveries have been made possible at all (stage setting). Of course, sudden unexpected discoveries may occur: we call this serendipity.

Dear Roland,

How do you eliminate? With testing, I presume? This would imply that for each idea eliminated there must have been a test, right? Do you eliminate ideas without mental or other tests?

Dear Manuel,

Try bird nest!

Cheers