Bah, sorry to write a book. Here's what I came up with.
Ok, first the references. Thank you for providing that information. I will try to look into it more when I get a chance, but I got a start.
For one thing, I noticed that the Journal of Paleontology by Cambridge has an open-access section, which is convenient for those who cannot afford the time or money to dive into this information full time. I found a few articles there which show pictures, which is good. To the extent that Cambridge is reporting eyewitness observations, I think I can trust them. I trust that the pictures are real. I trust that they really did find the fossils they say they found. As for their conclusions, I'm skeptical. I think the main reason I trust their observations is the same reason I don't trust their conclusions. They want to appear "socially good" to modern culture - they have an anti-slavery statement and an equality diversity and inclusion policy. To modern culture, lying about observations is bad. So I doubt they're doing that. But to modern culture, it seems brilliant to forecast events in the distant past based on data collected only in the past several hundred years. And even looking at a few articles told me they were doing that. I do not agree with this philosophy (though I did once), so I do not believe their reasoning is sound. Valid yes. Sound no.
This one looks promising:
The Evolution of Vertebrate Design Leonard B. Radinsky. It's available on Google Books in a limited sense, for a preview. Page 3 says:
We have two main sources for the evolutionary history of vertebrates. The fossil record and the diversity of living vertebrates. The fossil record is very incomplete, for it preserves evidence of only a very small percentage of all the species that ever existed, and it usually provides information only on their hard parts or skeletons. However, it is our only source of information about extinct species. The fossil record is set in the framework of geological time and therefore can provide direct evidence of actual evolutionary transformations at fairly specific periods through time.
This seems to indicate that fossils provide a small amount of evidence compared to all the species that ever existed... which I think is a fair way of looking at things. Instead of claiming they have "lots of evidence" compared to what a person might think to be a lot, they compare the volume of evidence correctly by weighing against all the species that ever existed. However, I don't share their confidence in the dating of events:
(on Page 4),
Fossils are dated by radiometric "clocks," that is, elements with unstable forms that change (decay) spontaneously and at a regular rate compared with other elements.
I don't know all the ins and outs, but I do know one thing: nobody mortal has directly observed any historical event prior to Adam. There is no record that I know of that could be used as an anchor point for the people who developed radiometric dating to test the accuracy of their method. So... all the anchor points they possibly could have used would be somewhere in the last, say, 8000 or so years. Now, when I'm forecasting the life of one of my company's products, or perhaps a wear part or a tool... I
love to assume that the rates will just be nice and linear. I'll even settle for exponential, as long as it follows a consistent curve. Then I can just curve-fit it into excel and poof: I've predicted the life of the product. Then I can take that product out in the field to a customer, expecting it to follow that same rate, and I can consistently without fail... fail. Every time. Never, ever, ever do things in the real world behave as I would expect them to. There is always
some lurking variable, some factor I didn't account for, some extra bit of grease or a hotter environment or less humidity or something. There's what you see in a lab... there's what you read about in a book... and then there's what happens when you go to test it. Proof, or lack thereof, is in the putting. But where's the putting for radiometric dating? Here's what I would do, if it were my job to test the dating method. I would find historical artifacts of known date. Documents or artifacts archaeologists had found, or items long-preserved, which stated the year of the writing. I would try to find a large number of these... several hundred per century for all of history would be ideal. I would test the various elements used in radiometric dating on each of the specimens. I would use measurements obtained from a lab setting as well (for what it's worth). Then I would generate a graph. Hopefully, I would find a nice-looking exponential decay curve. Then I would start at the present and go back to each data point, one at a time, to see if the half-life held constant. And then, if it did - if it held perfectly constant, and I could repeat this experiment, I would call myself an eyewitness that the half-life of these elements stays constant over the measured range of 8000 years. But I still would not dare to forecast backwards hundreds of millions of years. I might dare to forecast backwards by half the measured range, so that at least 2/3 of the data range on my graph was real, and 1/3 was an educated guess. But forecasting over a range tens of thousands of times greater than that of the data I measured? The error approaches infinity. So maybe they came up with a better way... but I can't see how they manage to test this.
