There is potential for broadening the scope of archaeological research and management to more fully integrate ecological heritage sites and resources (Cicinelli et al. Reference Cicinelli, Salerno and Caneva2018; Lindholm and Ekblom Reference Lindholm and Ekblom2019). However, the conservation and management of ecological heritage—biotic features of a landscape that hold cultural, educational, and historical significance—has largely been the domain of researchers outside the fields of archaeology and heritage management (Eliasson et al. Reference Eliasson, Knez and Fredholm2018; Hølleland et al. Reference Hølleland, Skrede and Bech Holmgaard2017). This is true whether it is heirloom crops, sacred groves, or traditional use sites. In part, this arises because of archaeology’s traditionally narrow focus on material cultural remains—such as stone tools, ceramics, features, and other manufactured and built heritage—to the exclusion of legacy ecosystems or plant communities that form part of those same heritage landscapes (Lepofsky et al. Reference Lepofsky, Geralda Armstrong, Mathews and Greening2020). The exclusion of living heritage or lack of recognition in archaeological contexts affects how archaeological heritage is defined, investigated, recorded, conserved, and honored. For example, failing to recognize Indigenous peoples’ broad definitions of heritage, which often includes ecological elements and reciprocity, can result in the diminishment and erasure of peoples’ histories, rights, responsibilities, and important inheritances (Apaydin Reference Apaydin2020; Laluk et al. Reference Laluk, Montgomery, Tsosie, McCleave, Miron, Russo Carroll and Aguilar2022).
Here, we consider a framework that combines archaeological and ecological indices for identifying and revitalizing living sites, a concept derived in part from living heritage approaches (Poulios Reference Poulios2011, Reference Poulios2014), which challenge inherently static and aesthetic-driven management practices that tend to limit their use by communities. We consider forest gardens—ecosystems in the Pacific Northwest dominated by perennial food plants that were cultivated by people in the past and continue to grow in the present—as living sites. The reason we treat forest gardens as living heritage sites is that these ecosystems are directly associated with archaeological features and sites, sometimes growing immediately on or adjacent to ancestral settlements (Armstrong et al. Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023); but they continue to grow, reassemble, and change every year, belying any static management or mitigation approach (e.g., Welch and Ferris Reference Welch, Ferris, Atalay, Clauss, McGuire and Welch2016). The ongoing use, concern for, and management of living heritage sites by communities is intricately linked to the continuity of the heritage’s original function—in this case, to support the production of food and transmission of knowledge, instill connections to place, and shape a future that honors the ancestors.
To assess the historical development, use, and value of forest gardens in Sts’ailes territory, we begin from the vantage point that ancestral Sts’ailes have long-term connections to their land that evolved over generations of living in and managing their surrounding ecosystems. To assess these connections in the context of forest gardens, we assemble multiple and overlapping lines of evidence spanning 2,600 years at the settlements of Seklwâtsel (Phillips site) and Yāçketel (John Mack site) in the Harrison Watershed of southwestern British Columbia (Figure 1). Our investigations were designed to explore the relationship between (1) Sts’ailes settlement histories, (2) people’s historical tending and harvesting of forest gardens, and (3) contemporary forest garden sites that continue to be used or valued and are central to Sts’ailes heritage and land-use management praxes.
Figure 1. The study area: (A) Location of study area in southwest British Columbia, Canada; (B) Chehalis IR 5 and surrounding area; (C) the study area showing the location of pithouses and plank houses from the two archaeological village sites, Seklwâtsel and Yāçketel. Map created by Morgan Ritchie.
We examine these Sts’ailes forest gardens as living sites, using a range of archaeological, sedimentary, and physiographic data, along with the memories and knowledge of Sts’ailes Elders. Together, these lines of evidence provide critical narratives from which we can infer the role of people in creating and tending occupied landscapes. Alongside Sts’ailes, our mutual vision has been to redefine how we and others perceive, investigate, and interact with living sites. Rather than limiting access to this living site, Sts’ailes has been “opening it,” seeking to revitalize the legacy forest gardens, generating greater awareness of their settlement history, and promoting the intergenerational transfer of traditional ecological knowledge.
Materials and Methods
Study Region
Sts’ailes is a Halq’eméylem-speaking Coast Salish Nation, whose Xa’xa Temexw (sacred earth, traditional territory) encompasses the Harrison Watershed as well as parts of the lower Lillooet River and Fraser Valley (Figure 1). Sts’ailes occupies a unique gateway position between two culturally and ecologically distinct regions: the coast and interior (Ritchie and Lepofsky Reference Ritchie and Lepofsky2020). Archaeological and oral historical evidence of the deep and intimate connection between ancestral Sts’ailes and their territory includes settlements, camps, resource harvesting areas, named places, transformer sites, and pictographs (Ritchie and Hatoum Reference Ritchie and Hatoum2020; Ritchie and Lepofsky Reference Ritchie and Lepofsky2020; Ritchie et al. Reference Ritchie, Ritchie, Blake, Simons and Lepofsky2024). The social heart of Sts’ailes is the confluence of the Harrison and Chehalis Rivers, where ancestors established and lived in 15 settlements over the last 3,000 years. Today, due to recent and ongoing settler colonialism, Sts’ailes people mainly reside in a single on-reserve settlement (Chehalis IR5) at this river confluence.
The settlements at the Harrison-Chehalis confluence were composed of both cedar plank houses along the banks of the Harrison and underground pithouses set back along aquifer-fed slough channels. Between and beyond the houses, Sts’ailes ancestors maintained small plants and trees that offered food and medicines, and they harvested the larger conifers for posts, planks, canoes, and fuel. The intricate channel networks served as social and physical divisions between settlements, gentle berths for canoes, and as places to collect salmon and other riverine resources. Many of these slough channels were modified and reformed by the past inhabitants to maintain and improve salmon habitat (Ritchie and Lepofsky Reference Ritchie and Lepofsky2020). Additional landscape modifications include the creation of fire-cracked rock terraces on which settlements were built (Ritchie and Lepofsky Reference Ritchie and Lepofsky2020), resource processing areas (Lyons and Ritchie Reference Lyons and Ritchie2017), and forest gardening (Armstrong et al. Reference Armstrong, Miller, McAlvay, Ritchie and Lepofsky2021).
