Urve Linnamae in Placentia Bay in 1970

From June 26 to August 3, 1970, Dr. Urve Linnamae and her crew carried out an archaeological survey of the Placentia Bay area of Newfoundland while she was under contract to the National Museum of Canada. Prior to this Dr. Linnamae conducted excavations at the significant Palaeoeskimo sites of Cape Ray Light in 1967 and the Pittman site in White Bay in 1967 and 1968. In 1975 she wrote the influential monograph The Dorset Culture: A Comparative Study in Newfoundland and the Arctic. Technical Papers of the Newfoundland Museum, No.1 in which she discusses ‘…the position of Newfoundland within the Dorset culture area and the nature of Dorset culture regional variation.’ Dr. Linnamae went on to become an associate professor in the Department of Archaeology at the University of Saskatchewan and retired from that position in 2003 (Linnamae 1971 & Biography).

The National Museum of Canada focused on this area of Newfoundland for several reasons including:

  • the lack of known archaeological sites in the area at the time;
  • an attempt to increase the known geographic distribution of the Dorset;
  • to define the ecological adaptations and relationships within this southern and environmentally different area of Newfoundland and the Dorset culture (Linnamae 1971).

In particular the survey focused on the islands of inner Placentia Bay and around Come by Chance and Arnold’s Cove. The team also briefly visited Old Perlican and Gooseberry Cove, Trinity Bay (Linnamae 1971).

During the survey the team found or revisited 12 archaeology sites. Six of these were very small sites and two were spot finds of just single artifacts. The visit to Old Perlican was for a revisit of a known site and to view the associated collection held by a local person. All of the sites had some precontact component, most commonly it was Dorset Palaeoeskimo. Several sites had either a Maritime Archaic or Recent Amerindian component.  Finally, four sites had a European component (Linnamae 1971). In this post I’ll tell you what Linnamae and her team found at four of these sites.

General area of Placentia Bay Archaeological Survey. 1 Old Perlican 2 Heart's Ease 3 Bordeaux 2 4 Bordeaux 1 5 Long Island Neck 6 New Grove 7 Great Brule 8 Tack's Beach 9 Dog Harbour 10 Unnamed Cove 11 Come-by-Chance 12 Little Brule (Linnamae 1971)
General area of Placentia Bay Archaeological Survey.
1 Old Perlican
2 Heart’s Ease
3 Bordeaux 2
4 Bordeaux 1
5 Long Island Neck
6 New Grove
7 Great Brule
8 Tack’s Beach
9 Dog Harbour
10 Unnamed Cove
11 Come by Chance
12 Little Brule
(Linnamae 1971)

The largest site found by Linnamae was New Grove (CkAm-01). This site is located in a small cove on the eastern shore of Long Island, the second largest island in Placentia Bay. The site consists of Maritime Archaic, Dorset Palaeoeskimo and European components. Unfortunately the European occupation seems to have disturbed the earlier occupations. As well, the site has been heavily eroded along the beach side by high seas and winter ice (Linnamae 1971).

PLATE I New Grove CkAm-01 Looking down at site area towards the North. Seated figure in midground is at Test Trench 3. PLATE II Looking at eroding bank edge at southern part of site. From the water Test Trench 3 is near the right side of the photograph. (Linnamae 1971)
PLATE I
New Grove CkAm-01
Looking down at site area towards the North. Seated figure in mid-ground is at Test Trench 3.
PLATE II
Looking at eroding bank edge at southern part of site. From the water Test Trench 3 is near the right side of the photograph.
(Linnamae 1971)

Linnamae opened four small test trenches along the eroding bank at the beach near the center of the cove. From these trenches and observing the eroding bank she determined that the occupation layer only minimally extended beyond the area of her trenches. Despite the impact on the site they recovered 2694 pieces of cultural material including 225 artifacts and a sample of charcoal from the Dorset component which returned a date of 1730±80 (Gak-3276) BP (Linnamae 1971).

They recovered 42 endblades most of which exhibited grinding on the basal element and sometimes over the whole endblade surface. All of these were made on a white/grey chert which later became known as Trinity Bay chert among archaeologists. We now know that geologically this chert is actually part of the Conception Formation and more accurately should be referred to as Conception Formation chert. The extensive grinding and white/grey chert are typical of Trinity/Placentia Bay Dorset occupations. Linnamae and her crew also recovered 18 microblades, 14 bifaces, two pieces of ground slate, four abraders, one endscraper and a fragment of a soapstone pot (Linnamae 1971).

Conception Formation chert typically exhibits a whitish-beige or brownish-beige weathering rind that can be chalky, however when freshly broken surfaces display a blue-grey (battleship) colour. This colour and the distinctive weathering constitutes a clear Conception Group signature (LeBlanc 2008: 59).

New Grove Artifacts a-h endblades i blade fragment j endscraper k-o microblades (Linnamae 1971)
New Grove Artifacts
a-h endblades
i blade fragment
j endscraper
k-o microblades
(Linnamae 1971)

The recovery of one endscraper from a Dorset occupation is unusual, Linnamae noted this in her report and that endscrapers are usually one of the most frequent artifact categories found on Dorset sites. Therefore her preliminary interpretation of the site was that hunting and the manufacture of hunting implements were the primary function of the site (Linnamae 1971).

In 2002 I revisited New Grove with a colleague. We found a few small artifacts on the beach and noted that the site is continuing to erode.

New Grove in 2002
New Grove in 2002

Long Island Neck (CkAm-02) is located on a partially grass covered sand bar near the northern tip of Long Island, just two kilometres north of New Grove. There are high rock outcrops on both ends of the sand bar so the site was limited to just the sand bar. With almost no grass cover on the north end of the bar the majority of the site was found on the south end. Even when the site was found by Linnamae in 1970 it had undergone heavy erosion (Linnamae 1971). 

Long Island Neck CkAm-02 Looking down at site towards the North. In immediate foreground is the remaining grassy area containing a buried occupation area
Long Island Neck CkAm-02
Looking down at site towards the North.
In immediate foreground is the remaining grassy area containing a buried occupation area.

Linnamae and crew opened two trenches in the southern area and recovered 470 pieces of cultural material including 25 artifacts and a charcoal sample that returned a date of 2240±210 (Gak-3274). The site has a possible Maritime Archaic and a Dorset component. The date however is too late for the Archaic and would be one of the earliest dates for Dorset on the Island of Newfoundland if correct. Interestingly, the Bordeaux 2 (CkAm-05) site was found by Linnamae in 1970 just outside Arnold’s Cove and just 9 kilometres from Long Island Neck. It had a single Dorset occupation dated to 1090 ± 90 (Gak-3275) making it one of the latest dates for Dorset on the Island of Newfoundland (Linnamae 1971 & 1975).

Nine of the recovered artifacts are endblades, three of which are ground on the basal half. Four microblades and a microblade core were recovered. As well there were four fragmentary bifaces and a piece of ground slate. Given the small assemblage it’s hard to say much about the site; Linnamae did suggest that it represented a repeatedly used small hunting camp (Linnamae 1971).

