The Indigenous History of New England, 6000 BC – 1700 AD
A Comprehensive, Excavated, Academically Rigorous Reconstruction
The history of the New England region, from the final collapse of the Laurentide Ice Sheet to the aftermath of King Philip’s War, is a record of Indigenous adaptability, cultural complexity, and endurance under environmental and human pressure. Reconstructed from excavated assemblages, palaeoenvironmental proxies, ethnohistoric texts, linguistics, and genetics, this narrative is not a prelude to English settlement. It is the deep history of the Wampanoag, Narragansett, Pequot, Mohegan, Massachusett, Nipmuc, Pocumtuc, and Abenaki peoples—and of the earlier communities from whom they descend.
The title’s start date of 6000 BC marks the Late Archaic florescence that produced the region’s first dense, archaeologically unmistakable cultural geography. The human story begins earlier. Any honest reconstruction has to start with ice.
Part I. Geological and Human Foundations (18,000 – 6000 BC)
1. The Laurentide Ice Sheet and Environmental Decompression
1.1 Chronology of deglaciation
At the Last Glacial Maximum the southeastern Laurentide Ice Sheet stood across Long Island, Nantucket Shoals, and Georges Bank. Ice thickness over the Boston Basin and the White Mountains is commonly modeled in the 1.5–2.5 km range in interior reconstructions; the southern margin itself was thinner, but the load was still sufficient to depress the crust and reorganize every drainage in the region. Cosmogenic-nuclide work on the terminal and recessional moraines now places the outer terminal complex around 26–25 ka and inner terminal ridges near 22 ka, with smaller recessional limits within about 50 km of the terminus dating to roughly 21–20.5 ka.
That is not the same as saying New England was ice-free. The major pulse of accelerated recession through the rest of the region does not begin until after atmospheric CO₂ rises near 18 ka, the conventional onset of Termination I. Varve chronology—Antevs’s early-twentieth-century New England Varve Chronology, recalibrated by Ridge and colleagues—then becomes the most precise inland clock. The North American Varve Chronology records thousands of years of seasonal deposition in Glacial Lake Hitchcock and related basins and ties ice-margin positions to Greenland ice-core climate with annual-to-decadal resolution.
The sequence, compressed:
- 21,000–18,000 BP. Maximum and near-maximum stand. Periglacial tundra and steppe-tundra south of the ice; megafaunal communities occupying the unglaciated fringe and, later, the newly uncovered foreland.
- 18,000–15,000 BP. Heinrich Stadial 1 aftermath; Hudson Valley lobe collapse; stagnation south of modern Hartford and Springfield; first large proglacial lakes in southern New England.
- 15,000–13,000 BP. Bølling–Allerød rapid retreat. Minor readvances (Portland, Emerson, Luzerne and related phases) interrupt an otherwise fast northward withdrawal. Tundra gives way to spruce parkland. This is also the window of regional megafaunal collapse—not a single afternoon of extinction, but a compressed interval in which habitat fragmentation, climate volatility, and, eventually, human hunting coincide.
- 13,000–11,700 BP. Terminal moraine systems already left behind to the south; final ice withdrawal from northern New England. The Charlestown–Sandwich–Buzzards Bay moraine belt and the Pineo Ridge–Pond Ridge system in downeast Maine fix the last major ice-marginal architecture. In the Champlain lowland, marine invasion produces the Champlain Sea; the maximum marine limit in northern Vermont reaches on the order of 250 m above modern sea level. Isostatic adjustment and the Lake Candona phase complete the first reorganization of the northern valleys.
Two caveats belong with the chronology. First, ¹⁰Be exposure ages near the terminal moraines and organic ¹⁴C dates from early lake basins can disagree by several millennia: inherited cosmogenic nuclides make some exposure ages too old, while lags between ice retreat and the first datable organics make some ¹⁴C ages too young. The disagreement shrinks north of the terminal zone. Second, “deglaciation” is not a single event. It is a century-scale and millennial-scale process with local stillstands, lake drainages, and marine incursions that created the habitats people later used.
1.2 Meltwater dynamics and catastrophic drainage
Glacial Lake Hitchcock occupied the Connecticut Valley from Rocky Hill, Connecticut, north toward St. Johnsbury, Vermont. Its varves are not only a clock; they contain paleoflood layers that record sudden inputs of meltwater. The Coveville–Northampton spillway breach—placed near 13,680 cal BP in recent varve work—is the most cited of these failures. When the lake dropped, the Connecticut River began cutting its present gorge. Downstream, the valley floor that later carried maize horticulture and Pocumtuc towns is in part a drained lake bed.
Comparable systems formed elsewhere. Glacial Lake Merrimack–Pemigewasset left the Hooksett–Concord–Manchester delta complexes and drained southward, producing giant current ripples and boulder trains in central New Hampshire. Glacial Lake Bascom connected, at times, to early Champlain Sea waters. In northeastern Maine, Glacial Lake Passadumkeag and related basins produced scabland tracts. Across the region the geomorphic signature is the same: outburst-flood scablands, boulder erratics, and ripple fields. They are the reason some of the best Early and Middle Holocene sites sit on terraces that did not exist until the lakes died.
1.3 Isostatic rebound and relative sea level
The crust, unloaded, rebounded unevenly. Modern uplift in coastal Maine is on the order of 2–6 mm yr⁻¹; southern Massachusetts is stable to slightly subsiding. Relative sea-level curves from Barnstable marsh (Massachusetts), Clinton (Connecticut), Portsmouth (New Hampshire), and Eastport (Maine) therefore do not tell one story. Maine’s coast emerged. Southern New England’s shelf drowned. That contrast determined where Late Archaic and Woodland shell deposits could accumulate, which Paleoindian landscapes now lie offshore, and why the Gulf of Maine and the southern bight cannot be treated as a single coastal history.
