Ethnoarchaeology as slow science

ABSTRACT In recent years, the purpose and objectives of ethnoarchaeology have been called into question. In this paper, we propose that ethnoarchaeology might best be considered a form of ‘slow science’ that works to counterbalance ‘big science/fast science’ approaches in archaeology. We consider the interpretative challenges facing archaeology and the risks posed by a shift to fast science approaches that emphasize large-scale, strategic and analytically focused ‘big data’ analyses. We draw on recent literatures that define ‘désexcellence’ and ‘slow science’ approaches, which forefront ethically driven and collaborative research, and suggest that ethnoarchaeology might be well positioned to redefine itself as a form of slow science. Doing so, however, requires redefining ethnoarchaeology’s field objectives and its relations to research subjects.


Introduction
In North America, ethnoarchaeology remains an ambiguous subfield within a broader four-field anthropology. On the one hand, ethnoarchaeology's existence seems natural: if archaeology and cultural anthropology function as part of the same discipline, sharing both subject matter and theory, then a hybrid subfield focused on these linkages seems unavoidable to the point of redundancy. However, the slow erosion of Boas' 'sacred bundle' of a four-field anthropology (Segal and Yanagisako 2005), the often contradictory roles ethnoarchaeology was given in its early years (compare, for example, Binford 1978;Gould 1980;Hodder 1982), as well as the expansion of work at the boundary between archaeology and cultural anthropology (Bruchac and Wobst 2010;Edgeworth 2006;Gosden 1999;Hamilakis 2011), have left many archaeologists wondering whether ethnoarchaeology isin Simms' (1992) memorable phrasingan 'obnoxious spectator, a trivial pursuit, or the keys to a time machine'. Among many who have practised or observed ethnoarchaeology, there has been a sense recently that the subfield needs a substantive rethink. This paper serves as a small contribution to that discussion, as well as a call to reconsider the ethical positioning of ethnoarchaeology within archaeology more broadly. We both conducted ethnoarchaeological work in West Africa earlier in our careers, and have now moved on to more conventional forms of archaeological research, one continuing in Africa (MacEachern) and the other in North America (Cunningham). The lessons learned from these experiences continue to inform our archaeology, in ways that are not always reflected in recent assessments and critiques of ethnoarchaeology. Our perspective also originates in an approach to ethnoarchaeology that coalesced at the University of Calgary in the 1980s and early 1990s, one inspired initially by post-processualism and then refocused into a broadly postpositivist concern with how analogy contributes to archaeological interpretation (David 1992; Kelley and Hanen 1988;Stahl 1993;Wylie 1985). Additionally, we also engage with ethnoarchaeology in the francophone world, which has a different focus than the Anglo-American tendency to associate ethnoarchaeology closely with 'middle-range theory' defined for a positivist epistemology (Cunningham 2009(Cunningham , 2013. We further emphasize the political implications of ethnoarchaeological work, situated as it is in an analytical space where distinctions between past and present societies and social relations may potentially become confused to the extent that modern populations are seen as representative of ancient cultural systems, and frequently associated with geographic spaces (like sub-Saharan Africa) where such confusions are endemic.
In what follows, we draw from a recent European literature to suggest that ethnoarchaeology might be conceptualized as a form of 'slow science', one that follows epistemological and ethical paths different from the 'big science/fast science' approaches that seem to be gaining ground in the discipline. We begin below by exploring the interpretative challenges archaeology faces as a discipline, and suggest that 'fast science' approaches often exacerbate these challenges. A refashioned view of ethnoarchaeology might help to ameliorate these problems.

Archaeology as capitalist ideology
If we want to speak in broad strokes about the value of ethnoarchaeology, it would seem necessary to discuss the objectives of the parent discipline to which ethnoarchaeology hopes to contribute. Archaeology is essential to global understandings of the past, but also faces an interpretative challenge that we believe that ethnoarchaeology may help to solve. The first part of this phrasethat archaeology is central to history makingis uncontroversial. As we teach in our introductory classes, archaeology is one of the only sources of information for about 97 per cent of humanity's history, assuming that we locate our rhetorical flourishes at the species level. In parts of the world without written histories, archaeological remains are one of the few sources for history-making that extends beyond living memory. Moreover, we know that the historical sources produced during that more recent 3 per cent of humanity's time on earth originated with small numbers of literate ritual or administrative elites that frequently used those texts to naturalize their privilege. Those living at the edges of the panoptic visions of these literate elites, whether they have been spatially proximate but socially marginal peasantries, slaves or women or spatially as well as socially marginal communities, can remain invisible within even rich literary traditions. Archaeology is thus not only essential for the study of areas without written history, but also serves as a fundamental counterbalance to written history. It is uniquely positioned to speak for the lived experiences of the marginalized and the invisible in the present as well as past (e.g. Gokee and De León 2014), and thus help make histories for those whose histories have been written by others.