Long Island Neck CkAm-02 a-f endblades g New Grove h blade i uniface jendblade k microblade core l knife fragment m microblade n side-notched point o biface p ground stone implement (Linnamae 1971)
Long Island Neck CkAm-02
a-f endblades
g from the New Grove site
h blade
i uniface
j endblade
k microblade core
l knife fragment
m microblade
n side-notched point
o biface
p ground stone implement
(Linnamae 1971)

I also revisited Long Island Neck in 2002. As at New Grove, the site was still eroding but not completely gone and we found a few small artifacts eroded out on the beach.

Long Island Neck, to the left is south, to the right is north.
Long Island Neck, to the left is south, to the right is north.
Close up shot of the south end of Long Island Neck.
Close-up shot of the south end of Long Island Neck.

Linnamae also found two sites (CkAm-04 & 05) on either side of Bordeaux Head which is a broad sandy point of land that separates Come by Chance from Arnold’s Cove. Bordeaux 1 (CkAm-04) was a very small site consisting of 13 artifacts only one of which, a retouched flake, was found in situ (undisturbed). Everything else was eroded out on the beach, including a partial Dorset endblade (Linnamae 1971). 

Bordeaux 2 (CkAm-05) was a more prolific site with more than 100 pieces of cultural material recovered including endblades and microblades from an in situ occupation layer. This layer also contained an arrangement of stones which Linnamae interpreted as a hearth. A carbon sample was taken and returned a date of 1090 ± 90 (Gak-3275) which, as stated above, is one of the latest dates for Dorset on the Island of Newfoundland (Linnamae 1971 & 1975).

Bordeaux 2 (CkAm-05) Looking along beach toward the North at Test Trench 1, which is located on the grassy slope in front of the trees. (Linnamae 1971)
Bordeaux 2 (CkAm-05)
Looking along beach toward the North at
Test Trench 1, which is located on the grassy
slope in front of the trees.
(Linnamae 1971)

In 2005 I was fortunate enough to visit the Bordeaux 2 site as well. Although no artifacts were found it is believe that the site is pretty much how it was when Linammae found it in 1971.

Looking at the Bordeaux 2 site.
Looking at the Bordeaux 2 site.

After this survey Urve Linnamae wrote her 1975 book The Dorset Culture- A Comparative Study in Newfoundland and the Arctic. For the most part this work is based on her excavations at the Cape Ray site near Port aux Basques and the Pittman site in White Bay. As stated earlier this was an influential work, particularly for anyone studying the Dorset culture on the Island of Newfoundland. For example, Doug Robbins in the abstract of  his 1985 MA thesis dealing with the Dorset at Stock Cove referred to Linnamae’s book as a landmark in the history of Newfoundland Dorset archaeology. While Cape Ray and the Pittman sites are at the centre of this work, the Dorset sites found during the lesser known 1971 survey are also incorporated and play a role in this significant work. 


LeBlanc, Sylvie
2008 Middle Dorset Variability and Regional Cultural Traditions- a Case Study from Newfoundland and St. Pierre and Miquelon. PhD, University of Alberta.

Linnamae, Urve
1971 Preliminary Report of an Archaeological Survey of Placentia Bay, Newfoundland.

1975   The Dorset Culture- A Comparative Study in Newfoundland and the Arctic.

Linnamae Biography, University of Saskatchewan
http://library2.usask.ca/spcoll/University%20Authors/UA2006-07/UA%202007%20Linnamae%20Bio.doc

Robbins, Doug
1985  Stock Cove, Trinity Bay: The Dorset Eskimo Occupation Of Newfoundland From A Southeastern Perspective. MA, MUN.

Labrador South Coastal Survey: 1992

Two weeks ago I told you about the 1991 season of the Labrador South Coastal Survey (LSCS). This was a two year archaeology project which started in 1991 and covered more than 600 kilometres of previously unexamined Labrador coastline. The 1991 survey area was between Cape St. Charles and Seal Island, Labrador. It was directed by Marianne Stopp, she was assisted by Doug Rutherford and crew. The 1992 survey started at Seal Island and finished at southern Trunmore Bay. It was again directed by Marianne Stopp, and this time she was assisted by Ken Reynolds and crew. More than 60 sites were found or revisited in 1991 and nearly 90 sites were found or revisited in 1992 (Stopp 1995, 1997, Stopp & Rutherford 1991, Stopp & Reynolds 1992).

Labrador South Coastal Survey. Yellow dots are sites recorded by the survey.
Labrador South Coastal Survey. Yellow dots are sites recorded from the survey.

Like the 1991 season, the 1992 season was funded through the Labrador Comprehensive Agreement which was administered through the Historic Resources Division of the Department of Municipal and Provincial Affairs, St. John’s, Newfoundland. Today this section of the division is known as the Provincial Archaeology Office and is part of the Department of Business, Tourism, Culture and Rural Development.

The geography of the 1992 survey area was characterized by fewer large bays and inlets than in 1991, but had a greater expanse of low elevation, sub-arctic tundra. This meant that more coastline was available for testing. The original mandate of the LSCS was to finish at Rigolet, but because of the greater available coastline surveying, southern Trunmore Bay to Rigolet would require another four to five week survey (Stopp & Reynolds 1992).

The 1992 survey ran through July and August and resulted in the recording of 76 new sites and a revisit of several previously recorded sites. The top three most common cultural occupations at the sites were 21 European, 14 Palaeoeskimo and 11 sites where the culture was Undetermined.

One of the possible European sites found (it may also be Inuit) is Creek 2 (FkBe-18) which consists of a two-room stone house northeast of Hare Harbour and west of Isthmus Bay. Test pits yielded kaolin pipe fragments, pearlware sherds, lead sprue, and fish, seal and bird bones. A smaller structure may also be a house which uses the natural bedrock bank as its back wall. Spalling of the rock suggests the fireplace may have been against this wall. The contemporary eroding shoreline just to the southwest of the structures contained large amounts of bone, ceramic and iron. The two features measure 10.43 m x 5.22 m and 5.14 m x 3.38 m (Stopp & Reynolds 1992).

This slideshow requires JavaScript.

Curlew Harbour 1 (FkBd-11) is one of the 11 sites where the culture could not be determined. The site has two stone features that were recorded in a raised shingle beach that is on the point of land between Isthmus Bay and Curlew Harbour. One feature is a rectangular cache composed of beach shingles with two large lintel stones covering its top. The inner chamber is 24 cm wide and 1.9 m long. Outside dimensions of the feature are 2.1 m x 3.3 m. Rocks have slumped to the sides and into the interior. The other feature is a cobble pit cache in the beach, located 70 m north of the cache, with its interior floor (1.3 m in diameter) lined with small beach stones. The pit measures 3.5 m x 3.5 m with a wall height of 0.75 m (Stopp & Reynolds 1992).

Curlew Harbour-1, FkBd-11, facing N, probable cache? in foreground (Stopp).
Curlew Harbour-1, FkBd-11, facing N, probable cache? in foreground (Stopp).
Curlew Harbour-1, FkBd-11, facing N, boulder pit (Stopp).
Curlew Harbour-1, FkBd-11, facing N, boulder pit (Stopp).

Porcupine Strand 1 (FkBg-07) is an Intermediate Amerindian site found in the deep blowouts of sandy, southern Trunmore Bay. Site elevation is about 5  masl and wind deflation has eroded out great bowls in which cultural material is found at, or near, sea level elevation.  The site has four loci. Lithic materials in those loci include Ramah & quartz crystal. Some calcined bone was also recovered (Stopp & Reynolds 1992).

This slideshow requires JavaScript.

Eagle (Grago) Island 1 (FiBa-01) is an Inuit or Innu site on the small island adjacent to the north shore of Eagle Island, in the Rocky Bay/Porcupine Bay confluence. It has a well defined oval tent ring (5.86 m x 4.4 m) alongside an equally well defined oval midden (2.5 m x 1 m). Tent ring test pits were sterile, while test pits in the midden yielded bird and fish bone, a European gunflint fragment, and 14 creamware sherds. These artifacts date the site to the late 18th or early 19th century, and the tent ring further suggests historic Inuit or Innu occupation.

Eagle (Grago) Island-1, FiBa-1, shovel pointing north at centre of tent ring test pit #1 (Stopp).
Eagle (Grago) Island-1, FiBa-01, shovel pointing north at centre of tent ring test pit #1 (Stopp).
Eagle (Grago) Island-1, FiBa-1, test pit #2 in midden (Stopp).
Eagle (Grago) Island-1, FiBa-01, test pit #2 in midden (Stopp).

The LSCS was significant for Labrador archaeology in terms of the amount of area covered and the large number of sites found. One of the more compelling sites is Huntingdon Island 5 (FkBg-03).  In 1992 the site was made up of two sod houses on the north shore of the small island in the mouth of Indian Harbour, on the west coast of Huntingdon Island. The larger of the two houses (7.6 m x 6.5 m) has rocks protruding above vegetation and the walls are slumped. The smaller house (6 m x 5 m) has an “L” shape with rocks throughout it (Stopp & Reynolds 1992).

Huntingdon Island-5, FkBg-3, facing N, sod house in foreground (Stopp).
Huntingdon Island-5, FkBg-03, facing N, sod house in foreground (Stopp).

Like several sites found during the LSCS, Huntingdon Island 5 has been revisited several times and has had extensive archaeological work carried out on it, in this case by Dr. Lisa Rankin of Memorial University and her Graduate students. Carrying out more extensive research has allowed them to show that the site is a multicomponent site with late precontact Amerindian, Inuit and European components. In fact it contains at least five sod-walled houses and several tent rings. Full scale excavation of some of the sod-walled houses has shown that they contain distinct bench areas, sleeping platforms, and lamp stand areas. Dr. Rankin has had several graduate students complete their thesis on research at this site. You can learn more about this site in Volume 12 of the Provincial Archaeology Office Annual Review for 2013.


Stopp, Marianne
1995 Long term coastal occupancy in southern Labrador. Unpub. ms., on file, Historic Resources Division, Dept. of Tourism and Culture, 53 pp.

Stopp, Marianne
1997 Long-term coastal occupancy between Cape Charles and Trunmore Bay, Labrador. Arctic 50(2): 119-137.

Stopp, Marianne & Ken Reynolds
1992 Preliminary Report of the 1992 Labrador South Coastal Survey.

Stopp, Marianne & Doug Rutherford
1991  Report of the 1991 Labrador South Coastal Survey.

Labrador South Coastal Survey: 1991

The Labrador South Coastal Survey (LSCS) was a two year archaeology project which started in 1991 and covered more than 600 kilometres of previously unexamined Labrador coastline. The 1991 survey area extended from Cape St. Charles to Seal Island (near Frenchmans Harbour), Labrador, and was directed by Marianne Stopp and she was assisted by Doug Rutherford. In 1992, the survey started at Seal Island and finished at southern Trunmore Bay, and was directed by Marianne Stopp and and she was assisted by Ken Reynolds. More than 60 sites were found or revisited in 1991 and nearly 90 sites were found or revisited in 1992 (Stopp 1995, 1997, Stopp & Rutherford 1991, Stopp & Reynolds 1992).

Labrador South Coastal Survey. Yellow dots are sites recorded by the survey.
Labrador South Coastal Survey. Yellow dots are sites recorded by the survey.

The Labrador Comprehensive Agreement provided the funding for the 1991 Labrador South Coastal Survey, which was administered through the Historic Resources Division of the Department of Municipal and Provincial Affairs, St. John’s, Newfoundland. Today this section of the division is known as the Provincial Archaeology Office and is part of the Department of Business, Tourism, Culture and Rural Development.

The various cultural occupations at the 1991 sites run the gamut of typical sites found in Labrador. The top three most common occupations were 20 European, 18 Palaeoeskimo and eight Maritime Archaic. The sites ranged in size from single spot finds of artifacts to the Pardy site which is listed at more than 60,000 m2.

The Pardy site extends along the northeast side of Spear Harbour for 250 m. It extends along a second beach terrace (18 masl) for the same distance, for a total area of 62,500 m2. The site is partially disturbed by the excavation of an historic period cemetery, trails, and wind erosion and was discovered through visual examination of blowouts. The blowouts yielded a large number of surface flakes and tools, and a sample was collected. Test pitting in the southeastern portion of the site indicates the possibility of two separate cultural levels, at depths of 8 and 15 cm below surface. The upper, later levels contain a predominance of Ramah chert, while a variety of lithic materials were noted from lower levels. The upper level is probably a Dorset occupation, with microblades and endblades recovered from the site. A charcoal sample was collected from the lower level for dating purposes resulting in a date of 5070 +/- 170 B.P. (Beta-48303), supporting the suggestion of a Maritime Archaic component at the site. This date is further supported by quartzite flakes in the lower levels of some test pits. Given the site size and artifact density, the site likely represents a major precontact occupation (Stopp & Rutherford 1991).

Photo mosaic of three slides showing the Pardy site along the shoreline in the foreground . (Stopp & Rutherford LSCS slides)
Photo mosaic of three slides showing the Pardy site along the shoreline in the foreground (Stopp LSCS slides).

The St. Francis Harbour Bight 1 site has both precontact and historic components. The precontact Dorset Palaeoeskimo component was wind deflated and the cultural material was collected from the surface. The site yielded endblades, bifaces, endscrapers and a quartz crystal core fragment. There was also a selection of flakes of chert, quartz crystal and slate collected. The artifact scatter was concentrated around a linear arrangement of stones, 3 m in length and 1.5 m in width (Stopp & Rutherford 1991).

Linear rock arrangement at the Dorset component of St. Francis Harbour Bight 1 (Stopp & Rutherford LSCS slides).
Linear rock arrangement at the Dorset component of St. Francis Harbour Bight 1 (Stopp LSCS slides).

The recent component, which was either Inuit or European, was made up of six sod houses, the dimensions of which averaged 5 m by 4.5 m. A test pit in one yielded a wooden button (possibly ebony), pipe bowl fragments, and refined white earthenware sherds. The houses are all located in close proximity to the present shoreline in three groups of two and offer a good vantage point for hunting seals in the nearby narrow tickle. A seal bone midden (St. Francis Harbour Bight 2), yielding artifacts dating to the same period, is located across the bight (Stopp & Rutherford 1991).

Photo mosaic showing the historic house pits at
Photo mosaic showing the historic house pits at St. Francis Harbour Bight 1 (Stopp LSCS slides).

