Re: Humphreys responds to the accusations

Glenn R. Morton (grmorton@waymark.net)
Fri, 02 Oct 1998 20:56:13 -0500

At 07:59 PM 10/2/98 -0400, Ed Brayton wrote:
>Dr. Humphrey's posted this to the CARM board sometime today:
>
>"The mistake here is not mine! The figure you
>show (4.5) from Merril & McIlhenny was *not*
>what I drew my figure from. I got mine from
>a different figure in the same section having
>500-year *global averages*. As every paleo-
>magnetist knows, intensities from a single site,
>especially from clays, are full of "noise". Also
>note corrections (for C-14 dating)I applied
>to the timescale.
>
>All the details are in my 1986 ICC paper. I
>suggest that you read it before jumping to wrong
>conclusions again. Also, you might consider
>following the usual journalistic practice of
>consulting the author before writing your
>article. It saves the journalist the
>embarassment of printing a retraction, which
>I now ask you to do. Please send me a copy.

Probably few on this or any list have the 1986 article cited by Humphreys.
I do. I checked on it to see if Humphreys was correct. Here is what that
paper says:

"Table 2 summarizes the results of 1167 archaeomagnetic field intensity
measurements as equivalent dipole moments averaged over 500 or 1000 year
intervals 'for the past 12,000 years', whihc I assume represents the
post-flood period. Figure 4 plots the number of samples / century versus
age. Notice that the slope changes more than tenfold very abruptly at 3500
years B.P. (1500 B. C.). It looks just as if the time scale abruptly
changed at that point. If therew was no particular bias in selecting the
1167 samples, the change of slope strongly implies a problem in dating
samples more than 3500 years old-- that the dating technique stretches out
the time scale, giving much greater ages than it should.

Corrected Carbon-14 Dates

"Almost all archaeomagnetic samples more than a few thousand years old are
either directly or indirectly related to radiocarbon measurements.
However, as young-earth creationists have long been pointing out, present
carbon-14 dating techniques take no account of the strong possibility that
the percentage of carbon-14 in the air was much smaller before flood."
Russell Humphreys, "Reversals of the Earth's magnetic Field During the
Genesis Flood," Proc. 1st ICC, (Pittsburgh: Creation Science Fellowship,
1986), pp. 113-126, p. 119-120

Humphreys lists the archaeomagnetic dipole moments in his table 2

years BP # samples Dipole Moment 95% error
(10^22 A m^2)
0-500 268 8.72 .17
500-1000 187 10.30 .27
1000-1500 205 10.90 .27
1500-2000 131 10.94 .37
2000-2500 89 11.10 .54
2500-3000 60 11.28 .63
3000-3500 43 9.64 .85
3500-4000 17 9.21 .90
4000-5000 34 8.87 .74
5000-6000 44 7.20 .57
6000-7000 36 6.73 .65
7000-8000 18 7.08 .66
8000-9000 15 8.61 1.17
9000-10000 14 8.26 1.25
10000-11000 5 6.76 1.17
11000-12000 2 8.36 ---

Humphreys then re-dates these archeomagnetic samples by assuming no C14 in
the immediate post flood atmosphere and scales the data between 4000 years
BP and the present. He claims that the curve he published is a least
squares fit to the data. Since I don't have a scanner, I will draw this as
best I can

| @
| @
| @ @
| @ @ @
| @ ** @ ****** @
| ************** @ *************@*
| * * *@ @ @ *
|* @ @ *
| * @ @ @*
| @
|
|
|
|---------------------------------------------------------------
4000 years BP 0 years BP

* the raw data
@ Humphreys curve

Two things are immediately obvious. First, contrary to Humphreys claim in
Impact 252 the archaeomagnetic field shows NO reversals. Yet in Impact 252
he draws a diagram that makes it look like the post flood magnetic field
reversed many times. The data doesn't support that. And the fluctuations
that the archaeomagnetic data show is not reversals but experimental error.
And there is no data showing any reversal in the post flood era. And
Humphreys didn't get the reversals shown in Figure 1 of his Impact article
from the archaeomagnetic data. Second the curve he derives from the data
(which he shows in Figure 5 fo this 1986 article and is the @ above) is NOT
a least squares fit. I don't know what it is, but the upswing at 4000 BP is
NOT supported by the data. It is bogus. My suspicion is that Humphreys
tried to least squares fit a third degree polynomial to a curve that would
be more accurately represented by a second degree polynomial. If this is
true, that too would be bad technique.

glenn

Adam, Apes and Anthropology
Foundation, Fall and Flood
& lots of creation/evolution information
http://www.isource.net/~grmorton/dmd.htm