It is very hard work to stop myself sharing all the pretty graphs I’m drawing but I’m resolved to wait until the conference itself to show my whole hand on this data. Yesterday was super productive largely because my quantification of the variation between readings from the same specimen meant that I asked for help from Wayne and learned a whole bunch in the process.
Ternary plots! A bit like that triangle below. This lets demonstrate how the ratios of copper: lead: tin appear in all specimens and allow v clear patterns to appear. 3 axes are way cooler than just 2 and a nice mental baby step towards more multi dimensional plotting stuff like Correspondence Analysis (CA)–that’s the cool type of data exploration used by Kris Lockyear to explore hoards. Unfortunately excel needs a plug in to draw these. Everyone’s favorite seams to be tri-plot. There is a fancy add-in called XLStat but it isn’t available it seems for the version of MSOffice my university gives me. BOO! Maybe I can figure out how to get a copy it looks like a great deal of fun.
Color issues! So I’ve known for a while (see below) that we’re not sure of the original surface look of aes grave, because of the high lead content, but there is some suggestion from unpublished data from the Balkans that high lead may have been used in the bronze age specifically for decorative and ritual objects because of the silvery surface color. This lead me to be v tempted by experimental archaeology to produce a few specimens to test.
Division of elements into intentional and unintentional! So I’d been mostly doing my thinking about ratios leaving the trace elements in, but this doesn’t account for original intent. There is no evidence they meant to put antimony, silver, iron, arsenic etc… into these things. They are impurities in the ores or bullion used. They tell stories but if what I want to recover is the recipe or recipes the Romans intended they only add noise. Removing them from some of my calculations has made a clearer picture.
Base-12! Ok this inspiration was my own not Wayne’s. Any target recipe was likely in Roman units of measure, likely weight and likely on a base-12 system. So we’re talking uncia and scruples once we get below a pound. This is proving a very useful means of thinking of how the data patterns might fit into Roman intention.
Casting SAND! This I owe to my v practical neighbors who know all about metal for industrial and artistic modern applications. I was running the idea of what it would take to try some experimental archaeology and we got to the material to cast in and we talked through clay and that can explode presenting safety issues as we learn what we are doing, so we thought safest to start with green sand, which means moisture-rich sand and is in no way green in color, NOT to be confused with the geological/gardening green sand. I asked how you make a shape you want and was told you just impress it same as clay. I don’t think this is what the Romans used for aes grave, but I wonder about some other aes formatum/italic aes grave. It would be fine for ramo secco and probably also the bronze shells. The later worry me. None in Haeberlin and I don’t know (yet!) of any hoards with them or site finds, a quick internet search shows Italian metal detectorists seem to turn them up but not sure I want to put scholarly weight on that. They also show up on the market but are nice collectors pieces so no surprise there. Cf. Vecchi 2014: p. 90 figs. 4-5.
It I do try some casting I think I’ll aim for this shell shape to start.
Today’s goal is to finalize visuals and start to make sure sequence of slides tells the big story of why investigating this what I did before, what comes later and what I know now. I want to try to narrate without script.
Liv Mariah Yarrow 