"Columbus and the City of Gold" 
 
by:   William F. Keegan 
 
in:   "Journal of the Bahamas Historical Society."  Vol. 6, No. 1, 
      pp. 34-39, 1984. 
 
      Despite the passage of almost 500 years, the particulars of 
Christopher Columbus' voyage of discovery remain unresolved.  
Since speculation on the voyage began, almost every island in the 
archipelago has been identified as one of the four visited.  
While each new reconstruction is presented as a "theory", they 
are in fact hypotheses that require objective testing. 
      Objective testing is complicated by a variety of factors.  
First, significant environmental changes have occurred in the 
past 500 years.  Shorelines have been modified by deposition of 
sediments, beach sand has hardened into rock, estuary outlets 
have closed and sediments have filled the ponds formed by such 
closures, and the vegetation has been modified by years of slash- 
and-burn ("casual") cultivation and the removal of economically 
important trees (e.g., dyewood, mahogany, Lignum vitae, etc.).  
Therefore, an exact correspondence between Columbus' descriptions 
and present environmental conditions should not be expected. 
      Second, while it is logical to assume that the trade items 
distributed by Columbus and his crew would be found at the 
settlements they visited, it is not logical to assume that the 
discovery of trade items proves that Columbus visited a settle- 
ment.  Exchange and other forms of communication within the 
Lucayan Islands and between these islands and the Greater 
Antilles is well documented.  Thus, objects brought by Columbus 
to one village were probably redistributed to uncontacted 
villages. 
      Finally, Columbus' journal must be interpreted with great 
care, especially when English translations are used.  As with all 
reports on a foreign culture, the reporter translates his obser- 
vations and communications into his culture's language.  The use 
of these reports requires an examination of the reporter's 
ability to communicate with the natives and the stimuli that 
motivated the report.  Columbus was motivated by self-interest to 
make his discoveries sound as appealing as possible.  As Sauer 
suggests:  "The means of communicating with the natives were 
obviously poor, and Columbus supplied what he did not understand 
from his imagination." 
      Previous attempts to establish the route of Columbus' first 
voyage have relied on the translation of Columbus' journal, 
creative attempts to replicate the voyage, and the promise of 
discovering Spanish trade items at Lucayan sites.  However, no 
single form of evidence is sufficient, and an integrated approach 
must be initiated.  Such an approach must first reconcile the 
location of Lucayan settlements reported by Columbus with sites 
discovered through archaeological surveys on all four islands.  
Next, the congruence of these settlements with environmental 
descriptions must be demonstrated; this probably will require 
geological interpretations of changes in coastal geomorphology.  
Finally, trade items should, but need not, be found at these 
settlements. 
      That approach was first suggested by Theodoor DeBooy over 
sixty years ago.  It is being implemented by the author, and the 
results will be reported following the completion of archaeo- 
logical excavations and surveys on Long Island, Crooked Island, 
and Rum Cay.  The purpose of this paper is to report on the 
possible discovery of a settlement that was never visited by 
Columbus, but was responsible for providing direction to his 
otherwise random voyage.  That settlement was "Samaot . . . the 
island or city where the gold is." 
      This paper will review what is known of Samaot from 
Columbus' reports, and then present archaeological evidence that 
supports my identification of its location.  Although originally 
skeptical of Morison's suggested route (i.e., San Salvador to Rum 
Cay to Long Island to Crooked Island to Cuba) my research has led 
me to accept it with minor modifications.  However, this account 
relies primarily on Fox's translation of Columbus' journal.  It 
is superior to Morison's by its inclusion of the original Spanish 
alongside the English translation.  This permits the examination 
of passages whose translations are unclear or disputed. 
 