1.4 Submerged paleolandscapes
The continental shelf retains paleo-river channels, incised valleys, drowned forests, and peat exposures. Mastodon and walrus remains from offshore contexts are evidence that the first people in the region were walking a coastline that no longer exists. Stellwagen Bank and Georges Bank have been the focus of acoustic survey. BOEM and NOAA mapping in the 2020s has improved the resolution of those drowned terrains. The archaeological yield remains small relative to the area, which is expected: preservation is patchy, survey is expensive, and much of the Paleoindian record of southern New England may simply be underwater.
2. Paleoindian Colonization (12,900 – 10,500 cal BP)
2.1 Routes and genetics
Two entry models still structure the continental debate: an interior ice-free corridor and a Pacific coastal “kelp highway.” For New England the relevant point is downstream of that debate. By the time fluted-point makers reached the Northeast, they were already descendants of the First Peoples founding population that separated from East Asian ancestors, paused in Beringia, and then split—most analyses now place the northern/southern Native American split south of the ice sheets around 17.5–14.6 ka.
Uniparental markers in living Indigenous people of the Northeast include Y-chromosome lineages in the Q-M242 and C-P39 families and mitochondrial haplogroups A, B, C, D, and X2a. X2a has a distinctive northeastern and Great Lakes visibility and is often invoked in discussions of early Holocene continuity. What the genetics do not yet do is provide published, high-coverage Paleoindian genomes from Bull Brook, Vail, or Debert that can be lined up, individual by individual, with modern Wampanoag or Abenaki descendants. Continuity at the continental and regional-haplogroup level is robust. Site-to-tribe matching remains, for this period, an inference rather than a sequenced fact.
2.2 Bull Brook I and II
Bull Brook, Ipswich, Massachusetts, remains the type site for Early Paleoindian aggregation in New England. Salvaged during sand and gravel mining in the 1950s, it produced one of the largest fluted-point assemblages on the continent: on the order of several hundred fluted points within a total formal-tool inventory of several thousand objects, arranged in a ring of loci typically reconstructed at about 15–20 m across. Spatial analysis by Robinson and colleagues reads the ring as a planned gathering of multiple family units or micro-bands, not a palimpsest of unrelated camps.
The faunal signal, where it survives as calcined bone, is consistent with a caribou-focused intercept strategy in a Younger Dryas parkland–tundra mosaic, with smaller mammals also present. The lithic signal is more decisive than the bones. The great majority of the stone—on the order of 95 percent in provenance studies—comes from sources 200–450 km away: Hardyston/Reading Prong jasper from Pennsylvania/New Jersey, Onondaga and Normanskill cherts from New York, Munsungan chert from northern Maine, and spherulitic rhyolite from Mount Jasper, New Hampshire. Local stone is the minority. That is a mobility-and-exchange system.
2.3 Fluted-point technology
Northeastern fluted points are Clovis-derived but not identical to western Clovis. Regional literature sorts them into Gainey, Barnes, Crowfield, and local expressions (Bull Brook, Michaud–Neponset, Debert). “Cumberland” labels in New England are now often treated as Barnes-related rather than as a separate mid-South type transplanted intact. Use-wear and residue studies, where they have been done, implicate caribou, beaver, and small mammals. Hafting involved basal grinding, and in some assemblages red ochre and resinous adhesives. The toolkit is specialized, curated, and expensive in transport cost—the material signature of people who moved far and met often.
2.4 The regional site network
South and west of Bull Brook: Templeton and Hidden Creek (Connecticut), Whipple (New Hampshire), West Athens Hill and Potts (New York), Reagen and Jefferson / Israel River sites (Vermont). North and east: Vail, Adkins, Michaud–Neponset, Lamoreau in Maine, and Debert in Nova Scotia. These sites share exotic lithics, fluted-point technology, and evidence of repeated seasonal reoccupation. Isolated finds on the outer Cape and the Gulf of Maine hint at a drowned coastal scatter. The Maine–Maritimes continuum is not a separate “culture” so much as the same mobile system operating on a larger ice-marginal landscape.
2.5 Social structure and ideology
The working model is fission–fusion. Spring and summer micro-bands on the order of 15–35 people; intercept-season or cold-season macro-band gatherings of 100–250 or more. Exotic lithics are the archaeological proxy for kinship and alliance: Pennsylvania jasper does not move to Ipswich unless both ends of the path already belong to the same social world. Ritual caching and red ochre are present early. Gendered labor is inferred from tool-kit structure—projectile manufacture and butchery versus hide-working and plant processing—not from graves, which are almost absent.
3. Early Archaic Transition (10,500 – 6000 cal BP)
The Younger Dryas ends. Early Holocene warming is interrupted by the 8.2 ka cooling event, visible in northeastern pollen and lake records. Spruce–pine woodland gives way, over the next two millennia, to mixed deciduous forest. The classic mid-Holocene hemlock decline—a regional pollen event centered near 5,400 cal BP, not immediately after the Younger Dryas—belongs to the later part of this interval and is still debated as pathogen, climate, or both. Some archaeologists have tied the Neville-to-Stark transition and a retreat from large interior wetlands to the 8.2 ka downturn. The causal link is plausible, not proven.
The specialized fluted-point complex does not survive the forest closure. What replaces it is a southern-derived bifurcate-base horizon—LeCroy, Kanawha, MacCorkle, St. Albans, and related Kirk corner-notched forms—dated in the mid-South between roughly 9,000 and 7,800 cal BP and present in New England mainly as the later members of that series. North of Long Island Sound these types are never as abundant as they are in the mid-Atlantic and Tennessee Valley. They are real, but they are a thin scatter. An older literature spoke of an Early Archaic “hiatus.” The hiatus was an artifact of visibility: acidic soils, rising seas drowning Early Holocene shorelines, and a toolkit that used more local stone and left fewer diagnostic points.