Our roughly 9 ha study area is in the midst of the Harrison-Chehalis confluence and bounded by water on three sides: Phillips Slough to the southwest, Meth’á:lméxwem (John Mack Slough) on the northeast, and the Harrison River on the southern edge (Figure 1, 2A–B). The northern boundary of the study area follows a walking trail and a secondary channel of Meth’á:lméxwem. Within the study area are two ancestral settlements that are separated by approximately 100 m of marshy, lightly forested area: the Phillips site (DhRl-77), or Seklwâtsel, and the John Mack site (DhRl-83), or Yāçketel (Figure 1, 2A–B; Ritchie et al. Reference Ritchie, Ritchie, Blake, Simons and Lepofsky2024). Together, these settlements encompass 21 pithouses and dozens of plank houses dating from approximately 2600 cal BP to AD 1850. After this time, residence in the study area persisted seasonally for fishing and ceremonies, given that Sts’ailes families built houses farther away from the river. We use a variety of historical-ecological methods to document the ongoing legacies of Sts’ailes use and occupation of this place over multiple millennia (Table 1).
Figure 2. Select photos of study area: (A) aerial view of study area highlighting Seklwâtsel center frame (photo courtesy of Robert Dash); (B) aerial view of study area with Seklwâtsel (left); (C) seasonal flooding of cleared Pacific crabapple orchard near Seklwâtsel; (D) recently cleared Pacific crabapple orchard as a result of Sts’ailes’ ecological restoration efforts; (E) current state of cleared and managed crabapple orchard (November 2024); (F) Simon Fraser University students gathering along Harrison River in study area; (G) Sts’ailes community members participating in archaeological field school in study area. (Color online)
Table 1. Methods, Results, and Inferences Used to Document the Biocultural Heritage of the Study Area.
Forest Structure, Age, and Composition
Initial surveys of Seklwâtsel and Yāçketel indicated that Pacific crabapple (Malus fusca) and beaked hazelnut (Corylus cornuta) were the dominant species around both ancestral village sites (Armstrong et al. Reference Armstrong, Miller, McAlvay, Ritchie and Lepofsky2021). Beaked hazelnut is a relatively large, multistemmed deciduous shrub that often grows on moist but well-drained sites with open forest canopies (Klinkenberg Reference Klinkenberg2020a; Pojar and Mackinnon Reference Pojar and Mackinnon2004). Charred beaked hazelnut shells have also been recovered from archaeological contexts on the Harrison River, including from a feature in the study area (Lyons and Ritchie Reference Lyons and Ritchie2017). Pacific crabapple is a small, deciduous, shade-intolerant tree native to the Northwest Coast. It prefers moist to wet open environments such as stream banks, swamps, and bogs (Klinkenberg Reference Klinkenberg2020b; Pojar and Mackinnon Reference Pojar and Mackinnon2004). In general, the study area has experienced minimal recent anthropogenic disturbance (e.g., logging, urban development) and therefore is an ideal location with which to trace long-term cultural ecological histories. Places that have been recently disturbed are easily identifiable by encroaching invasives and other seral species, often creating areas of impassable bramble. We avoided these areas in our plant surveys but took them into consideration when interpreting the current distribution of Pacific crabapple and beaked hazelnut trees.
To better account for the spatial distribution of Pacific crabapple and beaked hazelnut, we used high-precision GPS to map and enumerate all mature living and dead Pacific crabapple and beaked hazelnut trees in the entire study area. We focus on these two forest-garden indicator species (Armstrong et al. Reference Armstrong, Miller, McAlvay, Ritchie and Lepofsky2021, Reference Armstrong, Earnshaw and McAlvay2022, Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023) because they are easily identifiable and abundant in the study area. Furthermore, mapping and restoring the productivity of the Pacific crabapple trees along the Harrison River is of particular importance for Sts’ailes due to their importance in the past as food, medicine, and technology (Table 2; Wyllie de Echeverria Reference Wyllie de Echeverria2013). As larger life forms, these species also endure longer on the landscape than other forest-garden indicator species (e.g., Vaccinium spp., Rubus spp.) and provide greater opportunities to detect persistent ecological legacies that are part of the region’s landscape history.
To provide a more complete picture of the forest garden landscape, we walked the study site to record the presence of culturally important plant species. We avoided surveying edge areas, defined as a 20 m buffer around modern disturbances (such as trails and roads) and natural features (such as the edge of the Harrison River). Similarly, we avoided places with dense clusters of invasive species, such Himalayan blackberry (Rubus armeniacus), or what we presumed to be recently disturbed areas of virtually impassable thickets of blackberry species, Nootka rose (Rosa nutkana), black twinberry (Lonicera involucrata), and hardhack (Spiraea douglasii). The composition of these thickets was noted to the best of our ability by our surveying around the edges and gaining higher vantage points to view inside the thickets.
To assess recent changes in forest structure and composition, we linked changes visible in historic air photos from 1936 to 2003 with knowledge provided by six Elders and knowledge holders associated with the Phillips family. The interviews focused on participants’ lived experiences on the Harrison River, with a focus on the study area. We asked questions about the history of land-use activities as well as the location of remnant features such as homesteads, smokehouses, roads, and trails. Similarly, the historic air-photo time series (BC Provincial Air Photos 1936, 1949, 1963, 1968, 1993) provided insight into changing land-use patterns and forest structure in the recent past. In addition, to assess the effects of other local environmental conditions on the distribution of Pacific crabapple and beaked hazelnut, we examined the distribution of these species in relation to groundwater.
We counted tree rings from cores and transverse rounds (or cookies) to determine the age of the tallest trees in the study area to assess whether they are recent encroachments into the forest gardens. Around the two ancestral village sites, in the southern portion of the study area, where there are few coniferous trees (<20), all the conifer trees were mapped and cored using an increment borer. However, some cores were unsuccessful because of internal rot. Additionally, we selectively cored large deciduous species such as red alder (Alnus rubra), paper birch (Betula papyrifera), black cottonwood (Populus trichocarpa), and aspen (Populus tremuloides) with an increment borer. We also sampled Pacific crabapple and beaked hazelnut via a combination of tree cores and cookies. In the northern part of the study area, where the canopy is dominated by conifer trees, we judgmentally selected and cored trees to compare their ages with those found closer to the river and the ancient settlements. We did not attempt age estimates via coring for beaked hazelnut because they reproduce clonally, which makes it difficult to obtain establishment dates via tree ring analysis (see Armstrong et al. Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023).