Oh, and on page 10, they do mention that they are using the Linnean system of classification. I'm okay with this for practical everyday use, but I don't agree with how Linneaus classifies humans, and I don't think species can be said to share an ancestral relationship simply because they share a few distinctive characteristics. I don't even think a high-percentage DNA match implies an ancestral relationship, for that matter. It is just as reasonable to think that God made beasts instantaneously with a lot of the same components as people, as to think that he made the process take millions of years. One reason would be to remind us that we will die. When we see the beasts die, and we notice that they have eyes like us, and legs and arms, and sometimes even fingers... we remember that we are dust, and to dust we will return.
I'm wondering if the other sources would be quite right for me... easy to read isn't necessarily more credible. Usually if something has been tailored to non-experts, it has gone through a few extra mediators, putting it farther from the eyewitness observers who generated the information. Then again, I
am a non-expert, so what am I to do? This is a problem I keep finding. How can the non-experts trust the experts? If I don't
first trust the expert, then I cannot learn
from the expert. But if I am told that I must
first learn what the expert is teaching
so that I might trust him, then I am at an impasse. I must do A to get B, but I must have B to do A. It is a logical impossibility. Therefore, the means of trusting an expert must lie outside the realm of his expertise
. There must be a way for me to find out if an expert is really an expert
without learning what he has to teach. I must trust
before I can learn, or else I'm not really learning. Similarly, there must be a way for me to find out if a Pharisee is a wolf in sheep's clothing before I sit down and learn what he has to say about the law. Perhaps... I could know them by their fruit?
Science doesn't work like a standards organization. It pretty much by consensus.
I think you're referring to something other than what I am referring to when you say Science. I don't think science works by consensus at all. Although it is good if a truth is popular, the popularity of an expert opinion does not increase its credibility. Nor does the total volume of information written about it.
I think standards are very important to science, because science is inherently collaborative. Without some common medium through which information can be shared, your observations and my observations can never be combined, and each is left only to his own direct observations. I often refer to standards organizations for scientific information to do my work. This is information obtained by the scientific method and published and controlled by national or international standards. Examples are AISI or ANSI or ASTM or ISO or DIN. Others abound. These govern things like the chemical composition of materials, strength testing of mechanical components, grades of oil... anything related to any component in any product that someone might buy. Beyond products, there are also standards organizations to make building codes, health codes, and whatever other kind of code there might be.
Standards require continuous scientific testing, both in the lab and in the field. This testing is not speculative in any way, and it isn't hidden from view. Direct observers witness the testing... and I don't even know them! But I don't need to. Anyone - even a non-expert - can test the efficacy of the testing method very simply. You buy the product. You see if it works. Or in the case of building codes, you test it by seeing if the building falls down or not. Or in the case of health codes, you test it by... well... I'll spare you the imagery, but you get the idea. Trial and error, guess and check, proof in the putting. The average non-expert can do this. Say the organization tried to lie, or misreported, or made a mistake about how much Iron is supposed to be in AISI A2 Tool Steel or some other material. Everyone would eventually see the results... buildings would start collapsing... lawnmowers would start falling apart... and airplane tails would start falling off. Yes it would take time for the error to be caught. But sooner or later it would be noticeable. Then one of the pilots might stop flying the Airbus 300 and become the safety chairman and decide to investigate it.
But it's just like assessing the credibility of the guy who designs the airplanes you ride. You'll need a lot of prep before you can do that.
It doesn't take much prep to do this much: I don't trust the guy who designed the Boeing 737 Max. And I don't trust the guy who designed the Airbus 300. These are two aircraft that I know of with failures causing planes to fail mid-air, ending in lethal crashes. I know of at least 2 mediators who conveyed this information to me: one of the then-A300 pilots that I know personally, and the news. I find these sources credible within the sphere of these incidents. Therefore, I believe that there really were such things as eyewitnesses of these events. Therefore, I believe that these events happened. Therefore, I don't trust the people who made these planes. Therefore, I would not knowingly fly on one of these planes.