By about 8,200–7,800 cal BP the bifurcate series grades into the Neville point, defined at Dincauze’s stratified type site on the Merrimack at Amoskeag (Manchester, New Hampshire). Neville dates there fall roughly 7,800–7,000 rcbp; Stark points overlay and overlap them at about 7,600–6,400 rcbp; Merrimack stemmed points close the Middle Archaic near Merrimack stemmed points close the Middle Archaic near 6,000 rcbp. Neville is the northeastern analogue of Stanly; Stark of Morrow Mountain. Assemblages become quartz-, quartzite-, rhyolite-, and felsite-dominated. Ground-stone tools appear in force: early ground-slate points and gouges in Maine, adzes and gouges farther south. That is a forest technology—woodworking, wetland plant processing, canoe inference—not a tundra intercept kit. Settlement pulls onto newly stabilized river terraces and wetland margins. Diligent River (Nova Scotia) and Neponset (Massachusetts) paleosol sites, Robbins Swamp in northwestern Connecticut, and the Merrimack and Connecticut floodplains show the pattern: more sites, more local stone, less of the continental lithic ballet of Bull Brook. Regionalization has begun. It will define everything that follows.
Part II. Late Archaic New England (6000 – 1000 BC)
4. The Hypsithermal and a richer coast
The Hypsithermal (broadly 7,000–5,000 BP in New England pollen sequences) is not a uniform “optimum.” It is a long interval of relatively high summer insolation, oak–hickory–chestnut expansion in southern New England, and, in some reconstructions, a prairie-peninsula finger reaching the southern edge of the region. Cores from Rogers Lake (Connecticut), Linsley Pond and related Massachusetts basins, Berry Pond (Vermont), and Maine lakes fix the vegetation. Sea level is still rising, but more slowly. Brackish estuaries stabilize. Anadromous runs—alewife, shad, sturgeon, Atlantic salmon where the rivers allow—become predictable. Shellfish beds thicken.The Boylston Street / Back Bay weirs under Boston are not one mega-weir of 65,000 stakes. Re-excavation in the 1980s showed many smaller fence-like alignments, rebuilt across shifting tidal channels over about 1,500 years. The best radiocarbon brackets put construction near 5,200 BP and abandonment of that particular mudflat near 3,700 BP, when tide range and water depth outgrew small wooden stakes. They harvested late-spring anadromous fish in the intertidal zone. They are among the oldest preserved wooden fish weirs in eastern North America. The Damariscotta oyster heaps (Whaleback, Glidden) are among the largest shell monuments on the Atlantic seaboard north of Georgia. They are often dropped into Late Archaic narratives because Maine’s coastal Archaic is maritime and because oyster is the right animal. The dated heaps themselves, however, are largely Ceramic Period: radiocarbon on Glidden shells clusters around 2,400–1,500 BP. Earlier Maine shellfishing is real—Turner Farm on North Haven has shell and swordfish by about 5,000 BP—but the famous Damariscotta mountains of oyster are Woodland accumulations on an older coastal tradition. Putting them in Part II as “Late Archaic explosions” overstates the dating. Putting them in the longer story of marine intensification does not. Fire is the other landscape tool. Charcoal microstratigraphy in southern New England soils and ponds yields fire-return intervals on the order of 15–40 years in many Holocene sequences. Whether every peak is anthropogenic is the standing debate. The case for management is strongest where fire-sensitive mesic taxa recede, oak and berry-producing shrubs increase, and the pattern is spatially tied to occupation. The case against treating all charcoal as Indian burning is equally strong: lightning, drought, and the Hypsithermal itself burn forests. The honest position is mixed: people burned, climate burned, and by the Late Archaic the two were no longer fully separable.
5. Regional traditions
Neville–Stark / southern quartz industries (ca. 6000–4000 BC). Mystic Rockshelter and related Groton, Connecticut, locales; Assawompsett Pond (Lakeville, Massachusetts); cobble-tool reduction on quartz. This is the dense southern New England Middle-to-Late Archaic floor.Merrimack Valley / Gulf of Maine Archaic. Shattuck Farm (Andover) storage pits with chenopod; Heath Brook (Tewksbury) living floors with unusual organic preservation; and, above all, the Moorehead Burial Tradition. Moorehead cemeteries—once marketed as the “Red Paint People”—run from roughly 6,000 to 3,800 BP across a relatively tight 350 × 80 km belt from southern Maine into the Maritimes. Graves are often “boneless”: ochre stains, ground-slate tools, gouges, bayonets, and plummets survive after bone has dissolved. Where shell preserves bone, as at Nevin, the burial program is visible in full. These cemeteries are the first unambiguous corporate ritual landscapes in the region. Vergennes Phase (Champlain Valley Laurentian). Otter Creek points, ground-slate technology, the KI site and related Lake Champlain stations. This is the northern interior expression of the Laurentian tradition, with a stronger hunting and lake-edge signature than the southern quartz riverine sites.Narrow-stemmed / Orient (ca. 3000–1000 BC).Orient Fishtail, Brewerton, and late Stark-like forms; grass-lined seed pits (Westfield 19-HD-109); a Long Island–lower Hudson interaction sphere that will matter when soapstone and, later, wampum move.Susquehanna intrusion (ca. 1700–1300 BC). Broad-bladed Atlantic and Perkiomen points, cremation cemeteries, Orient-phase mortuary elaboration, and soapstone vessels whose raw material can be sourced to Rhode Island and mid-Atlantic quarries. The interpretive fight is unfinished: small-scale migration versus elite-driven acculturation versus both. Hybrid models are currently the most responsible. Skeletal violence indicators rise in some of these contexts. That does not prove a conquest horizon. It does mean the Late Archaic was not a pastoral of peaceful foragers.
