The Argument Clinic

Conversation Side Conversation
2673

Sorry, which is it? Is it a hypothesis or is it a prediction?

I do not understand this question. Predictive models do not test hypotheses. Sitting in an arm chair having a deep think is not a way of querying nature. Experiments are. I have attracted some heat for this in the past, so I shall not try and declare with any rigor some vocabulary, but put it rather in broad strokes as I see it. For me, the meaningful distinction is between theory and experiment, each as a discipline. Theoretical research consists of the development of new descriptions with the goal of finding ones that can predict more data than the current state of the art on the topic, or predict it with more precision. Experimental research consists of gathering raw observable facts, so as to test the outputs of theoretical research, or to motivate its expansion by isolating and confirming anomalies at the fringes of the current theories’ scope.

So, if you want to test a hypothesis, what you do is logically derive from it a proposition concerning an observable, and then see if that proposition actually holds. If it does, then the hypothesis has made a correct prediction, and therefore has some utility. If it does not, then the hypothesis has made an incorrect prediction, and therefore may at best have limited utility.

Respectfully, this sounds boring. Are we doing science or are we just stroking our beards? Why on earth would we stop at “just think about”? Personally I don’t care how much confidence or doubt gazing at the stars and having some deep ponder leaves me with. I care if the ideas I end up with are correct, and as comfy as my arm chair is, it will never tell me that, nor will my big wrinkly brain. Only nature can tell me if what I think about it is correct. So if I am to learn that, I need to find a way of communicating with it, a way to ask the question and to identify a reply as such, when one comes in.

This is why vagueries like “nature is powerful enough to form such-and-such” are such a waste of time. What sort of logic can I throw at it to end up with something I can test in a lab? What sort of observation could I ever hope to make that would conflict with it? What functional, usable technical understanding about a natural phenomenon do I have with this vapid string of words in the back of my head, over and above what I would have had without it?

See, if Newton’s laws of motion and universal gravitation are correct, then setting aside fluid resistances, all objects should fall at the same rate. That’s not trivial. There is no contradiction in assuming it to not be so, so if it turns out that it is so after all, that’s a vindication for Newton’s theory. It is not something just to think about, and it doesn’t matter whether we feel like our doubts of it tell us something. We can see more of the future with Newton’s theory than we could without it. It is useful. It works.

Desperate searches in one’s own head for some routine natural occurrence remarkable enough to give up and say “it’s magic” about is not a scientific or otherwise respectable intellectual engagement.

4 Likes
2674

It appears you have answered your own question here.

Mathematical models like in the Newton and Einstein case are predictive models. When you empirically test the hypothesis you are testing the integrity of the model. In the case of Einsteins model Eddington did the initial testing and the results helped validate the model.

What is important in testing these models is repeatability. This requires the forces of nature to be repeatable.

I agree with your statement:

The hypothesis that Perry Marshall made is that nature is powerful enough to form a transcription/translation type mechanism. At this point we don’t have either a testable model or a validation by experiment that could help claim the prize.

How would we describe the status of Perry’s hypothesis? I propose it is most charitably an untested hypothesis that may turn out to be untestable. I would propose that Perry has made a claim that is currently outside the scope of Science. Do you agree?

2675

How did I get dragged into this? :laughing:

Oh, THAT. You are so close …

Creation itself is not falsifiable.

Check.

What is falsifiable is if something observable is the direct product of creation or can it be explained by the laws of physics and chemistry.

Nope. Fixed it for you. You just had it too … Divine Creation is not falsifiable.

If someone showed a translatable code being formed in nature that would falsify that the transcription translation mechanism was a direct product of creation.

No, it would not, because Divine Creation is not falsifiable.

A Digression: In general, statements asserting a negative cannot be proven. Assertions like “a THING could not have evolved” are simply not made for this reason. We might present positive evidence in support of a statement like “this THING evolved”. You can pose hypotheses about the equivalence of two things, “A and B differ by no more than some amount”, but that’s still not assertive a negative.

The example of this mentioned previously is Newton believing that the movement of the planets was guided because it did not fit his gravitational model.

Newton certainly believed that God set the planets in motion in an orderly manner. But I have never seen that Newton doubted his own math because the data didn’t match up (I think the necessary data didn’t exist at the time). Have you got a citation for this?

