Humanizing Science and Philosophy of Science: George Sarton, Contextualist Philosophies of Science, and the Indigenous/Science Project

Abstract

A century ago historian of science George Sarton argued that “science is our greatest treasure, but it needs to be humanized or it will do more harm than good” (1924). The systematic cultivation of an “historical spirit,” a philosophical appreciation of the dynamic nature of scientific inquiry, and a recognition that science is irreducibly a “collective enterprise” was, on Sarton’s account, crucial to the humanizing mission he advocated. These elements of Sarton’s program are more relevant than ever as philosophers of science articulate research programs that take seriously the contextual factors, situated interests, and historical contingencies that shape the sciences we study. I trace the trajectory of a long-term philosophical engagement with archaeology that illustrates a succession of ways in which social, cultural, and political values configure inquiry, culminating in a program of collaborative research that raises the question of what role philosophers can usefully play when the challenges of humanizing scientific practice have centrally to do with navigating entrenched asymmetries of power.

1. “Science is our greatest treasure.”

The “science and values” turn in philosophy of science is often taken to be a new departure, and in many respects it is. But as History of Philosophy of Science (HOPOS) colleagues remind us, a century ago many founders of our field took for granted the deeply social, values-inflected nature of scientific inquiry, and argued that a crucial role for history and philosophy of science was to cultivate an appreciation of these dimensions of science among its practitioners. George Sarton, a pivotal figure the history of science, was one of these. In his famous 1922 lecture for the American Association for the Advancement of Science (AAAS), “The New Humanism” (published in 1924), Sarton articulated a vision for “humanizing” the sciences: “Science is our greatest treasure,” he argued, “but it needs to be humanized or it will do more harm than good” (1924, 33).

First and foremost, on Sarton’s account this humanizing project should bring to the sciences an “historical spirit,” a recognition that empirical inquiry is a continuously unfolding process, its open-ended contingency often leading to insights that were not anticipated, much less sought, by its practitioners. Recognizing the dynamic nature of science, he argued, could ensure that our current best science never hardens into a “final doctrine”: “fixed knowledge is dead knowledge” (1924, 27). It also requires a recognition that science is a collective enterprise, one that depends on the “combined efforts of all peoples” (11); a living, growing science will thrive only given an expanding web of “social ties” (24). Indeed, Sarton considered it a “vital necessity” that scientific inquiry should be informed by “the sympathetic study and understanding of the thoughts and ways of other peoples” (26), not only to enrich the resources on which scientists draw but to foster in them a “keener realization of one’s ignorance,” an ongoing reckoning with the limits of their own methods and understanding (21). In short, the fierce commitment to rigorous inquiry that Sarton prized above all else must be “tempered by … humility and charity,” a “broadness of outlook and boundless generosity” (21–22). ¹

In what follows, I first consider humanizing lessons that arise from taking philosophy into the field—the archaeological field. These reinforce insights that motivated a succession of contextualist philosophies of science; they illustrate how, in concrete and practical terms, empirical inquiry is configured by the values, interests, and assumptions of those who set the agenda, do the research, interpret and apply its results, and by the institutions and disciplinary cultures within which they operate. ² I then broaden the frame to consider lessons drawn from the forms of community-based and community-led collaborative practice that archaeologists have developed in partnership with Indigenous communities in recent decades. Challenging though such collaborations often are, they are proving to be a crucial means of transforming research practice in ways that both reinforce and extend Sarton’s program for “humanizing science,” and that illustrate many of the key points made by philosophers of science who advocate not just a more thoroughly social contextualism, but one that is normatively engaged: one that aims to articulate norms of socially responsible science. I return, in the conclusion, to the questions framed by Sarton about the role that historians and philosophers of science can play in fostering these humanly, socially engaged modes of practice. I argue that, to do this, we must be prepared to humanize our own practices; attending to and, indeed, actively engaging in the collaborative research programs taking shape in the sciences we study has reciprocal benefits for our own practice.

2. Humanizing vernacular positivism

To illustrate Sarton’s humanism in action, consider an example of philosophical debate that I encountered in the context of an archaeological field program when I worked at Fort Walsh, a Northwest Mounted Police (NWMP) fort in the heart of the Cypress Hills, southwest Saskatchewan. This was one of several National Historic Sites (NHS) where Parks Canada was running an ambitious field research program in the 1970s an 1980s. ³ As a philosophy student I was struck by how much attention the archaeologists I worked with gave to philosophical issues. Sarton’s “attitude of the philosophical scientist” (1924, 21) was clearly evident in ongoing debates about whether the NHS-mandated aims of the project were well served by the conventions of archaeological practice that informed our work as we surveyed and excavated, mapped, and recorded the surviving record of the NWMP presence on the site. North American anthropological archaeology was then in the throes of a “revolution,” as described by the advocates of “processual” New Archaeology who had trained our field director. ⁴ They rejected the “narrow empiricism” of “traditional” archaeology which, on their view, perpetuated a barely reformed antiquarian preoccupation with recovering and systematizing archaeological data as an end in itself. Ironically, however, the philosophical models of science invoked by the New Archaeologists were drawn from what came to be known as “Received View” philosophy of science, a form of logical positivism then widely reputed to be meeting its demise (Suppe 1977). What they advocated was a vernacular logical positivism, not unlike that which has been influential in a good many social sciences (Wylie 2002, 78–80). It was anything but a bulwark against the kinds of epistemic ideals—empirical foundationalism and inferential certainty—that underwrite the quest for fixed and final science Sarton warned against.