Tree cores were mounted and sanded using increasing grit sizes (Stokes and Smiley Reference Stokes and Smiley1968). The transverse cookies were sanded with 120 to 600 grit paper. For all samples, the Epson Expression 10000XL scanner scanned cores at 2400 dpi. Images were then imported to CooRecorder to identify ring boundaries (Maxwell and Larsson Reference Maxwell and Larsson2021). Samples were collected in 2020 near the end of the growing season, so latewood formation was present in both hardwoods and softwoods. For samples in which the pith was present, measurements were taken from bark to pith, with the last year of growth identified as 2020. For samples missing the pith, but still displayed concentric rings at the end of the core, we used the pith estimation feature in CooRecorder. A Velmex stage system and MeasureJ2X software were used for cookies and cores that required higher resolution (https://www.voortech.com/projectj2x/docs/V321/install.htm).
Archaeological Surveys and Analysis
Soil charcoal was analyzed as a proxy for historical forest composition (species presence) and fire history. Charcoal was extracted from judgmentally placed soil pits (∼30 × 30 cm) dug in areas with—and without—Pacific crabapple and beaked hazelnut as well as areas close to (minimum ∼10 m) and distant from (maximum ∼200 m) the ancient settlements. Pits were dug down to basal alluvial fan deposits. To maximize the amount of charcoal for identification, we collected bulk sediment samples (~0.5 L) from each stratigraphic layer, as well as any additional visible charcoal from the pit walls. These bulk sediment samples were then dried and charcoal was sorted from them. We described all stratigraphic layers to determine their depositional history (e.g., whether they were deposited during flood events or created through gradual soil pedogenesis).
From these pits, we extracted charcoal from both in situ cultural layers and surrounding silt matrices that would have been influenced by low-action water movement and mixing from tree roots. The in situ charcoal likely reflects cultural preference and availability of fuel woods. The charcoal from silt matrices was likely generated by forest burning (deliberate or accidental) or similar cultural activities. In all instances, we reason that the charcoal collected from each test pit represents a relatively local picture of forest composition. We recognize that many shrubs are likely to burn to ash and therefore will be underrepresented in our identifications. Finally, we infer that the depth of the charcoal corresponds to age, allowing us to analyze changes to forest composition through time. The validity of this proxy measure is supported by radiocarbon dates taken from a range of contexts from across the study area.
Following standard procedures for wood identification (Friedman Reference Friedman1978; Hoadley Reference Hoadley1990), we identified all charcoal specimens ≥2 mm from each stratigraphic layer within the soil pits. We chose this cut off because charcoal fragments smaller than this are difficult to sort or identify to species. For each charcoal fragment, we exposed clean surfaces (e.g., by snapping or using a razor blade) of at least two of the three identifiable sections (i.e., transverse, tangential, radial). We used a metallurgical microscope (max. magnification 80×) to ascertain any identifiable characteristics. For identifications, we first determined if the charcoal was deciduous or coniferous. If coniferous, we limited our identifications to taxa whose unique microscopic characteristics are relatively easy to identify (i.e., Taxus, Pseudotsuga, Picea, Pinus). All others were lumped into an “unidentified coniferous charcoal” category. If charcoal was deciduous, identification was attempted to the most specific taxonomic order. This basic set of identifications allowed determination of the relative percent abundance of deciduous versus coniferous woods across soil pits. Finally, we selectively radiocarbon-dated older (deeper) charcoal that we identified as Pacific crabapple that either came from layers whose charcoal suggested a transition from coniferous to deciduous trees, or was recovered away from the ancestral village sites where we had little prior knowledge of cultural use.
To explore the influence of people on the distribution of Pacific crabapple and beaked hazelnut, specifically, we examined whether these forest-garden indicator species (Armstrong et al. Reference Armstrong, Miller, McAlvay, Ritchie and Lepofsky2021) grew in identifiable patterns relative to the ancient settlements. Our expectation was that areas closer to settlements would be relatively more tended, which would in turn be reflected in a relatively greater abundance of cultural species associated with these settlements. The association of culturally important plants with settlement sites is well documented in many places around the world (e.g., Armstrong et al. Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023; Dambrine et al. Reference Dambrine, Dupouey, Laüt, Humbert, Thinon, Beaufils and Richard2007; Ross and Rangel Reference Ross and Rangel2011).
Results
Forest Structure, Age, and Composition
Both Pacific crabapple and beaked hazelnut are abundant in the study area and tend to be found in nonoverlapping clusters associated with ancient and modern settlements (Figure 3). Pacific crabapple (n = 320) are most densely distributed in the southwestern and southeastern corners of the study area alongside Yāçketel and Seklwâtsel. Beaked hazelnuts (n = 74) are most common along the northern end of the study area near the former Phillips family homestead. Conversely, away from the ancient and historic settlements, there are few Pacific crabapple or beaked hazelnut trees. This strong positive association between Pacific crabapple, beaked hazelnut, and ancient and modern settlements is typical of Northwest Coast forest gardens (Armstrong et al. Reference Armstrong, Miller, McAlvay, Ritchie and Lepofsky2021, Reference Armstrong, Earnshaw and McAlvay2022, Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023). Similarly, we note that the rare culturally important species in the study area (i.e., those represented by only one or two individuals) are also uniquely associated with the pithouses and modern homes (Figure 3). This suggests that people may have taken advantage of or had a role in the distribution of these plants. The sparse distribution of other culturally valued plants prevented us from discerning whether they tend to occur in association with Pacific crabapples and beaked hazelnuts or with settlements.
