THE ARCHEOLOGY OF COASTAL NORTH CAROLINA
HISTORIC INDIAN OCCUPATION
(Page 12) Today Indians of relatively pure stock are virtually non-existent in northeastern North Carolina. There are a few claimants to relationships with the Core Indians and the Croatan of Robeson County undoubtedly have some Indian components, but within the compass of our Banks study there is none that may be recognized as of Indian origin. It is difficult to understand and impossible to accept Speck's statement that one hundred surviving Indians were to be found on Roanoke Island in 1915 (Speck, 1916, p. 272; 1924, p. 188). That this region was once widely populated, though, is attested both archeologically and historically.
Coastal North America north of Spanish contacts is the best documented and first described by early English explorers. The English, as compared to the French and Spanish, were deplorably negligent in describing the native peoples with whom they had relations. However, it is the very area about the Outer Banks and immediate mainland that the best information is available. Several modern students have made the most of it, but Quinn's (loc. cit.) monumental accumulation and analysis of all the data pertinent to the Roanoke colony is a classic in historical research. Mook (1944) has made an exhaustive study of historical sources for the Algonkian tribes, his doctoral dissertation being more comprehensive than the portion published and cited here. He differs in certain details from Quinn but many problems relating to location of tribes are unresolvable on the basis of present information. Flannery (1939) has attempted a trait study of the Algonkians' material culture and social (Page 13) data. Swanton (1952) makes several references to Indians of the area in his encyclopedic treatment of North American Indian tribes. These are generally too brief to be of much value. More useful is the compendium on southeastern Indians (Swanton, 1946). Indispensible to any study of the Carolina country is Lawson's History (Lawson, 1937). However, the period of which he writes is one of Indians almost entirely enculturated with European traits.
In addition to these major sources there are innumerable references to neighboring areas or related materials. One of these should be cited, namely, Coe's Cultural Sequence of the Carolina Piedmont (Coe, 1952). Milling's Red Carolinians is useful but farther removed and does not include any Algonkian groups (Milling, 1940). Other sources are secondary; they usually are treatments of the above major citations.
Despite this seeming wealth of historical documentation there is an amazing lack of specific information about the native peoples encountered by the English colonists in the Carolina Sounds. This is true for even such items as geographical location of tribes, a fact that might have been expected to be of some political importance to the very earliest explorers. As a matter of fact, our best data do come from the very earliest, viz., Barlowe's discourse of the first Roanoke Voyage (Quinn, loc. cit., pp. 91-116); Lane's discourse (ibid., pp. 255-294); Hariot's, A Briefe and True Report (ibid., pp. 317-387); and, of course, John White's maps of the area.
From the above sources and modern analyses of place names and cultural traits, it may be established that the tribes inhabiting the Carolina Sounds from the Neuse River north were Algonkian speakers. (Page 14) This is clearly revealed in the names given to peoples and places by the natives themselves. Of course, the earliest Englishmen had no knowledge of the aboriginal language and frequently must have confused a tribal name and a place name. As an example, it is doubtful if the Roanoac Indians called themselves by this name, which means "northern people." Geary, the language authority for Quinn's monograph, suggests that the English may have named the tribes from their principal settlement (ibid., 1955, p. 870). There is an interesting alternative derivation by Gerard for the name Roanoac, namely, it means "rub," "abrade," "smooth," or "polish.", (Hodge, 1907-10; p. 392). The one arrow-shaft smoother or polisher found in the area was at Northwest Point on Roanoke Island, presumed site of the Roanoac Indian village!. (See Figure 13.)
Nothing in the historical documentation would suggest that any but Algonkian speaking aborigines were in the Sound area. This does not mean that the material culture of all Algonkians is the same; it quite obviously differs from Canada to Carolina, yet in an area so restricted as the one here under study, it might be expected that a high degree of cultural uniformity existed. Again, the early chroniclers suggest that the Indian tribes warred with one another and each was probably eager to take advantage of the presence of the English as allies. Lane hinted that the strained relations between the Secotan and the Pomouik (Pamlico) might be useful to the English (Quinn, loc. cit., p. 113). All of this indicates other than complete solidarity among the Indian tribes.
Epitomizing the known facts in so far as they relate to an archeological study, it would appear that all the area of the Carolina Sounds was inhabited by a single linguistic stock. Within this greater grouping were a number of smaller alliances. There are two references in (Page 15) Lane's account to Wingina (a chief or "king" of the Secotan) the domain of which extended from Roanoke Island to Secotan (on Pamlico River). There is reason to believe this was only rhetorical for when the time arrived that this same chief wanted to lead an uprising against the English not all the Indian groups would enter into the conspiracy. Thus it would seem there was a considerable amount of local autonomy.
The size of the villages is important. Barlowe said that the village on Roanoke Island consisted of nine houses (ibid., p. 107). White's drawing of Secotan shows only eleven houses. His picture of the arrival of the English in Virginia as engraved by De Bry indicates the villages are small and encircled by stockades (ibid., Figure 5). The accuracy of DeBry's engravings of White's drawings is obviously poor but an examination of the picture referred to above shows Roanoac village with eight houses, Pasquenoke has five, and Dasamonquepeue eight. It is probable that these representations are only conventionalized. Hariot found, "Their townes are small and neere the coast but fewe, some contain 10. or 12. houses: some 20. the greatest that we have seene have bene but of 30. houses: if they be walled it is only done with barks of trees made fast to stakes, or els with poles onely fixed vpright and close one by another" (ibid., p. 369). Hariot gives some dimensions of houses which imply their range of size to be from "36 ft. by 18 ft. to 72 ft. by 36 ft." (ibid., p. 370). It is probable that each village was populated by less than one hundred persons. Hariot mentions that the greatest "chief lorde" of which the English "had dealing with" commanded but eighteen villages and could raise only seven hundred or eight hundred fighting men. Unfortunately, none of the early observers indicates how many people resided in a single house but ten persons (Page 16) would probably be a generous number. Swanton has gathered together many references to show the small size of the villages of the coastal Algonkians (Swanton, 1946, p. 630).
Nevertheless, there is constant discrepancy between the population that a village so small could muster and claims set forth by Lane and others about the number of "fighting men". The obvious answer is found in Hariot's description of Secotan. "Ther are but few howses therin, saue those which belonge to the kinge and his nobles" (ibid., p. 631). The remainder of the peoples, who must have supported the king, lived scattered about the countryside in any area suitable for agriculture. Thus a single house with its family group would surmount a low, well-drained knoll about which the fields of corn where planted. Smith observed this in Virginia at Kegquouhtan (Kicotan), "Their houses are in the midst of their fields or gardens, which are small plots of ground. . . ." (ibid., p. 630). This condition is certainly borne out by the archeological investigations.
Next to nothing can be learned from the early accounts about pottery. White illustrates a vessel balanced upon a fire with the title "Their seetheynge of their meate in earthen pottes" (ibid., plate 54). An examination of a color facsimile of the original clearly suggests a pointed or nipple-bottom vessel. In the descriptive matter accompanying this illustration White says the pointed bottom was thrust into sand and the fire built around it very carefully.
Only the most minute scrutiny of these records gives any clues at all to the material culture of the peoples. Flannery has done, in part, such a search and now, with Quinn's work, a picture of the life activities of these coastal Algonkians may be summarized.
(Page 17) A common language was shared by all and within the Carolina Sound there were no linguistic barriers of any kind. At no time do any of the first colonists or explorers speak of using more than one interpreter. All the Indians were of a fairly pleasing (by European standards of the sixteenth century) physical make up with straight black hair, medium stature, and probably medium coloration. They lived in small groups although the village where the chief and his nobles resided was always the center of religious and political activity. The chief probably was supported through the efforts and contributions of his followers. There is strong suggestion that the office of chief passed down through the mother's side of the family.
