Prof. Felix Riede - "Past 'Cultures' at the Macro Scale"
Duration: 36 mins 23 secs
Share this media item:
Embed this media item:
Embed this media item:
About this item
| Description: | A recording of Prof. Felix Riede (Aarhus University) speaking on "Past 'Cultures' at the Macro Scale: An Archaeological Perspective on Culture and Culture Change." as part of the "Cultures at the Macro Scale" seminar series. |
|---|
| Created: | 2021-02-11 11:03 |
|---|---|
| Collection: | Culture at the Macro-Scale |
| Publisher: | University of Cambridge |
| Copyright: | Andrew Buskell |
| Language: | eng (English) |
| Keywords: | Cultural Evolution; Archaeology; Cultural Taxonomy; |
Transcript
Transcript:
0:00
Well, thank you very much for having me for this quite interesting seminar series about "Culture at the Macro Scale: Boundaries, Barriers and Endogenous change." Sometimes it's nice to talk about, you know, your own work... your own small studies. And sometimes it's really nice getting a bit more of a set task. And today's presentation is more like the latter.
0:26
So I've been asked to talk about "Culture at the Macro Scale". And what I'm going to do specifically is talk about past cultures at the macro scale. So what I'd like to do is provide an archaeological perspective on culture and culture change. The seminar organisers have posed a series of key questions to all the presenters, and they are: "How should cultural groups be distinguished from one another?", "How can cultural groups be identified from data?" and, "How should cultural barriers and borders be understood?" In addition to that, in my own invitation, I got a more specific task frame, and that was addressing cultural taxonomy; so the definition of cultures in archaeology and specifically in relation to genetic groupings. And this goes back to a paper that we recently published on reconciling archaeological and genetic taxonomies and why that is important—and I will get back to that work a little bit later on.
1:37
Now, thinking about culture change, it is unavoidable really to address... talk about this particular evocative and iconic illustration from Kroeber back in 1948. Here he contrasts effectively the Tree of Life—the patterns of biological evolution—on the left hand side as one of branching divergence with change in human cultures as more of a braided stream model: of reticulating networks that obscure and spin off different directions in terms of some kind of clear evolutionary pattern. However, there is no doubt really, that humans—biocultural creatures—they have a cultural makeup and there is tremendous, tremendous cultural richness across the world. But they also have a biological makeup.
2:45
So we are also animals, we are biological creatures, we have phenotypic traits, and we have specific genetic traits. For a long time, anthropologists and archaeologists were really concerned with defining their own discipline, and doing so in opposition to to biology. So back in 1952, Kroeber and Kluckhohn catalogued 164 definitions of culture. Twenty-two of those explicitly talked about a notion of tradition or inheritance, social heritage, or social heredity, it ... as it was called at a time, and some of them even reference processes or mechanisms such as learning.
3:37
But at the end of the day, most of these definitions are quite useless for archaeologists. Now, gene.. genetics, also deserves a little comment here that genetics is a pretty well understood transmission system. And it can take different kinds of forms. And when those folks in the social sciences, or humanities anthropologists, archaeologists, but also many, many lay people think about genetic transmission, what they normally think about is a large mammal transmission. But not all organisms have vertical transmission. Plants have a much more complex patterns of genetic inheritance. And let's not even talk about bacteria—other kinds of creatures—where any kind of horizontal oblique transmission is more the norm than the exception. So, large mammal genetics is perhaps not the kind of model system that we need to contrast cultural transmission with, when we think about how those two may be related.
4:48
At any rate, if you take a given human population as a biocultural phenomenon—that is one that composes a degree of cultural diversity and genetic diversity—and you add the dimension of time, there will be salient correlations emerging over time: there will be more or less connectedness between transmission patterns in biology and in culture. And if you add to this the spatial dimension, you will automatically and naturally end up with a patterned landscape with a variable distribution of the cultural diversity and the genetic diversity, the degree to which these are correlated map onto each other. How in the degree to which they leave "parallel tracks in time", as Rob Foley called it some time ago, is an empirical one.
5:46
And this is really where the cultural evolutionary approach comes in. And early in the 1980s initial steps—very significant steps—were taken by a Cavalli-Sforza and Feldman and more or less in parallel by Boyd and Richerson. And in terms of formalising the way in which we can approach the study of cultural evolution, these initial approaches drew extremely heavily on pop... population genetics and associated mathematical models.
6:18
More or less in parallel—but also somewhat separate—archaeologists, were b... were thinking about the application of evolutionary theory to culture change in a more deep time perspective. And initially, this really was through Dawkins, this notion as the extended phenotype. So here, material culture was very, very directly understood as the hard parts of the human phenotype, so very much like fossils. And if you look into volumes, published like these in the late 90s, early 2000s, you see, you come across a relatively hard rhetoric about a kind of transferal of evolutionary thinking to material culture.
7:13
In more recent times, all of this has coagulated into and developed into a really rather coherent framework for cultural evolution. (Sorry, I just have to get some coffee here.) Where the idea of the of material culture as the hard parts of the extended phenotype, it's more or less been given up in favour of the idea of dual inheritance or the parallel linked—more or less linked—trajectories of cultural and biological evolution. Ma.. mathematical models still play a major role. But a lot of hard thinking about conceptual issues epistemic issues, empirical issues has also occurred.
