Are Untestable Statements Necessarily "Bad Science"?

Independent lines of evidence are good! :smiley:

Specificity again. We hope to develop a general theory from specif8c examples. Exceptions to the resulting theory either show it to be wrong, point tge way to better theory, or both.


1 Like

I posted this in the thread on fine tuning, but I think it is likely to have relevance here.

Many people seem to think that in order for a hypothesis to be testable, it needs to be falsifiable. Sean Carroll has argued against this as a criterion for a good hypothesis.

This runs contrary to what some of my peers and I remember about what we were taught in our undergrad degree.

1 Like

I was at a talk on this subject a few months ago, where the example of the Human Genome Project was raised. Ask any man on the street what is the most important scientific achievement of the past half century, and there is a good chance this is the answer you’d get. But what was “falsifiable” about it?


This may point to a distinction between applied science and fundamental science. A distinction that can be quite fuzzy, I hasten to add.

Yes, acquiring data is in itself science. As often as not, scientific progress seems to be made without a real hypothesis in scope, or put another way, there is serendipity in the data itself. Significant figures are added to measurements, magnetic signatures correlated with isotope dating, the limits of analytical instrumentation being extended, more species added to the list of whole gene mapping; the sky being mapped across the spectrum, all data contributing more and more to the total picture, steadily filling the gaps of understanding, and making it easier for theorists to validate or redirect their ideas.

1 Like

Perhaps their methodology was falsifiable, or at least failable. Their methodology was novel and not guaranteed to work.

As somebody who is currently working on assembling and characterizing a novel genome, I’d say the process is more analogous to exploration than it is to hypothesis testing.

But the end result of this exploration is something that can be used for generating and testing hypotheses.


In my line of work, exploration IS hypothesis testing :slight_smile:

1 Like

The two are certainly not mutually exclusive. :slight_smile:

Yes, this is very important. And it is often missed by those who are not scientists.

No, science excludes them. All conclusions are provisional.

1 Like


I can accept this view … IF you give me the proper vocabulary to enforce it:

What kind of conclusion is E=mc2?

… compared to the conclusion I.D. supporters think can be made?

It started as a hypothesis, with evidence it became a theory, and because it is mathematical, we can also call it a law.

It’s also still provisional.


If you are going to forbid the word “proved”, then you need a word to replace “proved” if you expect to make a dent in the English usage!

My understanding is that E=mc^2 is mathematically proven from other assumptions of relativity. It is those other assumptions that are tentative.


Sure… I can roll with that.

But what do we do with @Mercer’s claim that even E=mc2 is not “proven”.

If we don’t settle on the right vocabulary, we aren’t going to be very persuasive with our narratives!

If “proven” is the wrong term, what is the best terminology for distinguishing between the truth of E=mc2 and the assertion that I.D. demonstrates the presence of God?

I am taking it to be consistent with my version. So, in my opinion, you should be able to roll with that, too.


Based on @nwrickert 's post, are you willing to say that E=mc2 is:

Provisional Yet Proved?

It would seem you are the one I am trying to please.

When the project began it was a question of effort. We had all of the methods necessary to sequence DNA, so it was just a question of how long it would take. Along the way, new technologies greatly increased our ability to quickly sequence DNA and stitch the reads together. The real stars of the show were the advancements in technology, IMHO.

And this takes us back to the question at hand. We can have ideas in science that are not currently testable, but are still part of science. As mentioned before, we had to build the LHC to find the Higgs, and part of the wonder inspired by that test of the standard model is the effort that went into building the technologies that ultimately led to the experiments.

1 Like

Perhaps we could say that E=mc^2 is mathematically proven within the GR model of Physics, which itself is provisional (as all Physics models are)?

The problem with ID demonstrating anything is that they don’t have a model with entailments, so there is nothing to prove.

1 Like