I think he already has, and I believe he’s right in saying that isochrons aren’t infallible. However, he has overlooked the element of cross-checking them against other dating methods as a means of overcoming the corner cases.
The article that I’ve found that gives the best and most comprehensive explanation of the nuances and intricacies of radiometric dating is Radiometric Dating, a Christian perspective by Roger C Wiens. It goes into a lot of detail about exactly what corner cases scientists run into and exactly how they work round them. It was written in 1994 and revised in 2002, so no doubt there’s a lot of new material that could be added from more recent research. However, I’m yet to see a YEC rebuttal of radiometric dating that raises an objection that it doesn’t address.
Of course they aren’t “infallible”. That’s not a scientific term. We deal in probabilities. In and of themselves, no single methodology is the endall. It is the consilience of great many methodologies and types of evidence that gives us confidence in scientific discoveries.
As to “I think he already has”, that was my original assumption. But the more I’ve read his comments, I realized that they are gleaned from science-denial websites without an understanding of the underlying science. Many of the scientists posting on these threads have identified dozens of such clues.
No apologies necessary. I think the restatements and explanations will be helpful to readers who are new to these concepts. One of the didactic advantages of forum threads is the opportunities for readers to hear the same concepts restated in multiple ways. Textbooks rarely provide that.
Obviously, the term consilience is relatively new even though the concept is quite old. I try to reinforce it as often as I can. It is so essential to an understanding of modern science and it is a reality which science-denialists don’t want to talk about.
For example, with radiometrics, they want their readers who are not well informed about science to think that radiometrics is some casual and iffy set of presumptions. They ignore the fact that multiple scientific journals have for decades now focused on excruciatingly detailed and careful methodologies to maximize the accuracy and reliability of the procedures. So when scientists published papers explaining why mollusk shells dated the minerals of the surrounding environment and not the age of the mollusk, science-denialists spun the articles as failures of radiometrics! That’s not just ignorance. That’s deliberate lying!
Yes, I do want to give my Christian brethren the benefit of the doubt. But we are not called to do that to such extremes that church discipline becomes impossible and that to call out dishonest propagandists is not a duty of Christ-followers who care about honesty.
When I say that we have been remiss in failing to identify the liars and charlatans among us, I’m pointing the finger at myself, not just others. Even after I left the Young Earth Creationist “creation science” movement, I was quite content to “go along to get along” and to pretend that the blatant dishonesty was simply science ignorance and not deceit.
That’s true. So a geologist who is expressing himself (or herself) carefully might say that a petrified tree in Yosemite is 50mya +/- 4.5my. Or better yet, after adding error bars for the dating of the non-sedimentary layers, s/he might say the petrified Yosemite tree is 50mya +/- 5my.
What the geologist would not say is that the tree dates to 7000 ya. That would be absurdly outside the boundaries established by the scientific investigation.
Actually Chris, the error range is 50my +/- 9my. This doesn’t even consider the potential for sediment to be deposited after the formation of the ash, but before its deposition in situ. This means that the error potential is nearly 20%, but certainly not less than 10%, not including the potential error in the actual dating of the ash layers themselves.
Here is one example of changing of radiometric dates, which changed certain layers from a later Jurassic to lower cretaceous. This agreed better with the biostratigraphical dating of various fossils in those layers, but this is a reduction of 10% in the estimated age :
The earlier radiometric dating methods for this region gave a Late Jurassic age, which was in agreement with the prevailing biostratigraphic evidence that was broadly accepted at the time. Typical dates determined from the lower Yixian Formation were given as:4 40K–40Ar: 137 ± 7 Ma, and 87Rb–87 Sr: 143 ± 4 Ma. Biotite crystals from a tuff in the Yixian Formation gave dates of: 40>Ar–39Ar: 145.3 ± 4.4 Ma and the combined isochron: 147.1 ± 0.18 Ma.
However, this dating was subject to revision in 1999 (table 1).4 …
…The later revised dating assessments for Liaoning Province provided dates that are consistent with an Early Cretaceous time frame: 4 40Ar–39Ar: 124.6 ± 0.1 Ma, 40Ar–39Ar: 125.0 ± 0.18 Ma (total heating and incremental heating analysis of sanidine and biotite crystals in the Jianshangou beds); 40Ar–39Ar: 128.4 ± 0.2 Ma basalt capping the Lujiatun beds; 235U–207 Pb: 125.2 ± 0.9 Ma from zircons in Jianshangou beds; and 235U–207 Pb: 121.1 ± 0.2 Ma from zircons overlying lava in Jianshangou beds. A date for the Tuchengzi Formation at the base of the Jehol Group was reported as: 40Ar–39Ar: 139.4 ± 0.19 Ma. The intrusive Basalt in the Jiufotang Formation from Inner Mongolia gave an age approximation of: 40Ar–39 Ar: 110.59 ± 0.52 Ma. Zhou, A., Barrett, P.M. and Hilton, J., An exceptionally preserved Lower Cretaceous ecosystem, Nature 421(6925):807–814, 2003 | doi:10.1038/nature01420. (Andrew Sibley)
You may well be right about that. I’m not omniscient, and possibly all objections are addressed somewhere, although merely being addressed doesn’t mean that the responses are adequate. (Many of the reponses on wikipedia to the 101 objections to old age are certainly completely inadequate, for example). But on the other hand, when I keep hearing that radiometric dating is accurate to +/- 0.1%, it is necessary to put that into context. Only the measurement of the daughter and parent compounds is accurate; the extrapolation to the estimation of the age of sediment is far from that.
Fascinating. The date was revised downwards by approximately 20 million years, or about 14%. Significant, but not extraordinary given that the original error bars were in the 3-5% range.