8:05
One important introduction for archaeologists, at least, that happened already in the early 90s, was the idea that perhaps we can use some of the tools of paleobiology to study also material culture change over time. So while we do have a kind of grand framework for how biological evolution and cultural evolution match up, (this here taken from from some of Alex Massoud his work), where palaeontology and evolutionary archaeology are really matched up as those disciplines that play a major role in addressing macro evolutionary change in the biological and cultural worlds respectively.
8:54
But this is not to say that cultures are like species. This notion, of course, has been around for a long time and it rears its ugly head again and again. But the analogy should not must not be understood in a direct kind of fashion. Rather, what needs to be appreciated is that there are significant epistemological and data structural similarities between paleobiology and evolutionary archaeology.
9:29
And this goes a long way back indeed, to the Cambridge philosopher, mathematician, chemist, mineralogist—you know, polymath—William Whewell, who in the mid in 1847, I believe, published a treatise where he would have the historical sciences classed together as one kind of way of understanding the world. So to him departments should be made up of geologists, historical linguists, art historians and archaeologists, because all of their data has an inherent temporal dimension. And it is about inferring patterns and processes from usually actually a fragmented record. (And quite interesting really.)
10:18
But also found... archaeological founding fathers such as the Swede Oscar Montelius here in the middle. A very amicable chap—who really developed the so called topological method—drew direct parallels to Darwinian evolution in his writing. And most recently, the ever optimistic Adrian Currie highlighted the epistemic and data structural similarities between palaeontology and, and archaeology, I think, to quite a lot of benefit.
10:52
But stop. This talk wasn't really meant to be a kind of an abridged discussion of the intellectual ancestry of evolutionary approaches to culture. I did want to do something that was a little bit more case-based—something that also addressed the questions said in this seminar a little bit more directly. So let's change track here slightly.
11:19
Our recent work on trying to understand cultures analytically at the macro scale, goes back some time, I have long been concerned with this, in fact, going all the way back to my Ph.D. But more recently, it was a publication by Eisenmann et. al, where they made a range of very practical suggestions as to how to think archaeological and paleogenomic clusterings together. And they came up with a kind of taxonomic system, which they call the mixed system where any kind of cluster genetic cluster/paleogenomic cluster is matched up with an archaeological term that is a hybrid of geography, chronology and sometimes also a kind of cultural labels. So something that perhaps refers to the actual content of the material culture.
12:26
The argument for adopting such a mixed system taxonomy was that it included.. that it offers brevity, coherence, accessibility, flexibility, and stability. And those are all very important aspects of any kind of taxonomic system. But actually, I argue, together with my co-author Steve Shennan and Chris Hoggard would actually prefer theoretical rusticity, transparency, interoperability and analytical power as the driving factors behind any kind of taxonomic system.
13:04
Our concern here, of course, was not talking to a lay public or telling compelling narratives about past processes in the first instance. Instead, our motivation was to make the notion of culture in an archaeological sense: actually analytically useful. In our ... problems, so to say, with the suggestions by Eisenmann et al. was that cultural labels—such as the "Bell Beaker" culture, the "Lyngby" culture, you name it—the archaeological record is populated, over populated, some would say, by these kind of named cultural groups. They are rarely unanimously defined. So if you ask ten archaeologists how they would define a given culture, you're likely to get definitions that actually vary tremendously.
14:02
Some of the taxonomic suggestions made by Eisenmann et al. also relate to subsistence practice labels such as Neolithic, for instance, but they can... they are not really well, they're rather... we would argue they're rather ecologically determined than determined by inheritance. And these can be confusingly assigned anyway.
14:25
So there's a lot of talk also in the paleogenomic literature about the "Pitted Ware" culture. Sometimes they're called Mesolithic, sometimes they're called Neolithic. It is definitely a cultural phenomenon that is chronologically framed by the Neolithic, that is by societies that use domesticates and pottery, but they adopt more or less a hunting and gathering kind of economy. It's a super fascinating cultural phenomenon precisely because it defies clear labels relating To subsistence practice, for instance.
15:03
In addition, sometimes even very sort of standard labels such as Neolithic would vary depending on what kind of research tradition you're from. So in Western Europe, the Neolithic is usually defined as something related to serial agriculture using domestic goods, whereas in a recent Russian research tradition, any kind of culture is effectively termed Neolithic, as soon as you get pottery. So, these are economic versus political or labour related definitions.
15:38
And finally, cultures are time transgressive phenomena. So chronological assignments also have their inherent problems. And partly because dating uncertainties and because in part because these dates would usually vary depending on where you are in space.
15:55
In contrast, genetic labels such as haplotypes, are so useful and so precise, because they allow inferences about historical processes. And this is rests in them being linked to an explicit process model of change, evolutionary theory, from which flows a nested hierarchical taxonomy that allows the division of any given sample into clusters. And the relative relatedness of these clusters is informative vis-a-vis the histories. And most of our cultural taxonomic systems use an archaeology don't have... are not underwritten by these kind of process oriented theories, and they do not result in nested taxonomic systems.
16:47
So, this is just one example here, where you see the final Palaeolithic cultures of Europe divided into a series of different kinds of cultural groups. It is quite unclear also from the original writing, whether these groups are to be understood in the same way as contemporary hunter gatherer groups understood for example, whether they would have a kind of conscious ethnic identity, whether there would be linguistic differences inferrable from these kinds of maps. Also, you cannot really read in any kind of detailed or informative manner, the real distribution of archaeological sites from such maps, it is impossible to back-engineer any kind of meaningful distribution pattern here. It's unclear what the empty spaces mean, also note how vastly different in spatial scale these cultural units are, but it is not clear whether they the smaller units are nested in any kind of way within the bigger units.
17:55
Okay. And admittedly, this is an old map, but it is, I would argue, still quite characteristic of many kinds of maps that you will see in the archaeological... archaeological literature. Now critiquing archaeological thinking on cultural taxonomy, or indeed on the notion of culture itself is not new. For a long time, archaeologists, especially, perhaps in the Palaeolithic have been arguing that many of our culture definitions are analytically dubious, and they're mostly grounded in what what has been called accidents of history. That is, particular kind of sociopolitical currents that characterise a given part of the world. And they've given rise to a specific kind of spatial temporal circumscription of archaeological phenomenon.
18:51
So arguably, at least the Palaeolithic record, but probably also, the record of later times is populated with what Barton and Neeley called "phantom cultures". And more radically, folks, such as John Shea have argued that a lot of these labelled cultures these named cultures (he calls them "NASTIES") should indeed be abandoned.
19:18
In the slide before us, just so, you saw a sample of just the international English speaking literature on the topic, but there is a vast undercurrent of more specific writings targeting more specific cultural groups at at national or regional scales. Now, the drawback of this literature which often is very, very, to the point and very much goes to the empirical foundations of these cultural groupings is that it is usually written in relatively obscure languages more difficult to find through library... international library systems or databases. And, of course leaves a relatively limited impact on practice.
20:07
So over the last few years, we have been trying to draw some of these patterns out and trying to publish new work on cultural taxonomy here specifically in relation to the late or final Palaeolithic in an open access fashion and in the English speaking literature. So there's no doubt that some of these pilot studies here are merely indicative of wider conceptual epistemological problems that that plague the the archaeological record of the Palaeolithic and other periods.
20:44
And the problem, of course, is that if our definitions of cultural taxonomic groups of cultures are not operational, then the kind of conceptual upscaling that we'd like to do in terms of archaeological interpretations about processes or interactions, social networks and so on, simply will stand and fall with the robusticity of these archaeological groupings.
21:10
Okay, so, cultural taxonomy is the kind of significant linkage between an empirical groundwork that has to do with how we actually look at at any kind of given artefact assemblage, how we excavate and so on, how taphonomic processes affect the archaeological record and subsequent archaeological interpretations at a more macro scale. But how then can we identify and define these kinds of archaeological groupings, these taxonomic operational taxonomic groups in archaeology?
21:50
So in in an older framework, one that is still very much alive and kicking you have the idea of the type. And topological thinking has long been critiqued in the biological literature as an undo idealisation of the material variability out in a world that is then being filtered in a vain search for for a kind of ideal representation of, say, a given animal or a given artefact class.
22:27
(Now, some of this historically, if you allow me to digression may have to do with the fact that the topological method was initially developed in in relation to mostly to metal.. metal work. And a lot of metalwork is done using moulds. So you do actually end up with a lot of artefacts that look extremely alike, as if they were conforming to an idealised representation. But actually, what is more likely the case is that a lot of these artefacts are actually made from a very limited number of moulds. Also, of course, in those sort of metal-using societies you often contend with much.. much very... very strongly normative patterns of cultural transmission so that folks actually stick very closely to a limited amount of variation. Had the... had this early thinking been developed in relation to for instance, lithic technology or early ceramic technology, it could have taken a very different turn because there the variability is much much greater in many, many more societies around the world and the past.)
23:37
But basically, the type of logic in a topological approach... you do have this kind of notion of community of practice, material culture is produced, but then you upscale or idealise, from the actual material culture to a topological idealisation and change from one type to another is effectively saltational.
23:56
In population thinking you really don't care about idealisation, you might want to compute an average kind of form say of an artefact class. But that's not really what's interesting in terms of analytical terms. So really, what you don't focus on is the actual material culture variability and how this can change over time as represented in our stratigraphic records. So here, change can be much more processional in nature. And this is of course, what we'd like to focus on.
24:32
So building on early work by Cavalli-Sforza, Feldman and MacDonald and others, I have modified this otherwise very common common table (you see it in literature quite a lot). It tries to divide different kinds of social transmission patterns up into the broad categories of vertical, oblique, and horizontal transmission. And it... and it also tries to to associate each of these transmission patterns with their predominant kind of processual results. So, whether you have a great deal of transmission fidelity, for instance, what within- and between-group variation would look like, and what sort of tempo of change you can expect.
25:17
And what you can appreciate is that, first and foremost, there is actually quite a bit of overlap between the different categories in what sort of expectations we would bring to the empirical record, depending on what kind of transmission dynamics we would expect. But fundamentally, I argue that cultural transmission in the long term is fundamentally driven by vertical and oblique transmission, and that phenomena such as social networks, which also play a very, very strong role in in the definition of cultures and how they interact, is fundamentally a form of horizontal transmission.
25:59
So, the time-axes both in terms of ontological time to transmission, the kind of intimate transmission of information from teacher to learner from an older generation to a younger generation, that then grows up to be the older generation to then continue transmit information to the younger generation is very important in this. So asking.. so, depending on what kind of questions you ask, whether there are diachronic or synchronic in nature, what kind of resolution you have in a record, different forms of processes and interaction terms may be important. So rather than opposing a kind of branching divergent patterns of evolution with a blending network-y type of process—so rather than opposing cult... biological and cultural evolution—I think both horizontal and vertical and oblique transmission play important roles.
27:01
The base figure for this particular slide is actually taken from the German pioneer of cladistics, Billy Hennig. So all the black arrows and a small black and white dots. And this was his effort to show the divergence, the branching evolution of two different populations. So here now, in our case, of course, we have to do with sort of a meta-population, mixed age community of practice, that over time will have varying kinds of interactions. Andy Roddick has called this "constellations of practice." So in reality, in the archaeological record, we can really rarely capture any kind of real community of practice, but we can make inferences from the material cultural record about these sort of temporarily stretched constellations of practice. And within these constellations of practice, you will have a mix of interaction terms: you have horizontal, vertical and oblique transmission. And at some point, there may be a process going on, that creates a division that hinders, say, horizontal transmission more and more so that you end up with a divergence of these populations. This can be socioeconomic strife, could be warfare, it could be climate change, and it could be some form of extreme event, often an interaction. All of... all of these and eventually, populations will diverge through to changes in the interaction dynamics. And you'll end up with two kind of units that empirically will be a we will be able to define differently on the basis of their content.
28:48
So returning to the final Palaeolithic of Europe, we tried to put some of this to a test. And an amazing paper published in the mid 90s. It was... the question was already raised, whether all the fantastic cultures that apparently populated the late Pleistocene, early Holocene, of Eastern Europe were, in fact real. And this paper, like so many of its ilk, had had no appreciable impact on archaeological practice. So people continue to use some of these cultural labels—even to.. to come up with new ones.
29:21
So the maps that you see here are all published at different times, although only in the 90s in the early 2000s. And fundamentally, they're based on the same empirical record and published lithic material of this tanged... this kind of tanged morphology. But obviously, workers divide this material up into very different kind of groups. So you see here, sort of a lumping taxonomy that will place every site that is characterised by the presence of tanged points into the so called Bromme culture. You'll see more of a splitting a high-level splitting taxonomy here that recognises two fundamentally different different groupings. And here, down here you have a more extreme splitting approach, where authors would want to recognise a series of different, very spatially very small scale units all based on the same material.
30:20
Now, we went to the literature, and sourced all the sort of assemblages that made up these kind of cultures, digitised the artefact drawings, and used that as our input data for subsequent analysis. Just note that the different colours denote the labels given to these assemblages in the literature.
30:45
This is what the material looks like you will see that they are relatively, moderately modified large tanged points—these are probably spearheads or dartheads for spears thrown with the help of spear-thrower. We also used an out group or comparison and that is a similar kind of point class from the much much later Pitted Ware culture in the same region.
31:17
We subjec... subjected all of these points to the geometric morphometric outline analysis using elliptical Four.. Fourier methods. And you can see that that the points cover a relatively a wide spectrum of forms: From kindof chunky to relatively elongated and slender. This kind of data can then be analysed using principal components analysis (you see it on the left here) which effectively flattens the variability along some major dimensional axes. But you can also use different kinds of cluster analysis methods that are in the same analytical family as what geneticists use to analyse a genetic data—in the same kind of analytical family that palaeontologists use to analyse species-level data. And there you will see higher level relationships between the terminal nodes within such clusters.
32:15
So, in an approach like this, we can always follow from say, a higher level relationship defined at this point to whatever artefact actually makes up this kind of cluster. So we have a clear and transparent pathway from a higher level definition that may correspond to a ki... kind of upscaled culture to the base data here at the bottom. Now, in our particular case, the interpretation was that actually a few if any of these traditional cultural labels actually work, so, we were inclined to reject all of them.
33:02
So, to sum up, the application of phylogenetic der... derivations of cultures in the sense of operational and analytical units will naturally place them within nested hierarchies that can you be informative of the histories. And importantly, you do not sacrifice any kind of relationship between the base data and upscaled cultural labels—should one care to define higher level nodes in in a kind of cultural terminology. So the promise of an evolutionary motivated phylogenetic approach of culture is that we can in principle, analyse multiple lines of cultural transmission—lithic, ceramics, houses. language, (if you have that kind of data), or different kinds of transmission systems--and biological transmission genetics (paleogenomics) in parallel, using similar maybe even the same kind of methods.
34:00
And ultimately, such co-phylogenies would highlight the convergences and divergences in these transmission histories would allow us to quantify this further and far downstream would also allow the application of comparative methods so a whole different set of tools that allows us to look at correlated changes and social organisation, adaptation, and niche construction.
34:27
So returning to the key questions posed for the seminars, how should cultural groups be distinguished from one another? My suggestion is that archaeologically, we can define them as temporally extended communities of practice we can capture them by material culture variability, carefully selected quantitatively and transparently analysed.
34:48
How can cultural groups be identified from data? Well, ideally by upscaling detailed material culture data through computational methods, Some of these are now readily available. And importantly, by not abstracting from base or data to any kind of cultural label, we just don't have to do that anymore. It was a necessity in the early years of archaeological practice, but it is no longer so.
35:17
And how should cultural barriers and borders be understood? Well, ultimately as divergence is created by physical social processes, that kind of increasingly hinder horizontal transmission social networking across communities, and thereby eventually, would lead to divergences in how material culture is made. And of course, specific environmental ecological conditions can create harder or softer boundaries depending on whether territories are defended or not.