This rewriting of history is often associated with the desire for accuracy and reliability in our interpretations. As feminists have shown (Wylie 1992(Wylie , 1995, history-making that strives for objectivity can become emancipatory because, as it doggedly pursues evidentiary support and logical coherency in its interpretations, it reveals that elite-centred narratives were often based not at all in careful historical analysis, but rather in ideological shortcuts buttressed by the power of their authors. The powerful do not need evidence on their side, because ideology tips the table in their favour. Histories channel the imagination of futures; getting histories right denaturalizes inequalities and can thus prompt individuals and communities at the margin to imagine and pursue alternative futures. Archaeology's traditional Enlightenment goalto write the global history of our speciesmight then also be the basis of broader emancipatory programmes. While people in all societies have their own forms of history-making (after Schmidt and Patterson 1995), archaeology's ambition to be an authoritative global discipline means that part of its responsibility is to ensure that its findings are commensurate with an understanding of the full breadth of contemporary cultures. One of the most significant challenges that archaeology faces is to think beyond the Eurocentric, neoliberal and often alienated systems of knowledge production within which archaeologistswhether academics or heritage managersall too frequently work. The gendered and ethnic diversity of archaeologists has been extensively examined, with some tendency towards greater diversity in recent decades, although there remain enormous class and racial imbalances in the possibility of community participation in archaeological research. Advocacy in areas such as heritage scholarship and indigenous archaeologies seems specifically engineered to expand diversity of viewpoints within archaeological interpretation. However, global economic trends tend to militate against this happy picture. The financial services company Credit Suisse (2015, 11-12) reported last year that 1 per cent of the world's population now owns 50 per cent of global wealth. The median global income sits at just US$3,210 and the half of the global population below this number owns less than 1 per cent of total global wealth. By contrast, the upper 10 per cent now control 87.7 per cent of global wealth and membership in this elite group begins at US$68,800. Piketty (2014) argues convincingly that the decrease in economic inequality in the Western world in the twentieth century may be an historical accident, with generational economic stratification increasing in the future. We must thus acknowledge that most professional archaeologists are part of a global economic elite whose relative wealth gives us an ability to interpretand, indeed, createglobal history for the vast majority of humanity.
There is, further, a fundamental disjunction between the goal of an emancipatory archaeology and disciplinary structures that increasingly seem designed to foster the antithesis of emancipation. This is not uniquely (or even primarily) an archaeological issue; rather, it afflicts intellectual work in many parts of the world where archaeology is taught and undertaken (see, for example, Graeber 2013Graeber , 2015. To a significant degree, archaeology increasingly aspires to become 'big science', on a model familiar from other research disciplines (Coleman and Kamboureli 2012). This involves a variety of different processes: (1) the investment of resources in large-scale, interdisciplinary, international projects, and 'strategic' grants that are steadily transitioning from very rare to virtually unobtainable (cf. Bond 2013; Blommaert 2015); (2) the privileging of advanced analytical techniques that are either unaffordable or so specialized that they are inaccessible to most researchers; (3) the global mobility of specialized investigators and heritage managers, with contingent local labour forces eking out a precarious living as either teachers or diggers (Hardy 2014;Möller 2014); (4) the increasing use of 'big data', with deployment of large datasets, often variably accessible and verifiable (Huggett 2012;Gattiglia 2015); and (5) increasingly unequal structures of publication, data access, evaluation and promotion.

Fast science and slow science
The positioning of archaeology as 'big science' may have considerable advantages for particular institutions and individuals, because of the resources and prestige that it can make available to them. On a comparative disciplinary basis, we already see the advantages of 'big science' in the rhetorical positioning of archaeogenetics vis-à-vis archaeology, in, for example, the frankly credulous reception given to historical pronouncements made by archaeogeneticists in Western media (MacEachern 2013). However, these increases in scale, inequality and 'capitalization' of the disciplinearchaeology as 'big science'may lead us gradually towards models of what might for want of a better term be called 'fast science', with rather different kinds of consequences.