Figure 3. Study area showing distribution of Pacific crabapple, beaked hazelnut, rare species, species represented by a single stand of trees (e.g., European cherry), and pithouses. (Color online)
We recorded 39 plant species in the vegetation plots and judgmental surveys, the majority of which are shrubs (Table 2). Of this total, nine are forest-garden indicator species (Armstrong et al. Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023). In fact, only three of the 12 forest-garden indicator species identified in a coast-wide study of forest gardens is not represented in the study area (soapberry [Shepherdia canadensis], Saskatoon berry/serviceberry [Amelanchier alnifolia], and oval-leaf blueberry [Vaccinium ovalifolium]). A large percent of these 39 species (97%) have known uses for food, medicine, technology, and other applications among Northwest Coast First Nations (Table 2), and several of them are species that are known to have been transplanted in the past (Turner et al. Reference Turner, Geralda Armstrong and Lepofsky2021). Based on our judgmental survey of the entire study area, six of the 39 species are uncommon within the study area (Figure 3), being represented by only one or two plants (lodgepole pine [Pinus contorta], blue elderberry [Sambucus cerulea], black swamp gooseberry [Ribes lacustre], oceanspray [Holodiscus discolor], and stinging nettle [Urtica dioica])—or in the case of European cherry (Prunus avium), a single stand of trees. Culturally important shrubs and herbs are rare, and they are much less common than is typical of other anthropogenic (forest) gardens in Sts’ailes territory and elsewhere on the Northwest Coast (Armstrong et al. Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023).
Table 2. Species Growing in the Study Area Today or in the Recent Past.
a Forest garden indicator species as per Armstrong et al. (Reference Armstrong, Miller, McAlvay, Ritchie and Lepofsky2021).
b Abbreviations: VP = Vegetation plots; IS = Informal survey; IN = Interviews; AR = Archaeobotanical remains, which includes charcoal remains identified as part of this project and reported in Lyons and Ritchie (Reference Lyons and Morgan Ritchie2022).
c Per Turner (Reference Turner1995); Turner et al. (Reference Turner, Deur and Lepofsky2013, Reference Turner, Geralda Armstrong and Lepofsky2021); and interviews from this study. Abbreviations:; F = Food; M = Medicine; T = Technology; S = Spiritual/ceremonial; RT = Known regional transplant.
d Introduced/invasive species.
Archaeological Surveys and Analysis
Evidence from sediment profiles and radiocarbon dates within the study area demonstrate that the Chehalis River alluvial fan was a dynamic fluvial environment, until around 2600 cal BP, when the landform stabilized, and forests began to establish (Figure 4; Supplementary Table 1). Layers with charcoal and fire-altered rock immediately overlying thick deposits of coarse sands, pebbles, and cobbles reveal that this newly stabilized landscape was rapidly inhabited by Sts’ailes ancestors. Fine silts deposited in the last 2,600 years from overbank flooding of the Harrison River are interleaved with anthropogenic materials, marking a significant transformation in the environment (Supplementary Table 2).
Figure 4. Radiocarbon dates from the study area. Dates from plank houses are black; pithouses are dark gray; charcoal layer is light gray; cooking pit is white; dates from Meth’á:lméxwem are depicted with a star. Dotted lines represent continuity.
The uneven distribution of soil charcoal highlights the concentration of human activity and use along the sloughs and riverbank rather than the inland sections of the study area (Figure 5). Evidence of fires comes from charcoal and fire-altered rock found in shovel tests of cultural features and sediments near the Phillips Slough (2736–2492 cal BP) and Meth’á:lméxwem (2499–2348 cal BP; STs 9, 10, 18, and 21) and from small amounts of more diffuse charcoal scattered unevenly in the soil near the Harrison River plank houses (1517–1354 cal BP; ST 2 and ST 3). In comparison, shovel tests from farther inland revealed fewer and smaller fragments of charcoal, and fewer fragments of fire-cracked rock. Our analysis of charcoal from the cultural features suggests that people had ready access to both conifer and deciduous forests, as would be expected of high-disturbance areas (i.e., both from the floodplain and cultural activities). In contrast, charcoal fragments from soils away from ancestral houses are predominantly coniferous charcoal, as would be expected of areas that are less influenced by flooding and anthropogenic disturbances.
Figure 5. The distribution and abundance of conifer, deciduous, and crabapple charcoal through time in relation to the modern distribution of Pacific crabapple and beaked hazelnut. Note the location of the charred hazelnut shell fragment found by Lyons and Ritchie (Reference Lyons and Morgan Ritchie2022) along the riverbank. Charcoal recovered from ST 17, 60–80 cm (n = 3) is not represented in this figure because the field notes, sediment profiles, and radiocarbon dates demonstrate that the charcoal is intrusive from the topmost layers. See Supplementary Table 4 for complete list of charcoal identifications. (Color online)
Our identification of Pacific crabapple charcoal from cultural features suggests that this culturally important tree has been growing in the study area for at least 2,600 years, and perhaps earlier as a seral colonizing species (Figure 5; Supplementary Table 3). Notably, the two oldest pieces of Pacific crabapple charcoal were recovered from areas close to the ancient pithouse settlements and the sloughs (ST 10, ST 20), where Pacific crabapple still grows today. The antiquity and continuity of this Pacific crabapple patch is reflected in the 130-year-old Pacific crabapple tree currently growing near Yāçketel (Figures 6 and 7 it is probably from the same cohort of Pacific crabapple as those represented by Pacific crabapple charcoal in the upper portion of ST10 ne charcoal.
Figure 6. Results of tree-ring analysis showing the Initial encroachment of deciduous trees, followed by coniferous trees, demonstrating shifting forest structure. Note that Pacific crabapple grows throughout the sequence. See Supplementary Table 2 for more detailed information of species and tree ages. (Color online)
Figure 7. Collage of historical air photos showing the Phillips homestead, historic smokehouses and modern homes, changes in forest cover, and the distribution of modern Pacific crabapple and beaked hazelnut. (Color online)
Building on the charcoal identification results, tree-ring analysis allowed us to reconstruct the progression of tree encroachment over the last century from deciduous trees to shade-tolerant conifer trees (Figure 6; Supplementary Table 4). By at least 110 years ago, the forest around the ancient settlements was a mixed deciduous forest of cottonwood, alder, and Pacific crabapple, followed by birch and aspen that became established within the last 80 years. The establishment of these—and other early successional species, such as red alder—after our oldest-dated Pacific crabapple suggests a shift in forest conditions approximately 110 years ago (ca. AD 1900). Although the absence in our sample of older deciduous trees other than Pacific crabapple may in part reflect the maximum life span of most deciduous trees, we note that cottonwoods can live for considerably more than 100 years and therefore should be present in our sample (Thomas and Podmore Reference Thomas and Podmore1953). Tree-ring analysis also indicates that recent encroachment of coniferous trees (e.g., Sitka spruce [Picea sitchensis], western redcedar [Thuja plicata], lodgepole pine) around the ancient village sites began less than a century ago, creating the mixed canopy forest that exists today.