Subsistence was a combination of agriculture and hunting and fishing. The staple agricultural product was maize and its planting was in two "crops", an early one harvested in mid-summer and a later one harvested in fall. No agricultural implement is known except the digging stick or dibble, although the account of Barlowe says they break "...a little of the soft turfe with a woodden mattocke, or picke axe" (Quinn, loc. cit., p. 105). Hariot observed "...the men with wooden instruments, made almost in forme of mattockes or hoes with long handles; the women with short peckers or parers, because they use them sitting, of a foote long and about five inches in breadth..." (ibid., p. 341). Flannery (loc. cit., p. 12) interprets this statement as meaning the dibble. There is no archeological evidence for stone or shell hoe blades.
Corn grown in the Banks area today does not attain a very large size. It is doubtful if yields near 20 bushels per acre could be realized now. Whatever they now are, the Indians of the sixteenth (Page 18) century probably equaled them, and no fertilizers were employed then either. Because Hariot was writing a propaganda tract he made glowing the promise of yields for Roanoke. Maize was not sufficient for subsistence but was supplemented by other vegetables. Sunflower seeds were used for "both a kinde of bread and broth." Kidney beans of perhaps at least two varieties, pumpkins, squash and gourds complete the list of known cultivated food plants. A number of wild food plants were utilized, of course, among which were grapes, berries of several kinds, acorns, chestnuts, walnuts, mulberries, and several edible roots. In the test excavation at Duck site, B5, many large specimens of tuckahoe, Poria cocos, were found. (See Figure 15.) The specimens were wholly carbonized and may date from the time of Indian occupancy. The form of this fungus, admirably fits the description of Hariot for one of the roots utilized by the Roanoke Indians but Quinn does not think "tuckahoe" fungus is implied (Quinn, loc. cit., p. 348). If these fungi grew wild, it does seem remarkable that they were concentrated in the area of an Indian village. This condition was most pronounced at B5 but it was observed at several other sites.
It is evident that both men and women worked in the fields although elsewhere it seems established that tending the fields was primarily women's work. Men assisted in clearing and harvesting. Since no fertilizers were used, the location of the corn field needed to be changed often. White's drawings would suggest that a corn field was to be found within the village area and this may have been the corn grown for the chief. However individual families probably raised their own corn although communal fields were known among other Algonkians. The size of a field which would yield enough corn for ten to thirty (Page 19) Indian families is not known, but the size of individual plots was that workable by one family and this was probably small, maybe under two acres. Archeological evidence clearly shows that nearly every well-drained low hill was the site of a family dwelling (see above). This largely accounts for the great number of small archeological sites in favorable areas and their total absence in others.
Although not a food plant, tobacco was an integral part of each planting. Hariot said it "is sowed apart by it selfe" and the powdered, dried leaves were smoked in clay pipes. Certainly tobacco was not introduced to England by any of the Roanoke group yet the form of the pipe may have been (ibid., p. 345).
Venison must have been of considerable importance in the diet of these Indians and deer probably constituted the major portion of this. Hariot considered the deer he observed to be as large as English deer which would have been the Red Deer. As suggested above, the deer of modern Hatteras Island are virtually pigmies but the bones recovered from the Cape Site, H 1, near Buxton were of average size animals. Rabbits were also important venison and next would be birds of several varieties. It is not known whether the beaver was present on the Banks but raccoon and opossum were plentiful. Bears probably did not frequent the Outer Banks but must have been numerous enough to have been a food item in the sixteenth century. A black bear, a victim of an automobile, was observed in 1954 near Lake Pungo and tracks were seen several times in that area. Bison are said to have frequented the area (Lawson, 1709, p. 115), but not one prehistoric specimen has been found in an archeological site. This is true not only for North Carolina but for all the United States east of the Mississippi River. (Page 20) Any specimens found are within the historic time period.
Quinn (loc. cit., p. 356) repeats some erroneous information passed on to him about a supposed Algonkian sixteenth or seventeenth century burial accompanied by a necklace consisting of 27 drilled teeth. "They comprised 18 Grey Wolf (Canis lupus lycaon Schrebes), 3 coyote (Canis latrans, not found east of the Alleghenies and therefore acquired by trade), 2 Raccoon, 2 Bear, 1 Puma (Felis concolor cougar) and 1 Red Fox (Vulpes fulvus, frequently thought to have been introduced from Europe). Together they indicate most of the larger animals killed by the Indians." The source is probably a newspaper account. The grey wolf, bear, cougar, and raccoon may be correct. The coyote cannot be since experience has shown it is impossible to differentiate wolf, dog, and coyote teeth except when they are in the skull (and then it is impossible in many borderline cases.) Red fox simply has not been shown to be present in eastern North America in pre-European times. This statement is based on personal efforts to check every reference to the red fox in archeological sites; each has proved to be a guess. Red fox and grey fox may be readily separated and identified only when dealing with whole skulls or skull caps.
Thus the animal assemblage was not imposing but generous. Smaller game such as squirrels, muskrats, marsh rats, even field mice, snakes, turtles, and lizards were constantly available and probably a part of the inevitable to "stew."
Birds included wild turkeys, doves, partridge, water fowls (the Banks are one of the great flyways for migratory ducks and geese) and undoubtedly anything that was big enough to eat. Hariot (ibid., p. 358) said "...of al sorts of fowle I have the names in the countrie language of fourescore and sixe….."
Despite the evidence that agriculture was a flourishing activity (Page 21) (at times) and that game animals were plentiful, it is most likely that the waters of the Sound supplied the most generous filling for the aboriginal larder. Barlowe may have been telling the first English language fish story of the Carolina Sound when he told of the first Indian fisherman with whom his party had contact, "....in lesse than halfe an howre, he had laden his boate as deepe, as it could swimme" (ibid., p. 98). Fish of many kinds were plentiful then as now, and were taken in every conceivable way, by nets, traps, weirs, lines, spears, etc. Shellfish, too, were often the most evident remains of the populace. Certain shell heaps on Colington Island are composed almost wholly of oyster shell. Oysters were found at the Cape Creek Site, H 1, on Hatteras Island, another suggestion that an inlet was near by. On Pamlico River oyster shells are absent from some extensive shell heaps, the shells being clams and mussels.
This summary could be greatly expanded especially in those traits not archologically significant. The intent here is to describe some aspects of the lifeways of these Indians of the Sound, especially aspects that might be archeologically preserved and may now be interpreted from their remains.
(Page 22) In view of the limited number of sites on the Outer Banks, every effort was made to enhance the information they yielded by closely relating them to mainland sites. Hence, the maximum time expended was along the mainland sides of the sounds. A few inland excursions were made to look for aboriginal remains that might be correlated with certain geologic phenomena, such as old beach ridges or terraces but this met with little or no success.
After a short orientation period, an archeologist will get the"feel" of an area and learn what spots are most likely to have aboriginal remains. The confidence in this "feel" becomes so strong with experience that one seeks some explanations for situations that are not fulfilled. Conversely, when archeological remains appear where not expected, again one must seek adequate explanation for their presence. (See the description of Gum Point site, p. 26.) Certainly, within the area of our study the Indian peoples lived--apparently always--on the shores of bodies of water. Since a total survey of the area was impossible, efforts were directed toward getting an adequate sample of all portions of the area and a detailed sample from selected localities.
Although the fact is often deplored by ethnologists, pottery is virtually indestructible. Only in rare circumstances are these carelessly-cast-aside fragments of vessels so altered by subsequent conditions as to render impossible the revisualization of the pot from which they came. (Page 23) Also--and again for unknown reasons--pottery from given times, within the same culture, and often over considerable geographic extent, is nearly identical. Perhaps this is because pottery-making is a comunity wide chore about which there were few secrets and generally was wholly public. Then too, primitives rarely consider the innovator with enthusiasm; doing things as the forefathers did was an ideal. Thus many kinds of pottery had long histories.
