The Beat Goes On

imageYou can read the latest, [Stephen Jones’] theory that the radiocarbon dating laboratories were duped by a computer hacker #5. I don’t know what to say.

Even though my theory at this early stage is entirely circumstantial, lacking as yet a `smoking gun’, by a process of elimination of "the impossible," my theory that the radiocarbon dating laboratories were duped by a computer hacker, "however improbable" it may seem to be, "must be the truth" . . .

Stephen then goes on to explain that the Arizona physicist Timothy W. Linick (1946-1989) may not have necessarily worked with a former KGB agent named Karl Koch by explaining he could have issued a software update to the other labs in Oxford and Zurich or (presumably for he doesn’t really say) another KGB agent could have entered the laboratories clandestinely to install the code in the control console computers. He then promises more in installment number 6 (which because some confusing reorganizing of his original posting is probably by now installment 14 or 15).

There is some interesting material about the carbon dating tests.

5 thoughts on “The Beat Goes On”

  1. I think the easiest explanation is that there were three hackers who had all been told by the KGB to aim for 1325 but the Oxford one made a mess of it and didn’t get the right date.

    1. Or just as likely he misread 1235 for 1325 and he was just about on target.

  2. Here we go again. But as I have said before; if this stuff is not discussed, then it can be taken as approved of, which I don’t.

    Stephen’s chart showing the non-overlapping ranges of the averages of the laboratories’ dates is wrong. He has blithely assumed that BP dates can be converted into AD dates simply by subtracting them from 1950, which is wrong to start with, but also he has not taken into account a particular anomaly in the calibration curve during the 13th century, whereby, for example, a BP date of 650 can indicate an AD date of either 1300 or 1375, but not 1325-1350. Properly calibrated, all three labs’ dates overlap, although it is true that Oxford’s overlaps the other two the least. The Arizona and Zurich samples overlap considerably more than indicated by Jones’s simplistic treatment.

    Although Riani & Atkinson have demonstrated a statistical chronological gradient across the sample, which may be due to minor contamination to an increased degree, but in no way invalidates the radiocarbon dates, and provides good evidence that the 12 BP dates produced by the labs were not randomly variant about 1325 as Jones suggests, and thus not randomly generated by a computer program. As it happens 1325AD corresponds to a BP date of almost exactly 600, so one would expect random numbers to cluster around that figure.

    Paradoxically, just as there is only a 5% chance that the Oxford and Arizona dates represent exactly the same age of cloth, so there is only a 5% chance of achieving the same result using a random number generator. If Jones is wedded to his spy theory, then he must accept that there is no discrepancy between the radiocarbon dates, in which case there is no need to explain their innacuracy by means of a random number generator!

  3. Stop Press: A little while ago Stephen either read this blog, or someone told him of the post above, and he inserted a comment into his latest posting to the effect that he realised his dates and chart with its non-overlapping bands were based on BP dates and needed calibrating. This he has now done, and amended his blog to show a new set of dates and a new chart which, curiously, still shows that the Oxford dates do not overlap either of the others. By using the tiny calibration diagram published in the Nature paper rather than an established calibration program, he has enabled himself to reinforce his convictions. This cannot be allowed to go uncriticised.

    It is not clear that Stephen really understands the calibrating process, particularly when it comes to how to deal with quoted errors. Actually anybody can do it by using OxCal, which is online, easy to access, easy to use, and free. Stephen tried to do it using the tiny published calibration chart in the Nature paper, deriving his dates simply by adding and subtracting the error from the mean as published in the Nature paper, and attempting to read off the calendar date.

    His results, and those from OxCal (to 95.4% probability) are as follows:

    Arizona BP dates: 646 ± 31
    Stephen’s calibration: 1290 – 1317
    OxCal’s calibration: 1281 – 1396

    Oxford BP dates: 750 ± 30
    Stephen’s calibration: 1214 – 1266
    OxCal’s calibration: 1222 – 1287

    Zurich BP dates: 676 ± 24
    Stephen’s calibration: 1286 – 1293
    OxCal’s calibration: 1275 – 1388

    Well, it’s only a few years here and there, what does it matter? But Stephen is a great pains to point out that by his calculation the Oxford dates still do not overlap either of the other two – which of course is untrue, and that the Arizona and Zurich dates only overlap by 25 years, which is also untrue. His finding of a minimum Oxford date of 1214 is grotesquely inaccurate and must surely be a typo, but his even more inaccurate maximum misreading has a serious effect on his later argument – I wonder if it could be deliberate fraud?

Comments are closed.

%d bloggers like this: