The Lost Colony of Roanoke

by Fullam, Brandon

  As noted, Strachey’s interviews with Machumps produced fragments of information about two “slaughters” that apparently involved Englishmen. The first of these had to have occurred in 1586 at Roanoke, and the second—the same slaughter reported by Smith—most likely happened in 1607 in the southeast area of present-day Virginia involving the Chesapeake tribe and the few Englishmen who dwelt with them. John White’s 1587 Lost Colony obviously could not have been part of the 1586 slaughter at Roanoke. Furthermore, if the few English victims of the 1607 slaughter were survivors of the 1586 event, then they also could not have been White’s Lost Colonists.

  All of this further supports the conclusions reached earlier, that the 1587 colony had ceased to exist after 1589. Many likely perished in a failed attempt to reach Newfoundland and almost all of any remaining colonists would have been victims of the hurricane and surge in 1589. Since the references to European styled houses and clothed men at various locations (with the possible exception of Pananioc) had nothing at all to do with the Lost Colonists, then where else could they have been? Furthermore, if the Lost Colonists were not involved in either of the events reported by Smith and Strachey—the 1586 “slaughter” at Roanoke or the 1607 “slaughter” of the Chesapeake tribe—then what became of them? The early 17th century Powhatan oral traditions contained memories of the 1566–68 Spanish presence in the Piedmont, as well as a victory by their fellow Algonquians over Englishmen at Roanoke in 1586, and of course the more recent slaughter of the Chesapeake tribe in 1607 along with a few of the aforementioned Roanoke survivors who had lived among them for more than twenty years. Noticeably absent from the Powhatan oral tradition was any memory whatsoever—with the possible exception of the clothed men at Pananioc—concerning the 1587 Lost Colony.

  Everything that the Powhatans related to Smith and Strachey was factually accurate, yet none of it had anything to do with the Lost Colony. It was only Smith’s and Strachey’s interpretation of that information that resulted in the false assumptions about the slaughter of the Lost Colony and the existence of survivors in faraway places. If the Powhatans related complete and accurate information to the English, and if we accept the premise that Indian oral tradition is an essentially valid and complete record of tribal history and memory, then logic dictates the following inescapable conclusion: Whatever happened to the Lost Colony was never part of the Powhatan oral history because the Powhatans had no contact whatsoever with the 1587 colony. This would be perfectly reasonable, of course, if the colony had collapsed decades before Smith and the Jamestown settlers arrived.

  As suggested earlier, the ramifications of this conclusion are critical for almost all past and present Lost Colony theories. If the Powhatans were not connected in any way with the fate of White’s colony, then obviously the centuries-old Powhatan–Lost Colony slaughter narrative, initiated by Smith, perpetuated by Strachey and Purchas, promoted by Quinn in the 1980s, and still held by many today, is logically impossible. All modifications of that theory must also be mistaken. As reviewed in an earlier chapter, a number of Lost Colony theories claim that White’s 1587 colony left Roanoke and settled in a variety of locations including present-day Bertie County, the Chowan River, Dasamonguepeuk, the Alligator River, or Weapemeoc. These theories usually conclude with the Lost Colony being attacked and destroyed by Powhatan’s warriors as they ventured northward in an attempt to make contact with the Jamestown settlers. Obviously if the Powhatans were not involved in any way with whatever happened to the Lost Colony, then all scenarios that involve the Powhatans in the Lost Colony’s demise are also logically negated.