This recent encroachment is visible in the historic air photos (Figure 7). In 1936, tree cover along the riverbank and the northwest corner of the study area was sparse. Through time, the forest has become visibly denser, especially along the edge of the river. Four areas of increased conifer encroachment can be seen via the historic air-photo series: the clearing around the Phillips’ homestead in the north, the two clearings associated with the smokehouses at each slough, and a major disturbance visible in the center of the study area, which is visible in the 1968 photo (Figure 7). Today, only the areas along the river remain free of trees and shrubs.
Interviews with Sts’ailes Elders and knowledge holders suggest that, in general, forest composition in the mid-twentieth century was similar to that of today. They recall 14 culturally significant native species that were harvested in the study area (Table 2). All are present today except Saskatoon berry. Elders shared that beaked hazelnut used to be plentiful near the river and in the center of the study area, where few continue to grow today. Remembered activities that would have influenced forest structure and composition included clearing for orchards and gardens via tree cutting and controlled fires. Vange Point (personal communication 2020) noted that her grandfather used to keep the area around his house “clear so that there was just grass growing all over.” Controlled fires were also used to maintain her grandfather’s orchards and clearings. He would usually “burn on the other side of their yard and he would burn it towards the river” (Vange Point, personal communication 2020). Interview responses from Phillips family members emphasize the large amount of work that went into maintaining these clearings and tending these food plants, from pruning fruit trees to using scythes to keep the ground clear.
Other memories shared by the Phillips family include accounts of frequent travel on a well-maintained trail that took them from their homestead to the Harrison River and family smokehouse. Access to the Harrison River and the smokehouse was essential for several reasons. Virginia Peters (personal communication 2020) remembered that “our road was the river,” given that there was no road back then to connect Sts’ailes to the outside world. The smokehouse was also a hub of activity and was used seasonally for hosting family, processing salmon and other resources, and serving as a refuge for giving birth or recovering from illness. The frequent visitation and use of the riverbank near the confluence of the Harrison River and Phillips Slough during the mid-twentieth century would have influenced the pattern of conifer growth because the family intentionally kept this area clear of underbrush (Darius Kelly-Lawrence, personal communication 2020). Today, there are no conifers in the area around the Phillips family smokehouse or in the area around the smokehouse near Meth’á:lméxwem. Instead, conifers are more common inland, where people have not actively managed the vegetation over the past century.
Aside from the effects of recent forest clearing, we note that the generally nonoverlapping distribution of crabapple and hazelnut in Yāçketel and Seklwâtsel is correlated with groundwater—that is, crabapple tends to grow on the moderately wet soils near the relatively low-lying areas around the river and sloughs, whereas hazelnut is found on higher, drier ground (Figure 8). Groundwater also contributes to the distribution of conifers in the study area, given that they are unable to become established on the wet, low-lying areas prone to seasonal flooding (Figure 2C).
Figure 8. Distribution of crabapple and hazelnut relative to elevation and exposure to flooding: (A) flood and elevation model showing that crabapples tend to grow around low-elevation slough channels, whereas hazelnuts prefer higher, drier ground. Note the low-lying, relic slough channels near the central riverbank extending northward, where few crabapple and hazelnut grow; (B) violin graph showing that crabapple and hazelnut have unique elevational distributions (p < 0.001). (Elevation map produced by Sts’ailes; Violin graph produced by Gavia Lertzman-Lepofsky.) (Color online)
Although wet soils along the riverbank and sloughs may be limiting hazelnut growth today, these conditions alone were not enough to prevent stands from becoming established in the past. Sts’ailes Elders noted that hazelnut used to be more abundant in the study area in general—including near the river—yet today, only a handful are left along the river and sloughs. We attribute this to the cessation of human management and the subsequent encroachment of coniferous trees leading to an increasingly closed canopy cover, which shaded out both crabapples and hazelnuts. Only the already established crabapples could tolerate the more shaded conditions. In contrast, in the area around the Phillips homestead, which was managed and cleared until 40 years ago, there is a large stand of hazelnuts. This stand became established in the drier, open area created when the fruit trees around the homestead were cut down.
Discussion
We began our investigation exploring the long-term influences of Sts’ailes people on the lived landscapes of Yāçketel and Seklwâtsel, with the understanding that these age-old relationships involved some degree of management of fire, soil, and plants. In other words, we assume that over countless generations, as the Sts’ailes ancestors lived in communities anchored in these places, they developed increasingly intertwined relationships with their natural world. The challenge that remains, however, is how to distinguish the evidence of traditional management practices from those spurred by natural processes—recognizing that the two are not mutually exclusive and that the former is more difficult to detect the farther we go back in time (Fowler and Lepofsky Reference Fowler and Lepofsky2011).
Our exploration of Sts’ailes relationships with Yāçketel and Seklwâtsel landscapes over nearly three millennia is bolstered by diverse sources of data and knowledge (Table 1; Figure 9). In using such an approach, the boundaries of archaeology—with its traditional focus on “material evidence of the past”—blur, as do the boundaries between “natural” and cultural ecosystem processes (Lepofsky et al. Reference Lepofsky, Geralda Armstrong, Mathews and Greening2020). This more expansive approach is better aligned with understanding and more fully respecting Indigenous heritage and connections to the land (Baumflek et al. Reference Baumflek, Cabe, Schelhas and Dunlavey2022). At the start of our historical narrative, our summary is more general, given that it relies on fewer kinds of data and less direct inferences. Later in time (nineteenth century onward), our summary is more robust, especially because we can augment our narrative with family stories, historic air photos, tree-ring ages, and vegetation survey data. These more recent data allow us to compile historical-ecological narratives at more precise decadal scales, whereas deeper-time data demand wider temporal ranges, focusing on centennial and millennial intervals. Our reconstructions also highlight the difficulties of pointing to distinctly human-related ecosystem processes and instead reflect how inextricably linked cultural histories—including traditional management systems—are with ecological histories.