6. Technology, plants, exchange, complexity
The ground-stone revolution is real: ulus, gouges, celts, plummets, bannerstones, grooved axes. Dugout and birchbark canoe origins are inferred from the toolkit and from later watercraft ethnography, not from a preserved Late Archaic hull in a New England pond. Atlatl weights and regional bannerstone styles mark hunting technology before the bow.Proto-horticulture is the quiet revolution. Chenopodium, little barley, sumpweed, erect knotweed, and sunflower appear in floodplain features. Seed-size increase and thin testa are the morphological flags of management. This is Eastern Agricultural Complex gardening, not Mexican maize. It is patchy, riverine, and easily missed if flotation is sloppy.Long-distance goods—Lake Superior native copper, Ramah chert from Labrador, Busycon shell moving inland—prove that the Late Archaic Northeast was never a closed basin. Dogs of the Eastern North American pre-contact lineage are present. Grave-good differentiation and interpersonal trauma increase. Status is emerging. It is not yet a chiefdom landscape.
Part III. Early and Middle Woodland (1000 BC – AD 1000)
7. Ceramics and Meadowood
Vinette 1 pottery—thick, grit-tempered, cord-marked inside and out, conoidal—enters the Northeast around 1000 BC. The idea of pottery is southern and interior; the earliest New England vessels sit in a Meadowood world. Meadowood (ca. 1000–500 BC) is an interaction sphere, not a tribe: Onondaga chert cache blades, blocked-end tubes, birdstones, and a mortuary program that reaches the Connecticut and Merrimack valleys. People are cooking in ceramic pots and burying some of their dead with objects that only make sense if Great Lakes and Atlantic networks are open.
8. Middle Woodland feasting and “Hopewell echoes”
Fox Farm (Rhode Island), Boucher (Vermont), Wheeler’s Site (Vermont), and a handful of other ceremonial stations produce mica, copper, galena, shark teeth, and platform pipes. These are Hopewell-world objects at the far northeastern edge of a midcontinental ceremonial economy. They do not make New England a Hopewell province. They show that between roughly 200 BC and AD 400–600 some communities hosted feasts and interred exotic goods in ways that participated, at a distance, in that economy. After AD 400–600 the long-distance pulse fades. Localized ceremonialism replaces it.Jack’s Reef corner-notched points appear in the same centuries. They are the strongest typological candidate for early bow-and-arrow in the region. The debate is not whether the bow arrives in the Woodland; it is how early and how completely it displaces the atlatl. Levanna triangles later become the commonplace Late Woodland arrow tip.Ceramics elaborate: pseudo-scallop shell, rocker-stamping, incised motifs. Point Peninsula styles dominate the north; Fox Creek / Kipp Island related forms the south.
9. Mortuary practice and rock art
Small accretional mounds at places such as Bellows Falls and Portsmouth, cremation arenas, Adena-style blocked-end tubes, stone cairns, and petroglyph panels (Dighton Rock, Machias Bay, Bellows Falls, Embossing Rocks) concentrate ritual in visible places. Many petroglyphs are now treated as Late Woodland to protohistoric rather than Archaic; motif dating remains argument, not clock. Red ochre continuity from Moorehead graves into Woodland burial is one of the longest symbolic threads in the region. Grave goods—copper beads, marine-shell gorgets, pipes, bannerstones—mark difference among the dead.
Part IV. Late Woodland (AD 1000 – 1600)
10. Maize and demography
Maize does not arrive in the Northeast as a single dated kernel that changes everything overnight. Microbotanical residues (phytoliths, starch in cooking residues) appear in the broader Northeast centuries earlier, in some cases before the Common Era. In New England the earliest secure macrofossils of Northern Flint (8-row) cluster after AD 1000. A commonly cited southern New England window is AD 1040–1220; Vermont’s Headquarters site on the Missisquoi has a direct maize AMS date calibrated around AD 1216; Burnham-Shepard and related Connecticut Valley sites document fourteenth-century and later maize, beans, and storage pits. A cluster of New England dates between about cal AD 1250 and 1450 marks the moment maize becomes archaeologically common, not the moment it is first known. The Three Sisters complex—maize, beans, squash—is adopted unevenly. Coastal southern New England and the Connecticut and Champlain valleys take it up more fully than the far Downeast and northern interior. Stable-isotope work shows a shift toward C₄ plants that is clearest after about AD 1300 in sampled series. That is dietary maize, not just a few cobs in a pit. Fish fertilizer, floodplain ridging, and slash-and-burn are the inferred agronomic package; direct field features are scarce.Population is the most abused number in this literature. A rise from something on the order of 20,000–40,000 around AD 1000 to a working early-seventeenth-century envelope of roughly 70,000–100,000 for greater New England is defensible as an order of magnitude. The figure 72,000–95,000 is a chosen point inside that envelope, not a census. Older high counts and older low counts both still circulate. What can be said without inflation: Late Woodland communities were numerous enough to nucleate, to fortify, and to sustain confederacies that Europeans later named as if they had always existed.
11. Villages and houses
Palisaded villages appear in the last pre-contact centuries: Guidos (Connecticut), Lucy Vincent (Massachusetts), Eddy (Rhode Island), and hilltop refuges in Nipmuc and Pocumtuc country. Longhouses of 15–30 m with hearth rows are documented in Rhode Island and eastern Massachusetts; multi-family wetu clusters remain the more common southern form. Nucleation is a political fact. People who farm stored maize and live inside palisades are defending something—stores, people, a name.