The New Archaeologists’ appeals to covering-law models of explanation and a hypothetico-deductive account of confirmation informed their programmatic arguments for a properly scientific approach to archaeology. By contrast to the limited ambitions of traditional archaeology, they insisted that the goal of inquiry must be to develop explanations of the particulars of cultural history and lifeways in terms of putative laws of long-term, large-scale cultural processes understood in “eco-materialist” terms. To this end, they called on their colleagues to invert the traditional practice of building “just so” stories to account for what they found in the archaeological record. Rather than treating these as the terminus of archaeological inquiry, archaeological research should be rigorously problem oriented; surviving material traces of the cultural past were to be valued, first and foremost, as a resource for testing explanatory hypotheses about human–environment interactions operating at the level of cultural systems. Processual New Archaeologists insisted that this reorientation—simultaneously ontological, epistemic, and methodological—constituted nothing less than a Kuhnian revolution. ⁵

In the context of grappling with the realities of fieldwork, the jointly epistemic and methodological debates I was party to at Fort Walsh in the mid-1970s called into question not only the norms of “traditional” archaeology as exemplified by the NHS research program, but also virtually every key tenet of the logical positivism endorsed by the New Archaeologists; they resonated with the genre of contextualist accounts of the theory-laden nature of evidence that were getting traction in post-positivist philosophy of science at the time. ⁶ We grappled with the implications for hypothesis-testing methodologies of recognizing just how much interpretive judgment goes into identifying and recording “data” that count as archaeological and putting them to work as evidence, even on a site as recent as Fort Walsh. For example, the NWMP were stationed at Fort Walsh for the express purpose of controlling the whiskey trade and yet one of the most prominent types of artifact we found were alcohol bottles: champagne, French wine, and cognac bottles in the commandant’s latrine; American beer and whiskey bottles in the officers’ latrines and under the floorboards of their barracks; and large quantities of high alcohol–content medicine bottles in a wide range of contexts including the footing trenches of the enlisted men’s quarters (Murray and Sciscenti 1977, 163, 175–78; Lunn 1979; Murray 1977). In contravention of their mission, the archaeological record delivered uncompromising evidence that large quantities of alcohol were present and presumably consumed by the NWMP stationed at Fort Walsh. Even so, it was not altogether obvious what to make of these material traces as evidence. Did they represent expropriated and illegally consumed contraband, or ongoing (also illegal) cross-border trade in intoxicants that now catered to the rank-differentiated tastes of the NWMP? The second commissioner to oversee Fort Walsh, A. G. Irvine, issued a general order in 1880 that “the most rigorous steps are to be taken for the stoppage in liquor traffic” including “intoxicating medicines” but, if anything, the archaeological evidence suggests that the volume of trade in these medicines increased over time (Sciscenti et al. 1976, 160–67).

Other lines of evidence indicate that although the NWMP were charged with implementing Canada’s Indian policy, which meant enforcing law and order and initiating treaty negotiations, they were heavily reliant on “Indian products” for survival. Indigenous butchering techniques were evident in the faunal assemblages we recovered and the beads we found embedded in the compacted subfloor deposits of the barracks suggest a widespread practice of amending NWMP uniforms with a range of climate-appropriate Indigenous clothing (Sciscenti et al. 1976, 236–41). Perhaps the boundaries between the NWMP and Métis were a good deal more porous than acknowledged in official reports and histories. In a similar vein, although Fort Walsh is described as the “largest and most heavily armed fort that the North West Mounted Police garrisoned during their early years in the West” (Parks Canada 2018), ⁷ it was essentially indefensible. It was located in a valley, visible from high ground on all sides, and surrounded by a gappy, unstable palisade that was constructed only after the first major buildings were erected (Scicsenti et al. 1976, 25). Excavation documented at least one meter-wide gap in the palisade with no evidence of a gate or footing trench (40), and the bastions seem not have been constructed with functional firing platforms; archaeological deposits inside the earliest of these suggest that it was used for grain storage (44).

In short, the material record of Fort Walsh ran counter to then-conventional accounts of NWMP life on the frontier, raising a number of questions about how it should be “read” as evidence. ⁸ The first insight that arose from taking philosophy into the field was that archaeological evidence is thoroughly a construct, but it is by no means a construct that obligingly conforms to expectation. This posed a challenge not only to the vernacular positivism of the New Archaeology but also to the constructivism of their “post-processual” critics who, a decade later, were invoking what Norwood R. Hanson described as the theory-laden nature of observation (1958) to support the conclusion that hypothesis-testing in archaeology is untenable. At their most uncompromising, they insisted that archaeologists inevitably “create facts” which reflect their expectations (Hodder 1983, 6), setting up such an intractable circularity that archaeological evidence must inevitably confirm their theories about the past; there is “literally nothing independent of theory or propositions to test against” (Shanks and Tilley 1987, III; see discussion in Wylie 2002, 172, 206). Neither of these views, articulated in abstract terms, do justice to the ways in which, even though archaeological evidence is undeniably an inferential construct, it can subvert expectations; sometimes it decisively challenges seemingly well-established claims about the past and brings into view cultural histories and forms of life that had never been considered. ⁹

When archaeological data get purchase as a defeasible but nonetheless consequential source of empirical constraint on claims about the past, I have argued that it is precisely because of its dependence on a wide range of background claims and assumptions that “warrant” inferences about its evidential significance; crucially, these include empirical and technical as well as theoretical warrants (Wylie 2011, 2020; Chapman and Wylie 2016, 33–39). Archaeological data never confront hypotheses as a wholly autonomous empirical “tribunal” of their truth, and rarely do they establish evidential claims with certainty, but by no means are they all or only arbitrary interpretive conventions. There are at least three aspects of evidential reasoning that account for the capacity of highly constructed archaeological data-under-interpretation to counter the threat of “fact-shaping” (Kourany 2020; see note 2). For one thing, many of the warrants that mediate inferences about the evidential significance of archaeological data are themselves subject to empirical constraint, which means that the security of evidential claims can be systematically adjudicated; in the case of dating techniques, for example, this involves an ongoing process of assessment and calibration (Wylie 2020; Bokulich 2020). Moreover, the theoretical commitments that do play a role in evidential reasoning (alongside other forms of background knowledge), are not necessarily, or even typically, the same as the theoretical assumptions that inform the reconstructive or explanatory hypotheses about past events, conditions, and processes that archaeologists use their evidence to support or to test. Combined with empirically grounded appraisals of the degree of security provided by mediating warrants, it is epistemic independence in this first sense—between the “theory” that ladens archaeological data-as-evidence, and theories that are the direct or indirect object of archaeological investigation—that archaeologists exploit when they make effective use of archaeological data as a source of empirical insight and constraint. In addition, archaeologists rarely depend on a single line of evidence; they typically enlist diverse ranges of background knowledge to constitute different types of data as evidence when building or testing archaeological hypotheses. ¹⁰ Epistemic independence in this second sense—between lines of evidence, each backed by its own set of warrants—puts archaeologists in a position to exploit strategies of triangulation and robustness reasoning (Wylie 2011, 381–89; 2020; Wimsatt [1981] 2012); diverse lines of evidence can be mutually constraining, exposing error as much as reinforcing one another. ¹¹

