Revealing the data behind the science of evolution

[Loudmouth]

I think you have it all wrong.

The conversations go more like this:

Zosimus: Darwinism suffers from the tacking paradox. Natural selection, one component of Darwinism, is not testable.

It completely is. When I spread a mixed population of bacteria on an agar plate containing specific levels of antibiotic, I can predict that the bacteria carrying the mutations conferring antibiotic resistance will quickly outnumber those without those mutations.

I can give you other examples, if you want. For example, we can look at Ka/Ks ratios in exons and introns.

I see that you are still ignoring the math. Why ask for these things if you are just going to run away from them when presented?

 

[Zosimus]

Well, given that you're speaking to someone on the other side of the planet via the same machine you can use for research, work, entertainment, recreation, nourishment, and mindless self-indulgence, I'm more than at a loss what you could be talking about.

Science works. This epistemology has shaped every single aspect of our modern world, largely for the better, to the point where to claim otherwise is simply absurd.



And yet:

Huh. Weird. It's almost as though the fact that we can spot a whole lot of errors and problems in research doesn't invalidate the strength of the research which does hold up, and in fact allows us to self-correct.

Of course, as is the case 9 times out of 10 this study is brought up, you don't seem to understand it.



Oh, by the way - this failing research? It's substantially increased cancer survival rates and survival durations for almost every cancer in the last 40 years.

But ultimately, I think that explaining the paper is a bit besides the point. I could take your interpretation of it at face value, and it still would be a very weak argument. After all, if this is how badly science is working now, and it still managed to build essentially every part of the world around us, what would happen if we got our act together? No, I'm sorry, the claim that some theoretical research does not pan out does nothing to diminish the fact that the scientific method works, works better than anything else anyone has proposed, and that as a result, your argument completely fails.

EDIT: misunderstood what you mean by falsification of natural selection. My bad. That said, if you want a criteria of falsification for natural selection: how about species evolving away from traits that would help them survive in their environment for multiple generations (for example, bacteria becoming less resistant to antibiotics in their immediate environment)? Get this happening to a significant degree, and it would be a fairly clear falsification of natural selection.

...Of course, that doesn't happen, because mutations that negatively affect survival tend to get weeded out of the population very quickly, but the fact that a theory is not false does not mean it is not unfalsifiable.

The most important thing for you to understood is simple: Correlation does not imply causation. Just because two things happen together does not mean that one causes another.

Do you think this image shows that greater margarine consumption in the US is causing a lower divorce rate in Maine?

Do you think this image shows that the more often Nicholas Cage appears in films, the more likely people are to drown in swimming pools?

What should we conclude from this image?

 

[Zosimus]

It completely is. When I spread a mixed population of bacteria on an agar plate containing specific levels of antibiotic, I can predict that the bacteria carrying the mutations conferring antibiotic resistance will quickly outnumber those without those mutations.

I can give you other examples, if you want. For example, we can look at Ka/Ks ratios in exons and introns.

I see that you are still ignoring the math. Why ask for these things if you are just going to run away from them when presented?

I think you need to get a dictionary and look up the meaning of the word tautology.

 

[The Cadet]

The most important thing for you to understood is simple: Correlation does not imply causation. Just because two things happen together does not mean that one causes another.

I understand that perfectly well, thank you. It's just that in almost every case where life has become better, we can point directly to one of the many, many fruits of science and say, "That. That's what caused it." To accuse me here of mistaking correlation with causation is absurd; akin to wondering if there's any correlation between being hit by a baseball bat and feeling pain.

 

[Loudmouth]

I think you need to get a dictionary and look up the meaning of the word tautology.

You first. Show how that is a tautology.

We can also discuss Poisson distributions and the fluctuation test, if you want.

https://en.wikipedia.org/wiki/Luria–Delbrück_experiment

 

[Loudmouth]

The most important thing for you to understood is simple: Correlation does not imply causation.

Then why do you keep asking for p values?

Why do you cite drug studies that use p values as evidence that medications don't work?

 

[Loudmouth]

A nested hierarchy is nothing more than a well-defined super set which consists of and contains the lower levels. A good example would be the US army, which is broken up into field armies, which are further broken up into Corps, which consist of divisions, which consist of brigades, which consist of batallions, which consist of companies, which consist of platoons...

When we look at the distribution of characteristics between those groups we find clear and numerous violations of a nested hierarchy. For example, we can find the same airplanes in a Marine and Navy unit while finding two different airplanes between two Navy units. Nested hierarchy violated.

Also, you haven't shown us the math. Where is the math showing the p values and bootstrap values for military divisions, like those I have supplied for the phylogenies found in biological species?

Perhaps a better question would be this: Why do you think that data are important? The answer to this will necessarily be that you are an empiricist. You believe that sense data are the key to understanding the world around us. How can you justify this claim?