12. Ethnogenesis and political geography, ca. 1400–1600
These are the names that enter the European record, already formed:
- Coast: Wampanoag / Pokanoket (Massasoit’s polity), Narragansett and Niantic, Massachusett, Nauset, Pennacook / Pawtucket.
- Interior: Nipmuc networks and the Quinebaug divide; the Pocumtuc maize-belt polity from Deerfield to Turners Falls; Mahican pressure on western Connecticut and Massachusetts.
- North: Eastern Abenaki (Penobscot, Kennebec, Androscoggin) and Western Abenaki (Sokoki, Missisquoi, Cowasuck).
- Pequot–Mohegan split (ca. 1450–1550), later personified in colonial sources by Uncas, is both a political fission and a linguistic marker within the Eastern Algonquian dialect continuum.
Wampum—quahog and whelk beads, with Narragansett and Long Island production especially visible—is not “Indian money” in the colonial sense until Europeans make it so. Before that it is treaty currency, diplomatic gift, and condensed political speech. Control of production is control of alliance.
13. War and diplomacy
The mourning-war complex—raid, captive-taking, torture of some, adoption of others, ritual replacement of the dead—is the northeastern war grammar in this period. Osteological series that have been published show projectile wounds and scalping trauma in adult males at rates often cited in the 12–18 percent range; those figures come from small, uneven samples and should be treated as order-of-magnitude, not a regional average. Seasonal truce gatherings and wampum diplomacy are the other half of the same system. This is institutionalized violence inside a diplomatic order, not primordial chaos.
Part V. Apocalypse and Reorganization (1500 – 1700)
14. The long prelude, 1497–1615
Cabot (1497), Verrazzano (1524), Gómez (1525), Gosnold at Cuttyhunk (1602), Pring (1603), Weymouth’s 1605 kidnappings, Champlain’s 1604–1609 reconnaissance and the first durable French–Abenaki alignments: these are not “first contact” so much as the moment the North Atlantic fishery and fur reconnaissance become a standing presence. Basque, Breton, Portuguese, and English crews are on the Grand Banks and in the Gulf of Maine for decades. Metal tools move ashore. So do rats, and so, almost certainly, do pathogens that do not yet have names in English letters. Kidnapping for pilots and display is documented; it is not incidental.
15. The Great Dying, 1616–1619
Coastal mortality from Massachusetts Bay to the Penobscot in these years is the demographic event that makes Plymouth possible. Eyewitnesses—Dermer, Gorges, Morton, later Bradford—describe emptied villages, unburied bone, and a disease that appeared to spare nearby Europeans. Fatality on the immediate coast is commonly reconstructed at 75–90 percent. Inland and northern rates were lower. Abandoned-village archaeology and the “miraculous clearing” language of English godly letters are two sides of one landscape.The pathogen is not known. Marr and Cathey (2010) argued for leptospirosis with Weil’s syndrome: jaundice, epistaxis, rodent-urine contamination of fresh water, a disease ecology that would hit wetu communities harder than shipboard Europeans. That remains a coherent clinical hypothesis. It is not confirmed by aDNA from 1616–1619 New England burials, because those genomes have not been published. Smallpox, plague, hepatitis co-infection, and other candidates remain in the literature. Secondary epidemics—1633–34 smallpox is the clearest—then cut into communities that had already been broken. English writers called the emptying providential. It was a virgin-soil catastrophe interacting with a century of prior coastal contact. Those two sentences can stand together.
16. Realignment, 1620–1675
Plymouth plants itself at Patuxet in 1620. The Massasoit–Pilgrim treaty of 1621 is a Wampanoag strategic act, not a surrender: a weakened coastal sachemate using the English against Narragansett power. Land deeds, cattle that eat unfenced planting grounds, alcohol, and the legal fiction that a sachem’s mark on an English instrument extinguished sovereignty rather than leasing use—these were the ordinary solvents of Native title between 1620 and 1675.
John Eliot’s mission, beginning in earnest in 1646, created a parallel geography of Christian Indian towns—Natick, Punkapoag, Hassanamesit, Wamesit, Nashobah, and others—inside the Massachusetts Bay legal order. The towns were not empty of Indigenous politics. Kinship, hunting rounds, and diplomacy with non-Christian relatives continued. Massachusett literacy, the 1663 Eliot Bible, and Native schoolmasters produced a documentary culture that is itself a primary source for the language and for land. They also produced a class of go-betweens whom English courts trusted more than sachems and whom many non-Christian kin distrusted as translators of dispossession. John Sassamon (Wussausmon) is the type of that class. Ponkapoag-born, a year at Harvard, schoolmaster at Natick, scribe to Wamsutta and then to Metacom, later preacher at Nemasket: he moved in both jurisdictions and was accused, in both, of serving himself. In late December 1674 he warned Plymouth’s Josiah Winslow that Metacom was preparing war. Winslow dismissed a single Indian witness. Within weeks Sassamon’s body was taken from the ice at Assawompset Pond. Murder, drowning, and land-title violence at the pond itself have all been argued; Lisa Brooks has insisted that the pond was already a contested deedscape, not a neutral crime scene. In June 1675 a mixed English and Praying Indian jury convicted Tobias, Wampapaquan, and Mattashunnamo. Two were hanged at once. Plymouth had claimed the right to try and kill Wampanoag men for a death in Wampanoag country. That claim, more than the corpse, was the casus belli. Swansea was raided the same month.
17. King Philip’s War, 1675–1678
Causes. Demographic rebound on the English side against a Native population still unrecovered from 1616–19 and 1633–34; cattle, fences, and courts that converted seasonal planting grounds into trespass; a theocratic legal theory in which sachem authority was custom, not sovereignty. Puritan writers framed the war as rebellion. Native participants framed it as defense of land and jurisdiction. The material conflict was over who could say what a deed meant.