Figure 9. Timeline of events in the study area and supporting evidence. The break in the timeline reflects the shift in our understanding of the historical ecology from the millennial to decadal time scales. AP = air photos; AR = archaeology; H = historical; IN = interviews; TR = tree-ring analysis.
2600 BP–1782 AD
Our understanding of lives lived on the Jack Mack and Phillips slough systems, and in the settlements of Yāçketel and Seklwâtsel, begins approximately 2,600 years ago, when the alluvial fan stabilized. This allowed for both human settlement and preservation of the paleoecological record associated with the settlement. Charcoal recovered from scattered cooking features near the Phillips Slough and Meth’á:lméxwem reflects a mix of predominantly deciduous species and some coniferous species, typical of a floodplain forest. The presence of Pacific crabapple charcoal in our samples indicates that Pacific crabapple was already established enough for its branches to be collected for fuel. At this time, people started living near the Phillips Slough in a small settlement consisting of two or more plank houses (Figure 4; Ritchie et al. Reference Ritchie, Ritchie, Blake, Simons and Lepofsky2024). Notably, the focus of human activity around the sloughs illustrates the onset of a settlement pattern that would be strengthened in the following millennia: an orientation toward the Harrison River. Our earliest insights into riparian forest history comes from soil charcoal, which indicates that earlier fires were concentrated along the waterways in the study area. Given that this is a wet and active floodplain where fires would not typically flourish, we infer that these fires are mainly associated with human activity in the ancient settlements.
Settlements began to expand in size some 800 years after the first plank houses were established in the study area. Around 1800 cal BP, plank houses were constructed near the mouth of Meth’á:lméxwem (Yāçketel), and soon after, at roughly 1600 cal BP, people built plank houses along the mouth of the Phillips Slough (Seklwâtsel). Over the next several hundred years, plank houses continued to be built, expanding along the river’s edge. The process of building plank houses involved recontouring the lower margin of the alluvial fan and creating raised level construction surfaces (Ritchie et al. Reference Ritchie, Ritchie, Blake, Simons and Lepofsky2024). It also involved cutting down and planking dozens of large cedar trees for each house. Diffuse charcoal collected from sediment layers near the plank houses from this time show that people were burning nearby forests, likely to make way for the expansion of the plank-house settlements. We hypothesize that the diffuse charcoal recovered from the soil pits near the plank houses represent either intentional or unintentional forest burning events associated with the daily life of the settlement and management of the forest. Interestingly, Pacific crabapple charcoal is only found in cultural burn features, not in these diffuse charcoal layers; forest burns were not directed at this species.
Over time, the settlements in the study area continued to grow. By approximately 850 years ago, we see charred red elderberry seeds and what has been tentatively identified as Pacific crabapple charcoal in a cooking feature just outside of the study area (across the Phillips Slough; Lyons and Ritchie Reference Lyons and Morgan Ritchie2022), which reflects the ongoing use of culturally important plants in this area. A slightly more recent cooking feature dating to around 650 cal BP with a charred beaked-hazelnut shell provides evidence that beaked hazelnut may have been growing in the study area and that people were harvesting them (Lyons and Ritchie Reference Lyons and Morgan Ritchie2022). These remains, though sparse, when taken with the persistence of Pacific crabapple charcoal in the sediments through time (ST 10, ST 20), suggests that Pacific crabapple and beaked hazelnut may have co-occurred in the forest at least 650 years ago along with other culturally important species, such as elderberry.
During the Little Ice Age (ca. 800–250 cal BP), we observe considerable changes to the settlements, in part influenced by cooler temperatures (Pitman and Smith Reference Pitman and Smith2012; Ritchie et al. Reference Ritchie, Ritchie, Blake, Simons and Lepofsky2024). Starting around 550 cal BP, inhabitants constructed and lived in semisubterranean pithouses located behind the plank houses. Seed and charcoal samples collected from pithouse floors in the study area dating to approximately 400 cal BP demonstrate that people processed and presumably ate red elderberry and salal (Gaultheria shallon), and that they used spruce, hemlock, western redcedar, alder, birch, and willow for fuel (Lyons and Ritchie Reference Lyons and Morgan Ritchie2022). Today, all these species continue to grow in the study area, with the exception of salal. That the salal and elderberry were bearing harvestable fruits suggests that they were growing in relatively open forest—a forest structure consistent with our expectations around the two settlements. This is supported by paleoethnobotanical evidence from a cooking feature along the riverbank where the charred remains of grasses, sedges, and western hemlock were identified (ca. 651–540 cal BP; Lyons and Ritchie Reference Lyons and Morgan Ritchie2022).
Taken together, these isolated but overlapping data indicate continuity in settlement and land use over several millennia beginning around 2,600 years ago. Burning and harvesting of plants increased after around 1,800 years with the construction of new plank houses along the river’s edge. Similarly, clearing and plant management was necessarily expanded with the establishment of pithouses farther away from the river around 550 years ago.
AD 1782–1950s
From the late eighteenth century onward, as the resolution of our data increases, we witness dramatic shifts in the use of the landscape around Yāçketel and Seklwâtsel. In 1782, the first smallpox epidemic devasted First Nations communities throughout North America. At this time, the population of Sts’ailes plummeted, resulting in dramatic social and ecological effects. By 1800, among Coast Salish groups more broadly, entire families and more than three-quarters of the population were lost to disease (Boyd Reference Boyd1999:135, 137–138). At this time, Sts’ailes people left their villages in the Harrison Watershed to congregate at a single settlement downriver of the study area (Ritchie et al. Reference Ritchie, Ritchie, Blake, Simons and Lepofsky2024).