What I encountered in practice at Fort Walsh was the paradox that as enigmatic as archaeological data are, they also manifest a stubborn concreteness that, to paraphrase Lorraine Daston (2008, 11, 13), sometimes reasonably inspires confidence that they bear trustworthy witness to pasts (or aspects of the past) that may not be otherwise documented. ¹² In archaeology, as elsewhere, evidential claims are always defeasible, but the “tangle” of interdependent, mutually constraining evidence of different kinds makes it possible to assess degrees of credibility that lie between the deductive certainty prized by New Archaeologists and the radical contingency on which their critics insisted. Archaeology at its best is an ongoing, iterative process of eliciting evidential friction and constraint. ¹³ Recognizing this, I would argue, is a matter of humanizing the vernacular positivism that was embraced by processual New Archaeologists. ¹⁴

3. Humanizing philosophy of science

Other challenges to this vernacular positivism that were less discussed in archaeological contexts in the 1970s bring into sharp focus more profoundly humanizing lessons for philosophy of science that advocates of the “values in science” program have since explored. In the final year I worked at Fort Walsh, we successfully lobbied Parks Canada to support an archaeological survey of the park property which I directed (Wylie 1978). Up to that point, we had strict instructions to confine our attention to the fort itself: our mandate was to “uncover the precise configuration of the original Fort buildings” and recover “as many artifacts as possible” so that the Fort could be restored “to its exact original specifications and materials”—not a “vague copy of an Ontario fort from a different period” (Halstead 1974). Expanding the scope of our fieldwork program to the park property as a whole fundamentally reframed our understanding of life in the Cypress Hills during this volatile period (Wylie 2017a, 123). ¹⁵ We documented dense concentrations of Indigenous encampments—First Nations tipi rings, lithic scatters, and cairns on the bluffs and benches overlooking the Fort—as well as clusters of pit features along Battle Creek and historic trails, likely associated with Métis trading posts and households, that had gone unrecognized, undocumented, and unprotected. A visitor reception center had been built in the middle of a major encampment site, and utility roads transected lithic scatters and cellar pits. Parks Canada has since made First Nations and Métis history central to the management plan for Fort Walsh (Parks Canada 2013) and, on the official website for Fort Walsh, the NWMP are now described as “the advance guard of wholesale invasion and settlement” of the Canadian prairies (Parks Canada 2018). ¹⁶ But apart from a recognition that the catalyst for establishing a NWMP fort in the region was the infamous Cypress Hills massacre of 1873 (see note 3), this Indigenous history was not taken into account in the archaeological investigation of the site when it was being developed as a national historic park.

This lack of attention to First Nations history and heritage was by no means unique or inadvertent; it was explicit in the priorities set by Parks Canada for its archaeological research program at Fort Walsh in the 1970s. The relevant philosophical insight here, central to the “aims approach” to values in science proposed by Kristen Intemann (2015), is that the values and interests that set the research agenda configure all aspects of research practice; their influence cannot be circumscribed by stage of inquiry or role. ¹⁷ Reviewing these aspects of a “contemporary program for a ‘contextualized’ philosophy of science” (Kourany 2010, vii), Ingo Brigandt (2015) provides a systematic assessment of reasons to doubt not only that social values can be excluded from contexts of inquiry and theory evaluation as required by an ideal of “value-free” science, but also that their influence can be limited in various ways—to the choice of background assumptions in what he describes as “inference from evidence” analyses, or to indirect as opposed to direct influence on the choices scientists in the course of a research program, as proposed by Heather Douglas (2000). ¹⁸ I concur with Brigandt’s conclusions; a much broader role for “values” in science must be recognized that includes pragmatic and value-driven aims which, on his account, configure domain-specific “conditions of adequacy” that inform theory evaluation (Brigandt 2015, 343). The “new worries about science” discussed by Janet Kourany (2020, 7–9; see note 2) have to do with exactly this. “Fact-shaping,” as she refers to it, arises from the value-informed choice of aims that determine which facts will be considered relevant and will be gathered, a process that can, she argues, “preclude the discovery of other facts” (2020, 8). Crucially, this has path-dependent implications for the capacity of future research to generate counterevidence that could call into question assumptions about the subject domain which inform the orienting aims of a research program. In the case of the archaeological work at Fort Walsh, Kourany’s point about fact selection applies directly. In the first instance, the questions of interest to the NHS were limited to the NWMP at Fort Walsh, which meant that the archaeological record we had been creating as we worked within the confines of the fort compound systematically marginalized evidence of the presence of Indigenous peoples that the NWMP had been charged with policing and “settling.” ¹⁹ The heritage we were investigating at Fort Walsh was that of the Canadian settler state, and the NHS research program was itself an extension of the settler state policies and practices we were documenting archaeologically. It was only when the aims of inquiry were reframed and the scope of the NHS research program expanded beyond the fort that “facts” situating the NWMP in the context of a much more complex history could be recognized—a history of Indigenous survivance in which, rather than reproducing eliminationist narratives, First Nations and Metis peoples are understood to have been a dominant force in the region and to have played an active role negotiating their rights and interests with a diverse cast of incomers.