This was falsified by Einstein as his model accurately predicted the motion of the planets.

What is falsified? Newtonian mechanics? Newton’s belief in God??

Though I appreciate the credit, the idea is not original to me.

… where Einstein falsified the motion of the planets was direct evidence against Divine guidance it did not falsify creation of the solar system or the universe.

Whoa! Triple negatives make my head hurt (… or maybe it’s just my sinuses :sneezing_face: ).
Einstein falsified Newtonian Mechanics, showing that a modified theory was needed at large scales (Relativity). Nothing about Divine guidance here.

1 Like
2676

By willingly responding.

bait
bait640×267 24.1 KB

The bait only works if you bite it.

2 Likes
2677

Being the direct product is because between divine creation (assuming a created universe) and what you are observing are the laws of nature. If you can identify the laws of nature as the direct cause how would you claim that divine creation was the direct cause?

2678

Until someone can present a logical derivation of a proposition regarding observables from it, I’m comfortable calling it scientifically uninteresting[1]. And that’s if I’m generous enough to grant that it is even a “hypothesis” in that it might entail propositions like this. What’s your point?

  1. I shall assume for the sake of brevity, that the claim in question actually is Marshall’s, instead of systematically replacing references to it as such with its actual contents in my messages. ↩︎

1 Like
2679

Hi Gisteron
I think I can agree with scientifically uninteresting. Thanks for the discussion.

2680

NOT what I wrote. I’m not writing it out again.

2681

Removed comment by author

2682

Good grief!

2683

Fair enough. I will just restate my position.

Divine creation is not ultimately falsifiable but you can eliminate it as the direct cause as Einstein did through his general relativity model that was ultimately tested by Eddington.

In this case it can be used as a null hypothesis as random is commonly used.

2684

Yep not a good post. Thanks.

2685

So no point to waste anybody’s time with it, then.

Not falsifiable but eliminate-able? What is this gibberish?

If general relativity “eliminated” (as opposed to falsified, obviously, since ideas do not falsify one another, but maybe this new elimination thing you invented just doesn’t have reasonable limitations like that, so what do I know) divine creation (what ever that is) as the direct cause (of what ever), then, again, why is anyone still wasting anybody’s time with it?

What is your point?

1 Like
2686

Eliminate it means to dismiss it as a direct cause. If the forces of nature are adequate to explain the observation there is no need to invoke Divine creation as a cause. This however does not eliminate Divine creation as the cause of the origin of the forces of nature. Divine creation can be falsified as a direct cause but not as the ultimate cause as Dan has pointed out.

Hypothesis testing deals with direct causes and not ultimate causes. This is why Divine creation would be best used as a null hypothesis and as an alternative in certain cases where using random or no cause can be misleading.

2687

Respectfully, this sounds like scientifically inconsequential sophistry. We are not looking for personal intellectual gratification, for some ill-defined “explanation” or “causes”. We are looking to produce models that allow us to foresee the outcomes of future experiments more reliably than we can without said models. Whether philosophers are happy to call such models explanations, how they are comfortable to map their logic of causation onto it[1], whether they find it adequate in some subjective manner is completely uninteresting. Something either works or it doesn’t.

And from what I can tell so far, the “divine creation” you are suggesting is not “best used as” anything[2], because as much as I prompted to present what the claim even is scientifically, and how it can be logically treated to open it up to scientific inquiry, I have no indication that there even exists an answer to these very basic matters. “Divine creation” seems to me so far to be a label without there being anything it actually refers to. I’m happy to leave it to philosophers to explain to you, if they so please, how utterly this precludes intellectual treatments of this babble in general, but I can with some confidence say that wastes of time like this cannot move science forward in any way. Something this good-for-nothing cannot be “best” for anything.

  1. Perhaps someone who read Hume may explain in more sophisticated terms than I’d employ. If between two highly correlated event pairs \left(A_1,B_1\right) and \left(A_2,B_2\right) the events were linked causally in one but not the other, I know of no means to even begin investigating, let alone to find out, which is which. ↩︎

  2. Not just anything you like to make up gets to become a null hypothesis by arbitrary choosing. The least a candidate statement has to accomplish is being a statement of some kind in the first place. Next it has to be some measure of plausible given the state of theory and data on the subject. That’s far from all, but seems already like too much. You personally being emotionally invested in something having a place in the discussion is insufficient to actually give it one. It might have one, but so far you are going miles out of your way to ensure it looks like it does not. ↩︎

2688

HI Gisteron
I agree with this statement.