Salt Pond Ridge 1 was a quarry/possible occupation site and the largest and richest of the Maritime Archaic sites recorded, with an extensive flake scatter. White and red quartzite and quartz crystal were abundant on a raised terrace. It is within easy walking distance of two further Maritime Archaic sites, Spear Harbour 1 and Spear Harbour 3, as well as two cobble features in a raised beach ridge on the opposite shore of Salt Pond Ridge 2 which may also be Maritime Archaic (Stopp & Rutherford 1991).

Photo mosaic of part of the Salt Pond Ridge site
Photo mosaic of part of the Salt Pond Ridge 1 site (Stopp LSCS slides).

A sample of the surface scattered cultural material was collected yielded a red quartzite stemmed biface, an asymmetrical, convex base quartzite non-stemmed biface, a distal biface tip and biface preform, flakes of clear quartz crystal, quartzite and red quartzite, and a clear quartz crystal preform and core. One test pit produced 36 quartzite flakes, 13 clear quartz crystal flakes, and 7 red quartzite flakes. The site is in close proximity to two other Maritime Archaic sites (Stopp & Rutherford 1991).

Salt Pond Ridge flakes (Stopp & Rutherford LSCS slides).
Salt Pond Ridge flakes (Stopp LSCS slides).
Salt Pond Ridge artifacts (Stopp & Rutherford LSCS slides).
Salt Pond Ridge artifacts (Stopp LSCS slides).

The final site I’ll discuss from the 1991 survey is Great Caribou Island 1. This site included two sod houses, middens, pit features in raised cobble beaches, two collapsed stone fox traps on a raised cobble beach, and flake scatters (within sod houses and elsewhere in the cove). The site was recorded following information given by residents of Caribou Run-Indian Cove (Stopp & Rutherford 1991).

The site has a precontact component and an undetermined cultural component, meaning the archaeologist couldn’t say for sure who made the component or when. The undetermined component was made up of 12 cobble pit beach features. No artifacts were discovered in the features, which probably represent storage pits or caches. The features averaged 141 cm by 134 cm and 59 cm deep (Stopp & Rutherford 1991).

Measuring one of the Great Caribou Island 1 pit features
Measuring one of the Great Caribou Island 1 pit features (Stopp LSCS slides).
One of the Great Caribou Island 1 pit features
One of the Great Caribou Island 1 pit features (Stopp LSCS slides).

The recent component is made up of two sod houses, both of which have been extensively disturbed by locals looting the houses for artifacts. Several artifacts were discovered within the looters’ back dirt, including three seal phalanges, two flow blue pearlware sherds, two kaolin pipe stem fragments, and one European gun flint fragment. The ceramics indicate a 19th century occupation (Stopp & Rutherford 1991).

Great Caribou Island 1 historic sod houses
Great Caribou Island 1 historic sod houses (Stopp LSCS slides).
Historic artifacts from Great Caribou Island 1
Historic artifacts from Great Caribou Island 1 (Stopp LSCS slides).

In two weeks I’ll discuss some of the sites found during the 1992 season of the Labrador South Coastal Survey.


Stopp, Marianne
1995 Long term coastal occupancy in southern Labrador. Unpub. ms., on file, Historic Resources Division, Dept. of Tourism and Culture, 53 pp.

Stopp, Marianne
1997 Long-term coastal occupancy between Cape Charles and Trunmore Bay, Labrador. Arctic 50(2): 119-137.

Stopp, Marianne & Ken Reynolds
1992 Preliminary Report of the 1992 Labrador South Coastal Survey.

Stopp, Marianne & Doug Rutherford
1991  Report of the 1991 Labrador South Coastal Survey.

The Bank site, Terra Nova National Park: Part 1

The Bank site (DdAk-05) was found by Dr. James Tuck during his 1979 survey of Terra Nova National Park. The site is inside Chandler Reach and is strategically situated at the juncture of three major channels, offering fine views down Clode Sound to the west, Chandler’s Reach to the east, and Goose Bay to the south. At that time of its discovery Tuck recovered two lithic artifacts, 98 lithic flakes and described the site as containing a predominantly Dorset Palaeoeskimo occupation. Most of the material culture was found in an eroding bank and a lack of evidence from test pits suggested the site was mostly eroded. The two identifiable artifacts recovered were a microblade and “an asymmetric knife with one straight and one convex edge and a deliberately blunted tip.” (Tuck 1980:37)

Despite the erosion present at many of the sites found in 1979, several sites were considered important enough to warrant further investigation. This was carried out in 1980 and during this time the Bank site was revisited and six more Dorset Palaeoeskimo artifacts and 43 unmodified flakes were recovered from the eroding bank. Three of the artifacts were endblades or endblade fragments, one was a scraper, one was a possible asymmetric knife or biface and the last was a bifacially retouched flake (Sawiki 1980: 96-71). The 1980 work in Terra Nova National Park formed the basis of a Master’s thesis for Anna Sawicki. The eroding nature of the Bank site and the small amount of recovered cultural material meant there was minimal interpretation of the site in Sawicki’s thesis (1983).

The Bank site around 1980 (Sawicki 1983)
The Bank site around 1980 (Sawicki 1983)
The Bank site around 1980 (Sawicki 1983)
Artifacts from the Bank site. a-c: Endblades d: Scraper e: Microblade f, g: Knives (Sawicki 1983)

In 1992 Dr. Fred Schwarz returned to the Bank site in part under contract to the Canadian Parks Service, and in part for Memorial University of Newfoundland under the terms of a research grant to the University by the Canadian Parks Service. Prior to his re-visit the general thoughts on the site were summarized by Schwarz as: The site thus appeared to be a typical Dorset site, with little potential for advancing our understanding of Newfoundland culture history, and, owing to its advanced state of erosion, little potential even for revealing anything new about the Dorset period (1993:6)

This attitude changed in 1991 when a Maritime Archaic stemmed projectile point was exposed and recovered during a routine monitoring visit. The projectile point was unusual for Newfoundland and its closest stylistic affinities were to be found in the Early Archaic in southern Labrador, dated to ca. 7200 BP. This find suggested that the site contained evidence of an Early Archaic occupation dating 1700 years earlier than any yet known on the island. The erosion of this now significant site prompted further investigations. Excavations were planned for the summer of 1992. In planning for this excavation the site was visited and more cultural material was visible in treefalls on raised ground beyond the eroding bank. It was also abundant in treefalls within the bog to the north. The latter also raised the possibility of preserved organic artifacts, a rare find on archaeology sites in the province. The Bank site had become much more important and much larger than previously believed (Schwarz 1993:6-7).

Excavations began on August 26 and continued until October 2. Archaeological operations at the site consisted of a combination of excavation, surface collection, screening, and test-excavation. The main excavated portion of the site, over 100m2, was stratified, with three distinct cultural layers extending across the site, in addition to more localized cultural features and deposits lying above and within the principal strata. During this work evidence was found for a small Maritime Archaic component, a major Dorset occupation, a Groswater occupation and Recent Indian (Beothuk ancestor) components (Schwarz 1993).

View West across excavations at the Bank Site (Schwarz 1993).
View West across excavations at the Bank Site (Schwarz 1993).
View East Across excavations at the Bank Site (Schwarz 1993).
View East Across excavations at the Bank Site (Schwarz 1993).