At a broader level, Brigandt’s point about conditions of adequacy also applies to the archaeological program at Fort Walsh, albeit in quite general terms. The orienting conventions of responsible archaeological practice at the time were complicit in perpetuating the legacy of displacement, appropriation, and denial originally enacted at Fort Walsh. The vernacular positivism that informed the Fort Walsh fieldwork, as much as the narrow empiricism of traditional archaeology, foreclosed active consideration of the political interests and social values embodied in the NHS research agenda; both were predicated on a conviction that the goals of archaeological science transcend any “parochial” situated interests and that sufficiently rigorous pursuit of these goals could be counted on to secure empirical results that are “value-free.” ²⁰ The second lesson I draw from taking philosophy of science into the field is that the epistemic contextualism catalysed by early critiques of Received View philosophy of science must itself be humanized in the senses now being explored by aims theorists; it must be reconceptualized in robustly social and normative terms. With respect to Sarton’s humanizing mission, the point is that if philosophers of science are to play a role in counteracting the stultifying tendencies that concerned him—the risk that current best practice and scientific knowledge will harden into “fixed and final doctrine”—they must be prepared to scrutinize the purpose-specific goals and associated norms of practice that determine what counts as properly scientific practice in disciplinary research programs. ²¹

4. Humanizing archaeology

By the turn of the 1990s, a series of tectonic shifts were taking place within archaeology and in its contexts of practice that brought these more robustly social and normative contextualist insights to the fore. Internal critiques, as well as challenges from a growing number of descendant communities intent on regaining control of their cultural heritage, were fundamentally changing the conditions under which archaeologists operate, provoking for many a reckoning with the role that archaeology had played in transacting and legitimating oppressive political agendas, including the eliminationist policies and practices of settler states. Indigenous and Aboriginal activists were winning significant legal battles, especially with respect to the repatriation of human remains and the protection of sacred and ancestral sites. In some contexts, this took the form of federal legislation, for example, the 1984 Aboriginal and Torres Strait Islander Heritage Protection Act in Australia and the 1990 Native American Grave Protection and Repatriation Act in the US. In Canada, the protection of heritage and burial sites that are not on federally owned land is a provincial and municipal responsibility, and provisions for their protection vary widely. Whether enacted on a broad or local scale, the enforcement of heritage legislation is notoriously uneven and, even when implemented, it typically has little effect on the questions archaeologists address or their research methodology.

Faced with the threat of externally imposed constraints on their practice, some archaeologists doubled down in their defense of the value-free ideal associated with the “narrow empiricism” of traditional archaeology and the self-avowed positivism of the New Archaeology. ²² They reject outright any demand that they be held accountable to the interests of nonarchaeologists, especially when these embody culturally distinctive ways of understanding and valuing cultural heritage. Some are explicit on the point that there is nothing of archaeological relevance to be learned from the “thoughts and ways of other peoples,” as Sarton might put it (1924, 26), ²³ and that nonscientific interests should not constrain the pursuit of scientific questions; a properly scientific understanding of the human cultural past is in the interests of, if not of interest to, all people, society as a whole, and should not be held hostage to the demands of special interest groups. Most fundamentally, the concern here is that any capitulation to the interests of descendant communities—any move to humanize archaeology in this sense—risks compromising the integrity of archaeological science. Unless a hard line is taken against the incursions of outsiders and their potentially biasing influence, there is no stopping the slide into corrosive relativism. Framing their objections in abstract terms, these internal critics articulate a defence of the value-free ideal as essential to scientific practice that echoes Paul Boghossian’s philosophical arguments against forms of liberal pluralism that presuppose a “doctrine of equal validity” (2006, 2). On Boghossian’s account, the epistemic norms typical of contemporary Western science, norms that exemplify “our own classical picture of knowledge” (19), uniquely approximate “absolute, practice-independent facts” that determine what should count, for all, as true beliefs and norms for their justification (110). ²⁴

At the same time, a growing number of archaeologists have responded to Indigenous demands for control over their cultural heritage with a commitment to reorient archaeological research so that it benefits nonarchaeological stakeholders, especially Indigenous communities when it is Indigenous history and heritage they study. ²⁵ They seek out community-based and community-led collaborations, working in partnerships that reframe their research agendas, directing their efforts to questions that are relevant to and informed by the knowledge and expertise of Indigenous communities. A small but growing number of Indigenous archaeologists, as well as non-Indigenous archaeologists who work for Indigenous cultural and historic preservation offices, have been pivotal in building these working relationships. By contrast to critics who feared that any such engagement would be epistemically compromising, those engaged in such efforts point to a range of ways in which a commitment to values of social justice and accountability has significantly raised the bar epistemically. ²⁶

4.a Nineteenth-century legacies: Archaeology at c̓əsnaʔəm

These shifts in the aims that orient archaeological research go hand-in-hand with critical appraisals of archaeology’s colonial entanglements that often take the form of historical genealogies of specific research programs. These are, I suggest, one means of fostering a measure of the humility advocated by Sarton. They make explicit the purpose-specific contingency of taken-for-granted assumptions about the nature of the subject domain that configure the aims of inquiry, and throw into sharp relief the limitations of the methodological conventions and epistemic norms of justification—the domain-specific conditions of adequacy—associated with them. To illustrate what this involves, consider a case from the Indigenous territory where I now live and work: ²⁷ a history of the Musqueam Band’s engagement with the generations of archaeologists who have excavated the “Great Marpole Midden.” In These Mysterious People (2006, 2016), Susan Roy details the process by which this famously rich and expansive Musqueam settlement at the mouth of the Fraser River—c̓əsnaʔəm—became, for archaeologists, a regional type-site in terms of which they defined the characteristics and developmental sequence of precontact Coast Salish cultures. ²⁸ Archaeologists date its earliest occupational levels to at least four thousand years ago (from 2400 BP). The Musqueam understand c̓əsnaʔəm to have been continuously occupied by generations of their forebears; despite decimation of the community by smallpox in the nineteenth century, it continues to be one of the most important ancestral villages and burial grounds in their traditional territory.

