[Aztlan] Empirical alignments

[David Hixson]

For some, Lloyd's last post may have appeared encoded, please forgive this re-posting of his last email. -- Dave.


Lloyd Anderson wrote:
----------------------------------------

John Major Jenkins asks:

<<Would you allow into the discussion, as a piece of evidence, the empirical alignment of, say, the Izapan ballcourt   to the rise point of the December solstice sun? If that can be identified as evidence then we can move the discussion forward. If it can’t please explain why.>>

It is not up to me to "allow" or "disallow" this.  All facts are allowed, if really facts.
All reasoning can be considered.

What I *can* point to is some attempts to make the question of alignments 
more empirical, not a matter of wishful thinking, selecting just the ones
one wants and shading their alignment slightly (how much is OK?)
to make them seem more exact.

Statistical treatments can do this.  I am much less confident on uses of
single alignments.  That is a more difficult problem, where multiple 
lines of *independent* evidence must converge.  Not merely assumptions
and secondary assumptions to shore up the initial assumptions.

On the one hand, Anthony Aveni took all of the "E-shaped" structures
thought to be used for view lines for solstices (ends) and equinoxes (middle).
He determined a difference between earlier and later that looks statistically
very good (not perfect) distinguishing use of the solstices and equinoxes
vs. use of the zenith passages which approximately divide the Mayan 365 days
into ranges 105 and 260.  I don't remember seeing a test of significance,
but the patterning looked very strong to me simply as histograms 
divided by dates as he did.
I trust the degree of pattern skewing more than significance tests,
because the latter depends too much on sample size, which is of somewhat
limited relevance where we have tiny samples of ruins accidentally preserved.

On the other hand, a fellow named Brad Schaefer who is very very
good, points to the large number of instances where claims which
looked good at a single example prove to not work when one looks at several
sites, not because they are not the same, but because the same *methodology*
does not work for them.  He uses a lot of statistics.  I don't always agree with
him, this he has mis-defined the nature of the problem in one case at least
(which greatly affects statistics)
but he is the kind of person you need to satisfy on alignments, not me.

One reason we don't all jump to discuss Izapa is because 
we can't all be experts on everything, and don't presume
we are competent to judge Izapa if we do not know it well *in depth*.

If you can simplify your presentations, and show consistency of patterning,
a pattern which can be seen on one or two pages maximum,
then people can chew over your *data* (and reject some of it, of course,
as happens to everyone).  That's the way it is *supposed* to happen in 
collegial science.  

And evaluate whether you have asked the right
questions, or whether the patterning you claim is real or strong enough.
Partly what you get back is that you ask supposed questions which already
have your answers built into them, they aren't really questions,
or you forget that you are choosing only the examples you like 
(as pointed out by Jorge).  All of those actions undermine your claims
to support your hypotheses.

Science is not merely consensus.   It is consensus among those who
understand a field with ideally extensive opportunities for deliberative
discussion and multiple perspectives (though not about whether gravity
exists in the ordinary person's experience of it).  And again ideally,
with no neglect of unpopular ideas merely because they are unpopular,
but if there is to be neglect, it has to be justified to a plausible degree
or better.

Consensus on a falsehood does not make a falsehood true.

But I can no longer participate as much in these discussions for a time.

Best wishes,
Lloyd

Lloyd Anderson
Ecological Linguistics
PO Box 15156
Washington DC 20003
ecoling at aol.com
202-547-7683