35:48
These are all empirical, specific instantiations of general processes that would need to be evaluated ultimately, on a case by case basis, what basis what we do need is a theory driven precessional model for how these kinds of cultures how these sort of processes come about, and how they unfold.
36:09
Thank you very much.
Well, thank you very much for having me for this quite interesting seminar series about "Culture at the Macro Scale: Boundaries, Barriers and Endogenous change." Sometimes it's nice to talk about, you know, your own work... your own small studies. And sometimes it's really nice getting a bit more of a set task. And today's presentation is more like the latter.
0:26
So I've been asked to talk about "Culture at the Macro Scale". And what I'm going to do specifically is talk about past cultures at the macro scale. So what I'd like to do is provide an archaeological perspective on culture and culture change. The seminar organisers have posed a series of key questions to all the presenters, and they are: "How should cultural groups be distinguished from one another?", "How can cultural groups be identified from data?" and, "How should cultural barriers and borders be understood?" In addition to that, in my own invitation, I got a more specific task frame, and that was addressing cultural taxonomy; so the definition of cultures in archaeology and specifically in relation to genetic groupings. And this goes back to a paper that we recently published on reconciling archaeological and genetic taxonomies and why that is important—and I will get back to that work a little bit later on.
1:37
Now, thinking about culture change, it is unavoidable really to address... talk about this particular evocative and iconic illustration from Kroeber back in 1948. Here he contrasts effectively the Tree of Life—the patterns of biological evolution—on the left hand side as one of branching divergence with change in human cultures as more of a braided stream model: of reticulating networks that obscure and spin off different directions in terms of some kind of clear evolutionary pattern. However, there is no doubt really, that humans—biocultural creatures—they have a cultural makeup and there is tremendous, tremendous cultural richness across the world. But they also have a biological makeup.
2:45
So we are also animals, we are biological creatures, we have phenotypic traits, and we have specific genetic traits. For a long time, anthropologists and archaeologists were really concerned with defining their own discipline, and doing so in opposition to to biology. So back in 1952, Kroeber and Kluckhohn catalogued 164 definitions of culture. Twenty-two of those explicitly talked about a notion of tradition or inheritance, social heritage, or social heredity, it ... as it was called at a time, and some of them even reference processes or mechanisms such as learning.
3:37
But at the end of the day, most of these definitions are quite useless for archaeologists. Now, gene.. genetics, also deserves a little comment here that genetics is a pretty well understood transmission system. And it can take different kinds of forms. And when those folks in the social sciences, or humanities anthropologists, archaeologists, but also many, many lay people think about genetic transmission, what they normally think about is a large mammal transmission. But not all organisms have vertical transmission. Plants have a much more complex patterns of genetic inheritance. And let's not even talk about bacteria—other kinds of creatures—where any kind of horizontal oblique transmission is more the norm than the exception. So, large mammal genetics is perhaps not the kind of model system that we need to contrast cultural transmission with, when we think about how those two may be related.
4:48
At any rate, if you take a given human population as a biocultural phenomenon—that is one that composes a degree of cultural diversity and genetic diversity—and you add the dimension of time, there will be salient correlations emerging over time: there will be more or less connectedness between transmission patterns in biology and in culture. And if you add to this the spatial dimension, you will automatically and naturally end up with a patterned landscape with a variable distribution of the cultural diversity and the genetic diversity, the degree to which these are correlated map onto each other. How in the degree to which they leave "parallel tracks in time", as Rob Foley called it some time ago, is an empirical one.
5:46
And this is really where the cultural evolutionary approach comes in. And early in the 1980s initial steps—very significant steps—were taken by a Cavalli-Sforza and Feldman and more or less in parallel by Boyd and Richerson. And in terms of formalising the way in which we can approach the study of cultural evolution, these initial approaches drew extremely heavily on pop... population genetics and associated mathematical models.
6:18
More or less in parallel—but also somewhat separate—archaeologists, were b... were thinking about the application of evolutionary theory to culture change in a more deep time perspective. And initially, this really was through Dawkins, this notion as the extended phenotype. So here, material culture was very, very directly understood as the hard parts of the human phenotype, so very much like fossils. And if you look into volumes, published like these in the late 90s, early 2000s, you see, you come across a relatively hard rhetoric about a kind of transferal of evolutionary thinking to material culture.
7:13
In more recent times, all of this has coagulated into and developed into a really rather coherent framework for cultural evolution. (Sorry, I just have to get some coffee here.) Where the idea of the of material culture as the hard parts of the extended phenotype, it's more or less been given up in favour of the idea of dual inheritance or the parallel linked—more or less linked—trajectories of cultural and biological evolution. Ma.. mathematical models still play a major role. But a lot of hard thinking about conceptual issues epistemic issues, empirical issues has also occurred.
8:05
One important introduction for archaeologists, at least, that happened already in the early 90s, was the idea that perhaps we can use some of the tools of paleobiology to study also material culture change over time. So while we do have a kind of grand framework for how biological evolution and cultural evolution match up, (this here taken from from some of Alex Massoud his work), where palaeontology and evolutionary archaeology are really matched up as those disciplines that play a major role in addressing macro evolutionary change in the biological and cultural worlds respectively.