The c̓əsnaʔəm settlement first attracted attention as an archaeological site in 1884 when road improvements exposed cultural material and burials—in contemporary terms, Musqueam belongings and ancestral remains ²⁹—and it has since been relentlessly “mined” for artifacts and human remains by “archaeologists, collectors, and treasure hunters” (University of British Columbia Library n.d.). Under Franz Boas’s direction, Harlan Smith undertook what was understood to be “salvage anthropology” at c̓əsnaʔəm in the late 1890s with the aim of documenting the migration of populations and the diffusion of cultural traits within the region and between the Pacific northwest and northeastern Asia (Roy 2016, 35). ³⁰ Like most anthropologists at the time, they operated on the assumption that Indigenous cultures had disappeared, or were rapidly disappearing, and that the cultural histories of interest to archaeologists lie beyond the reach of Indigenous oral traditions so would be of little interest to contemporary Indigenous people in the region. In the case of c̓əsnaʔəm and the Musequeam, these assumptions were amplified by claims that had been published in 1895 by Charles Hill-Tout on the basis of material he collected from c̓əsnaʔəm for the Art, Historical, and Scientific Association of Vancouver. He maintained that “contemporary Aboriginal residents were relatively recent newcomers to the area”; the original inhabitants of c̓əsnaʔəm, a pre-Salishan people who had occupied villages throughout the Lower Fraser River region, had been “displaced or exterminated” by a “hostile people” who had migrated into the region from the interior (Roy 2016, 32, 107,113). Hill-Tout’s interpretation was informed by the racialist assumption that “primitive peoples such as our Indians [are] deeply conservative” (n.d., as quoted by Roy 2016, 115). Their cultures are static, bounded, and unchanging so that observed differences in material culture, over time or across a region, must be attributed to differences in cultural tradition and identity which were, in turn, associated with racially distinct populations. Where c̓əsnaʔəm was concerned, Hill-Tout based his theory of population discontinuity on the claim that the shape of crania recovered from the site changed over time; he identified the hostile intruders as a distinct race of “broad-headed people,” in contrast to earlier inhabitants characterized by “long skulls” (Roy 2016, 53).

Smith, working for Boas, interpreted these differences in cultural terms; he took long-type skulls to be the result of a practice of head-binding and, in his field notes, he reported a continuum in skull types; between the extremes of broad and narrow, he found “every conceivable intermediate form” (1898 correspondence with Boas, as cited by Roy 2016, 52). Nonetheless, in his report to Boas, Smith reiterated Hill-Tout’s displacement hypothesis, concluding that c̓əsnaʔəm had been occupied by “two distinct peoples” which Boas interpreted as evidence of population migration. Although Boas emphasized the dynamism of local cultures—“the people of the North Pacific coastal region no longer appear to be unchanging, ahistorical entities” (Boas 1908, as quoted by Roy 2016, 35)—he too reproduced Hill-Tout’s claim that contemporary Musqueam were descendants of an intrusive population. Contra established Musqueam tradition and understanding, the original occupants of c̓əsnaʔəm were not their ancestors.

Whatever Boas’s and Smith’s intentions were as anthropologists, this discontinuity thesis served to legitimate setter-colonial political interests and agendas in a number of ways. Roy observes that, in addition to dissociating contemporary Musqueam from key villages and sacred burial grounds in their traditional and ancestral territory, it normalized the violent displacement of one population by another as a “natural process in all human settlement” which, in turn, legitimated ongoing processes of Indigenous dispossession by the Canadian state and settler population in the region (2016, 33). ³¹ As such, it is recognizably a variant of eliminationist ideologies by which Indigenous territories have been declared terra nullius in settler-colonial states around the world, in this case not by insisting that the land is literally unoccupied but by asserting that the territorial claims of Indigenous people resident in the region lack historical credibility and therefore cannot be accorded legal standing. ³² The “two-race model” introduced by Hill-Tout was reproduced and made vivid by midcentury cranial reconstructions created for display in the Vancouver City Museum (77), and it repeatedly resurfaced in legal rights and title cases well into the 1990s (32). This “commitment to disconnectedness,” as Roy describes it (115), is explicit in the 1948 news article from which she takes her title:

Who were these mysterious people who lived long ago at Sea Island at the mouth of the Fraser River? … They were not Indians certainly. (70)

4.b Borden and the Musqueam Indian Band

Crucially, for present purposes, these assumptions about Indigenous culture and history, and the settler-colonial interests they embody, set the conceptual framework for archaeological research in the region through most of the twentieth century. A call in the mid-1940s for “systematic, scientific” archaeological work on lower mainland and Fraser delta sites was taken up by a UBC-based scholar, Charles Borden, who later developed the site designation system still in use across Canada. ³³

Borden undertook a more rigorous and ambitious program of excavation than had been typical of his predecessors at a number of Musqueam sites from 1947 through the 1970s. Although these were often salvage operations prompted by encroaching urban development, his goal was to establish a standardized spatiotemporal system of Coast Salish cultural phases for the region. He was especially concerned that his predecessors had not systematically documented the stratigraphic sequence of cultural phases—the temporal distribution of distinctive assemblages of belongings and ancestral remains—that was presupposed by the narrative of displacement. Over decades, his fieldwork served to establish the central importance of c̓əsnaʔəm as a regional type site. However, despite building a richer and more detailed empirical basis for a site-specific and regional typology and chronology, and despite a growing body of evidence that occupation at c̓əsnaʔəm had been continuous for at least four thousand years—described by Roy as “complicat[ing]” his “recent settlement theory” (2016, 136)—Borden’s culture-historical phase system for the region emphasized discontinuities. He posited a series of discrete, temporally and spatially bounded precontact cultures, recapitulating the conventional wisdom that an earlier “Marpole” culture (450 BCE to 500 ACE) had been replaced by a later immigrant culture, designated in his system as Whalen Phase II (Borden 1970). The existence of this phase was called into question in the early 1990s and is no longer widely used (Thom 1992), ³⁴ but the basic architecture of the system persists, setting the terms in which archaeological sites and assemblages are described and compared, and also reported under the British Columbia Heritage Conservation Act that Borden was instrumental in establishing. ³⁵ Roy makes the case that this program of archaeological research played a role in establishing and sustaining the settler-colonial “civic narrative” of dispossession that legitimated the dissociation of the Musqueam from their ancestral settlement at c̓əsnaʔəm and from their heritage in the region as a whole (2016, 33, 150). ³⁶