  It is worth pausing here to recall the misunderstood line of inductive argument discussed in the Preface—proving a negative—since the Powhatan slaughter myth is one more in a long line of examples. Some of the most important clues to the Lost Colony mystery have been uncovered by similar instances of “proving” negatives. For example, it can be reasonably concluded that the 1587 colony did not relocate to the fort symbol location in Bertie County because, in addition to the documentary and logistical arguments provided earlier, the First Colony Foundation’s physical investigation found no archaeological evidence to indicate that the Lost Colony had settled there. Croatoan is another example. A number of early English artifacts have been found at Hatteras, a known contact location. However, from the contemporary documentary evidence and the extensive archaeological work conducted there, it can be reasonably concluded that the 1587 colony did not resettle at Croatoan upon leaving Roanoke. From the search of the Chesapeake by Vincente Gonzalez in 1588 it can be reasonably concluded that the Lost Colony did not relocate there either. From Francisco Fernández de Écija‘s interviews with the local Indians at Rio Jordon in 1609 it could be reasonably concluded that no English colony other than the one at Jamestown existed at that time. And like Carl Sagan’s previously mentioned dragon, we can reasonably conclude that a Powhatan slaughter of the Lost Colony did not happen because—if it had—there would have been legitimate evidence of its occurrence in Powhatan tribal memory.

  Since it can now be said that the basic information related by the Powhatans—including the two slaughter events—was essentially accurate, but misinterpreted by Smith and Strachey, then all versions of the Powhatan–Lost Colony slaughter scenario are invalid. All of Smith’s and Strachey’s suppositions (except maybe Pananioc) about the “lost company of Sir Walter Rawley” between 1607 and 1611 can be disregarded, in fact, because they are completely unrelated to the Lost Colony. The 1587 colony was never slaughtered by Powhatan’s warriors at the Chesapeake or as they ventured northward to make contact with the newly established Jamestown colony.

  Could such an event have occurred elsewhere in the vast Powhatan chiefdom? By the time of the Jamestown settlement in 1607 the Powhatan domain was indeed vast, having expanded to about thirty sub-tribes, each of which paid tribute to Wahunsunacock, Chief Powhatan, as their paramount leader. His realm stretched throughout the inner coastal plain from as far as present-day Washington, D.C., in the north all the way south to perhaps present-day Chesapeake, Virginia, and westward to the fall line. The existence and slaughter of an English colony anywhere in the Powhatan chiefdom would probably not have occurred without Wahunsunacock’s sanction and certainly not without his knowledge. Slaughters were conducted by Powhatan in 1607 and 1608, such as that of the Chesapeakes and the Payankatanks, and he was not reluctant to speak freely about them. A slaughter of an English colony anywhere in the Powhatan chiefdom would certainly have been part of recent tribal history and would have been related, as were the other slaughters, to Smith and Strachey during their inquiries with the Powhatans.

  It is also unlikely that a slaughter of the 1587 Lost Colony could have occurred beyond Powhatan’s domain. The territory to the south was occupied by other tribes belonging to the same Algonquian linguistic group, but they were not associated with the Powhatans. At the time of the initial English-Indian contact by the Barlowe-Amadas expedition in 1584, the Algonquians were one of the most populous and widespread native language groups in North America, extending all the way to the province of Quebec. The tribes encountered by the English in 1584, however, represented the southernmost extension of the Algonquian groups along the Atlantic seaboard.1 Anthropologist Frank Speck, specialist in the Algonquian and Iroquoian peoples, wrote that the Carolina Algonquians were comparatively recent intruders into the region and formed the last offshoot of the general Algonquian migration southward along the Atlantic Coast.2 It stands to reason, then, that the range of overall Algonquian oral tradition extended at least as far south as the geographic limits of Algonquian migration, which was towards the southern reaches of Pamlico Sound, near the area of the Pamlico and Neuse Rivers.

  It is obvious that the account of a specific event in Algonquian oral history would naturally be most detailed and accurately recalled within the tribal community where the event took place or in which its tribal members participated. Symbols woven into beadwork or painted on skins
, as well as tangible markers such as “heaps of stones” or branches marking locations on pathways, were used by the Algonquian tribes to commemorate important events.3 Physical markers such as these would help keep the memory of an event in sharper focus and also tend to “localize” the oral traditions of a particular tribe.