At the end of the 1992 work  the Maritime Archaic component was extremely small consisting of no more than seven artifacts including two chipped and ground slate axes, one side-notched projectile point, two possible tapered-stem bifaces, one possible lanceolate biface base and the Early Archaic projectile point found in 1991. In fact those seven artifacts were widely scattered across the site and showed no horizontal clustering which suggested to Schwarz that the Archaic occupation was never very intensive or was mostly eroded away. The 1992 excavations yielded no other evidence for an Early Archaic occupation, the remaining pieces all conforming stylistically to the well-documented Late Archaic occupation of Newfoundland (Schwarz 1993).

Maritime Archaic Artifacts Recovered at the Bank Site: a-b) Projectile points c-d) Possible tapered-stem bifaces e-f) Ground slate axes (Schwarz 1993)
Maritime Archaic Artifacts Recovered at the Bank Site:
a-b) Projectile points
c-d) Possible tapered-stem bifaces
e-f) Ground slate axes
(Schwarz 1993)

The Groswater component was slightly larger with 34 diagnostic artifacts recovered and 23 specimens that may be Groswater. The Groswater diagnostics include five semi-lunate inset sideblades, five sickle-shaped gravers, ten side-notched endblade bases and three complete examples, eight fine bifacially-retouched serrated endblade tips and midsections, two multiple-notched endblade bases, and one chipped and ground chert burin-like-tool. While Groswater artifacts were found in other sites in the area, the Bank Site Groswater assemblage is unusual for the high level of workmanship evident in the tools. In fact the workmanship was so fine Schwarz referred to the collection as “a strong, if not pure, component of Groswater artifacts in “Phillip’s Garden West” style.” (1993: 45) However, given the small sample size little else could be said about their occupation beyond it seems to have involved the full range of activities associated with a residential base-camp (Schwarz 1993).

Early Palaeo-Eskimo (Groswater) Artifacts Recovered at the Bank Site a-j) Side-notched endblades k-l) Multiple-side-notched endblade bases m-p) Serrated biface tips q-r) Hooked gravers s-w) Scrapers x-bb) Sideblades cc-dd) Side-notched knives ee) Large ground biface ff) Chipped and ground burin-like tool (Schwarz 1993)
Early Palaeo-Eskimo (Groswater) Artifacts Recovered at the Bank Site
a-j) Side-notched endblades
k-l) Multiple-side-notched endblade bases
m-p) Serrated biface tips
q-r) Hooked gravers
s-w) Scrapers
x-bb) Sideblades
cc-dd) Side-notched knives
ee) Large ground biface
ff) Chipped and ground burin-like tool
(Schwarz 1993)

The Dorset component was by far the largest at the site. It is represented by 1355 (93.4%) of the artifacts, by two distinct cultural strata (and elements of a third), and by five of the six cultural features identified at the site. Two of those features were interpreted as habitation structures with other features within the structures. One habitation structure was a semi-subterranean house with central axial feature flanked by cleared living spaces and was roughly oval in shape. The second habitation structure was similar to the first but less clearly defined. Both structures were backed by a midden area (Schwarz 1993).

The Little Passage complex, Recent Indian component at the Bank Site was also small, and highly localized, but distinctive according to Schwarz. Most of the Recent Indian component consisted of comer-notched projectile points and triangular bifaces. There were also a few undiagnostic pieces which Schwarz feels were Recent Indian. The diagnostic Recent Indian artifacts were strongly associated with Feature 1, a roughly linear deposit of fire-cracked rock and charcoal with lenses rich in calcined bone fragments measuring 2m by at least 5m. Similar linear hearth features found in Newfoundland and Labrador have been interpreted as the remains of communal feasting structures, because they are similar to shaputuan structures erected for mokoshan ceremonies, or “eat-all” feasts documented for the Innu of Labrador-Ungava. There were also seven Recent Indian biface tips, sixteen corner-notched projectile points and fifteen triangular bifaces localized to the Feature 1 area (Schwarz 1993).

Recent Indian Artifacts from Operation 10A2 at the Bank Site: a-m) Little Passage points (a-e are of Ramah) n-s) Triangular bifaces (n-q are of Ramah) (Schwarz 1993)
Recent Indian Artifacts from Operation 10A2 at the Bank Site:
a-m) Little Passage points (a-e are of Ramah)
n-s) Triangular bifaces (n-q are of Ramah)
(Schwarz 1993)

As unusual as Feature 1 was (In 1992 it was just the second such feature identified on the Island, today there are still just a handful of these features on the Island), the Recent Indian artifacts were more interesting for several reasons including because they were a very limited range of functional types; corner-notched projectile points, triangular bifaces, and perhaps one sidescraper. Schwarz believes the projectile points were from a chronologically-tight assemblage dating to the early-middle portion of the Little Passage Complex stylistic continuum ca. AD 1200. Missing from the assemblage are the lanceolate bifaces, endscrapers, linear flakes, etc. which might attest to a broader range of domestic activities on-site. Add to this the fact that diagnostic Recent Indian artifacts were closely associated with a single deposit, Feature 1. The Recent Indian occupation of the site was likely limited in both function and duration, to a single brief occupation. However, what is most distinctive and interesting about the Recent Indian artifacts is most were made from Ramah chert – a lithic material only found in northern Labrador and relied upon heavily by Labrador Recent Indians. While Ramah does infrequently occur on Newfoundland Recent Indian sites, it’s usually found as debitage. Finished Ramah artifacts are rare. Even today this site stands out; no other Recent Indian site on the Island has as many finished Ramah chert Recent Indian artifacts. The whole Recent Indian component suggests a brief, specialized occupation involving communal food preparation and the conspicuous consumption of valued objects (Schwarz 1993).

Charcoal and Firecracked Rock Deposit (Feature 1) in 10A2J. (Schwarz 1993)
Charcoal and Firecracked Rock Deposit (Feature 1) in 10A2J. (Schwarz 1993)

Since 1992 the Bank site has been revisited several times by three different archaeologists working with Parks Canada to monitor the erosion of the site. The next blog post will deal with those revisits.


Thank-you Lynne for providing the colour images.

Sawiki, Anna
1980 Archaeological excavations in Terra Nova National Park.

1983 Palaeo-Eskimo Occupations in Bonavista Bay, Newfoundland. MA, MUN.

Schwarz, Fred
1992 Archaeological Investigations at the Bank Site, Terra Nova National Park, Bonavista Bay, Newfoundland.

Tuck, James
1980 An Archaeological Survey of Terra Nova National Park.

Land-Use Applications and Jonathon and David Islands, Labrador

The Provincial Archaeology Office (PAO), in its capacity as a regulatory agency, determines the need for historic resources impact assessments through the review of land-use applications submitted by both government agencies, and the private sector. Collectively the four PAO staff members have more than 80 years of experience processing these applications. Over the last five years, on average, more than 2600 applications were processed per year. Those applications are often initiated by the private sector and come to the PAO through various government agencies or in some cases agencies within government initiate the applications. In either case the various agencies include Crown Lands, Environmental Assessment, Mineral Exploration, Quarries, Aquaculture, Interdepartmental Land Use Committee, Municipal Affairs, Forestry and Agrifoods Agency, Business, Tourism, Culture and Rural Development, Roads, Water and Sewer projects coming from engineering companies, Service NL and other projects. Once it is determined that the area doesn’t have archaeological potential or had already been surveyed, then the applications are processed fairly quickly.