At the same time, a distinctive feature of Borden’s practice was the relationship he developed with a number of prominent Musqueam families and community leaders. When he first proposed to run a UBC field school on Musqueam Reserve lands he faced stiff opposition. He ultimately secured permission through a process of negotiation that involved an effort on his part to inform Musqueam community members about his archaeological aims and practices, and he later encouraged their participation in his field projects. The Musqueam were by no means passive recipients of archaeological wisdom in this process; they engaged in what Roy describes as deliberate strategies of intervention by which they put the tools, evidence, language, and authority of archaeology to work for their own purposes (Roy 2016, chap. 5). Although Borden seems to have assumed, with Smith and Hill-Tout, that Indigenous oral history could provide little insight into deep-time cultural history, ³⁷ he did come to recognize that contemporary Musqueam have a profound sense of connection to their ancestral villages and territory. He encouraged the Indigenous community members who worked with him to pursue their own interests in archaeology ³⁸ and when, in the early 1970s, he was invited by the Musqueam Band to organize salvage excavations in an area where they planned a housing development, he celebrated the fact that this project was not initiated “for our own gain” but in response to “a request which has come from them and as part of the Museum Project conceived by them” (correspondence cited by Roy 2016, 141). ³⁹

As Roy points out, this level of engagement with an Indigenous descendant community was virtually unheard of at the time (2016, 134). Borden did clearly privilege archaeological questions and interpretations as authoritative; “no doubt,” Roy observes, he “had difficulty reconciling Aboriginal historical tradition and contemporary residency with his archaeological classification of cultural phases in the Lower Fraser Valley” in the face of mounting evidence that Musqueam settlements were continuously occupied, consistent with Indigenous oral history (Roy 2016, 135; see note 34). Nonetheless, he instituted practices that anticipate some elements of contemporary collaborative practice in archaeology where this requires active consultation, consent, reciprocity, and the participation of descendant communities. In this respect, Roy notes, Borden’s practice represents a “turning point in the colonial culture of archaeology” (Roy 2016, 13, 151).

4.c Indigenous activism

By the 1990s, the association of c̓əsnaʔəm with the Musqueam was “automatic and unquestioned” (Roy 2006, 69); the “civic narrative” of dissociation that had been sustained in and, in part, by archaeology through the Borden era was discredited and there had been what Roy describes as a “shift towards mainstream recognition of First Nations cultures, histories, and geographies” (2016, 117). This she attributes to the emergence of an “anti-colonial, or reclamation culture” in Canada (2006, 93; 2016, 123,142), as well as the disciplinary reorientation she documents in terms of Borden’s practice when in the 1950s and 1960s “salvage ethnography and archaeology responded to a sense of moral obligation that recognized the demoralizing effects of colonialism on Aboriginal cultures” (2016, 124). This reclamation-era “salvage” archaeology, which persists in many contexts and is, in some, embedded in heritage protection legislation, might have been “tempered by [a degree of] … charity,” to invoke Sarton, but in privileging the interests of the settler state and the disciplinary research agenda of archaeology it was not humanized in the stronger sense he suggests when he urges a stance of “humility” with respect to the limitations of one’s own science, and an active appreciation that the “thoughts and ways of other peoples” are scientifically relevant (1924, 21, 26). It is this latter shift in the values and aims of archaeological research that distinguishes the collaborative modes of practice that have taken shape in archeology since the 1990s, and that embodies the type of rationale for socially responsible science articulated by advocates of a robustly normative contextualism in philosophy of science: that scientific inquiry must be accountable to “sound social values, as well as sound epistemic values” (Kourany 2010, vii). ⁴⁰

A thriving tradition of collaborative archaeology that goes substantially beyond Borden’s community outreach is now well established at the University of British Columbia. In recent decades, the “sense of moral obligation” Roy describes in connection with the midcentury “reclamation” culture had been reframed in terms of an explicit acknowledgement by the Canadian Archaeological Association (CAA) that “Indigenous Peoples have an inherent and unique relationships with their archaeological heritage,” which entails that archaeologists must recognize and respect “Indigenous approaches to protection, conservation, and interpretation of that heritage”: they must “learn and respect” Indigenous cultural protocols, and they must “make every effort to engage, cooperate, collaborate, and/or partner with the relevant Indigenous peoples and communities” whenever their work concerns “Indigenous sites or sites that include an Indigenous component.” ⁴¹

In Canada, this was a response to intensified Indigenous activism culminating in the “tumultuous ‘Indian summer’ of 1990” when Elija Harper initiated the First Nations’ challenge to the Meech Lake Accord and the Mohawk Nation of Kanestake set up a blockade at Oka to forestall the extension of a golf course onto ancestral lands they regard as sacred (Coulthard 2014, 115–17). The Royal Commission on Aboriginal Peoples (RCAP) was created in 1991 with a mandate to “investigate and propose solutions to the challenges affecting the relationship between Aboriginal peoples … [the historic nations of Canada], the Canadian Government and Canadian society as a whole” (RCAP 1996). This report was followed by another, “Gathering Strength,” which set out recommendations for implementing the RCAP principles of “mutual respect, mutual recognition, mutual responsibility, and sharing” (Department of Indian Affairs and Northern Development 1998). A decade later, in 2008, the Canadian Supreme Court mandated the creation of a Truth and Reconciliation Commission (TRC) as one component of the Indian Residential Schools Settlement Agreement. Charged with documenting the history and legacy of the IRS system, the TRC issued its final report in 2015, along with ninety-four calls to action intended to “advance the process of Canadian reconciliation” (TRC 2015, 1). This report is a searing indictment of Canadian settler-colonial eliminationist policies and practices which, the Commission concluded, could only be described as a program of deliberate cultural genocide (TRC 2015). As contentious, uncertain, and radically incomplete as this process has been, the commitment to “surface truths” and to seek some form of (re)conciliation with Indigenous communities represents a significant step beyond the midcentury “reclamation” culture that Roy describes, at least with respect to public awareness of Indigenous demands for justice. It remains to be seen what tangible actions will follow. ⁴²

