On Sun, 22 Jul 2012 21:20:00 -0700, Nivalian wrote:
> On Sunday, July 22, 2012 9:53:10 PM UTC-4, chris thompson wrote:
>
>> Welcome to talk.origins!
>>
>> Since you bring it up right away, how can you support the idea of a
>> 6000 year old earth? That is contradicted by everything we know of
>> geology and radiometric dating.
>
>
> Good question. I know the difference between facts and interpretation
> of the facts; and how worldviews drive their interpretations of those
> facts. Once you look deeper and understand certain dating methods and
> how they work you will notice something is not right. I also know a
> little about how science funding works, which is more political than
> science.
>
>
I'd like to hear more about the science funding.
Here's what it looks like from my side of the fence.
Last week I finished the process of pitching my grant application for
"Lab-Directed Research and Development" (LDRD) funding. This is a pot of
money that's controlled locally at the lab, as opposed to National
Science Foundation (NSF) funds or Office of Science funds at the
Department of Energy. I'm asking for $1.5 million over three years,
which puts it on the lower end size-wise. That'll fund half of my time
and fully fund two Ph.D. students.
Any lab employee can submit a proposal. For the first round there's only
a two-page proposal needed, so *lots* of proposals get submitted. There
are two good rationales: 1) "This work with get us on the cover of
Nature" (and we get on the cover of Nature pretty frequently), or 2) "We
can use this money to secure a larger stream of outside funding". (Being
able to pitch 1 *and* 2 is pretty sweet.)
Competing for these funds are a population of postdocs and younger
scientists there who were not only good enough to get in and finish a top-
tier Ph.D. program, but who were also good enough to get hired on based
on their Ph.D. work. Nearly all of them will be submitting one or more
proposals. At the first cut this year, the total dollars in the
proposals exceeded the total dollars available by a factor of five.
This is seen (and seen correctly) as one of the easiest ways to get
funding. Grants that are competitive nationally tend be be
oversubscribed on the order of 20x.
I made the first cut: the second cut was only oversubscribed by a factor
of two. Then came the seven-page full proposal and a 30-minute
presentation to a *very* skeptical audience.
And at this point I'm told that, yes, politics does play a role. *All*
of the proposals that get this far are solid, have a reasonable chance of
success, further the mission of the lab, and will likely lead to either a
cover story in Nature or increased funding (or both).
So other factors start creeping in as tiebreakers. If there's agreement
that one hotshot postdoc might leave for academia if her proposal doesn't
get accepted, that might put a thumb on the scale. If another researcher
has had a few LDRD grants in the past but only had middling success, then
that will count against their current proposal.
I'll find out in about a month if I got funded or not.
As far as politics goes, it's reasonably transparent and is widely seen
as fair. If you're ideas are consistently good and your work is
consistently great, you'll (eventually, probably) get funded.
Well, so what?
I can tell you (having had the opportunity to observe this process and
similar ones up close and personal) that if you have any evidence at all
that there's even a 10% systemic error in isochron dating techniques,
that'll get funded --- it'll sail right through. That's Nobel-prize-
quality work, and yes, you'll definitely get at least one cover of
"Nature" out of it.
You've got a population of some of the brightest young scientists in the
world, they're all looking to make their reputation, and you've got an
idea that will do that and much, much more.
So why isn't this idea getting any traction?
The simplest explanation is that not only is "radiometric flaws" a bad
idea, it's been around long enough that we know exactly why it's a bad
idea. The technical term is "cherrypicking", and I've never seen an
argument in this area that relies on anything else.
Let me make you an offer. You convince me that best-practice radiometric
dating has systemic error >10% and I will make sure you get funding. I
will, of course, be asking geologists and nuclear chemists for assistance
(since those are the people who will ultimately be evaluating the work).
I don't need proof, I just need enough data to make a plausible story
that there's something there worth spending money to investigate.
I believe I've read all of the creationist geologists who have written on
this topic recently (the name Mike Snavely comes to mind). The kindest
description I have for that work is "obviously wrong". (This may have
something to do with why they've never been on the cover of Nature.)
One other note: I also do peer-review for journals, so I know precisely
how much politics are involved in getting a paper published. It ain't
hard (hey, I've done it, how hard can it be?). But the work does need to
be more than fan-dancing, and creationist aren't exceeding even that low
bar. Not even Mike Behe was willing to submit his creationist work to
peer-review; he's not an idiot, he knew where his holes were.
If you can do even modestly better, then fame, fortune and scientific
plaudits await.
<snip>