I’m still in Jordan but I finally got that pesky thesis finished and into the multiple revision process. I’ll have to say I am fairly happy how it turned out. There were a few times when I wasn’t too sure if it was going to come together. Now, I don’t mean get it finished on time because the submission deadline isn’t actually until June 15th. I was worried about how to effectively portray all the variables I needed on one graph. However, thanks to the great help of a friend in my lab, the Computational Archaeology Lab at McGill, I was able to figure it out.
Since it’s been a while, before I tell you about the graph I better review the thesis. The whole point is to test current descriptions of the prehistoric transition from chipped stone to ground stone technology on the southwest coast of British Columbia, Canada. Specifically, I am looking at whether or not the concept of phase (prehistoric periods of time for which all sites get grouped together) has an impact on interpreting archaeological change. So, I analyzed the above mentioned transition using prehistoric periods and then again without them, using strictly carbon-14 dates, to see if there was a difference in the temporal and spatial pattern of change. The tricky graph I mentioned was the final one, placing all the assemblages (75 of them) on one graph that had time and space (coordinates) represented as continuous axes as well as including the proportion of chipped stone, ground stone, and faunal tools for each assemblage. That makes 6 variables for one graph…and voila….kind of
What we came up with is this graph. It is a basic scatterplot with the Y-axis being time, the X-axis being space, and the assemblages coloured using an RGB composite where Red = ground stone proportions, Green = faunal tool proportion, and Blue = chipped stone proportion. This gives every assemblage a unique colour based on the combination of the three colour proportions. There were two tricky parts. First, the spatial axes is actually two variables, the latitude and longitude. Luckily our sites are positioned on a fairly east/west axis so we reclassified them as the distance from the mouth of the River. Zero kilometers is the river mouth and everything East is inland up the river while everything to the West is in the delta or coast islands. You can see the division between deltaic sites and coastal island sites clearly on the graph between 30 and 50km W. The other tricky part was the legend but thanks to some helpful colleagues we got it figured out properly. What the graph shows is a pattern that goes against the current descriptions of the transition and suggests that continuous axes of time and space are more appropriate than archaeological phases for understanding change through time. I’m sure more info is needed but that will have to come later.
For now, enough technical mumbo jumbo. I’ll get you up to speed on the work here in Jordan very soon!