It is in this context that Indigenous communities are demanding respect for principles of self-determination with respect to their cultural heritage. In 1993, frustrated with the slow pace of rights and title cases and faced with yet another threat to c̓əsnaʔəm, the Musqueam Band purchased the Fraser Arms Hotel; a permit for development of this property had been approved even though it had been designated a national historic site in the 1950s (Roy 2016, 131). Again, in 2011–2012, they organized a vigil that ran for over two hundred days and successfully turned back plans for condo development on another area of c̓əsnaʔəm where ancestors had been buried and were being disturbed by construction. ⁴³ The Musqueam Band currently employs a number of archaeologists whose mandate includes actively monitoring development projects and, although it is a constant struggle that is not well supported by provincial Heritage Act legislation, they routinely intervene to document, manage, and protect ancestral sites throughout their ancestral territory. One turning point in current relations between UBC and the Musqueam Band was the role that Musqueam scholars and knowledge keepers played in planning and producing a 1996 exhibit of archaeological material from southwestern British Columbia for the Museum of Anthropology (see note 35). Under pressure from the Musqueam Band, their representatives were involved as “full partners,” not just invited to consult on the details of interpretation once the project had been conceptualized. Co-curators Margaret Holm and David Pokotylo (1997) describe this collaboration as transformative. The result was a significantly different exhibit than originally envisioned; Written in the Earth was framed in terms of First Nations heritage as much as archaeologically defined cultural histories (1997, 39). Reflecting on this process, Holm and Pokotylo remark that it “required us to reassess our professional values, terminology and cultural biases in order to respect the philosophy and perspectives of First Nations” (39). This point generalizes to collaborative partnerships outside museum contexts; the advocates of community-based and Indigenous-led archaeological research often remark that, to translate good intentions and abstract policies into effective working relationships, archaeologists must relinquish their role as “sole experts” (41); control over the goals and products, conduct, and authority of archaeological research that have long been jealously guarded as a disciplinary prerogative must be redistributed among partners (Atalay 2006; Wylie 2015, 194).

4.d Indigenous/Science

At the University of British Columbia these principles are central to a number of ongoing research programs that have been developed in partnership with Indigenous communities in the region, ⁴⁴ and they were the motivation for forming “Indigenous/Science: Partnerships in the Exploration of History and Environments,” a research cluster founded in 2018 that builds on this tradition of practice. Its goals have been to convene a network of UBC-based researchers from a range of fields—archaeology, ethnography, education, legal studies, linguistics, history, molecular and chemical analysis, geology, applied ethics and philosophy—who share a commitment to find ways as researchers to take up the TRC calls to action and “do more than just talk about reconciliation” (TRC 2015, 21; Indigenous/Science 2018). ⁴⁵ In the spirit of cultivating partnerships centered on questions that Indigenous communities identify as relevant, representatives of the network reached out to a number of regional First Nations and asked if the expertise of its members might be useful to them. In the event, all the communities contacted identified projects of interest, most to do with identifying the materials used to produce Indigenous belongings, determining where these may have originated and traveled, and reconstructing on this basis patterns of settlement, cultural interaction, resource and landscape use.

When Indigenous partners set the research agenda, the questions they raise do not presuppose—indeed, they often call into question—the assumptions about Indigenous cultures and histories that have framed archaeological inquiry from the time of Boas’s Jessup expedition through Borden’s midcentury salvage archaeology. ⁴⁶ This includes the preoccupation with temporal and spatial discontinuity, and the presumption that Indigenous cultures are tightly bounded, highly conservative sociolinguistic units tied to discrete geographical locations. It is telling that Andrew Charles, a Musqueam archaeologist who worked with Borden, is quoted by Roy as observing that “culture is always in transition”; the lens through which he viewed the Musqueam history of occupation at c̓əsnaʔəm is one of dynamic continuity. His understanding of Musqueam’s relationship to their ancestral territory is similarly expansive; rather than localized to a particular village or cluster of settlements, Musqueam connections extend through a “far-reaching network of genealogical ties” that connect them to other Indigenous communities throughout the region and to their ancestors (Roy 2016, 135). ⁴⁷ One of the Indigenous/Science projects now under way in partnership with the Musqueam Band, “Splitting Obsidian,” documents these regional relationships in archaeological terms.

4.e Splitting Obsidian

This project was initiated by Rhy McMillan, then a PhD student in geological sciences, working with Dominique Weis and the research team affiliated with the UBC Pacific Centre for Isotopic and Geochemical Research (PCIGR) which Weis directs. In a meeting with Musqueam Band representatives in 2018, McMillan and Weis outlined several types of analyses they could undertake of obsidian belongings held on behalf of the Musqueam Band in an archaeological repository at UBC. Obsidian, a form of volcanic glass, is especially durable and chemically stable so, in principle, its chemical profile could be used to identify the most likely sources of the material from which the Musqueam belongings were made (McMillan, Amini, and Weis 2019). One technique McMillan could offer was nondestructive but relatively low-resolution, while another that would allow for a more accurate multivariate analysis of trace element composition and lead isotope ratios would require the extraction of physical samples from the belongings. ⁴⁸ Holm and Pokotylo had made a request twenty years earlier to “drill small holes in a few artifacts” so they could use Accelerator Mass Spectrometry (AMS) radiocarbon dating to determine the age of material they proposed to include in the “Written in the Earth” exhibit. In the process, they confronted a fundamental difference between the way they, as archaeologists, and their Musqueam partners value “heritage artifacts” (Holm and Pokotylo 1997, 36). From a Musqeuam perspective, whatever information might be gained by destructive testing was not worth the risk of compromising the intrinsic value and spiritual integrity of these ancestral belongings.

Mindful of these concerns, McMillan and the PCIGR team proposed the use of split-stream laser ablation, a component of the second technique, which would make it possible to remove samples so small that the damage is barely visible to the eye; it results in craters on sample surfaces of just 89 micrometres in diameter. After consultation with the head archaeologist and Musqueam knowledge keepers and elders it was agreed that, given the nature of the belongings to be analyzed and the minimal samples required, this degree of destructive testing would be acceptable. To interpret the results of this analysis, McMillan and his colleagues drew on a substantial body of baseline data that provide geochemical profiles of the magma reservoirs that feed clusters of volcanoes along the northwest coast from Alaska down to Oregon, as well as the Yellowstone volcanics. They are careful to specify that although they cannot render a definitive positive verdict about where a sample must have originated, they can eliminate obsidian sources given these profiles. Their conclusion is that the obsidian belongings they tested could not have originated in the immediate vicinity of ancestral Musqueam settlements; they most likely came from volcanic flows in the Snake River Plain in southern Idaho, as much as one thousand kilometers to the south and east (McMillan 2020; McMillan, Amini, and Weis 2019). ⁴⁹

