On 2006-08-20 09:17, Salopian wrote:
Joking aside, I reckon archaeo-astronomy may end up as a respectable science, but ONLY if it has a big element of statistics and mass regression analysis in it. They're starting an undergraduate course on it at...(west wales, name escapes me). Good for them. But if they're going to do it properly I bet they do button throwing on day one, else they'll be barking
(up the wrong tree.)

On 2006-08-20 10:26, Salopian wrote:
Well basically its a recognition that any observation can be due to not one but very many possible variables so a "survey" isn't useful unless you can tease out which observations are subject to which variables. Ideally, you keep doing surveys over and over until you understand which specific observations are the true determinants of your observations, and to what extent, at which point you can say yes, variable X is definitely the one that is the main reason for the observations. I spent many years applying it to research in the field of house prices, with some progress. Once you develop some coefficients relating to the true effect of variables such as double glazing and second bathrooms on the whole market value of a house then you get a handle on them, to a much better degree of accuracy than estate agents' flim flam.

It's a little known fact, IMO, that house valuation is just about the only field of human endeavour that has never made progress (apart from archaeo-astronomy and gameshow hostessing of course!)

On 2006-08-20 11:29, Salopian wrote:
it's not meant to be complex. If a single long barrow points to another site or a river or both it can't be reasonably claimed that there's a deliberate river alignment, though it tends to be on internet sites. So you need to spit the analysis between barrows near just near rivers and barrows just near other sites - and if the former show 70% river alignment and the latter show 5% "other site" alignment you have a reason to say the first barrow, most likely, points to the river, not the other site. Which is a bit of progress compared with before.

Multiply that process to scenarios where there aren't just two possible causations but dozens and you'll realise what painstaking work it is - but also what a wide open science is there, waiting to be explored and how very much more we could know if we approached it right (IMO). I envy those undergraduates. They'll be the first generation to be able to talk about site alignments with a bit of authority.

On 2006-08-21 00:09, mithra wrote:
You have to realize that some of this Archaeoastronomy is pants.
For a start off most of the star patterns we see today did not apply back then....nor would patterns such as 'The Big Dipper' (USA) or, as we call it, 'The Plough' have meant anything to them - even if they'd been apparent at the time, which they weren't - just random insignificant stars, which looked nothing like they do today!
Most so called 'rock art' is naturally formed - so the patterns have no significance whatsoever. They simply liked these stones and are only available in specific regions.
I have a professional astronomical program given me by a astronomist i.e. you cannot get it elsewhere. I can find out what the constellation patterns were for any given time and if you take 19 'markers' on a stone you really can find more than 'a few' star patterns which would've applied over 5,000 years ago.
Ancient man did not carve constellations these on stones. Mithra.