Here again is my view.

Biology is where I see Divine creation is sometimes a better null hypothesis than random or no cause as a null hypothesis. In the case of universal common descent as the current biological model it will go unchallenged with random as the null hypothesis. This is because there are many similar genetic traits that species share. What then gets swept under the rug is explaining the differences. Since random is the standard null hypothesis explanations like convergent evolution are easily accepted even though they are highly speculative.

2689

Clearly you don’t, as you go on to demonstrate:

Maybe you do, and the reason for this is that you do not understand the character of biology in particular or science in general, nor what a null hypothesis is. “Divine creation” is not a statement. “Random” is not a statement. “No cause” is not a statement. In general, stringing words together is not the same as making statements, let alone ones that can undergo or be employed within any sort of scientific discussion, concerning biology or any other field of science.

“Explaining” in general can mean anything. Your objection only exists, because someone’s personal proprietary understanding of that term was such as to warrant them saying to you that some thing explained another, but your personal proprietary understanding of that term is such that it should give them no such warrant. Mainstream theoretical discourse concerning the difference in genetic traits between different organisms currently stands in conflict with none of the available data on the subject and correctly predicts the outcomes of numerous experiments. That you personally do not find this good enough to qualify as genetic differences being “explained” is entirely inconsequential. The scientific enterprise is under no obligation to try and satisfy your individual intellectual sensibilities in particular.

It’s not. “Random” is not a statement. So it cannot and does not serve as a null hypothesis, standard or otherwise.

And this sort of gibberish is exactly why I’m so liberal in rejecting the language of “explanation” until at least after the terms have been rigorously defined. Convergent evolution is an observation. An independently verifiable datum, a nature-given fact. Some previously far divergent lineages of organisms can and do sometimes end up with very similar morphological features. This is something that undeniably happened many many times in the history of life. This is easily accepted by any and all who have no irrational desire to reject reality. What may at some time have been an open question is exactly how convergent evolution works mechanistically. That it happens is neither speculative nor in dispute any more than what the colour of the sky happens to actually be is speculative or a matter that needs debating.

2690

With all due respect this problem of accepting convergent evolution as a well demonstrated hypothesis is very problematic especially if you see genes shared uniquely between animals with distant relationships relative to the tree of life. I carries with it many untested assumptions such as highly similar gene sequences can be arrived at independently through known evolutionary mechanisms.

I am not sure if you have seen this paper or not.

https://bio-complexity.org/ojs/index.php/main/article/view/BIO-C.2018.3/BIO-C.2018.3

2691

I did not suggest that convergent evolution is or should be accepted “as a well demonstrated hypothesis”.

Lineages whose histories have demonstrably been separate for some time can and do, as a matter of fact, sometimes end up with very similar features. This is not a hypothesis formulated to account for some fact and from whence predictions may be derived. This just is a fact. The name we give to this phenomenon is “convergent evolution”.

How or why this happens is a legitimate scientific question, and that known evolutionary mechanisms can fully account for all such convergences is a non-trivial suggestion one should be absolutely welcome to question. But accepting that organisms can and do end up with similar features without inheriting them from one another or a particularly recent common ancestor who had them is not problematic whatsoever, no matter if there exists a model to account for it.

Likewise, there is a legitimate discussion to be had as to how or why it is that at daytime the sky looks blue at large angles away from the sun’s direction, but that it looks blue is a fact. It is not a hypothesis, nor is there any problem with accepting it as the fact that it is, even back when there was no functional understanding of the behaviour of waves or what gases are.

1 Like
2692

Hi Gisteron
I think your analysis is correct so we have common ground. The problem in my opinion is the name we give it implies a reproduction relationship between different species which ends up being question begging.

If you use random as the null hypothesis it supports a reproductive relationship. If you use divine creation as the null hypothesis it will bring a competing hypothesis in the mix which is the potential for separate trees.