This is an exciting result for all concerned. The geosourcing study undertaken by McMillan and the PCIGR team bears witness to a network of connections spanning the region that Musqueam community members are well aware of, but that settler conceptions of property and property ownership have systematically obscured. ⁵⁰ Roy traces the legacy of these assumptions in her genealogy of archaeological research at c̓əsnaʔəm, and they are foundational to the cultural typologies and culture-historical sequences that configure archaeological accounts of Coast Salish prehistory. ⁵¹ In this case, as in many others, Indigenous-led collaborative practice brings a distinctive set of values and interests to bear, humanizing archaeology in several senses. Most obviously, it reconfigures the aims of research by directing attention to new questions about the cultural histories of obsidian. This, in turn, required the geochemistry team to reassess standard geosourcing practices. The selection of a minimally destructive sampling technique was informed by Musqueam protocol, but the multivariate and multiproxy approach to geosourcing represents a methodological innovation. The laborious, technically demanding process of using multiple isotope ratios, rather than relying on a single trace-element proxy, might not be needed to address the questions that typically interest geologists and archaeologists intent on constructing regional cultural histories on a broad scale. However, it provides a level of resolution that is especially well suited to the questions about regional social networks and cultural histories that were the focus of a Musqueam-defined research program. This shift in aims is the motivation for several follow-on projects in which McMillan is now engaged with the Musqueam and other regional First Nations. Alongside the refinement of geochemical techniques, McMillan and several colleagues have developed an open-source statistical tool designed to support multivariate analysis of the data generated by these methods (McMillan et al. 2022). And because fine-grained baseline data are required for geosourcing projects that address questions about social networks and cultural histories, McMillan is currently engaged in field surveys designed to identify geological sources of obsidian and other types of tool-stone in the region.

In short, far from encouraging pernicious forms of value influence that pose a threat to the “factual basis of science”—the “old worries” about fact construction that Kourany addresses—the shift in cultural and political values that motivates archaeologists to build collaborative partnerships with Indigenous communities often enhances their own practice with respect to its evolving “epistemic values.” As Sarton would expect, learning about the “thoughts and ways of other people” (1924, 26) has the salutary effect of bringing a critical lens to bear on taken-for-granted goals and foundational assumptions, directing attention to a range of questions that had not been considered. And as aims theorists would expect, when purposes are redefined, researchers have to reassess what counts as evidence and recalibrate established methodological norms for gathering and analyzing it, a process that as often as not throws into sharp relief the limitations of research-as-usual, including conditions of adequacy of the kind that concern Brigandt. For decades, the results of archaeological research had been putting pressure on the “pre-understandings” (Bell 2015) that have structured regional archaeology and ethnology since the late nineteenth century; these insistently contrary “facts” have made a difference. But centering attention on questions that matter to the Musqueam Band and designing joint projects to address them—ones that are informed by Indigenous values, interests, and expert knowledge—promotes a much “keener realization of one’s ignorance” (Sarton 1924, 21): where it resides, how it has arisen, what interest-specific limitations it builds into archaeological inquiry, and what possibilities lie beyond the horizons of established research programs.

Conclusion: Reciprocal humanizing

I suggest that Sarton’s New Humanist ideals are today, almost exactly one hundred years on, more salient than ever, and that the modes of engagement with Indigenous partners that archaeologists have been exploring in recent decades reinforce many of the reasons he gave for “humanizing the sciences.” I draw two philosophical lessons from these examples of collaborative practice in archaeology.

One is that they illustrate the central insight articulated by advocates of an “aims approach” to understanding the pervasive, constitutive role of values in science: that the constellation of cultural values and social interests that set a research agenda configure all stages and aspects of the research process, including the domain-specific assumptions and methodological norms of practice that define what counts as rigorous, credible scientific inquiry. Taking stock of these values, constructing critical genealogies that document originating aims and their fact-shaping effects, to use Kourany’s terminology (2020), and, crucially, bringing other values into play, can make visible the limitations of taken-for-granted goals and norms of practice, reframing them in highly productive ways. The second is a specification of the humility that Sarton hoped his humanism would instill in the sciences: that if anything is central to, and defining of, scientific practice it is a commitment to hold its framework assumptions open to critical scrutiny—including ontological commitments and epistemic conditions of adequacy. To put this point in terms that call into question one influential line of philosophical thinking about science, we should not assume, with Boghossian (2006), that the norms captured by a “classical [Western, scientific] picture of knowledge” define what will or should count as authoritative norms of justification for scientific claims in all contexts, for all time. Ideals of epistemic transcendence are untenable; it is time to recognize the situated nature of scientific inquiry as well as its considerable power, and take responsibility for its aims-inflected partiality.

I add to these lessons learned from archaeological practice a point made in the early sections of this paper: if philosophy of science is to be fit for purpose—if it is to play a role in humanizing scientific practice in the expansive sense now envisioned by advocates of a thoroughly normative as well as social contextualism—it must itself be reciprocally humanized. The insights articulated by the various contextualist research programs that have taken shape in the last fifty years must be reflexively applied to our own confident presuppositions about what counts as philosophy as much as to what we assume about the sciences we study (Sample 2022). It is crucial to keep sharply in focus the contingency of goals and norms of that have taken on the status of “fixed and final doctrine” within philosophy. This calls for a resolutely nonideal program of philosophical inquiry in Charles Mills’s sense (2005), along lines currently being explored under the rubric of “socially relevant philosophy of science” (Fehr and Plaisance 2010) and captured by Katie Plaisance and Kevin Elliott’s taxonomy of “engaged philosophy of science” (2021). It requires that we cultivate in philosophy the virtues of “humility and charity” Sarton advocated for the sciences: that we learn from “the thoughts and ideas of other people,” especially critical voices within philosophy and those whose ways of investigating and understanding the natural and social world diverge from our own, and that we attend to values and interests that have typically not been addressed by scientific inquiry. Finally, if we are to do this well, we will have to build our own collaborative partnerships with colleagues in cognate fields of science studies who have the expertise necessary to construct critical genealogies of scientific research programs, and with researchers and stakeholders who are engaged in transformative criticism of the sciences we study.