Subject: Norman/Setterfield How they did it. Posted: 21 May 88 03:42:21 GMT About a week a

---
Master Index Current Directory Index Go to SkepticTank Go to Human Rights activist Keith Henson Go to Scientology cult

Skeptic Tank!

From: livesey@solntze.wpd.sgi.com (Jon Livesey) Newsgroups: talk.origins Subject: Norman/Setterfield - How they did it. Posted: 21 May 88 03:42:21 GMT Organization: Sun Microsystems, Inc. - Mtn View, CA About a week ago, I posted a fairly long, three part review of the Norman/Setterfield paper which purports to show, via a review of historical speed of light measurements, that the speed of light has been 'decaying'. I raised a number of questions about their methodology and goals, in particular, why anyone would believe the speed of light was decaying when extremely precise measurements over the past twenty years show that it is not. I also questioned their treatment of data, noting that they include some data and exclude other data without explaining why, but I said that I did not have time to do an extensive review of this matter. This week, I found time to review their largest set of data, consisting of 63 values for the speed of light derived from the Bradley Aberration Method. They have plotted 13 of these values in an accompanying graph. They have some doubt about a value which was reworked by Struve, plotting it, but then omitting it from their calculations. Their graph does indeed show a decay in the speed of light, so naturally, I was interested to know why. I have now done my own very rough and ready plots, and when you have seen them, you will understand as well as I do how Norman/Setterfield come to claim that their data shows what they claim it shows. In the first graph, I have plotted only the values Norman/Setterfield chose to plot, consisting of a set of 13. For reference, I have included a horizontal line at the currently generally accepted value for c. Norman/Setterfield do not do this, but instead show a zero mark at a lesser velocity. As you see in the first graph, some are above and some below the current value for c. ************************************************************************ ** ** * * 1783 * * * ** 300400 ** * * * * 1841 * ** ** * * 1841 * * * ** ** * * * * ** ** * * * * ** 300000 * 1843 ** * * * * ** ** * * * * 1883 * 1907 * ** ** - - - - - - - - - - - - - - - - - - - - - * * 1909 * ** ** * * 1926 * * * 1914 * ** 299600 * 1909 ** * * 1935 * * 1922 * ** * 1916 ** * * * * 1908 * ** ** * * * * * * ************************************************************************ 1750 -x- 1950 As you can see, this graph does indeed show a very impressive decay in the speed of light from the eighteenth century to today. If you like, you can draw a straight line diagonally down the graph, and get a decay rate, and Norman/Setterfield do exactly that. They also produce a curve that they claim fits these points, and which they later extrapolate backwards to get a very young Universe. However, I said that there are actually 63 data points in this table, so what would happen if we were to plot all of them, including the ones that Norma/Setterfield do not plot in their report? Here is that plot: ************************************************************************ * * * 300600 ** * * * * * ** ** * * * * * * * * * ** * * * * * ** ** * * * * * * 300200 ** * * ** * * * * ** * ** * * * * * * * * * * * ** * * * * * ** * ** * * * * * * * * 299800 * ** * - - - - - - - - - - - - - - - - - - - - - * * ** * * ** * * ** * * * * * * ** * * * * * ** * * * * * * * * * ** * ** * * * * * * * * * * 299400 * ** * * * * * * * * * * * * * * * ************************************************************************ 1740 -x- 1950 Gosh! Merciful heavens! Our wonderful declining line has gone away, or at least it has been swallowed up in the data. What we see instead is something that is familiar to anyone who has had to deal with historical data. That is, the data is all over Hell's Half Acre. What is more, look at the left of the graph, and you will see that there is almost a hundred years gap between the first two values and the rest of the graph. There really is nothing very spectacular about this data now. There are about as many data points below the current value for c as there are above it, and although there are some high values a long time ago, there are also some very high values quite recently. If this graph tells us anything, it tells us that this method of determining c is not too great. Is there anything left to look at? Well, just one thing. What would we see if we looked at the data Norman/Setterfield chose to plot against the background of the data they chose not to plot. Here is all the data again, only this time I have shown the data they chose to plot with 'N' and the data they did not plot with '*'. ************************************************************************ * * ** ** * N * * * * * 300400 * ** * * * N * ** ** * N * * * * * ** * * ** * * * * ** * ** * * * * * * * * 300000 N * ** * * * * * ** * ** * * * * * N N * * * ** * - - - - - - - - - - - - - - - - - - - - - * * *N * * ** * * ** * * N * * * ** N * * 299600 * ** * * N * * * * N * ** N ** * * * * * * * * * * * * ** * * * * * * * ************************************************************************ 1740 -x- 1950 Now we can see quite clearly what is happening here. The authors have achieved their aim by the following method: for the early years, take only the high data values, for the middle years, take the middle data points, and for recent years, take only low data points. You could do the same with any data, as long as it is sufficiently scattered. This method has many advantages, of course. It effectively conceals the fact that in the middle years and up to quite recently, people using the Bradley Aberration method were regularly getting values as much as 1000 Km/sec different, but just as regularly they would quote their error bars as +- 150. In other words, we know that their estimations of their own accuracy were quite bogus. Secondly, it effectively conceals the fact that there have been some very high determinations of c by this method recently. Finally, by omitting the conflicting data points, it conceals the undoubted fact that some of the data were not the best in the world. Some of these data points, for example, come from Pulkova Observatory. I searched around to find something about Pulkova, and when I finally found a reference to it in a very large Astronomical Encyclopaedia, it said that it was the oldest observatory in the Russian Empire (!) but that in this century it had been abandoned because the observing conditions there were so bad! It goes without saying that in their report, Norman/Setterfield have a graph which corresponds to my first one, but not to my second and third. As usual, I just tell this stuff as I find it. You will have to draw your own conclusions. jon. (End of text)

---

E-Mail Fredric L. Rice / The Skeptic Tank