How do you justify the use of the same empirical methods in the studies you claim are evidence that medical treatments don't work?

 

[Loudmouth]

Incorrect. The probability may be rationally updated, but that updating process will depend entirely on the prior probabilities. It is entirely possible that confirmations may result in a theory being less true.

The current probability of evolution being wrong are 1 in 100000000000000000000000000000000000000. Funny how you ignore probabilities when they are inconvenient for your argument.

 

[Loudmouth]

Sure. Data can reduce you from an infinite number of theories to an infinite number of theories. Some infinities are larger than others. What's your point?

Name a single theory other than the theory of evolution which predicts a correlation between phylogenies based on morphology and DNA sequence.

If you can't, then it appears we have eliminated those other theories.

 

[46AND2]

When we look at the distribution of characteristics between those groups we find clear and numerous violations of a nested hierarchy. For example, we can find the same airplanes in a Marine and Navy unit while finding two different airplanes between two Navy units. Nested hierarchy violated.

What he is referring to is how soldiers are organized in units, which does follow a nested hierarchy pattern. Soldier A is in, say, squad B, Platoon C, Company D, Battalion E, etc. While soldier B might be in squad A of Platoon C, Company C, Battalion E, etc. These two soldiers are in the same battalion, but not the same company, platoon, or squad.

 

[46AND2]

If the matter is critical to you, can you afford to be agnostic?

So you believe...JUST IN CASE he's real? You think your God can't see through that?

 

[46AND2]

Nor do I. That's why I don't dispute natural selection. It's not a scientific theory. I simply remain agnostic.

Do you agree that both genealogy of human families, and the tree phylogeny which loudmouth provided, disregarding the mechanism for the moment, both exhibit a nested hierarchy structure?

 

[pat34lee]

Can you show that either creationism or intelligent design would predict a nested hierarchy?

I don't have to disprove unfalsifiable hypotheses.

Given a sufficiently imaginative creator, anything is possible, even likely.

 

[46AND2]

Given a sufficiently imaginative creator, anything is possible, even likely.

Even if some of those nested hierarchies are observed through remnants of inserted viruses? Or broken genes?

 

[Loudmouth]

What he is referring to is how soldiers are organized in units, which does follow a nested hierarchy pattern. Soldier A is in, say, squad B, Platoon C, Company D, Battalion E, etc. While soldier B might be in squad A of Platoon C, Company C, Battalion E, etc. These two soldiers are in the same battalion, but not the same company, platoon, or squad.

The problem is that those are names. The phylogenies in biology are based on observable characteristics. In those military units, the characteristics are tactics, weapons, vehicles, and so on. You can make anything into a nested hierarchy by simply naming them so. It gets a bit tougher when you have to use the objective characteristics of each group.

 

[pat34lee]

Even if some of those nested hierarchies are observed through remnants of inserted viruses? Or broken genes?

I have no idea what you're talking about. Please make your case.

 

[Loudmouth]

Given a sufficiently imaginative creator, anything is possible, even likely.

Which means that a non-nested hierarchy is just as possible. Therefore, there is no reason that the process of creationism or intelligent design should produce a nested hierarchy, just as humans do not force their creations into a nested hierarchy of shared and derived characteristics.

The only process that we know should produce a nested hierarchy is evolution.

 

[Loudmouth]

I have no idea what you're talking about. Please make your case.

If you randomly picked 100 words out of the Oxford Unabridged English Dictionary, and a friend of yours did the same without watching you pick your words, what are the chances that you would pick 99 of the same words?

 

[46AND2]

The problem is that those are names. The phylogenies in biology are based on observable characteristics. In those military units, the characteristics are tactics, weapons, vehicles, and so on. You can make anything into a nested hierarchy by simply naming them so. It gets a bit tougher when you have to use the objective characteristics of each group.

Sure, but the point if the original question was to demonstrate what a nested hierarchy IS. Not whether it is as significant as those in evolution. And these CAN be identified by characteristics, anyway. Different squads may be responsible for different tasks. Can there be other units at other posts which are responsible for the same task? Yes, of course. But that doesn't change the fact that the particular battalion we are looking at is organised in an hierarchical manner.

The point is that a nested hierarchy is simply a particular way in which a group is organized. We hadn't yet arrived at the point of demonstrating how it is significant in evolution.

 

[46AND2]

I have no idea what you're talking about. Please make your case.

Some of the nested hierarchies we see in genetics deal with identifiable remnants of viruses, and also with genes that don't function in certain species. An omnipotent creator certainly COULD build his creatures with these genes in place, but are you okay with the idea he did? What would that mean about those inclusions? Why would he design our genomes in such a way that it APPEARS as though we have common viral remnants with other species?