That's because you don't understand how the calibration process works.
Tree rings are the standard. We can test individual rings, because only the outer ring is still "alive" absorbing the atmospheric C14. So, when we test the individual rings, we can know what the atmosphere was like when that ring was the outermost ring, collecting the C14.
Except that beyond about 10,000 years the trees are now silica and cannot yield truly accurate C-14 assessment.
Then, we verify this calibration by comparing the adjusted carbon measurement to lake varves, coral, ice cores, speleothems, and even artifacts of known age. What we find is, for example, the 4000th tree ring contains approximately the same ratio of C14 as the 4000th lake varve. Furthermore, the true C14 ratio does not stray from the tree ring chronology by more than about 800 years throughout the entire tree ring chronology of 13,000 years.
Yes this works fine up to and a little beyond such a date but comparisons between Radio-Carbon years and known historical years show a variance in conclusion such that at 21,000 RC years we differ by 3,000 years with other historical analyses. And by 50,000 RC years other methods demonstrate there to only have been 36,000 years (and the farther one goes back in time the broader the variance becomes).
We can DETERMINE the initial content of the daughter isotope when we plot the data on a graph called an isochron. A couple of the methods can usually be assumed to have close to zero initial daughter because of how the isotopes are trapped and/or escape from the rocks. But this assumption is not necessary. We can, and do, check it to make sure, using the isochron or age spectrum methods.
Now I realize most of this opinion is coming out of my head and old notes I have collected over the years and you are correct in that I know little about the technicalities of Isochron dating (my field of knowledge is biology) but I have been told by Professor Daniel Schrag of Harvard University in a discussion forum a few years back that there are other reliable, intelligent, published scientists who question the reliability in some cases and that in quality scientific presentation their views should be (and among real scientists is) respected even if disagreed with ( IMO contrary data or evidence of possible error in findings or in what results were expected, should not be discarded but reported along with the concluding scientists own position as well as their reasons for why they disagree with the possibly dissenting information
which is the case with most scientists and what they publish). The sad thing for me is that in forums like these dissenters are not respected or their views really looked at for what they suggest.
Also I once read (and noted) a discussion forum where one researcher named Bale (sorry I no longer have a specific reference, but read his assessment and then comment) who says that in isochron dating, ages are obtained from the slope of a line based on isotope ratios measured for different minerals of the same age. He said, The theory is that, although the different minerals have different initial amounts of the radioactive parent, the same percentage of the parent will decay in each case, with the result that the measured isotope ratios will fall on a line but that there are different isochron systems.
G. Brent Dalrymple, Ph. D., "Radiometric Dating, Geologic Time, and the age of the Earth: A Reply to 'Scientific' Creationism," February, 1982, published by the U.S Geological Survey, discusses the Rb-Sr isochron, which plots the ratio Sr(87)/Sr(86) as a function of the ratio Rb(87)/Sr(86) ratio:
"When a rock is first formed, say from a magma, the Sr(87)/Sr(86) ratios in all of the minerals will be the same, regardless of the rubidium or strontium contents of the minerals, so all of the samples will plot on a horizontal line." (page 32)
However, Brooks, James, and Hart (C. Brooks, D.E. James, and S.R. Hart, "Ancient Lithosphere: Its Role in Young Continental Volcanism," Science, Vol. 193, September 17, 1976, pages 1086-1094, have found that this statement is not always true. They studied 30 examples given in the peer reviewed literature, correcting where necessary
So according to isochron theory, the resulting diagrams, called pseudo-isochrons, should always be horizontal lines. However, the authors report:
"Correlation theory and regression analysis indicate that most of these psuedoisochrons have slopes significantly different from zero at confidence levels up to 95 percent (in some cases up to 99.9 percent) and that they define excess "ages" ranging from 70 million years to more than 3000 million years." (see page 1087)
So Bale asked we please note that the upper limit of error magnitude would give a newly formed rock an "age" 75% of the believed age of the earth. The paper states that the cause of the error (which is really just data which disagrees with their expected results and is not contamination at all which in SOME cases is an excuse used to throw out or disregard the contrary findings) could be contamination (could be does not equal IS), but also states a number of reasons for believing it is not, putting forth the idea that the results represent the age of the underlying mantle that was the source of the magma. The conclusion of the authors is:
"One serious consequence of the mantle isochron model is that crystallization ages determined on basic igneous rocks by the Rb-Sr whole rock technique can be greater than the true age by many hundreds of millions of years. This problem of inherited age is more serious for younger rocks , and there are well-documented instances of conflicts between stratigraphic age and Rb-Sr age in the literature." (see page 1093).
This study shows the apparent results gave too ancient of a date for SOME SP<samples (thus demonstrating probable inadequacies within this very excellent method). However, knowing these discrepancies, the authors of the study still contended that the dates represented real dates for the earth's mantle (thus, the probable possibility was excluded from the conclusion).
See also Calculation of 230 TH/U Isochrons, Ages, and Errors shared by , K, R, Ludwig (Geological Survey, Mail Stop 963, Denver, CO 80225-0046, USA),and D. M. Titterington (University of Glasgow, Department of Statistics, Glasgow G12 8QW, United Kingdom) where it is pointed out that when analytical errors are responsible for the scatter of points on a 230Th-234U-238U isochron diagram, the isochron should be fitted by a technique that weights the points according to their analytical errors and error correlations, and either takes into account the presence of some of the same data in two coupled XY isochrons or (equivalently not specifically) uses a single, three-dimensional XYZ isochron. Which they know and honestly admit is a method based on maximum-likelihood estimation (thus not on established indisputable fact). Therefore other conclusions are possible, even probable, and may in fact be the case. (all parentheses mine)
While it is assumed that the rate of decay has been constant, we have very good reason to think so. Contrary to your claim, we have not observed ANY fluctuations in decay rates from any isotope used for radiometric dating, despite rugged attempts to do so. There have been VERY tiny fluctuations in isotopes which are not used for radiometric dating, but these fluctuations are less than 1%, not the orders of magnitude required to explain a 6000 year old earth.
And I also do not believe in a 6,000 year old earth, but I want you to know your comments are noted, I appreciate them, and I will explore these comments more fully.
As for the system remaining closed. You are wrong again, we do not assume that, we test for it. When a system has not remained closed, it is generally very obvious when we plot an isochron. The only way an isochron can remain a straight line, is if the system has remained closed. If there was diffusion of daughter element, or contamination of parent element, then the data points will not fall on a line. These are not considered good dates, and are not published as such.
Here I do have an issue which would be that because something does not fit the model or the expected results this does not necessarily imply contamination unless can show how and when and by what means something was contaminated
at least define what the contamination actually was and how it occurred (and theoretical speculation on this would be fine so long as it was admitted and understood to be theoretical speculation). And I did not say the system remains closed.
Thanks
Paul