8:54
But this is not to say that cultures are like species. This notion, of course, has been around for a long time and it rears its ugly head again and again. But the analogy should not must not be understood in a direct kind of fashion. Rather, what needs to be appreciated is that there are significant epistemological and data structural similarities between paleobiology and evolutionary archaeology.
9:29
And this goes a long way back indeed, to the Cambridge philosopher, mathematician, chemist, mineralogist—you know, polymath—William Whewell, who in the mid in 1847, I believe, published a treatise where he would have the historical sciences classed together as one kind of way of understanding the world. So to him departments should be made up of geologists, historical linguists, art historians and archaeologists, because all of their data has an inherent temporal dimension. And it is about inferring patterns and processes from usually actually a fragmented record. (And quite interesting really.)
10:18
But also found... archaeological founding fathers such as the Swede Oscar Montelius here in the middle. A very amicable chap—who really developed the so called topological method—drew direct parallels to Darwinian evolution in his writing. And most recently, the ever optimistic Adrian Currie highlighted the epistemic and data structural similarities between palaeontology and, and archaeology, I think, to quite a lot of benefit.
10:52
But stop. This talk wasn't really meant to be a kind of an abridged discussion of the intellectual ancestry of evolutionary approaches to culture. I did want to do something that was a little bit more case-based—something that also addressed the questions said in this seminar a little bit more directly. So let's change track here slightly.
11:19
Our recent work on trying to understand cultures analytically at the macro scale, goes back some time, I have long been concerned with this, in fact, going all the way back to my Ph.D. But more recently, it was a publication by Eisenmann et. al, where they made a range of very practical suggestions as to how to think archaeological and paleogenomic clusterings together. And they came up with a kind of taxonomic system, which they call the mixed system where any kind of cluster genetic cluster/paleogenomic cluster is matched up with an archaeological term that is a hybrid of geography, chronology and sometimes also a kind of cultural labels. So something that perhaps refers to the actual content of the material culture.
12:26
The argument for adopting such a mixed system taxonomy was that it included.. that it offers brevity, coherence, accessibility, flexibility, and stability. And those are all very important aspects of any kind of taxonomic system. But actually, I argue, together with my co-author Steve Shennan and Chris Hoggard would actually prefer theoretical rusticity, transparency, interoperability and analytical power as the driving factors behind any kind of taxonomic system.
13:04
Our concern here, of course, was not talking to a lay public or telling compelling narratives about past processes in the first instance. Instead, our motivation was to make the notion of culture in an archaeological sense: actually analytically useful. In our ... problems, so to say, with the suggestions by Eisenmann et al. was that cultural labels—such as the "Bell Beaker" culture, the "Lyngby" culture, you name it—the archaeological record is populated, over populated, some would say, by these kind of named cultural groups. They are rarely unanimously defined. So if you ask ten archaeologists how they would define a given culture, you're likely to get definitions that actually vary tremendously.
14:02
Some of the taxonomic suggestions made by Eisenmann et al. also relate to subsistence practice labels such as Neolithic, for instance, but they can... they are not really well, they're rather... we would argue they're rather ecologically determined than determined by inheritance. And these can be confusingly assigned anyway.
14:25
So there's a lot of talk also in the paleogenomic literature about the "Pitted Ware" culture. Sometimes they're called Mesolithic, sometimes they're called Neolithic. It is definitely a cultural phenomenon that is chronologically framed by the Neolithic, that is by societies that use domesticates and pottery, but they adopt more or less a hunting and gathering kind of economy. It's a super fascinating cultural phenomenon precisely because it defies clear labels relating To subsistence practice, for instance.
15:03
In addition, sometimes even very sort of standard labels such as Neolithic would vary depending on what kind of research tradition you're from. So in Western Europe, the Neolithic is usually defined as something related to serial agriculture using domestic goods, whereas in a recent Russian research tradition, any kind of culture is effectively termed Neolithic, as soon as you get pottery. So, these are economic versus political or labour related definitions.
15:38
And finally, cultures are time transgressive phenomena. So chronological assignments also have their inherent problems. And partly because dating uncertainties and because in part because these dates would usually vary depending on where you are in space.
15:55
In contrast, genetic labels such as haplotypes, are so useful and so precise, because they allow inferences about historical processes. And this is rests in them being linked to an explicit process model of change, evolutionary theory, from which flows a nested hierarchical taxonomy that allows the division of any given sample into clusters. And the relative relatedness of these clusters is informative vis-a-vis the histories. And most of our cultural taxonomic systems use an archaeology don't have... are not underwritten by these kind of process oriented theories, and they do not result in nested taxonomic systems.
16:47
So, this is just one example here, where you see the final Palaeolithic cultures of Europe divided into a series of different kinds of cultural groups. It is quite unclear also from the original writing, whether these groups are to be understood in the same way as contemporary hunter gatherer groups understood for example, whether they would have a kind of conscious ethnic identity, whether there would be linguistic differences inferrable from these kinds of maps. Also, you cannot really read in any kind of detailed or informative manner, the real distribution of archaeological sites from such maps, it is impossible to back-engineer any kind of meaningful distribution pattern here. It's unclear what the empty spaces mean, also note how vastly different in spatial scale these cultural units are, but it is not clear whether they the smaller units are nested in any kind of way within the bigger units.