Coalition.Metacom’s alliance was never a single command. Wampanoag, Nipmuc, and Pocumtuc fighters formed the core. Narragansett participation hardened after the English refused to treat neutrality as neutrality and after the December assault on the Great Swamp fort. Mohegan and Pequot auxiliaries fought with Connecticut. Many Praying Indians were interned, used as scouts, or killed by both sides.June 1675, Swansea. Opening raids on Plymouth’s exposed towns.
Summer–fall 1675, Connecticut Valley. Bloody Brook, the burning of Deerfield and neighboring towns, and the temporary collapse of the English river frontier.19 December 1675, Great Swamp. About a thousand Massachusetts, Plymouth, and Connecticut militia, with Mohegan and Pequot allies, crossed a frozen wetland in southern Rhode Island and stormed an unfinished Narragansett palisade of several acres. The fort was burned. Colonial losses are the firmest numbers: on the order of seventy dead and 150 wounded, many dying on the winter march out. Narragansett deaths remain estimates—several hundred to as many as a thousand when non-combatants and those who died of wounds and exposure in the swamp are included. “600+” is the figure that secondary literature has frozen. Warriors were present and fighting. Women, children, and the old died in the fire and after it. The destruction of winter stores turned a tactical English victory into a famine weapon. Canonchet survived and took the Narragansett remnant into open war. May 1676, Turner’s Falls. An English surprise against a Pocumtuc fishing encampment at Peskeompskut. Combatants and non-combatants died together. The valley polities did not recover as military powers.August 1676. Metacom killed near Mount Hope by a Native soldier in Benjamin Church’s company. Southern fighting winds down; Maine and the eastern theater continue into 1678.English losses of several hundred dead, dozens of towns attacked, and a high per-capita military death rate among colonial men are part of the record. They do not balance the Native dead. They explain why the English remembered the war as existential even as they won it.Northern diaspora. Schaghticoke, on the Hoosic near the Hudson, became an early gathering place for Pocumtuc, Nonotuck, Agawam, and other valley families after 1676. Missisquoi on Champlain, Cowasuck on the upper Connecticut, Odanak / Saint-François and Bécancour on the Saint Lawrence absorbed further streams. Odanak became a multi-community Wôbanaki town whose seventeenth- and eighteenth-century population included people from many New England places. Western Abenaki ethnogenesis after the war is the political consolidation of older Abenaki communities with southern refugees under French mission and alliance pressure. Ties to homelands were not cut; they were stretched. Aftermath. Surrender did not guarantee safety. Linford Fisher’s work on surrenderers shows that people who came in to avoid enslavement were enslaved, executed, or bound out at rates close to those of captured fighters. Destinations included Barbados, Jamaica, Bermuda, the Azores, Cádiz, and North Africa. Plymouth shipped 178 captives on a single vessel to Cádiz in October 1675. Barbados and other islands soon tried to bar further New England Indians—not from abolitionist principle, but from fear of importing a war. Local term slavery and forced apprenticeship of children kept others inside New England households. Philip’s wife and young son were among those sold or transported. “Over 2,000 enslaved” is an upper-end figure. Published reconstructions more safely say many hundreds shipped overseas and more held locally, with a combined captive population that may reach or exceed a thousand depending on how surrenderers, children, and later eastern-war captives are counted. The legal fact is not in doubt.What remained on the ground was the beginning of the reservation order: shrunken, surveyed, and legally encircled lands in the Narragansett country, at Mashpee and Aquinnah, in Mohegan and Pequot territories, at Natick and other praying-town remnants. English policy after 1676 treated survivors as subjects and sometimes as slaves, not as treaty partners of the 1621 type. Cultural erasure was attempted—language, jurisdiction, arms, movement. It was incomplete. Incomplete is not the same as harmless.
Research Frontiers (2015–2025)
Ancient DNA and the continuity problem
Scheib et al. (2018) published 91 ancient genomes from California and from southwestern Ontario. The Ontario (ASO) cluster sits with Algonquian-speaking populations and with other northern Native American ancestry. That is the correct regional frame for New England. It is not a published demonstration that named genomes from RI-1000, Fox Farm, or Swantown Hill are direct matches to living Wampanoag, Narragansett, or Mohegan individuals. Those site-level claims, where they exist, sit in tribal-control, restricted, or still-unpublished space. What can be said without outrunning the published genomes: founding Native American ancestry persists in the Northeast; Algonquian-speaking populations sit on the northern Native American branch; living southern New England tribal communities also carry living southern New England tribal communities also carry post-1700 European and African admixture at levels that vary by family and reservation history. Seventeenth-century bottlenecks were severe on the coast; interior and northern communities were hit later and differently. Pre-contact diversity was greater than the post-war remnant. How much greater is still a research question.Continuity is an archaeological, oral-historical, and political fact. It does not require a press-ready genome from every type site. Community consent, not only laboratory capacity, now gates legitimate sampling. That is a scientific condition, not an obstacle to it.
Linguistics
Proto-Eastern Algonquian reconstruction (Goddard, Siebert, Pentland) remains the backbone. Pequot–Mohegan divergence tracks the political split. Massachusett and Narragansett form a dialect continuum visible in deeds, the Eliot Bible, and Roger Williams. Language revitalization is now part of the same history, not a coda to it.The Wôpanâak Language Reclamation Project, founded in 1993 by Jessie Little Doe Baird with Mashpee, Aquinnah, Assonet, and Herring Pond citizens, rebuilt a spoken community from the colonial documentary corpus plus Algonquian comparative method. Mae Baird, raised in the language, is the documented first native speaker of the revived generation. Abenaki programs at Odanak and in Vermont and New Hampshire, and Mi’kmaq education in the Maritimes, follow different documentary paths to the same end. Fluency cohorts in the 2020s are historical events. They do not need to be footnoted as if they were excavation reports.