The metaphor of 'fast science' seems to have been deployed in English first in the late 1990s and used to describe the rapid advances in genomics of the time, especially with the Human Genome Project (Fortun 1999). Here we use the term 'fast science' in a rather different manner, in juxtaposition to a variety of calls for a re-imagining of academic work, calls (by anthropologists and ethnoarchaeologists, among others) for a movement of 'désexcellence' or 'slow science' (see below) (Alleva 2006;Gosselain 2011;Stengers 2012). 'Fast science' is one operational outcome of the academic trajectories described above, in archaeology as in other disciplines; it is probably not the only such outcome, but, as we shall see, striking examples do exist. 'Fast science' is managerial, competitive, data-centric, technocratic and alienated from the societies it serves and studies. It involves, analytically, the collision of datasetsthe assumption that if data are available on topics that might possibly be related, then statistical comparison of those datasets will yield useful insights. This is dangerous. These approaches tend to dismiss the importance of on-the-ground complexities in their quest to generate grand, high-impact-factor syntheses. The degree to which such complexities are subsumable within attempts to generalize is of course highly variable. One of the particular challenges associated with big data, in archaeology and elsewhere in the social sciences, is that such approaches tend to move problem orientations from the study of causation to the discovery of correlation (Calude and Longo 2015;Gattiglia 2015), in circumstances where entirely spurious correlations may easily be generated (Roberts and Winters 2013). This is a serious problem in circumstances where long-standing stereotypes are associated with particular social situations or relationships. In these cases, an alienated 'fast science', concerned only with manipulating datasets, may wittingly or unwittingly perpetuate such stereotypes, simply because there is no necessity for demonstrating the links in a causal chain. As Africanist researchers, we are very much aware of the stereotypes that Western intellectuals, media and publics have held concerning African societies over the last two centuries (MacEachern 2015), and worry that the deployment of 'big science/fast science' models of research may in some cases serve to disseminate obsolete and damaging stereotypes about the continent.
This has, in fact, already happened in other social science disciplines. Richard Lynn, a psychometrician, periodically in collaboration with the political scientist Tatu Vanhanen, published a set of works that purported to establish 'national IQ' for a variety of countries around the world and to correlate those measures with a variety of social and economic outcomes Vanhanen 2002, 2006;Lynn 2008). This directly implicates archaeology and ethnoarchaeology not only because it denigrates many of the modern communities we work with and is used to justify policies that would be inimical to their interests (Vanhanen 2013); Lynn and colleagues have also distorted human evolutionary and archaeological data to support their claims (MacEachern 2006). Lynn has a long history of such 'racial science' studies (Lynn 1991(Lynn , 2003(Lynn , 2006, and most of these works on 'national IQ' were privately published by a press affiliated with an American white supremacist organization called the National Policy Institute (Southern Poverty Law Center 2010). They are thus not peer-reviewed. The data used in these analyses are of extremely poor quality (Mackintosh 2007;Wicherts et al. 2010; Wicherts, Dolan and van der Maas 2010a, 2010b). Among many African examples, we may note that the 'national IQ' for Equatorial Guinea was derived from a sample of students attending a school for handicapped children in Spain (Fernández-Ballesteros et al. 1997), that for Ethiopia was derived from a group of displaced orphans in Israel (Kaniel and Fisherman 1991; see also Berhanu 2007) and the testing of South Africans described by Lynn as '[t]he best single study of the Negroid intelligence' was described by the researcher in question (Owen 1989) in the original publication as unsuitable for the derivation of IQ scores.