Generally larger projects require more time to process the application. Large ground disturbing projects such as a new mine almost always require archaeological assessment. Even smaller projects with less ground disturbing potential such as water and sewer projects, the construction of a cabin or mineral exploration sites may require archaeological assessment if they are in areas with historic resource potential. This means the person/organization submitting the application has to retain the services of a consulting archaeologist. The archaeologist would then apply to the PAO for a permit to carry out the assessment which involves carrying out fieldwork at the site in question. Such was the case for two proposed mineral exploration sites on Jonathon Island and David Island north of Nain in 1995.

In 1995 there were more than 50 known archaeological sites within 20 km of Nain and if you extend that selection perimeter out to within 50 km of Nain, which includes Jonathon Island and David Island, the number of recorded sites in 1995 jumps to more than 360. Given the large number of known sites in the area, the potential for historic resources on Jonathon Island and David Island was very high and the call for archaeological impact assessment was more than justified.

Known sites within 50km of Nain are yellow dots, sites outside this radius are red dots.
Known sites within 50km of Nain are yellow dots, sites outside this radius are red dots.

The proponent for the mineral exploration project hired an archaeologist to conduct the assessment. The archaeologist found no archaeological sites in the immediate area of the proposed drill holes. However, a number of sites were identified outside the main drilling foci, but within the broader study areas. Evidence for a Maritime Archaic and Pre-Dorset presence were found on Jonathon Island. On David Island there was a series of Labrador Inuit tent rings and cache features as well as two possible early Maritime Archaic pit houses (Hood 1995). All of these areas were delineated and to be avoided by the proponent; once that was done the proposed drilling was able to proceed without any danger to historic resources.

In total seven new sites were found as a result of the impact assessment, four on Jonathon Island and three on David Island. I recently came across some slides from three of the sites found during the 1995 survey of David Island and one from Jonathon Island.

View to the west over David Island 1, 1995 (Hood)
View to the west over David Island 1, 1995 (Hood)

The three David Island sites range from find spots of flakes and a biface fragment, to a larger site with multiple lithic scatters and a tent ring to the largest site, David Island 1, that has seven tent rings, three caches and another structure that consists of a small semi-circle of rocks built up against an outcrop (Hood 1995). David Island 1 is an Inuit site with a precontact component (possibly Dorset) and is about 4500 min size. The site is located at the southeastern corner of David Island and the cultural features are 4-8 masl (metres above sea level). The seven tent rings are made up of a ring of rocks used to hold down the outside skirt of a tent. The rings are described as circular, sub-rectangular and oval and average just over 20 m2 in size, the smallest being just 3.5 mand the largest is 55.25 m2.

David Island 1 tent ring 2, 1995 (Hood)
David Island 1 tent ring 2, 1995 (Hood)
David Island 1 tent rings 3 & 4, 1995 (Hood)
David Island 1 tent rings 3 & 4, 1995 (Hood)

The three caches at David Island 1 consist of large flat boulders arranged so that they form a storage area for goods or food. They average about 0.84 mand about 0.5 m high.

David Island 1 cache 1, 1995 (Hood)
David Island 1 cache 1, 1995 (Hood)
David Island 1 cache 2, 1995 (Hood)
David Island 1 cache 2, 1995 (Hood)

David Island 3 is a probable Maritime Archaic habitation site that was found on the southern shore of Eastern Harbour which is on the southern end of the island. The site was composed of two possible boulder pit-house features that are approximately 30 masl.

Structure 1 is a 4 m (north-south) by 3.5 m (east-west) oval, lichen-crowberry filled depression within a field of head-sized boulders. The floor of the depression is ~ 25-40 cm below the surrounding rocks. There are no visible interior constructions, but there is one rather large boulder embedded in the floor near the front (seaward) side of the feature. No artifactual material was observed (Hood 1995).

David Island 3 possible Maritime Archaic pithouse 1, 1995 (Hood)
David Island 3 possible Maritime Archaic pit-house 1, 1995 (Hood)

Structure 2 lies 3 m west of Structure 1 and slightly up-slope. It exhibits a cleared, circular, vegetation-filled depression measuring 3.5 m in diameter, with the “floor” at 20 cm below the tops of the surrounding rocks. No interior features or artifactual materials were visible (Hood).

David Island 3 possible Maritime Archaic pithouse 2, 1995 (Hood)
David Island 3 possible Maritime Archaic pit-house 2, 1995 (Hood)

Based on Hood’s previous experience and the experience of other archaeologists who have worked in the area, these features are believed to be early Maritime Archaic pit-houses and likely dated to 6000 BP or earlier. In fact, this past summer another Maritime Archaic pit-house was excavated in Labrador to the south of this area and a radiocarbon date of 6720-6560 cal. BP was recorded based on charcoal recovered from the structure (See Jolicoeur, Brake, Fitzhugh & Davies in PAO Review for 2015).

Of the four sites on Jonathon Island, three of them had evidence for a Maritime Archaic occupation and the fourth had evidence for a Pre-Dorset occupation. Most of the sites were artifact spot finds or lithic scatters of flakes related to making stone tools. One site had a small tent ring and six small lithic localities over an area of ~45 m by 25 m in size. Another consisted of only two Ramah chert flakes associated with about five head-sized rocks arranged in a semi-circle, possibly forming a tent ring. The third site was made up of one quartz and one slate flake. The fourth site contained the only evidence of a Pre-Dorset occupation found on the Island and consisted of a black chert biface fragment, probably stemmed with a retouched impact spall on the tip. A piece of crystal quartz was noted on the surface about 10 m from the biface, but it was uncertain whether it was culturally modified (Hood 1995).

The location of Johathon Island 5, this site contained the Pre-Dorset black chert biface, 1995 (Hood)
The location of Johathon Island 5, this site contained the Pre-Dorset black chert biface, 1995 (Hood)

The historic resources assessment of Jonathon and David Islands is a good example of the assessment system working properly and to the benefit of everyone. The company was allowed to proceed and the PAO was able to protect the historic resources and we all learn more about the past of Newfoundland and Labrador.


Hood, Bryan
1995 Archaeological Resource Evaluation of Noranda Mines Mineral Exploration Areas at Jonathon and David Islands, Nain, Labrador.

The Samms’ site: A Groswater site in Norris Point

About a year ago I told you about my participation in the Northern Peninsula Heritage Inventory during the summer of 2000. I led the survey of a large area of the Northern Peninsula extending from Crémaillère Harbour in the south to Quirpon Island in the north and as far west as Raleigh. Prior to the start of that project I spent four weeks excavating units at the Samms’ site in Norris Point.

The site was on the property of a former private residence that had been converted to a Memorial University of Newfoundland Biology Field Station. In August 1999 a six-person archaeological crew under the direction of M.A.P. Renouf tested the field station lawn and potato garden and found a small number of Groswater Palaeoeskimo artifacts. In 2000 the testing was followed up with more intensive and longer-term testing. Our objectives were to test the site to determine the range of cultures present, the extent of any disturbance, and site size. Our results confirmed the presence of a Groswater Palaeoeskimo site, along with some 18th century and 19th century European material. However, the Samms’ site turned out to be extensively disturbed, and therefore no further excavation was warranted (Renouf, Bell & Hull 2001).