Over the next hundred years, the Sts’ailes world continued to change under various colonial processes that in turn influenced their connections to Yāçketel and Seklwâtsel. Sts’ailes families moved back and began living permanently in and around Yāçketel and Seklwâtsel after the official boundaries for the reservation were established in 1870 (Sproat Reference Sproat1879:38). By 1894, with as few as 124 Sts’ailes people surviving (Ritchie et al. Reference Ritchie, Ritchie, Blake, Simons and Lepofsky2024), use of the areas around Yāçketel and Seklwâtsel was greatly diminished, especially when compared to the once thriving, densely settled villages of the previous two millennia.
This time corresponds to some of our strongest evidence of Sts’ailes stewardship of the landscape. Tree-ring data suggest that conifer trees began to encroach into the southern portion of the study area around 90 years ago. We surmise that this encroachment is due to the decline in human presence and, consequently, the decline in both active (burning and clearing) and passive (weeding and trampling) management of the forest. Declining diversity of culturally important species such as Saskatoon berry and stinging nettle in the study area within the last century likely also reflects the decline in human presence. Elders remembered that beaked hazelnut used to be more abundant, especially near the river, but today, they are predominantly found in the drier portions of the study area, farther from the river. It may be that on its own, without human management and presence, hazelnut could not flourish in these wetter portions of the study area; in the past, this culturally important tree likely benefited from both active tending and the open conditions along the river and sloughs associated with past human use and settlement. Pacific crabapple growing in and around the ancient settlements today are likely legacies of the pre-encroachment forest. However, many of the Pacific crabapples are in poor condition and are not fruiting. The persistence of culturally important plants on the landscape despite over a century of human activity is a trait indicative of other forest gardens in British Columbia (Armstrong et al. Reference Armstrong, Miller, McAlvay, Ritchie and Lepofsky2021, Reference Armstrong, Earnshaw and McAlvay2022, Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023; Fisher et al. Reference Fisher, Shackelford, Hocking, Trant and Starzomski2019; Trant et al. Reference Trant, Nijland, Hoffman, Mathews, McLaren, Nelson and Starzomski2016).
Despite the dramatic social and ecological changes of the eighteenth and nineteenth centuries, Sts’ailes people continued to manage and use the vegetation within the study area until the 1950s. Importantly, the very naming of the slough systems themselves—with the John Mack and Phillips family names—reflects the ongoing connection of Sts’ailes people to this landscape. The waterways were still a central focus of their daily life and were accessed via well-maintained trails. At the mouth of the slough carrying their name, the Phillips and Mack family smokehouses were a nexus for land-based family activities. Away from the river and sloughs, as a part of their homestead, the Phillips family cleared large swathes of the forest using controlled burns and other tools to create orchards and gardens composed of not only native plants but also newly introduced European cultivars (e.g., plums, cherries). However, even after the establishment of the homestead and its associated fruit orchards and vegetable gardens, the Phillips family continued to harvest native species from the study area for food, medicine, textile, and other needs. This includes important forest-garden indicator species, including Pacific crabapple and beaked hazelnut, as well as rosehips, salmonberry, blackberry, thimbleberry, stinging nettle, western redcedar, red alder, and cascara. As Boyd Peters (personal communication 2020) remembered, his grandparents referred to Phillips Slough as their grocery store.
Increasing colonial disruptions, including the mid-century peak of forced attendance to residential schools and the decimation of salmon population, led to a sharp decrease in human activity in the study area from the 1950s onward, severing many peoples’ connection to this place. This disconnection began with the cessation of the Phillips and John Mack smokehouses, both of which disappeared from the landscape by the 1950s. In the late 1960s, the area was commercially logged for cottonwood trees (Gerald Phillips, personal communication 2021), illustrating a significant shift in how Sts’ailes people interacted with the forest. By the 1970s, the Phillips homestead was replaced by modern homes that were oriented to the road. The orchards and clearings were left to grow over. Thickets of wild rose, blackberry, and spiraea became established in previously disturbed and open spaces near the river and sloughs, making access to the area difficult, especially as Elders aged. In turn, this decreased access to the waterways and meant that people spent less time harvesting traditional plants from the study area.
Yāçketel and Seklwâtsel Today
Our compilation of diverse types of knowledge and data speak to the continuity and transformation of the relationship between Sts’ailes people and their cultural landscapes of Yāçketel and Seklwâtsel. This long-term and intimate relationship is represented today by the combined snippets left in the paleoecological and ecological records, in archaeological features and deposits, and in peoples’ memories. The ecological legacies of Pacific crabapple and other culturally important species growing around the villages today reflect this deep time relationship. These legacies are also reflected in Sts’ailes people’s more recent practices of keeping important plants near at hand in orchards and forest gardens while simultaneously harvesting and tending “wild” foods from periphery forests and landscapes.
Today, the relationship between Sts’ailes people to this cultural landscape is being actively upheld and reawakened through efforts of Sts’ailes community members to celebrate, conserve, and revitalize these living sites of heritage. Sts’ailes maintains easy walking and boating access and open spaces around the ancestral houses and forest gardens so that they can be used by community members and visiting relatives for cultural tours, school excursions, healing, and harvesting for food and ceremony (Figure 2D–G). By maintaining clearings with brushing, pruning, and burning—the same practices that would have promoted food production a thousand years ago—Sts’ailes is prioritizing the resilience of culturally important species, places, and traditions, as well as promoting reconciliation by sharing their living heritage with others.
Broadening the scope of archaeological research and management to include living sites that continue to grow, bear fruit, and connect people through activities such as pruning or harvesting confers meaning and value that might not otherwise allow for intergenerational transfer of both cultural and ecological knowledge. The cultural importance of the places where we worked and that we helped to clear is enhanced by the ability of community members of all ages to visit again and reconnect with those who came before. The clearing, tending, and use of the fruit and nut trees is seen as a sign of respect to the ancestors. Elders who were born in the study area and who frequented it during their childhoods have a strong desire to pass on cultural teachings to younger generations through hands-on experiences such as harvesting, which reinvoke integrated systems of knowledge and practice. This transfer of knowledge is crucial to maintaining Sts’ailes’ cultural identity because “these things should not be forgotten. [They are] a part of who we are and what we belong to” (Patricia Charlie, personal communication 2021). Given this imperative, it is especially significant that Sts’ailes children are actively learning about this place and their ancestors with their parents and grandparents while they assist with ongoing archaeological investigations at Yāçketel and Seklwâtsel. (Figure 2F–G). By advocating for a living heritage approach, we are embracing the second of Poulios’s (Reference Poulios2014) concepts whereby the Sts’ailes community is deciding this place’s value and dictating its use.