Despite the poor quality of the data and results that would seem, minimally, to be worthy of careful questioning and further examinationthat 'national IQ' for sub-Saharan African countries indicates that African populations labour on average under moderate to severe mental deficit -Lynn's data have been deployed in an extraordinarily large number of social science correlational studies by other researchers, well over 200 in the last ten to twelve years. These purport to measure the relationship between 'national IQ' and factors as varied as entrepreneurship, parasite load, levels of corruption, criminality, unpaid parking tickets and manufacturing capacity (see, among many others, Eppig, Fincher and Thornhill 2010;Jones and Nye 2011;Potrafke 2012;Daniele and Ostuni 2013;Salahodjaev 2015). Some of these analysts may well have had poor opinions of Africans to begin with, but what is perhaps more striking is the example of those authors who managed to evade as irrelevant or uninteresting the issue of whether these populations, and other poor peoples around the world, are actually mentally deficient in the first place, possibly because they had no experience of, context for or engagement with the communities from which these datasets were derived. For most of these studies, these populations appear to be merely components in those datasets, useful only in their utility for comparing with other datasets and in generating publications. This is 'fast science' par excellencedata-centric, alienated and generating absurd and (within the broad corpus of the research) internally contradictory results (Dickins, Sear and Wells 2007; Wicherts, Dolan and van der Maas 2010a).
Archaeology's shift to value 'big science' approaches has fortunately not been nearly so pernicious as have been these psychometric excursions into 'national IQ', but it does share an emphasis on large-scale correlations rather than understanding the underlying causes of specific material patterns. For example, Barbara Mills and colleagues' (Mills et al. 2013(Mills et al. , 2015 analysis of changing regional networks in the southwest has been perhaps one of the most insightful recent applications of a big science approach in archaeology. Using an interregional GIS based on the massive Coalescent Communities Database, Mills and her colleagues have shown how social networks changed by analysing ceramic styles and obsidian exchange relations. Consistent with current understandings of the local culture history, the pattern generally shows the diminishing of relations between sites in the north associated with depopulation after AD 1300 and an expansion of connections among southern networks associated with Hohokam sites. However, because the networks are defined primarily on the basis of similarities in ceramic attributes, the project invites a consideration of enduring typological questions once the focus of the Ford-Spaulding debate (e.g. Ford 1954;Spaulding 1954). The issues that plague stylistic analysis more generally (Cunningham 2003) here are supposed to be satisfied by the sheer quantity of data used in the analysis, and by a 'communities of practice' approach that replaces the culture historical assumption that similar pots equal similarity in mental templates with the assumption that similar pots result from similar 'practices' in a productive sequence.
Both, however, rely ultimately on sherd identifications and thus on the deeply held belief that decoration is fundamentally diagnostic of social propinquity because it is 'adjunct' to the ecological constraints that might otherwise affect variability (after Sackett 1977). Thus, 'pottery' (as a class of archaeological evidence) and 'decorative style' (as a unique set of variability in pottery) can be treated as if they have consistent causal antecedents that do not vary for different cultural groups or at different time periods. Pottery styles can thus be compared in order to identify social networks. Since the inception of ethnoarchaeological work, however, the adage that pottery correlates in any direct or universal way with social variability has been extensively studied and frequently critiqued (e.g. Arnold 2000; David, Sterner and Gavua 1988;Dietler and Herbich 1989;Hodder 1982;Longacre and Stark 1992). The networks study represents a high-water mark for recent 'fast science' approaches in archaeology and is a substantive contribution to southwest archaeology. That it suffers from a similar lack of interest in exploring the origins of the data it uses as do the 'national IQ' studies described above, and attempts to solve questions about the quality of data by increasing its quantity (cf. Mills' earlier ethnoarchaeological work: e.g. Mills 1995), shows the ingrained problem with 'fast science'.

'Slow science' and ethnoarchaeological work
One reaction against this kind of 'fast science', and against the alienation and capitalization of academic work in the West in the late twentieth and early twenty-first centuries, has now appeared in a focus on 'slow science' or 'désexcellence' (Alleva 2006;Gosselain 2011;Stengers 2012). The latter might be roughly glossed in English as 'anti-excellence'which in this case indicates a refusal to pursue excellence defined only according to a specific set of management criteria (i.e. those of 'big science/fast science'). Instead, excellence is for these researchers defined by other epistemic and/or ethical values. We note here, as one of our reviewers has very usefully pointed out, that the terms 'slow science' and 'désexcellence' both encompass a variety of meanings and political stances, ranging from more or less quietist attempts to improve the work-life of individual faculty members in fully corporatized North American universities (e.g. Berg and Seeber 2016), to calls for more collective academic disobedience and new teaching and research strategies in Europe (L'Atelier des Chercheurs 2016), to much more wide-ranging critiques of academic knowledge production and its collaboration with neoliberal politico-economic systems (Stengers 2012). We recognize this diversity of approaches and of meanings, but believe that a number of positive elements, held to be central to the academic endeavour, can be found within many such discussions of 'slow science'/'désexcellence': (1) a recognition of the ethical and human consequences of scientific research and of the human relations involved in such research; (2) a recognition of the need for contemplation, collaborative learning and careful thought in the course of research; (3) some concern for the communal aspects of such research, for mutual cooperation and support by colleagues and students; and (4) a critique of bureaucratization and the inappropriate intrusion of practices and concepts from business and management into the academy, whether in research and/or in teaching.