Norris Point is located in Bonne Bay, on the west coast of the island of Newfoundland. The site was located near the active beach on the southeast end of a peninsula that sticks out from the north shore of Bonne Bay. The peninsula divides Bonne Bay into the South and East Arm. Less than 200 metres to the west and at a slightly higher elevation than the Samms’ site is the location of a once-rich Maritime Archaic Indian and Palaeoeskimo site, known as Norris Point 1 (DjBl-02). While the Samms’ site was found by Renouf in 1999, Norris Point 1 had been known to archaeologists since the late 1920s. Unfortunately, like the Samms’ site, Norris Point 1 was heavily disturbed by European activities including looting (Renouf, Bell & Hull 2001). In fact the late Dr. Elmer Harp recorded in his field notes from 1949 that local boys would sell buckets of artifacts from the Norris Point site to tourists (Harp 1949). Interestingly, the Groswater Palaeoeskimo were first identified on the island of Newfoundland at Norris Point 1 by archaeologist Paul Bishop in 1973 (Bishop 1974).

Location of the Samms' site (DjBl-09) in relations to Norris Point 1 (DjBl-02).
Location of the Samms’ site (DjBl-09) in relation to Norris Point 1 (DjBl-02).
Looking over part of the Samms' site. The upper terrace in the back ground, to the left of the white house, contains the Norris Point 1 site (Renouf).
Looking over part of the Samms’ site. The upper terrace in the back ground, to the left of the white house, contains the Norris Point 1 site (Renouf).

Site testing took place from June 22 – July 14, 2000. The main objective was to test for undisturbed deposits which might warrant future full-scale excavations. With a crew of three, we excavated 21 mthroughout the property along a site grid. Initially excavation proceeded by troweling but as the extent of the site’s disturbance became apparent, we proceeded by shovel-shining. All backdirt was sifted through a 1/4″ screen. In almost all excavation units the levels were very badly disturbed. For example in one unit we recovered a white plastic egg at 34 cm below surface, a side-notched Groswater biface at 39 cm below the surface and a small piece of window pane glass at 43 cm below the surface. In another unit we recovered the base of a Groswater biface at 30 cm below the surface and the tip of the biface at 68 cm below the surface (Renouf, Bell & Hull 2001).

Tbe trowel is poinwg to a Groswater biface showing in the profile below a plastic egg and above a piece of glass.
The trowel is pointing to a Groswater biface
showing in the profile below a plastic egg and above a piece of glass.

A typical Groswater tool kit on the island usually consists of small and finely-made stone tools fashioned most often made from fine grained Cow Head cherts. It would usually include side-notched endblades (box-based), sideblades, scrapers (sometimes referred to as eared scrapers), microblades, side-notched bifaces and chipped and ground burin-like tools.

We recovered nearly 6000 flakes and 105 Groswater artifacts including microblades, bifaces, and chipped and ground burin-like-tools. The 105 artifacts were mostly made up of microblades, bifaces, and utilized flakes. Interestingly, the site contained just one endblade and two scrapers which is atypical for a Palaeoeskimo site. For example, the nearby Norris Point 1 site contained nine side-notched endblades, five side-notched endblade fragments and 30 scrapers (some of which may have been Dorset)(Bishop 1974). We also recovered a lot of chert cores and numerous primary, secondary and tertiary flakes. The predominant tools recovered suggest activities associated with animal butchering and other food-processing activities, while the chert cores and flakes suggest tool making activities and biface retouch and resharpening took place at the site as well. With this in mind we suggested that the Samms’ site was a low-lying and sheltered butchering station associated with the higher and more exposed main camp at nearby Norris Point 1. It is also possible that the two sites were occupied at different years or different seasons (Renouf, Bell & Hull 2001).


Bishop, Paul
1974  Final Report: 1973 Excavations at Norris Point, Gros Morne National Park.

Harp, Elmer Jr.
1949  Elmer Harps’s 1949 Journal Entries for Newfoundland and Labrador: Visited Areas and Sites

Renouf, Priscilla, Bell, Trevor, & Hull, Stephen
2001  Excavation of the Samms’ Site (DjBl-09): A Groswater Palaeoeskimo Site in Norris Point, Newfoundland.

Groswater Palaeoeskimo update

Shortly after distributing this post last week I was contacted by two different archaeologists both of whom pointed out things in this post that required correcting, which I am happy to do. After all, the point of this blog is to distribute information about this province’s past and I want that to be as accurate as possible. The corrections will be included in the post as red text.

Dr. William Fitzhugh (1972) first defined the Groswater Palaeoeskimo based on the excavation of seven sites in the Groswater Bay area of Labrador. It seems they were a highly mobile group that preferred to live along the coast. Based on site location they had a settlement subsistence pattern that exploited inner bay/inner island areas. While they focused on marine resources, terrestrial resources were also important. Their sites suggest that they were occupied by small groups.

The sites that Dr. Fitzhugh used to define the Groswater were found on the outer islands in Groswater Bay and they produced very small assemblages, maybe a dozen or so stone tools. Many of the tools were well used and broken, and no structures were found. The sites were interpreted as summer marine mammal hunting stations by small highly mobile groups. This resulted in a somewhat narrow understanding of Groswater culture. However, our understanding became much clearer with the discovery of the large Postville Pentecostal Groswater site in 1977 discussed in further detail below.

Currently there are 84 Groswater Palaeoeskimo sites recognized in Labrador. Interestingly there are more Groswater sites on the island, 97. The Groswater are the only precontact culture for which this is true. It is not clear if this is a function of Labrador needing more survey work or if it is actually a cultural preference. Did they prefer the Island weather or the available food or was it something else entirely?

Groswater Palaeoeskimo sites in the province.
Groswater Palaeoeskimo sites in the province.

The Groswater culture first shows up in the archaeological record just over 3000 BP at sites in Labrador. It is last recognized in the archaeological record on the Northern Peninsula of Newfoundland around 1800 BP.

Renouf 2003 summarizes Palaeoeskimo dwelling structures in Newfoundland and Labrador. In Labrador, Groswater habitation structures tend to be small oval surface structures, with mid-passage axial features, box hearths and slab pavements. But, in Newfoundland, there are no clearly defined axial features, only a single example of a box hearth (perhaps not, see below) and no slab pavements. The shape of these dwellings is variable, including oval, rectangular and bilobate. Most Newfoundland Groswater structures are defined in several ways including perimeter rocks, a discard perimeter, a ring of postholes or a perimeter of humus and sand. There also tends to be interior and exterior pit features associated with the dwellings (Renouf 2003: 386-387).

As stated above, there is only one box hearth on the island that is clearly associated with a Groswater occupation based on associated artifacts. It was found at Cow Cove on the Baie Verte Peninsula by Dr. John Erwin. The hearth was comprised of four partially upright slabs enclosing fire cracked rock (Erwin 2003). However, box hearths may have been found at L’Anse aux Meadows. Along with the Norse occupation, the L’Anse aux Meadows site contained numerous precontact aboriginal occupations including Maritime Archaic, Palaeoeskimo and Recent Amerindian. Unfortunately, according to Wallace 1989, the occupations are disturbed. ‘An infinite number of disturbed tent floors, tent rings and fireplaces were excavated on the south shore of the bay by Bengt Schonback for Parks Canada in 1974 and 1975. The features are hopelessly intermingled, one disturbed by the other.‘ Over the years several box hearths were found at L’Anse aux Meadows, none of them are clearly associated with a Groswater occupation, in fact one has a date of AD 670+/-100 (Qu-363) which suggests it is Middle Dorset. However, box hearths are rare on Middle Dorset sites. So, given the amount of disturbance at L’Anse aux Meadows it is not entirely impossible these features are Groswater, but we may never know for sure.