Decolonizing practice
NAGPRA repatriation, tribal review of excavation, and collaborative projects—Mashpee Wampanoag with Harvard, Narragansett with Brown, Mashantucket Pequot Museum research, Wabanaki review in Maine—are now the condition of legitimate fieldwork. “Decolonizing” here has a concrete meaning: who consents, who interprets human remains, who stores sacred objects, and whose questions get funded. It is not a synonym for abandoning stratigraphy.Traditional Ecological Knowledge belongs in fire-history and fishery reconstructions because the charcoal curves and the weir stakes are incomplete without the knowledge systems that produced them. Integrating that knowledge is how one distinguishes a drought-driven charcoal peak from a burning regime, and a “natural” estuary from a harvested one.
Open debates
Three arguments should remain open in the text rather than closed by rhetoric.Population. A working contact-era interval on the order of 70,000–100,000 for greater New England is responsible. The figure 72,000–95,000 is one informed point inside that interval. Older high-count models and older low-count models are both still cited; neither should be treated as settled. Radiocarbon-sum and site-density work shows Late Woodland growth and no pre-1500 collapse comparable to some interior regions.Political form. “Tribe,” “sachemship,” and “confederacy” are English and anthropological words laid over nested, seasonal, and negotiated authority. Coastal polities look more centralized in the sources than many interior networks; that may be source bias as much as political fact. The named map of 1600–1675 is real. It is also a late, compressed, and partly colonial artifact of a longer ethnogenesis.Continuity versus rupture after 1676. Southern New England military confederacies were broken. Communities were not extinguished. Reservations, praying-town remnants, diaspora towns at Schaghticoke and Odanak, and later federally and state-recognized nations are the continuation, under constraint, of the same history. Genetic continuity at the regional-ancestry level, linguistic reclamation, and uninterrupted residence on pieces of the old geography all argue against a clean rupture. The reservation survey line, the slave ship, and the internment of Christian Indians at Deer Island argue against a clean continuity of sovereignty. Both sentences are required.
Conclusion
Eight thousand years is a long occupancy to treat as a prologue. The Laurentide retreat made the ground. Paleoindian aggregations at Bull Brook and its sister sites show that the first people in this region already moved stone across hundreds of kilometers and gathered in numbers large enough to leave a planned ring of hearths. The Early Archaic did not empty the country; it changed the toolkit when the parkland closed. The Late Archaic made coasts, falls, weirs, ochre cemeteries, and long-distance copper and Ramah chert into a single interaction field. Woodland pottery, Meadowood blades, and the later Hopewell-edge exotics pulled New England into continental ceremonial economies and then released it again into local ritual. Maize after AD 1000 did not invent politics here. It intensified them: palisades, wampum, named sachemships, mourning war.
The seventeenth century did not discover this world. It broke the coastal demographic spine in 1616–1619, occupied the emptied ground, converted deed and cattle into jurisdiction, and in 1675–1678 destroyed the southern military confederacies that had taken shape in the Late Woodland. Enslavement, internment, reservation lines, and the northward refugee stream were not afterthoughts. They were the English solution to a people who had not agreed to become subjects.Two errors close most popular versions of this history. One is the vanished-race story: as if Mashpee, Aquinnah, Narragansett, Mohegan, Mashantucket, Nipmuc, Schaghticoke, Missisquoi, and Odanak were footnotes to a completed extinction. The other is the unbroken-idyll story: as if 1676 were a weather event after which nothing essential changed. The excavated record supports neither. Continuity of ancestry, language family, place knowledge, and political identity is real. Continuity of sovereignty, population, and the pre-war map is not. A reconstruction that is willing to be brutal has to hold both.
What the last decade of work adds is not a new origin myth. It is constraint. Varves and exposure ages have tightened deglaciation. Residue and provenience studies have made Bull Brook a social site rather than a point type. Direct AMS dates have pulled maize out of charcoal-proxy fog. Pathogen diagnosis for 1616–1619 has not been closed and should not be written as if it had. Ancient DNA has placed the Northeast on the northern Native American / Algonquian-related branch without licensing invented site-to-surname pedigrees. Community review has made some older collection practices impossible, which is a gain.
The archaeological sequence from a Bull Brook winter ring to a palisaded wetu to a praying-town meetinghouse to a reservation council is not a story of disappearance. It is a story of people who were already ancient in this landscape when English letters arrived, who absorbed an epidemic and a war designed to end them as polities, and who remain the authority on what that sequence means. The task of a reconstruction that calls itself rigorous is to keep the ice, the weirs, the maize, and the swamp in the same frame as the living nations named in the first paragraph—and to stop writing as if the last of those were a coda.