17:55
Okay. And admittedly, this is an old map, but it is, I would argue, still quite characteristic of many kinds of maps that you will see in the archaeological... archaeological literature. Now critiquing archaeological thinking on cultural taxonomy, or indeed on the notion of culture itself is not new. For a long time, archaeologists, especially, perhaps in the Palaeolithic have been arguing that many of our culture definitions are analytically dubious, and they're mostly grounded in what what has been called accidents of history. That is, particular kind of sociopolitical currents that characterise a given part of the world. And they've given rise to a specific kind of spatial temporal circumscription of archaeological phenomenon.
18:51
So arguably, at least the Palaeolithic record, but probably also, the record of later times is populated with what Barton and Neeley called "phantom cultures". And more radically, folks, such as John Shea have argued that a lot of these labelled cultures these named cultures (he calls them "NASTIES") should indeed be abandoned.
19:18
In the slide before us, just so, you saw a sample of just the international English speaking literature on the topic, but there is a vast undercurrent of more specific writings targeting more specific cultural groups at at national or regional scales. Now, the drawback of this literature which often is very, very, to the point and very much goes to the empirical foundations of these cultural groupings is that it is usually written in relatively obscure languages more difficult to find through library... international library systems or databases. And, of course leaves a relatively limited impact on practice.
20:07
So over the last few years, we have been trying to draw some of these patterns out and trying to publish new work on cultural taxonomy here specifically in relation to the late or final Palaeolithic in an open access fashion and in the English speaking literature. So there's no doubt that some of these pilot studies here are merely indicative of wider conceptual epistemological problems that that plague the the archaeological record of the Palaeolithic and other periods.
20:44
And the problem, of course, is that if our definitions of cultural taxonomic groups of cultures are not operational, then the kind of conceptual upscaling that we'd like to do in terms of archaeological interpretations about processes or interactions, social networks and so on, simply will stand and fall with the robusticity of these archaeological groupings.
21:10
Okay, so, cultural taxonomy is the kind of significant linkage between an empirical groundwork that has to do with how we actually look at at any kind of given artefact assemblage, how we excavate and so on, how taphonomic processes affect the archaeological record and subsequent archaeological interpretations at a more macro scale. But how then can we identify and define these kinds of archaeological groupings, these taxonomic operational taxonomic groups in archaeology?
21:50
So in in an older framework, one that is still very much alive and kicking you have the idea of the type. And topological thinking has long been critiqued in the biological literature as an undo idealisation of the material variability out in a world that is then being filtered in a vain search for for a kind of ideal representation of, say, a given animal or a given artefact class.
22:27
(Now, some of this historically, if you allow me to digression may have to do with the fact that the topological method was initially developed in in relation to mostly to metal.. metal work. And a lot of metalwork is done using moulds. So you do actually end up with a lot of artefacts that look extremely alike, as if they were conforming to an idealised representation. But actually, what is more likely the case is that a lot of these artefacts are actually made from a very limited number of moulds. Also, of course, in those sort of metal-using societies you often contend with much.. much very... very strongly normative patterns of cultural transmission so that folks actually stick very closely to a limited amount of variation. Had the... had this early thinking been developed in relation to for instance, lithic technology or early ceramic technology, it could have taken a very different turn because there the variability is much much greater in many, many more societies around the world and the past.)
23:37
But basically, the type of logic in a topological approach... you do have this kind of notion of community of practice, material culture is produced, but then you upscale or idealise, from the actual material culture to a topological idealisation and change from one type to another is effectively saltational.
23:56
In population thinking you really don't care about idealisation, you might want to compute an average kind of form say of an artefact class. But that's not really what's interesting in terms of analytical terms. So really, what you don't focus on is the actual material culture variability and how this can change over time as represented in our stratigraphic records. So here, change can be much more processional in nature. And this is of course, what we'd like to focus on.
24:32
So building on early work by Cavalli-Sforza, Feldman and MacDonald and others, I have modified this otherwise very common common table (you see it in literature quite a lot). It tries to divide different kinds of social transmission patterns up into the broad categories of vertical, oblique, and horizontal transmission. And it... and it also tries to to associate each of these transmission patterns with their predominant kind of processual results. So, whether you have a great deal of transmission fidelity, for instance, what within- and between-group variation would look like, and what sort of tempo of change you can expect.
25:17
And what you can appreciate is that, first and foremost, there is actually quite a bit of overlap between the different categories in what sort of expectations we would bring to the empirical record, depending on what kind of transmission dynamics we would expect. But fundamentally, I argue that cultural transmission in the long term is fundamentally driven by vertical and oblique transmission, and that phenomena such as social networks, which also play a very, very strong role in in the definition of cultures and how they interact, is fundamentally a form of horizontal transmission.
25:59
So, the time-axes both in terms of ontological time to transmission, the kind of intimate transmission of information from teacher to learner from an older generation to a younger generation, that then grows up to be the older generation to then continue transmit information to the younger generation is very important in this. So asking.. so, depending on what kind of questions you ask, whether there are diachronic or synchronic in nature, what kind of resolution you have in a record, different forms of processes and interaction terms may be important. So rather than opposing a kind of branching divergent patterns of evolution with a blending network-y type of process—so rather than opposing cult... biological and cultural evolution—I think both horizontal and vertical and oblique transmission play important roles.