'Slow science' as we understand it can thus serve as a call for more engaged, critical, humane academic work. In an analytical sense, and in all these different approaches, it stands opposed to 'big science/fast science' as described above. Archaeology as 'slow science' would embrace an ethical stance that a number of archaeologists now see as fundamental to the discipline (e.g. Lydon and Rizvi 2010), including a long-term approach to research, practical knowledge on the ground (and in it), social engagement (both with fellow workers and with communities within which we work) and critical reflections on power relations in the past and the present. Gosselain (2011) notes an appreciation of the métier of the artisan as one way that researchers might recapture a more humane and satisfying engagement with the work that they do, as through such work challenging tasks are accomplished with materials, skills are gradually improved and a close relationship with the world of the craft is maintained. We entirely agree, and would further claim that the craft of archaeology itselfin a direct analogy to the work of artisans in other fields, in our fundamental engagement with obdurate earthremains under-emphasized and under-studied, even two decades after Shanks and McGuire (1996) wrote on the topic. More broadly, if archaeological interpretation has drawn perhaps too heavily on a Eurocentric baseline in the past, a baseline that in other disciplines is becoming increasingly insulated from critique by 'fast science' methodologies that stress correlation over causation and alienation over engagement, then the critical research advocated by slow science practitioners will lead to better histories. Hence, ethically engaged scholarship does not mean an abandonment of archaeology's historical desire to work towards greater degrees of 'objectivity'.
Ethnoarchaeology seems to us well-positioned to contribute to a 'slow science' approach that is concerned fundamentally with the quality of the data archaeologists use in their interpretations and with ethical engagements with the communities among whom we work. To some degree, this depends upon the basic characteristics of the discipline itself. Ethnoarchaeological work is intimately connected with living humans, and with the richly social worlds that those people in habit: it reminds us that the ultimate goal of archaeology is not an understanding of artefacts, but rather of people. It usually involves long-term engagements with particular communities, often on a scale quite different from archaeological field seasons. This makes it harder for investigators to avoid issues of power relationships, both within those communities and between such communities and larger societies. This may involve particular critiques of neoliberalism and of its consequencesas, for example, with the local effects in Cameroon of Structural Adjustment Programmes imposed by Western financial agencies that MacEachern witnessed in the 1980s. Finally, ethnoarchaeology is an accessible subdiscipline of what is probably the most expensive of the social sciences/humanities, one available to both Western and non-Western researchers who may not have access to the funding sources required by other, 'bigger' and 'faster' forms of archaeological research.
In more intellectual terms, one of ethnoarchaeology's fundamental roles in archaeology is to address the analogical myopia of archaeological theory. As Wylie (1985) and Kelley and Hanen (1988) outlined in some detail, analogical reasoning relies on the borrowing of information from a better-known source context to help interpretation in a second subject context that is usually less well understood. In archaeology, analogical interpretations typically rely on a belief that, based on observed similarities in material culture between a source and subject context, there are also similarities in the actions that produce those patterns. These might include analogous forms of subsistence, social institutions such as prestige exchange systems or descent patterns or similar ideological structures. Testing operates on both sides of the analogy to assess whether the original interpretation is valid. In ethnographic settings, testing assesses the explanation given for human actions (usually anchored in social theory) and the proposed entanglements between those actions and material patterns. In archaeological settings, testing searches for those 'other patterns' that should be present in material culture if the analogue cited in an interpretation is appropriate. While it is now normal for archaeologists to draw analogues from archaeological models, most interpretations trace back to knowledge archaeologists distil from contemporary contexts, ranging from their personal experiences and hunches, to explicit uses of broad range theory. In all these cases, archaeology relies on comparisons once made to modern human experience to interpret material remains, and ethnoarchaeology is a specialized approach dedicated to improving upon that knowledge.
