Sherlock:

How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?

Sam Vimes:

… he distrusted the kind of person who’d take one look at another man and say in a lordly voice to his companion, ‘Ah, my dear sir, I can tell you nothing except that he is a left-handed stonemason who has spent some years in the merchant navy and has recently fallen on hard times,’ and then unroll a lot of supercilious commentary about calluses and stance and the state of a man’s boots, when exactly the same comments could apply to a man who was wearing his old clothes because he’d been doing a spot of home bricklaying for a new barbecue pit, and had been tattooed once when he was drunk and seventeen and in fact got seasick on a wet pavement. What arrogance! What an insult to the rich and chaotic variety of the human experience!

[…]

The real world was far too real to leave neat little hints. It was full of too many things. It wasn’t by eliminating the impossible that you got at the truth, however improbable; it was by the much harder process of eliminating the possibilities.

For my money, Sam Vimes is the better detective. Here he has nailed the concept of equifinality, the idea that many different paths could lead to the evidence that we find. Vimes would make a good archaeologist. The problem, though, for archaeology (or history, or anthropology, or… or… or…) is that we don’t really grapple with the idea of equifinality in our writing. This is why history, archaeology, digital humanities, needs an experimental mind set. Through experimentation, we can whittle down the possibilities. Scott Weingart has an excellent post on this very issue, through analogy with astronomy:

Astronomers and historians both view their subjects from great distances; too far to send instruments for direct measurement and experimentation […] historians are still stuck looking at the past in the same way we looked at the stars for so many thousands of years: through a glass, darkly. Like astronomers, we face countless observational distortions, twisting the evidence that appears before us until we’re left with an echo of a shadow of the past. We recreate the past through narratives, combining what we know of human nature with the evidence we’ve gathered, eventually (hopefully) painting ever-clearer pictures of a time we could never touch with our fingers.

In which case, what we need is a laboratory for running different micro might-have-beens (I distrust uber-do-everything-social-models, because with so many moving parts, how the hell do you know what’s going on?). Fortunately, Scott and I have just published one over in the Journal of Archaeological Method and Theory. (It will be part of a special issue on archaeological networks analysis; but JAMT does a thing called ‘online first’, so you can see articles before they’re pulled into a particular issue.)

The Equifinality of Archaeological Networks: an Agent-Based Exploratory Lab Approachhttp://dx.doi.org/10.1007/s10816-014-9230-y.

The agent based model itself is available on figshare at http://dx.doi.org/10.6084/m9.figshare.92953.

 

In this article using this agent based model, we examine a theory of Roman economic history; we work out the kinds of generative processes for networks such a vision of the Roman economy might use; we model these; and we compare with known archaeological networks. More or less; you’ll have to read it for all the nitty gritty. Happily, if you’re paywalled-out, here’s a pre-print (the open access fee for Springer is $3000 US !!!! and I ain’t got that kind of cash – but that’s the subject for a post for another day).

Abstract

When we find an archaeological network, how can we explore the necessary versus contingent processes at play in the formation of that archaeological network? Given a set of circumstances or processes, what other possible network shapes could have emerged? This is the problem of equifinality, where many different means could potentially arrive at the same end result: the networks that we observe. This paper outlines how agent-based modelling can be used as a laboratory for exploring different processes of archaeological network formation. We begin by describing our best guess about how the (ancient) world worked, given our target materials (here, the networks of production and patronage surrounding the Roman brick industry in the hinterland of Rome). We then develop an agent-based model of the Roman extractive economy which generates different kinds of networks under various assumptions about how that economy works. The rules of the simulation are built upon the work of Bang (2006; 2008) who describes a model of the Roman economy which he calls the ‘imperial Bazaar’. The agents are allowed to interact, and the investigators compare the kinds of networks this description generates over an entire landscape of economic possibilities. By rigorously exploring this landscape, and comparing the resultant networks with those observed in the archaeological materials, the investigators will be able to employ the principle of equifinality to work out the representativeness of the archaeological network and thus the underlying processes.