A handbook for new diggers?

[Dinosaur]

Educating people on what can/should be taken, how, why, what other techniques could be applied, new techniques etc etc, what information can be derived, when to ignore the 'expert', a certain amount of 'canniness' with funding, some politics.....sampling is fun! [despite all the bedtime background reading]

[kevin wooldridge]

Where's Seedy girl when you need her?

If you ask me................depends on what question your asking............as in what is it you are digging and why.

Your sample strategy should be tailored to the project.

I think I agree....but maybe 'tailored' isn't the word you want to use' as this suggests a pre-determined bias that might confound the purpose of the sample. Perhaps 'sampling strategy should be determined by the research aims of the project, but may be constrained by the available resources'......

[BAJR]

Simple question... when to take a bulk sample... why..... and how.

[Dinosaur]

A decision on whether to sample often comes down to how far you've got to lug it back to the vehicle - sad but true :face-crying:

Have been doing a job recently where mass-sampling has been out of the question simply because of the 20 minute walk back with each tub (=4 trips per 4-tub/40L sample...tried taking two at a time but that seemed a fast route to permanent disablement), which if nothing else meant no archaeologist on site for 40 minutes on each round trip = 1/3 day per sample..... Luckily the ones that did get lugged off seem to be crackers although still awaiting full analysis, so still not lost my knack of spotting the good stuff

[Kel]

Appropriate sample size, from the correct place, for the planned analysis.

I remember an environmental archaeologist wearily describing how someone brought him several buckets of soil, when what he needed was a sample the size of a matchbox from a very specific location. Plus a site post-ex plan which had specified a one litre soil sample from each context. Where a context had contained less than one litre, this had been made up from the spoilheap...

[Dinosaur]

Oh, right, I won't bother making any comments about contamination then....:face-stir:

[Jack]

Simple question... when to take a bulk sample... why..... and how.

When you want to:
a) look for material to radiocarbon date
b) recover small bones (especially fish bones)
c) recover other palaeoenvironmental evidence - including snails, charred plant remains, other shells etc. etc.

The caveat is, however, the information recovered may be useless if you don't understand the formation processes of the context you are sampling...........

It can be complicated. For instance. There is no point sampling a context for charred material to date if its say, from a pit, dug into a ditch, dug into a buried soil.....and it looks like the context in question has derived from the re-deposited fills of the earlier features......unless, you've got a big dump of say charred grain that looks like its one event and not re-deposited earlier material.

Also, as Dino said you gotta be canny on strategy and interpretation.
Take the example of a well-dated (from pottery for example) pit in a settlement with a 'black looking' deposit that contained charred grain in it. The pit seems to have been back-filled quickly, with a narrow date range of pottery, no inter-cutting, etc.
The supervisor takes 4 tubs from this pit and say 2 tubs (100%) of the fill from another similar pit but doesn't sample any of the other features in the settlement because they didn't look like they had much stuff in them and to save money.
In interpretation, the supervisor states that the area where the two pits were represents a crop-processing area. And uses the analysis from the 4 tub sample (as it had more grain) to calculate the ratios of crops used by the settlement inhabitants, finding that they relied of oats more than say wheat and barley. In his discussion he goes on to link this with the altitude of the site and contemporary palaeoclimatic data.

Unfortunately, years later the site is redeveloped and the previous area is re-dug along with the rest of the settlement using a comprehensive sampling strategy, taking samples from most features (several from some of the ditches) in an equally distributed pattern across the whole settlement.
From the analysis it turns out that the former 'crop-processing' area is just part of a background of charred grain across the whole settlement, the spikes of charred grain/chaff/seeds were focused around most of the roundhouses. Furthermore the analysis of the much larger sample of material relating to the main phase of occupation contains a much higher proportion of wheat than oats. Radiocarbon dating also showed that the pits from the previous excavation were from a later phase............turned out that the three bits of 1st century pottery used to date the pits were small, abraded and redeposited..........

Sorry that was a bit of a ramble...................my point being, there are no hard and fast rules on when to take a sample, it depends on what your final objectives are.

[Dinosaur]

I remember an environmental archaeologist wearily describing how someone brought him several buckets of soil, when what he needed was a sample the size of a matchbox from a very specific location. Plus a site post-ex plan which had specified a one litre soil sample from each context

Am impressed by specialists who can reconstruct past environment, diet etc from only 1 fishbone and a couple of charred seeds, must save loads of money, although of course they're shooting themselves in the foot commercially, the usual mantra in evaluation reports is "this looks interesting, process the other 30L of the sample and then I'll do some really expensive analysis" }

[P Prentice]

Simple question... when to take a bulk sample... why..... and how.

simple answer - ask your specialist .... and if you havent got one you shouldnt be digging in the first place

[P Prentice]

I think I agree....but maybe 'tailored' isn't the word you want to use' as this suggests a pre-determined bias that might confound the purpose of the sample. Perhaps 'sampling strategy should be determined by the research aims of the project, but may be constrained by the available resources'......

i think jack is correct and 'constrained' means the reseach design was inadequate