27:01
The base figure for this particular slide is actually taken from the German pioneer of cladistics, Billy Hennig. So all the black arrows and a small black and white dots. And this was his effort to show the divergence, the branching evolution of two different populations. So here now, in our case, of course, we have to do with sort of a meta-population, mixed age community of practice, that over time will have varying kinds of interactions. Andy Roddick has called this "constellations of practice." So in reality, in the archaeological record, we can really rarely capture any kind of real community of practice, but we can make inferences from the material cultural record about these sort of temporarily stretched constellations of practice. And within these constellations of practice, you will have a mix of interaction terms: you have horizontal, vertical and oblique transmission. And at some point, there may be a process going on, that creates a division that hinders, say, horizontal transmission more and more so that you end up with a divergence of these populations. This can be socioeconomic strife, could be warfare, it could be climate change, and it could be some form of extreme event, often an interaction. All of... all of these and eventually, populations will diverge through to changes in the interaction dynamics. And you'll end up with two kind of units that empirically will be a we will be able to define differently on the basis of their content.
28:48
So returning to the final Palaeolithic of Europe, we tried to put some of this to a test. And an amazing paper published in the mid 90s. It was... the question was already raised, whether all the fantastic cultures that apparently populated the late Pleistocene, early Holocene, of Eastern Europe were, in fact real. And this paper, like so many of its ilk, had had no appreciable impact on archaeological practice. So people continue to use some of these cultural labels—even to.. to come up with new ones.
29:21
So the maps that you see here are all published at different times, although only in the 90s in the early 2000s. And fundamentally, they're based on the same empirical record and published lithic material of this tanged... this kind of tanged morphology. But obviously, workers divide this material up into very different kind of groups. So you see here, sort of a lumping taxonomy that will place every site that is characterised by the presence of tanged points into the so called Bromme culture. You'll see more of a splitting a high-level splitting taxonomy here that recognises two fundamentally different different groupings. And here, down here you have a more extreme splitting approach, where authors would want to recognise a series of different, very spatially very small scale units all based on the same material.
30:20
Now, we went to the literature, and sourced all the sort of assemblages that made up these kind of cultures, digitised the artefact drawings, and used that as our input data for subsequent analysis. Just note that the different colours denote the labels given to these assemblages in the literature.
30:45
This is what the material looks like you will see that they are relatively, moderately modified large tanged points—these are probably spearheads or dartheads for spears thrown with the help of spear-thrower. We also used an out group or comparison and that is a similar kind of point class from the much much later Pitted Ware culture in the same region.
31:17
We subjec... subjected all of these points to the geometric morphometric outline analysis using elliptical Four.. Fourier methods. And you can see that that the points cover a relatively a wide spectrum of forms: From kindof chunky to relatively elongated and slender. This kind of data can then be analysed using principal components analysis (you see it on the left here) which effectively flattens the variability along some major dimensional axes. But you can also use different kinds of cluster analysis methods that are in the same analytical family as what geneticists use to analyse a genetic data—in the same kind of analytical family that palaeontologists use to analyse species-level data. And there you will see higher level relationships between the terminal nodes within such clusters.
32:15
So, in an approach like this, we can always follow from say, a higher level relationship defined at this point to whatever artefact actually makes up this kind of cluster. So we have a clear and transparent pathway from a higher level definition that may correspond to a ki... kind of upscaled culture to the base data here at the bottom. Now, in our particular case, the interpretation was that actually a few if any of these traditional cultural labels actually work, so, we were inclined to reject all of them.
33:02
So, to sum up, the application of phylogenetic der... derivations of cultures in the sense of operational and analytical units will naturally place them within nested hierarchies that can you be informative of the histories. And importantly, you do not sacrifice any kind of relationship between the base data and upscaled cultural labels—should one care to define higher level nodes in in a kind of cultural terminology. So the promise of an evolutionary motivated phylogenetic approach of culture is that we can in principle, analyse multiple lines of cultural transmission—lithic, ceramics, houses. language, (if you have that kind of data), or different kinds of transmission systems--and biological transmission genetics (paleogenomics) in parallel, using similar maybe even the same kind of methods.
34:00
And ultimately, such co-phylogenies would highlight the convergences and divergences in these transmission histories would allow us to quantify this further and far downstream would also allow the application of comparative methods so a whole different set of tools that allows us to look at correlated changes and social organisation, adaptation, and niche construction.
34:27
So returning to the key questions posed for the seminars, how should cultural groups be distinguished from one another? My suggestion is that archaeologically, we can define them as temporally extended communities of practice we can capture them by material culture variability, carefully selected quantitatively and transparently analysed.
34:48
How can cultural groups be identified from data? Well, ideally by upscaling detailed material culture data through computational methods, Some of these are now readily available. And importantly, by not abstracting from base or data to any kind of cultural label, we just don't have to do that anymore. It was a necessity in the early years of archaeological practice, but it is no longer so.
35:17
And how should cultural barriers and borders be understood? Well, ultimately as divergence is created by physical social processes, that kind of increasingly hinder horizontal transmission social networking across communities, and thereby eventually, would lead to divergences in how material culture is made. And of course, specific environmental ecological conditions can create harder or softer boundaries depending on whether territories are defended or not.
35:48
These are all empirical, specific instantiations of general processes that would need to be evaluated ultimately, on a case by case basis, what basis what we do need is a theory driven precessional model for how these kinds of cultures how these sort of processes come about, and how they unfold.
36:09
Thank you very much.