  Yet it is also clear that there was a shared, though less detailed, oral history among chiefdoms in the same overall Algonquian language group. Perhaps an apt metaphor to explain the transmission of oral tradition among tribes within a language group would be the ripple effect in physics, whereby concentric waves expand from the point of origin, gradually diminish, and eventually disappear. As an example, the Algonquian warriors from Secota, Aquascogoc, and Dasamonguepeuk who participated directly in the assault and routing of Grenville’s contingent at Roanoke in 1586 would retain clearly defined memories of the planning, details, and perhaps even individual warriors involved in the event. The Powhatans to the north, however, knew of the “slaughter at Roanoke,” but had only vague knowledge and virtually no specifics concerning this occurrence.

  Moreover, it is also probable that the Powhatans would have known something about the existence of an English colony even in the distant territories of another Indian language group. The Algonquian Powhatans obviously knew of clothed men—who turned out to be Spanish—at Ocanahonan and Pakrakanick in the Siouan Piedmont region, and they would have known of Europeans elsewhere as well. The previously mentioned Indians interviewed by Francisco Fernández de Écija in 1609, almost certainly Siouan, had remarkably detailed information about the English colony at Jamestown, in what was Powhatan/Algonquian territory 300 miles away. The Powhatans would very likely have been similarly informed about another, more recent European presence than the one at Ocanahonan and Pakrakanick four decades earlier. Since the Powhatans had been relating accurate information to Smith and Strachey, there is no reason to doubt that they would have told Smith and Strachey about the existence of another colony or about a major event such as the slaughter of a European colony … if one had occurred.

  It has been suggested that the Lost Colony may have met its end at the hands of the Tuscarora,4 known to the Roanoke voyagers as Mongoacks or Mangoags. This suggestion, though, is another outgrowth of the unlikely theory that the Lost Colony headed west, perhaps to the Chowan, and encountered the Tuscaroras, who dwelt even farther west. There is no evidence that an English colony or any colony of whites was slaughtered there. As already mentioned, in 1654 the Tuscaroras still retained a century-old memory which they related to Francis Yeardley about a conflict with Europeans, but the conflict involved the Haynoke tribe, not the Tuscaroras, and the Europeans were the Spanish, not the English. Surely the Tuscaroras would have retained even more vivid memories if they themselves had participated in the slaughter of a colony of Europeans.

  Locations of the Indian villages that participated in the 1586 assault at Roanoke beyond Powhatan territory (courtesy Michael Gayle).

  The Tuscarora, part of the Iroquoian language group, dwelt along the upper regions of the present day Roanoke, Neuse, Tar, and Pamlico rivers. In the “History of the Tuscarora Indians” portion of his 1881 volume, Tuscarora Chief Elias Johnson spoke of the time “Before the discovery, by Columbus” when the Tuscarora “had many years of enjoyment and peaceful possession of their domain, consisting of six towns on the Roanoke, Neuse, Taw [Tar] and Pamlico Rivers in the [present] State of North Carolina.”5 There is no mention of any massacre of whites prior to the beginning of European settlement near the Pamlico River in the 1690s, which are referred to as “the first colonies.” The first reference to contact with whites was the following:

  A little previous to these disorders, it seems that there were some white men, as our tradition states, with long coats and wide brimmed hats, visited several nations of the Indians in that neighborhood, and appeared to be very friendly toward them, wishing them success in everything, and told them that those settlers who were on the borders of their lands and constantly encroaching and committing depredations on the Indians, were not of the government, but were merely squatters, who settled of their own accord, and if they were cut off, there would be none to avenge them, and were advised to do so.6

  The “disorders” referred to are the capture of John Lawson, his execution, and the coordinated massacre of settlers along Neuse, Pamlico, and Trent Rivers, which initiated the Tuscarora War in 1711.

  The previously mentioned “Legend of the Coharie” did speak of a skirmish with a hostile tribe as the mixed group of hurricane survivors ascended the Neuse in what would probably have been 1589. If the event is at all accurate, the hostile tribe could historically have been the Tuscarora, but the 1587 colony had already ceased to exist. There could only have been a few assimilated colonists at most in this mixed group, and in any case the encounter would hardly have been considered a memorable event to the Tuscaroras. In the final analysis there is no evidence at all that the Lost Colony was ever slaughtered by anyone. The evidence does indicate that after September of 1589 there was simply no colony left to slaughter. At some time after the summer of 1588 at least part of the colony probably attempted to reach Newfoundland, but failed, and whatever remained of the mainland colony would likely have been swept away in the hurricane surge in September of 1589.