Ethnoarchaeology is thus dedicated to expanding the 'analogical consciousness' of archaeologists (David 1992, 352). This is necessary because all archaeologists are at risk of affirming interpretationssuch as rationalist economic theorythat reflect their upbringing, but is perhaps especially pertinent for those who are members of urbanized and affluent industrial societies. As alluded to earlier, one solution to this problem has been to foster other history-making and recast archaeology as but one of many forms of inquiry. While this is certainly an improvement over a narrow positivism, real pluralism often is more complicated (see Conkey 2007). Archaeologists in general, and the discipline's 'big men' in particular, have privileged access to history-making because of their close association with publishing, research funding and professional employment in comparatively well-funded and reputable academic institutions. Interpretations fostered outside these contexts struggle for influence, and risk being ghettoized as a form of 'alternative history' that is nurtured for ethical reasons but otherwise cossetted in ways that limit its ability to challenge mainstream narratives (after West and Martin 1999). Ethnoarchaeology works to introduce diverse standpoints at the coalface of interpretation, by archaeologists who recognize that the background knowledge they have available to use when making history is insufficiently diverse to meet both their ethical and epistemological goals. They then seek to ameliorate these deficiencies by expanding upon the source-side analogues that are available through ethnographic research.
This refocusing of ethnoarchaeology as a form of 'slow science' means removing at least two of the positivist legacies that plague the subfield. The first of these is the notion that ethnoarchaeology is properly limited to 'middle-range' research and ethnoarchaeologists should spend most of their time identifying material correlates using an 'outside' or 'partitive' engagement with their research subjects (Binford and Sabloff 1982). The correlate mandate for ethnoarchaeology is a derivative of the desire for an ever 'faster' science that believes its datasets should have simple causes and wide reaching uniformityreflecting, for example, a sense that ceramic decoration should everywhere be able to tell us about social propinquity. The behaviouralist legacies of ethnoarchaeology have been discussed elsewhere (Cunningham 2009), but the idea that ethnoarchaeology is properly an observational rather than a dialogical form of inquiry is an all-toofrequent caricature that needs to be abandoned. Ethnoarchaeology as 'slow science' begins with the need to have socially engaged forms of research.
A second issue is more challenging. If archaeology relies on understandings of modern lifeways for its interpretations, then one of our fundamental conceptual challenges is developing an understanding of modern cultural variability and its contributions to archaeological analysis. It means revisiting assumptions about the relations between archaeologists and the contemporary 'others' among whom they hope to do ethnoarchaeology, the relations between archaeologists and ancients that create interpretative challenges and the relations between contemporary 'others' and ancients that give them enough commonality to make ethnographic engagements desirable. The question of relations has been explored previously in discussions about analogy under the rubric of 'boundary conditions' (Stahl 1993) and 'arguments of relevance' (Wylie 1988), but we would suggest that it needs to be a core component of a refigured ethnoarchaeology.
Relations between archaeologists, 'others' and ancients are most often understood through a temporalized understanding of cultural variability that is so widespread that it approaches common sense. Stated crudely, it suggests that archaeologists struggle to interpret ancient settings because their own experiences in the West leave them ill-equipped to understand pre-modern practices that might be found in the archaeological record. They thus travel to a global South to study populations that they assume possess vestigial practices that parallel those in the archaeological record but have been lost to Westerners. As Johannes Fabian (1983) has noted for anthropology and Ann Stahl (1993) for ethnoarchaeology, temporal metaphors like these define our understandings of cultural variability. They also undergird a global continuum of 'developed' and 'undeveloped' spaces (Ferguson 2006)with 'undeveloped' spaces subject to the distancing stereotyping mentioned above. Critiques of this view are well established and show how the definition of 'others' supports the rationalist and modernist aspirations of Euroamerican scholarship (e.g. Said 1978;Latour 1993) by downplaying the deep global connections that produced both capitalism (Wolf 1982) and a heterogeneous modernity (Knauft 2002).
Ethnoarchaeology's 'others' are thus not vestiges available for easy comparisons with archaeological parallels, but contemporary peoples who have experienced a different, frequently more impoverishing, pathway to that followed by most Western societies into the modern capitalist world-system. This observation does not deny the real diversity of cultural systems and logics that exist among societies that may be found in all areas of the worldboth those sometimes described by Westerners as 'traditional' and those that are called 'modern'. Those systems are not simply the results of diverse encounters with modernity; rather, they are very frequently also the consequences of cultural trajectories that involve different histories and understandings of the world, and so may inform and enlarge the understanding of ethnoarchaeologists in significant ways. Such cultural systems are never, however, 'traditional' in any straightforward sense: the peoples engaged with such systems all, in one way or another, inhabit the modern world and share its history, just as do the most cosmopolitan of Western academics.
