And again...is archaeology science?

And we have a resounding “NO!” from Willie Mc Loud of Life in the Universe. In fact archaeology’s not even history either. Harsh.

This is old ground, and I don’t have the burning interest in the topic that I used to, but  I was interested in what Loud had to say because he isn’t an archaeologist, but a physicist, one with, apparently, an interest in archaeology. Now I do think if you are going to criticise a discipline, it behooves you to get up to speed on the basics of that discipline, at least enough that the criticisms make sense. Loud stumbles in this regard. For example, his criticisms of archaeological dating made no sense to me until I realised he was unaware of the difference between dendrochronology and radiocarbon dating. Mistakes like that are a red flag, and make it hard to exercise the principle of charity.  But we must.  

Loud has some definitional problems.  His take on the goals of archaeology, and history for that matter, is pretty limited. His interest is Mesopotamian archaeology, and I suspect he, as with many buffs, is interested in archaeology insofar as it can fill in gaps in a rather conventional history. In his examples he judges archaeology on whether it can identify the reign dates of Naram-Sin, the occurrence of a historically recorded “great rebellion” against the aforementioned Naram-Sin, the existence of the Akkadian Empire, or the presence of Assyrian colonies in Anatolia.  This is “pots=peoples” archaeology failing at “big men, big battles” history.  Archaeology does not excel at this kind of history.  In fact history does not excel at this kind of history.  

His other definitional problem is what he means by “science”.  He’s a little slippery here, conflating “empirical science” (which is in reality all science) with experimental science.  An experimental science is one whose findings come from controlled repeatable experiments.  By this definition, psychology is a science, but astronomy, geology, evolutionary biology, and, yes, archaeology are not. Obviously not all sciences are experimental sciences—there is an important distinction to be made here between experimental and historical science, in which researchers use  pre-existing evidentiary traces to discriminate among hypotheses (e.g. Cleland 2002). If archaeology is a science, it is a historical science, not an experimental one (setting aside the subfield of experimental archaeology).  
       
Loud also criticises archaeology for problems that are general in science.  He notes, for example, that “archaeological data is provisional in the sense that there is always the possibility that new data will be found that could (even dramatically) alter the picture.” I assume he means archaeological interpretations or theories are provisional, rather than “data”.  Regardless, it seems an unfair observation, since provisionality and a willingness to revise theories in the face of new data are hallmarks of science. 

He is careful to note that scientists that scientists approximate reality rather than discover it, writing that “although reality exists, we do not know - and would probably never know - what it truly is like.  There is absolutely no way to establish whether any model describes reality as it really is - the best we can hope for is that models provide reliable results when tested against reality. ”  Of course this phrasing raises the question of how one “tests” models against an unknowable reality. What we actually do is test models against human interpretations of the effects (”evidence” or “data”) of that unknowable reality.  This is an important distinction, since he then goes on to state that archaeologists cannot test models since the reality they are approximating no longer exists. There are a lot of criticisms I can make of Loud’s idea that the particular reality under investigation needs to be coterminous with the researcher to be available for testing. But the simplest one is that it doesn’t, not in the historical sciences. For examples, nobody deals with long-gone realities like astronomers do.  The effects of the bygone reality are still present, and that is what we test against.  I also wonder at the practical difference between testing against the effects of an unknowable extant reality, and testing against the effects of an unknowable extinct reality. The practical difference between the two is non-existent.  

The question of whether archaeology as a whole is science is not one I worry much about anymore.  A lot depends the questions the archaeologist is asking—some variation no doubt of Christopher Hawkes’ “Ladder of Inference”.  When humans first migrated into the Americas is a scientific question.  How a group of people uses material culture to assert (or deny) group identity is less so, but nonetheless still worth exploring.  We can’t just look where the light is strongest.  Also archaeology, like history, is a form of social memory and thus embroiled in contemporary politics in a way that physics and geology are not.  

For the most part I agree with Loud that the kind of archaeology and questions he is interested in are not science. I just disagree with his reasons, and that his experience of archaeology is sufficient to take on the whole field.  Maybe a more limited title?

----------------------------------------------------------
References cited

Cleland, Carol E.

2002    Methodological and Epistemic Differences Between Historical Science and Experimental Science. Philosophy of Science 69(3):447–451.

Loud, Willie Mc 
2012    A critique of archaeology as a science.  Life in the Universe.  Posted 19 August 2012.