Although this research was not focused on highlighting Sts’ailes views and definitions of heritage, during interviews and discussions with the Elders and knowledge holders, it was clear that conventional archaeological definitions of heritage did not fully describe the relationships that people have with the objects, houses, and places that were tended and used by their ancestors. Elders often felt spirits particularly around the plank houses and pithouses. During fieldwork, Elders said prayers and offered explanations of our activities to inform the ancestors and to provide spiritual protection. People often referred to “Letsemot”—”everything is connected”—which expresses the relationship between the living the and dead, the physical and spiritual realms, and all things. Although “heritage” is not a term that is used often by Elders or knowledge holders, there is the definite understanding that people have been entrusted by those who came before, and that it is their responsibility to look after everything for future generations. This responsibility relates to clean water and good soil, as well as to ensuring that teachings from the ancestors are passed on. Important places such as Yāçketel and Seklwâtsel are intended to be used, just as they have been for millennia. To stop using the place is to lose connection with it. By recognizing how Letsemot is woven into Sts’ailes ways of being and thinking, this study advocates for more respectful ways to “do” archaeology—both in research and in cultural resource management (CRM) practices—while supporting larger policies such as the Declaration of the Rights of Indigenous Peoples Act.
The Future
At Yāçketel and Seklwâtsel, as in other historical-ecological and culturally significant settlements in the region (Armstrong et al. Reference Armstrong, Miller, McAlvay, Ritchie and Lepofsky2021, Reference Armstrong, Earnshaw and McAlvay2022, Reference Armstrong, Lyons, McAlvay, Ritchie, Lepofsky and Blake2023), there are profound connections among past and present environments, human actions, and ancient and current settlements and communities. Such linkages speak to the inseparability of environmental and archaeological heritage and the efficacy of reimagining archaeological sites and their surroundings as living sites or living heritage. This holistic perspective that connects the past, present, and future is in line with the Sts’ailes worldview, which sees the interconnectedness of human generations, of all living things, and the physical and spiritual realms. The recognition of such sites as living heritage, and an Indigenous community’s central role in creating and maintaining them, also initiates opportunities to enrich archaeological interpretations, engage broader publics, and contribute to powerful conservation goals. Unfettered by either provincial or federal heritage legislation, Sts’ailes people will continually redefine their use of this living heritage landscape, foregrounding principles of reciprocity, respect, and generosity that guide all interactions with both each another and the natural and supernatural worlds.
Ultimately, the recognition and protection of living sites such as the ecosystems surrounding Yāçketel and Seklwâtsel require a shift in how archaeologists perceive material culture, landscapes, and continuing Indigenous connections to them. In British Columbia, as in many places, current heritage legislation emphasizes protection of archaeological remains such as artifacts/belongings and features to the exclusion of other kinds of heritage such as unique plant species, stewarded forests, and long-managed berry crops (Kelly et al. Reference Kelly, Jesse, Lia, Candace, Raini, Sheriden, Sean and Lepofsky2024; Lepofsky et al. Reference Lepofsky, Geralda Armstrong, Mathews and Greening2020). Equally problematic is that the legislated “protection” of cultural heritage may still preclude Indigenous people’s access or engagement if within private property or environmentally restricted areas. Incorporating a more holistic understanding of landscapes that includes the ongoing connection of Indigenous peoples to their biophysical world is key to recognizing and honoring the full breadth of their long-term connections to place, territorial rights, and control and ownership of cultural inheritances. Thinking of these forest gardens as living sites of heritage may provide an avenue for moving away from rote and compliance-driven archaeological practices and toward stewardship practices that more closely align with some Indigenous communities’ own epistemologies, interests, and futures.
Acknowledgments
This project was conducted with the full support of and in collaboration with Sts’ailes, and we would like to express our deep gratitude to the community for allowing us to work, study, and learn on their unceded traditional territory. We are especially grateful to the Elders and knowledge holders who shared their memories and experiences with us: Boyd Peters, Virginia (Ginny) Peters, Patricia Charlie, Vange Point, Gerald Phillips, and Willie Charlie. We would like to acknowledge those who assisted with fieldwork for this project, including Darius Kelly-Lawrence, Emily Purcell, Kiran Basran, Jerram Ritchie, and Dr. Ken Lertzman. We are thankful to Kiran Basran and Sts’ailes for facilitating the GIS analysis, mapping, and creation of figures in this article. We are grateful to Gavia Lertzman-Lepofsky for providing analysis and graphs relating to flood modeling and mapping. Finally, we would like to thank Jenny Berg for processing our tree cores and cookies and for undertaking the tree-ring analysis. Unless otherwise noted, all figures are courtesy of the authors.
Funding Statement
This project was financially supported by Sts’ailes, the Social Sciences and Humanities Research Council (SSHRC) of Canada Grants, the Community Engaged Research Initiative graduate fellowship (Simon Fraser University), the Dean’s Graduate Fellowship (Simon Fraser University), the British Columbia Graduate Scholarship, the Patricia Gallaugher Award in Coastal Sciences (Simon Fraser University), and a Graduate Fellowship through the Department of Archaeology at Simon Fraser University.
Data Availability Statement
All data used in this study have been attached to the submission.
Competing interests
Patrick Morgan Ritchie is employed by Sts’ailes.
Supplementary Material
The supplementary material for this article can be found at https://doi.org/10.1017/aaq.2025.7.
Supplementary Table 1. Radiocarbon dates from charcoal collected in the study area. Calibrated using Calib 8.20 (Stuiver and Reimer Reference Stuiver and Reimer1993) and InCal 20 (Reimer et al. Reference Reimer, Austin, Bard, Bayliss, Blackwell, Bronk Ramsey and Butzin2020).
Supplementary Table 2. Description of shovel tests.
Supplementary Table 3. Charcoal identifications by context.
Supplementary Table 4. Tree-ring age database.
Open access