Within the Cape Hatteras area one often found the archeological remains just at the water's edge, an expected situation, of course. However, wave action has a seriously destructive effect on the "virtually indestructible." Wave action in this coastal area is always effective and sometimes becomes a tremendous force. As a consequence, at a few sites the potsherds were so badly worn as to render them almost useless for typological studies and in some instances the actual midden soil of the site was gone, only the heavier stones and sherds were to be found in the water.
These and other factors were constantly presenting themselves and all field techniques were ordered by them. Because a few sites gave evidence of least disturbance, they were chosen for most concentrated work. At these sites stratigraphic data were the sought-for information. In most cases, cultural materials were collected by three-inch levels from top to bottom in one or more test pits. As a result of these efforts, greater confidence may be had in the order of introduction of artifacts, especially pottery types.
Areas of Concentration
(Page 24) Because of the two-fold aim of the survey, namely, to depict the prehistory of coastal man and to unearth any evidence of the "Lost Colony," the immediate Cape Hatteras area received as minute an examination as feasible. Hatteras Island was surveyed thoroughly, whereas it was quite obvious that Ocracoke Island merited only a short visit. Since much of the main body of Hatteras is dune sand, some sites probably are buried. On Bodie Island a site that had been visible and described in 1938 was now buried and impossible to locate. On Roanoke Island the land surface of 1585 near the site of the fort was under more than ten feet of drifted sand. Colinton Island with its large shell heaps seemed a promising locality and, of course, Roanoke Island, although already the subject of certain archeological investigations, was thoroughly studied.
Currituck peninsula gave every evidence that it was an old beach ridge that had been superceded by the present beach line. If this were true, some age differences in archeological content might be apparent between the two areas. Hence, Currituck received a close examination.
Upon seeking sites of considerable extent and depth along the mainland shores of Albemarle and Pamlico Sounds certain facts were seen immediately, and the most obvious of these was that the north shores of these estuaries were rather heavily populated in prehistoric times and the south shores seemed to have been occupied to a limited extent only. Curiously, but perhaps for the same reasons, this is true today. Only within the past decade or two has there been any serious attempts to make homesites along the south side of Albemarle and Pamlico. Local residents account for this by the fact that the prevailing summer breeze is from the south; thus the north shore is blown free of mosquitoes and the south shore is quieter, (Page 25) more humid, and insect ridden. Of course, the winter storms are from the northeast, hence the southern shores of the sounds would be more exposed. Despite the lack of sudden violence characteristic of the summer hurricanes, the northeasters are bitter and harsh enough to require humans to be in a well-sheltered place.
A different interpretation of this may be entertained. Older residents along the Pamlico and Albemarle attest to the amount of attrition of the shore during the last century. Discounting a considerable portion of this, one still must accept the fact that the shores are being reduced. Thus it is probable that archeological sites once lined the south shores of these estuaries but have subsequently been removed. There certainly are several large middens well inland from the shore. However, it is difficult to see how this fact applies to the south shore only. Certainly the effects of hurricanes are more pronounced on the north shore. During the fall of 1954 a 30-foot slice was removed from a site (P 1) visited during the preceding summer and more was removed during the violent attack of hurricane "Connie" in August of 1955. It seems quite evident that storms have materially altered the shores of both Albemarle and Pamlico in the 250 years since there was an extensive Indian population in the area.
Hence, early in our survey it was established that the maximum amount of information would be recovered by exploiting two sources of information: first, extensive surface collecting was done; and second, any site that showed considerable depth of midden was stratigraphically tested. Several areas offered both possibilities and they were selected for maximum attention. On Albemarle Sound it was hoped that some sites would show traits (Page 26) that would relate the area to Virginia archeological remains. The Bandon Site, A 12, was most revealing in stratigraphy and a good chronological sequence could be demonstrated. At the other extreme, along the Neuse River, no site of considerable depth of midden was found, but one site produced a wealth of pottery that permitted judgments of time of occupation. The distribution of all sites located is given in Figure 2.
Carolina Algonkian Project