  Of all the Smith and Strachey references, the single potentially legitimate clue to the Lost Colony was the previously mentioned “men apparelled” at Pananiock. Unlike Ocanahonan and Pakrakanick, which were located in the Carolina Piedmont and were related to Juan Pardo’s excursions, Pananiock was situated between the Neuse and Pamlico Rivers. The only Europeans who could have been near that location in the latter part of the 16th century were the Grenville-Lane colonists of 1585–86 and the 1587 Lost Colonists themselves. Since the raw, uninterpreted information provided by the Powhatans was historically accurate in the other instances, the possibility exists that the references to “men appareled” at Pananiock may have been accurate as well. We do not know for sure whether Lane’s 1585–86 colonists explored Pananiock. Grenville’s week long excursion through Pamlico Sound in 1585 included visits to several villages, including Secota, and “diuers other places.” Ralph Lane reported that exploration from Roanoke had proceeded as far south as Secota, which was just across the Pamlico River from Pananiock territory. It is not clear, though, whether he was referring to Grenville’s excursion of Pamlico Sound or to another exploration conducted to that area afterwards.

  We also do not know with certainty that the 1587 colonists visited Pananiock, but it is a very likely possibility. As noted previously, the territory of Pananiock would have been a suitable choice for a mainland settlement in 1588. In addition to its geographic accessibility directly across from the Wokokon inlet, the Pomouik/Pananiock tribe had been at “mortall warre” with Wingina and may well have seen the 1587 colonists as welcome allies against a common enemy. Archaeologist William G. Haag made the following remark about the coastal region of the Pamlico River: “If the colonists left Roanoke willingly, this river might be one they would look upon with favor as a new homesite.”7

  Pananiock has already been discussed as a potential Lost Colony location, but not in terms of its credibility as part of Powhatan oral tradition. Since the references to clothed men at Ocanahonan and Pakrakanick have been shown to be historically accurate (but misinterpreted), it is likely that there were also clothed men at Pananiock. If so, it is possible that Wowinchopunck’s 1608 claim about clothed men at Pananiock could have been a recollection of the 1587 Lost Colonists. It is worth noting, too, that the Pananiock references appeared in early 1608, before Smith learned about the Powhatan “slaughter.” Once Smith’s misinterpreted slaughter report reached the Council, along with its equally erroneous tale of Lost Colony slaughter survivors at Ocanahonan and Pakrakanick, Pananiock became more or less irrelevant. Attention was exclusively focused on the more distant Ocanahonan and Pakrakanick where there were “men cloathed,” and “people with sho
rt Coates and Sleeves to the Elbowes” and “howses built with stone walles, and one story above another, so taught them,” as Strachey would later write, by those English who escaped the slaughter at Roanoke. The report seemed convincing, and it was quickly concluded that these must be Lost Colonists who managed to survive Powhatan’s slaughter.

  Overshadowed by Ocanahonan and Pakrakanick, and further diminished perhaps by Smith’s distrust of Wowinchopunck, the Pananiock references were unfortunately marginalized. Those references, however, deserve reconsideration. On the Zúñiga Map was the notation “Here the King of Paspahegh reported our men to be and wants to go.” Smith’s line in A True Relation was “We had agreed with the king of Paspahegh to conduct two of our men to a place called Panawicke [Pananiock], beyond Roanoke, where he reported many men to be apparelled.” If the presence of these men at Pananiock is accurate, they could have been 1587 colonists, in which case it would represent a significant step forward in the search for the Lost Colony’s mainland settlement location. It could either indicate that the colonists relocated to the mainland coastal area specifically between the Neuse and Pamlico Rivers in 1588, or that the survivors of the hurricane surge sought refuge there in 1589. “Many men … appareled” would suggest the former rather than the latter possibility.