That many practices involved in such cultural systems look like ancient forms should not lead us to assume that they are in some way survivals from an ancient past. For example, Malian women in the Inland Niger Delta continue to make pottery using techniques that have deep historical roots (LaViolette 2000;Mayor 2010), yet their production cannot be studied as if it was somehow removed from modernity. Recently, potters have shifted from decorating their water jars with time-consuming incised and stamped designs to the use of painting, because they needed to reduce labour investments in ceramic production (Cunningham 2013). The change emerged as their once-itinerant iron-worker husbands abandoned their own craft production and took up farming, because cheap imported steel such as leaf springs had destroyed local iron smelting. Potters continued to make ceramics, in part because it is one of the few economic activities available to women in the Delta, but shifted their decorative tendencies to accommodate new demands on their domestic labour generated by their husbands' adoption of agricultural production. Similarly, the town of Mata Ortiz in Chihuahua, Mexico, produces beautiful jars in a style that recalls the nearby fourteenth-century political and economic centre of Paquimé. The similarities invite the illusion, especially from foreign collectors, that the townspeople have continuities with Paquimé's ancient inhabitants. Official histories forefront Juan Quezeda as an innovator who saw a market for craft work and developed a village-wide industry (Bartra 2003). Recent interviews, however, show that seminal potters started out looting nearby sites for pottery they could sell to American collectors and museums, and then took up ceramic production to create replicas that they passed off as archaeological pots once local sites were fully mined (Silva and Kelley 2015).
In the cases of both Mali and Mata Ortiz, ceramics are reproduced because they address the economic circumstances of people living within the heterogeneous spaces of the capitalist world system. Craft work, peasant farming and even varying degrees of hunting and foraging do not exist because of cultural inertia, but are purposefully reproduced each generation because they meet the needs of communities that make use of them. Given the expanding wealth gaps noted above, the hallmarks of 'modern' prosperity will most likely remain an illusion for much of the global South. We might expect that, under these circumstances, craft production, subsistence farming and hinterland foraging will continue or even expand, rather than disappearing. The key is thus to recognize that ethnoarchaeology studies 'others' not because of their traditionalism, but because their cultural perspectives and experiences within capitalism have been much different than the contexts in which archaeologists routinely operate.

Conclusion
In this paper, we have tried to stress two delimited points. First, those with the privilege of doing archaeology as a profession are challenged by the narrowness of their experiences. They most often come from a very particular position within the broad reach of modern capitalism, and have too often drawn from celebratory ideologies of the modern West to explain ancient practices. As Nic David has suggested, we need to expand our source-side 'analogical consciousness' if we hope to produce histories that do more than affirm our privilege. Ethnoarchaeology continues to have a vital role to play in that intellectual expansion. Second, 'big science/fast science' approaches to twenty-first-century archaeological research impose significant dangers for the discipline, in terms of both the structural damage they can inflict upon the archaeological community and the interpretative ambiguities associated with their analysesespecially in those regions, often dismissed as 'undeveloped' or 'developing', where ethnoarchaeology frequently takes place. 'Slow science' critiques offer an explicit counter to the first of those dangers and an implicit counter to the second, through their calls for real social relationships and contemplation as part of the research process. We suggest that ethnoarchaeology can thus be a key component of a 'slow science' approach to archaeology, an approach that acknowledges its aspiration and perhaps even its responsibility to write global histories, but one that remains concerned about its ethical stance and the limits of its intellectual machinery, and takes measures to ameliorate the conditions of both.

Disclosure statement
No potential conflict of interest was reported by the authors.

Notes on contributors
Jerimy Cunningham is an associate professor in the Department of Anthropology at the University of Lethbridge. His research specializes in North American archaeology and West African ethnoarchaeology. His current research explores the emergence of social complexity in the Casas Grandes Region from Chihuahua, Mexico.
Scott MacEachern is chair of the Sociology and Anthropology Department at Bowdoin College. His research specializes in African archaeology and ethnoarchaeology. His current research involves the study of state formation and ethnicity in Iron Age Central Africa.