Now can you please explain how corrections of 20% or less to a tiny minority of results justifies the claim that thousands of other results that require no such correction could be in error by more than four orders of magnitude?
Could you please provide links to the articles you cite? I’ve no idea which Wikipedia article you’re referring to here so I can’t tell whether your statement has any substance or not.
I’ve said this several times to you already. Error bars and extrapolations are not estimates. Scientists do NOT guess them; they MEASURE them. They do NOT blindly extrapolate graphs; they establish a firm theoretical and experimental basis for their extrapolations. And they do NOT use unscientific hand-waving terms such as “far from accurate”; they express far-from-accurate-ness as a NUMBER.
Seriously, this is the first thing you learn in a first year undergraduate physics lab class. It’s observational science 101.
Yet another batch of reasons why we don’t look to non-scientists for claims that the entire science academy is caught up in fraudulent definitions of scientific terms. (Yes, where there is the sin nature there can be fraudulent behaviors in fallen humans That’s why science uses careful methodologies to expose such frauds. With non-science…not so much. That is why we see very different patterns of behaviors from the contrasting groups. )
I tended to take these shortcomings among my American students as standard-issue until I taught for a time in the UK. I realized that the USA has some of the worst mathematics-illiteracy in the world. I don’t know if this is a consistent pattern, but I rarely had to explain these issues to the Indian and Japanese students, for example. There seemed to be much more “integration of knowledge” in their grasp of physics and mathematics.
Of course, those nations don’t have the problem of science-denying creationist websites actively undermining the basics. But my experiences were long before the Internet had expanded the disinformation influences of YECism which we take for granted today.
These revisions were made not because of a detection of error in the method, but because of a desire to match to the biostratigraphal data. If there is no inconsistency with biostratigraphal data, there is no need nor any desire to re-examine or adjust measurements. So how would you know if an adjustment could be made or not? The point is that a method is revealed or validated by its exceptions. And the point is also that error ranges are significantly outside of the 0.1% when other methods besides radiometric analysis are considered, and when sedimentary layers are estimated.
Error bars are derived from the variances within a group of repeated measurements. Extrapolations however, are not. Graphs are extrapolations from limited data, and sometimes various graphs(lines) are statistically possible from the same data, depending on how much data is available. Although there is usually a “best fit”. Nevertheless, even an extrapolated line (even a statiscally mathematically derived line or curve) still can have a starting point or an end point, and these starting points can vary in real time, depending on what is being examined; this is particularly true in the natural biological world.
As an aside, you learn about error bars and variances in statistics courses, of which I have taken both a beginner and an advanced course. This type of statistics applies not just to physics, but also to chemistry and biology, and even to the social sciences.
So perhaps you understand the difference between statistical experimental significance and biological significance?
@johnZ: Nothing you’ve said here addresses my main point above. I asked:
Anomalies and corner cases do crop up from time to time in any discipline as complex and nuanced as dating methods, and they do invite explanation, which is usually forthcoming. However, in order to falsify the entire discipline, they must be sufficiently frequent, sufficiently large, and sufficiently unexplained to demonstrate that they are more than just isolated corner cases.
Doesn’t have any bearing on the point I was making. Extrapolations are still governed by mathematical equations. You can still put a number to how well they do or do not hold. And this number is still calculated, not estimated.
I don’t know how this could be stated any more clearly.
Ditto.
Before telling all the world’s scientists that the carefully developed, meticulously devised and tested methodologies are all wrong, it is absolutely necessary to learn the basics. Please.
Really? I’m curious as to how you can have such deep knowledge that allows you to be so certain of the motivations of others. Can you even name the people who have this alleged desire?
It’s a lot like your claim that the definition of evolution is fraudulent, I suspect…
So how would you know the desires of others, particularly given the context your inability to distinguish between a measurement and an extrapolation? That would seem to require much more understanding than that required for answering your question.
@benkirk: You seem to have ended up ascribing things @johnZ said to me. Not sure how…some bug or other in the forum software perhaps?
I think he understands the difference between a graph and a line. What he seems to be trying to do is play down the theoretial and experimental rigour involved in the scientific method.
I’m not so sure JohnZ does know the difference between a graph and a line or curve fitted to the data. He’s clearly trying to play it down, IMO because he can’t be bothered to study the method to gain any understanding–only for attacking.
Yes, you can calculate the line, but the determination of outliers? The impacts of an addition of new data?
But this is beside the point. The point is that the small error ranges in measurements are incorrectly correlated to small errors in dating of sedimentary layers. This is an error in extrapolation which cannot be calculated, since it is essentially not measureable, but is only extrapolated from boundary layers.
In addition, the variability in measurements themselves, from measuring different components of a particular rock, has also at times been quite large.
You are placing your confidence in the precision of mathematics, but forgetting about the necessity for assumptions before the mathematics can be utilized. The setting of boundary conditions and other parameters determines the usefulness of the mathematics. That’s why I also asked you about the difference between experimental statistical significance and biological significance. In an experiment, you can eliminate many variables to find significance for one, but in nature, you cannot eliminate these other variables.
Thankyou. But graphs are of different types, and can include bar graphs, line graphs, curvilinear graphs, three-d bar graphs, etc., etc. I am not playing down anything. Especially not the scientific method. However, today I just read a news article on how statistics was rarely applied to forensic evidence, and in some cases might not be able to be used for evidence such as bite marks, bullet casing patterns, hair samples, etc. The fact that numerous people have been falsely convicted on this evidence, and later released after 10 years of incarceration, is having an impact on the acceptability of the evidence, even though the numbers of false convictions are relatively small compared to the total. There is nothing wrong with questioning a theory, or questioning the rigor.
You see to believe that geologists have not set boundary conditions and other parameters. Why do you believe this? Which conditions and parameters have they not set?
