JB: From your emphasis, I gather you favor Christopher Stringer's Out of Africa theory over the multiregional approach to human evolution of Milford Wolpoff.

RENFREW: If you're looking at the Southeast Asia hominids it does look, to the very casual eye, and indeed to the very specialized eye, when Wolpoff and his colleagues get going, that there are some local features that you see in Homo Erectus that seem to be carried through into Homo Sapiens. The choice is a fairly simple one: The idea that Homo Sapiens emerged out of Homo Erectus in Africa is the easiest idea to grasp, because if you're a geneticist, you really don't like these things happening in different places simultaneously. The other idea is the notion of some universal sort of global gene pool in or through which Homo Erectus gradually evolved over a wide area, into Sapiens is a little difficult. The bottom line is that nearly all geneticists now feel that in terms of the genetic evidence, and not just the mitochondrial DNA, that Out of Africa, for Homo Sapiens makes sense.

But if you're starting from the skulls, there are some local features in Homo Erectus that seem to be matched later by local features in some of Homo Sapiens hominids, and there's also the matter which I've not seen much discussed, that in the western part of its territory, that is to say Africa and western Asia, Homo Erectus makes hand axes. In southeast Asia there is a pebble tool tradition. The curious thing is that when Homo Sapiens comes into play there, the pebble tool tradition still continues in southeast Asia, although hand axes give way to other things in the West. So although the genetic evidence probably will prove conclusive, and that the Chris Stringer and Paul Mellar's view may well win out, it has to be said that Wolpoff and his colleagues still point to pieces of evidence that do support their case, and are difficult to place in any other framework.

JB: As technologies change, will some of these positions be rendered obsolete.

RENFREW: One or other will have to give way. I have to say, perhaps because I've got colleagues in Cambridge that very much think along the lines of the Stringer position, that I predict that his approach will prevail.

JB: Let's talk a bit about how you got to where you are in terms of your science.

RENFREW: I started out at Cambridge doing natural sciences when I was an undergraduate, so I did Part One Natural Sciences, and then I went on to do Part Two Archaeology. I decided halfway through I really wanted to be an archaeologist. So that was straight archaeology; it wasn't the discipline that's emerged more recently, which is sometimes called archaeological science, which means the application of scientific techniques to archaeology.

But then I got into a very interesting problem which involved prehistoric trade. There was one commodity, a stone called obsidian, a volcanic glass, which was very widely traded way back in neolithic times. Obsidian is quite widely used in the world, but it's found in very few places. So when you have a good source, the obsidian tends to be traded over hundreds or even thousands of miles from that source. The scientific question was how can you characterize the material from one source as related to another source. If you find a piece of obsidian dated at 8,000 BC at Jericho, and you're a thousand miles away from a source, can you really say where this piece of obsidian came from? We found a way to do this by using trace element analysis. We originally used optical emission spectroscopy, but more recently neutron activation has been used. This was one of the first successful characterization studies that allowed the reconstruction of trading patterns.

JB: To what end was this currency used?

RENFREW: It wasn't currency; it was supply, really. They had a need for the raw material for making lithic artifacts, and this was the best material, so it was very widely traded, exchanged, or transported. But your question is a very sound one - sometimes it's the indicator of contacts that are taking place of which you'd not otherwise have knowledge. If you get a quantitative analysis you do falloff maps and so on, you can get a measure of interaction, and the obsidian may not have been the sole motivation for that interaction, but it's something accompanying it. I began to talk of trade as action at a distance, and began to get a measure of the extent to which communities were interacting. This is just a rather abstruse, but very concrete, measure of that interaction.

JB: What pictures could you draw from that?

RENFREW: It shows, for instance, that at the beginning of the Neolithic period, the beginning of farming in the Near East, just about everywhere was in contact with everywhere else. There is no early farming village in the Near East that doesn't get obsidian, even though the obsidian sources are hundreds of kilometers to the north. Obsidian from Melos, which is an island in the Aegean, is found way back before farming, 10, 12, 13 thousand years ago, so this meant that the Paleolithic hunter-gatherers must have been traveling in boats. Similar evidence for early seafaring has now been found in the Pacific, and it gives an indication that people were much more efficient at seafaring than had been imagined.

Then I got interested in radiocarbon dating and its interpretation. I've never made any scientific contributions to that area in the sense of technical contributions, but because I was interested, and I had some sort of broad grasp of the impact of tree ring dating upon the radiocarbon chronology, I saw very clearly how the tree ring calibration was right and how the objections made by European archaeologists had been founded on some doubtful assumptions. They argued that this couldn't be right because it's not what we already know. I saw very clearly how their arguments were flawed, and was able to clarify some things.

Otherwise my contributions are based just on taking an interest. For example, in terms of molecular genetics, there aren't many archaeologists very interested in molecular genetics in relation to the languages. It helps not to be intimidated by the undoubted complexities of the scientific methods. Ultimately many of the problems, as we were saying a moment ago, are problems of interpretation. You were rightly saying that in molecular genetics, the difficulties are not really so much, or not so often, in the molecular genetics itself, but in the interpretation of the data. And there sometimes a different viewpoint is a helpful one, I think.