Several Groswater sites contain architectural remains (Reader 1997; Auger 1984; Loring & Cox 1986). Perhaps the best known architectural remains at a Labrador Groswater site come from the Postville Pentecostal site. In 1976 Groswater lithics were discovered near and under the local Pentecostal church by a Smithsonian field crew. The rain drip-line off the roof of the church had excavated a shallow trench that was literally paved with Groswater lithic debitage and artifacts. When the Smithsonian crew looked under the church they could see that it had been built on pilings which had minimally damaged the site. They learned that this church was to be removed and a new one built in the same location in 1977. So, they made arrangements to fly-in to the village in 1977 and excavate the site before construction of the new church. The project was a joint Newfoundland Museum and Smithsonian Institution project co-directed by Brenda Clark and Dr. Stephen Loring (with Philip Hiscock and Eric Loring as part of the team). Based on the material recovered along the coast by Dr. Fitzhugh, the crew expected to find a tent ring and a few dozen artifacts but were astonished by the richness and density of the site once it was opened up. They excavated or partially excavated several mid-passage type dwellings with hearths from a total of ten identified structures. They recovered nearly 2000 lithic artifacts including 25 box-based points, 43 side-blades, 61 notched bifaces, nearly 800 chert microblades and more than 100 quartz crystal microblades. A majority (56%) of the artifacts were made of a mottled reddish-brown, green, grey and tan Ordovician chert which likely comes from the island of Newfoundland. So the site was huge with an extensive lithic collection (Loring & Cox 1986).

Perhaps the most important thing about the Postville site is the story the lithics tell. Groswater sites, all the way up at the northern tip of Labrador at Nunaingok, contain distinctive box-based points made of Newfoundland chert, while the Groswater assemblages on the Island have some Ramah (from northern Labrador). This is evidence of the existence of an impressive exchange of raw materials and information throughout the Palaeoeskimo world and an impressive degree of mobility! Loring and Cox made the argument that maritime adapted people, living essentially in a linear, i.e. coastal, environment, need to construct social mechanisms that enable them to have access to distant resources and neighbours should the local resources become compromised (Loring & Cox 1986).

Perhaps the best known architectural remains at a Newfoundland Groswater site come from the Factory Cove site, near Cow Head. The Factory Cove site was found by James Tuck in 1976 and it was excavated by Reginald Auger in 1981. Auger uncovered several features including a tent ring with an outline of stone measuring 4×4 metres; a bilobate dwelling that contained a mid-passage hearth; and a lean-to dwelling.

Excavation of Factory Cove in 1981 (Auger 1984)
Excavation of Factory Cove in 1981 (Auger 1984).
Factory Cove in 2015.
Factory Cove in 2015.

The Groswater lithic industry includes the typical plano-convex, box-based endblades used in harpoons, unnotched endblades, a variety of bifaces, chipped and ground burin-like tools, ovate and circular sideblades, corner spurred unifacial endscrapers and microblades. The Groswater people typically used several different types of lithic raw material, the most common being Cow Head chert, however, soapstone, quartzite, nephrite and slate have also been found in Groswater sites.

The Groswater tool kit is based on flake reduction. The first step to make a Groswater tool is to detach a large flake from a core. Regardless of the shape of the flake it was thinned until it was suitable for further modification. Only then did the flaking for the final shape take place. In the case of bifacially flaked knives, the end product was dictated by the shape of the thinned blank.

The first excavation I was part of was near Cox’s Cove on the west coast of the island with David Reader. We estimated that the site had eight Groswater houses, two possible Dorset houses, and a very late Little Passage-Beothuk house as well as several middens with excellent faunal preservation. In 1997 we excavated an oval Groswater house measuring approximately 5×5 metres. The house was surrounded by a mound of discarded fire cracked rock, lithic material and faunal remains.

Excavation of the outer perimeter wall of a Groswater house at Parke's Beach (Reader).
Excavation of the outer perimeter wall of a Groswater house near Cox’s Cove (Reader).

This excavation had such an impact on me that I wrote my Honours Thesis on Groswater unnotched endblades. The Groswater made distinctive side notched endblades for hafting, so much so that they are often referred to as ‘box-based’ endblades. Examples a & b in the photo of Groswater endblades above are good examples. However, triangular or unnotched endblades often are part of a collection from Groswater sites. Examples l, m & n in the photo of Groswater endblades above are good examples. For my Honours Thesis I searched 75 unnotched endblades from the Factory Cove and Postville sites for use-wear using a microscope. I was curious to know if these artifacts were tools or just a stage to becoming ‘box-based’ endblades. The use-wear I found suggested unnotched endblades were used as cutting and or scraping tools; they were much more than just one of the steps to a finished product.

I have always personally found that this was an interesting culture to study and there is still so much to learn. For example, Anton looked at the relationship between the Groswater and the preceding Early Dorset and concluded that while contemporaneous, both groups tried to avoid each other through a division of land use. However, a recent genetic study suggested that the people of these archaeologically recognized Palaeoeskimo cultures were related. So why the avoidance? Why the different tool kits? Was it just a cultural difference, i.e. they were basically the same people but they just preferred to live differently? How much of a role did environment play in these differences? Is Groswater just a regional variant of Early Dorset and the later Dorset?  So many questions yet to be answered.


Anton, Elaine
2004 St. John’s Harbour 5 HeCi-30 and an Examination of Groswater and Early Dorset Relationships in Labrador.  MA, MUN.

Auger, Reginald
1984  Factory Cove: Recognition and Definition of the Early Palaeo-Eskimo Period in Newfoundland. MA, MUN.

Erwin, John
2003 A Groswater Palaeoeskimo feature from Coachman’s Cove, Newfoundland. Études/Inuit/Studies, 27(1-2),435-449.

Fitzhugh,W.W.
1972 Environmental Archaeology and Cultural Systems in Hamilton Inlet, Labrador. Smithsonian Contributions to Anthropology, 16, Washington.

Hull, Stephen
1997 A New Perspective on Groswater Palaeoeskimo Unnotched Endblades. Hons., MUN.

Loring, Stephen & Steven Cox
1986 The Postville Pentecostal Groswater Site, Kaipokok Bay, Labrador. Palaeo-Eskimo Cultures in Newfoundland, Labrador and Ungava. Reports in Archaeology No. 1, Memorial University of Newfoundland, pp 65-94.

Reader, David
1997 Archaeological Excavations at Parke’s Beach, Bay of Islands, 1996: Groswater and Dorset Palaeoeskimo and Beothuk Components.

Renouf, M.A.P.
2003  A review of Palaeoeskimo dwelling structures in Newfoundland and Labrador. Études/Inuit/Studies, 27(1-2):375-416.

Wallace, Birgitta
1989 Native occupations at L’Anse aux Meadows. DRAFT.