Working bibliography
Part I
Antevs, Ernst. 1922. The Recession of the Last Ice Sheet in New England. American Geographical Society Research Series 11.Antevs, Ernst. 1928. The Last Glaciation with Special Reference to the Ice Sheet in North America. American Geographical Society Research Series 17.Balco, Gregory, and Joerg M. Schaefer. 2006. “Cosmogenic-Nuclide and Varve Chronologies for the Deglaciation of Southern New England.” Quaternary Science Reviews 25: 2608–2621.Balter-Kennedy, Allie, et al. 2024. “The Laurentide Ice Sheet in Southern New England and New York during and at the End of the Last Glacial Maximum: A Cosmogenic-Nuclide Chronology.” Climate of the Past 20: 2167–2191.Bourque, Bruce J. 2001. Twelve Thousand Years: American Indians in Maine. University of Nebraska Press.Bradley, James W., Arthur E. Spiess, and Robert E. Boisvert. 2008. “What’s the Point: Modal Forms and Attributes of Paleoindian Bifaces in the New England-Maritimes Region.” Archaeology of Eastern North America 36: 119–172.Byers, Douglas S. 1954. “Bull Brook—A Fluted Point Site in Ipswich, Massachusetts.” American Antiquity 19 (4): 343–351.Halsted, Christopher T., et al. 2024. “A Critical Re-analysis of Constraints on the Timing and Rate of Laurentide Ice Sheet Recession in the Northeastern United States.” Journal of Quaternary Science.Meltzer, David J. 2009. First Peoples in a New World. University of California Press.Ridge, John C. 2004. “Quaternary Glaciation of Western New England.” In Quaternary Glaciations—Extent and Chronology, Part II, ed. Ehlers and Gibbard, 169–199. Elsevier.Ridge, John C., et al. 2012. “The New North American Varve Chronology.” Quaternary Science Reviews 48: 1–20.Robinson, Brian S., Jennifer C. Ort, William E. Eldridge, Adrian L. Burke, and Bertrand G. Pelletier. 2009. “Paleoindian Aggregation and Social Context at Bull Brook.” American Antiquity 74 (3): 423–447.Robinson, Brian S., Jennifer C. Ort, and Willard A. McIntosh, eds. 2008. Early Maine Archaeology. University of Maine Press.
Part II
Chilton, Elizabeth S. 1999. “Mobile Farmers of Pre-Contact Southern New England.” In Current Northeast Paleoethnobotany II, ed. John P. Hart, 157–171.Dincauze, Dena F. 1972. “The Atlantic Phase: A Late Archaic Horizon in the Northeast.” Man in the Northeast 4: 35–57.Dincauze, Dena F. 1976. The Neville Site: 8,000 Years of Human Activity at Amoskeag. Peabody Museum Monographs 4.Funk, Robert E. 1993. Archaeological Investigations in the Upper Susquehanna Valley, New York State. Persimmon Press.Newby, Paige E., and Thompson Webb III. 1994. “Radiocarbon-Dated Pollen and Sediment Records from near the Boylston Street Fishweir Site in Boston, Massachusetts.” Quaternary Research 41: 214–224.Perri, Angela R., et al. 2021. “Dogs Accompanied Humans during the Second (Not First) Wave of Americas Settlement.” PNAS 118 (6): e2025036118.Ritchie, William A. 1980. The Archaeology of New York State. 2nd rev. ed.Snow, Dean R. 1980. The Archaeology of New England. Academic Press.Taché, Karine, et al. 2019. “Susquehanna Tradition Revisited.” Archaeology of Eastern North America 47: 45–68.
Parts III–IV
Bendremer, Jeffrey C. M., and Robert E. Dewar. 1994. “The Advent of Prehistoric Maize in New England.” In Corn and Culture in the Prehistoric New World, ed. Johannessen and Hastorf, 369–393.Bendremer, Jeffrey C. M., Elizabeth S. Chilton, and Robert E. Dewar. 1991. “A Grass-Lined Maize Storage Pit and Early Maize Horticulture in Central Connecticut.” North American Archaeologist 12 (4): 325–349.Bragdon, Kathleen J. 1996. Native People of Southern New England, 1500–1650. University of Oklahoma Press.Calloway, Colin G. 1990. The Western Abenakis of Vermont, 1600–1800. University of Oklahoma Press.Hart, John P., and William A. Lovis. 2013. “Reevaluating What We Know about the Histories of Maize in Northeastern North America.” Journal of Archaeological Research 21 (2): 175–216.Lavin, Lucianne. 2013. Connecticut’s Indigenous Peoples. Wesleyan University Press.Little, Elizabeth A. 1995. “The Late Woodland Diet on Nantucket Island and the Problem of Maize in Coastal New England.” American Antiquity 60 (2): 351–368.Little, Elizabeth A. 2002. “Kautantouwit’s Legacy: Calibrated Dates on Prehistoric Maize in New England.” American Antiquity 67.
Part V
Anderson, Virginia DeJohn. 2004. Creatures of Empire. Oxford University Press.Brooks, Lisa. 2018. Our Beloved Kin: A New History of King Philip’s War. Yale University Press.Cave, Alfred A. 1996. The Pequot War. University of Massachusetts Press.DeLucia, Christine M. 2018. Memory Lands. Yale University Press.Fisher, Linford D. 2017. “‘Why Shall Wee Have Peace to Bee Made Slaves’: Indian Surrenderers during and after King Philip’s War.” Ethnohistory 64 (1): 91–114.Mandell, Daniel R. 2008. Tribe, Race, History. Johns Hopkins University Press.Marr, John S., and John T. Cathey. 2010. “New Hypothesis for Cause of Epidemic among Native Americans, New England, 1616–1619.” Emerging Infectious Diseases 16 (2): 281–286.
Frontiers
Atalay, Sonya. 2006. “Indigenous Archaeology as Decolonizing Practice.” American Indian Quarterly 30 (3/4): 280–310.Cipolla, Craig N. 2013. Becoming Brothertown. University of Arizona Press.Goddard, Ives. 1980. “Eastern Algonquian as a Genetic Subgroup.” IJAL 46 (2): 156–163.Moreno-Mayar, J. Víctor, et al. 2018. “Terminal Pleistocene Alaskan Genome Reveals First Founding Population of Native Americans.” Nature 553: 203–207.Reich, David, et al. 2012. “Reconstructing Native American Population History.” Nature 488: 370–374.Scheib, C. L., et al. 2018. “Ancient Human Parallel Lineages within North America Contributed to a Coastal Expansion.” Science 360 (6392): 1024–1027.Wôpanâak Language Reclamation Project. https://www.wlrp.org/.