The Location of Samaot 
      Columbus first mentions Samaot in his report for Tuesday, 
October 16th, upon arrival at Fernandina.  However, he states 
that Samaot previously had been "asserted by those of the island 
San Salvador and Santa Maria."  At Fernandina, Columbus resolved 
to reach Samaot by the shortest possible route.  Despite the 
confusion he creates by reporting a series of different travel 
directions, the final result was a course to the southeast and 
east until he reached the southeast cape of Fernandina on the 
night of October 17th.  Continuing his voyage to the southeast 
and east Columbus arrived at the fourth island which he named La 
Isabela, but which the men that he brought with him from San 
Salvador called "Saomete." 
      Given the native designation of La Isabela as Saomete, the 
first task is to establish the modern name of the island at which 
Columbus landed.  Using Morison's reconstruction as a guide, 
Fernandina becomes Long Island and La Isabela is Crooked Island.  
The recent discovery of two sites on the northeast coast of Long 
Island provide support for this reconstruction.  Additional 
evidence is cited below. 
      Travel between those islands matches that described by 
Columbus.  From the southeast cape of Long Island he sent his 
ships to the south-southeast, to the east-southeast, and to the 
southeast.  After three hours they saw an island to the east, and 
arrived there before midday (about five hours sailing time).  
Crooked Island is located due east of the southeast cape of Long 
Island at a distance of 50 km (30 miles).  Provided with values 
for time and distance, Columbus' rate of travel would have 
equaled about six miles per hour.  In the journal, Columbus' rate 
of travel in another context is reported as between four and 
eight miles per hour.  Thus, the passage between Long Island and 
Crooked Island is consistent with the journal description. 
      Upon arrival at Crooked Island Columbus describes a rocky 
islet with rocks outside it to the north and between it and the 
mainland.  This description accurately portrays Bird Rock with 
the end of the barrier reef to the north and patch reef between 
it and the island.  Columbus also describes an extensive promon- 
tory to the northeast, near which he was unable to anchor because 
the water was too shallow.  Such a promontory occurs near the 
northeastern end of Marine Farm Salt Pond, and the water before 
this promontory is indeed too shallow to permit ships like those 
of Columbus to anchor near shore.  The physical descriptions of 
the northwest cape (Pitts Town) and the western side of the 
island as far as the southern end of Long Cay are sufficiently 
accurate to preclude detailed comparisons. 
      Columbus was informed that the population of Saomete resided 
"inland".  In this context inland ("mas adentro") refers to away 
from that shore across land, and not to the interior of the 
island.  This translation is supported by Columbus' contention 
that Crooked Island was a separate island from Saomete, and that 
a third island might be located between them.  The difference of 
opinion between Columbus and the natives concerning which island 
was Saomete may have resulted from Columbus' inability to distin- 
guish the name for the island from the name for the "city".  
Several spellings of the name appear in the journal Samaot, 
Samoet, Saomete, Saometo), which suggests that Columbus had 
difficulty understanding native denominations for locations. 
      The use of one name in two related contexts is fairly common 
in linguistic systems.  It provides a means of denoting relation- 
ship (e.g., parent-child), ownership (e.g., landowner-parcel of 
land), and forms of political integration (e.g., headman- 
village).  Fewkes reports this practice for the Taino Arawaks of 
the Greater Antilles.  He states that provinces generally were 
designated by the same name as that of the cacique (chief).  
Thus, Columbus may have been confused by the use of one name for 
both the island and the village of the paramount chief.  
Alternatively, two similar names might have been used with 
differences limited to a change in word ending or a slight change 
in pronunciation.  This would be similar to our use of the words 
city and state to distinguish between the two New Yorks.  
Therefore, Samaot (a.k.a., Samoet) probably refers to the 
village, while Saomete (a.k.a., Saometo) refers to the island on 
which the settlement was located or the province of the paramount 
chief. 
      The best evidence for the location of Samaot comes from the 
native's directions to the settlement from the southwest cape of 
Crooked Island (Long Cay).  Columbus reports that he was directed 
to sail across the lagoon (the Bight of Acklins) by following a 
course to the "northeast and to the east towards the southeast 
and south".  This course closely matches the configuration of 
Crooked Island and Acklins Island.  It suggests that the natives 
used the orientation of the lee shore as a guide for reaching the 
settlement.  Although dependent upon the distance traveled along 
each leg of the suggested route, it is unlikely that Samaot was 
located north of Spring Point, Acklins Island. 
      An alternative route would have been to sail to the south 
and approach the settlement from that direction.  Columbus 
rejected that route because he believed it to be too long.  The 
distance along the suggested route is approximately 58 km (35 
miles); travel along the southern route for a similar distance 
would have brought Columbus to the vicinity of Binnacle Hill, 
Acklins Island.  Therefore, a southern boundary near Binnacle 
Hill seems appropriate. 
      In summary, the descriptions from Columbus' journal 
adequately correspond to the geographical configuration of Long 
Island, Crooked Island, and Acklins Island.  Specific references 
to the northwest point of Crooked Island (near Pitts Town) match 
present physical characteristics:  The deep water that occurs 
before a sandy beach, the location of Bird Rock and associated 
reef, the extensive promontory, and the lagoon in which iguanas 
were captured (Marine Farm Salt Pond).  Finally, Columbus' 
intended route across the Bight of Acklins indicates that Samaot 
was located on Acklins Island between Spring Point and Binnacle 
Hill. 
 
Cultural and Archaeological Evidence 
      In July 1983, I directed an archaeological survey of Crooked 
Island and Acklins Island.  The details of this survey are 
presented in a report to the Government.  Of significance for 
this paper are the Lucayan sites discovered that correspond to 
the ones reported for the northwest cape of Crooked Island and 
those in the area defined for Samaot.  These sites will be 
reviewed in turn. 
      From the moment Columbus arrived at Crooked Island he was 
preoccupied with reaching Samaot.  As a result he spent little 
time ashore and provided few details concerning the island's 
inhabitants.  After two days of unsuccessful attempts to reach 
Samaot, Columbus returned to Bird Rock and anchored.  He went 
ashore and found a single, deserted Lucayan house. 
      That house would have been in the vicinity of Pitts Town 
Landing.  Our survey of this area did not uncover any evidence of 
prehistoric activity.  However, there is a low probability of 
finding the remains of a single house in an area that has since 
been disturbed by the construction of an airfield, residences, 
and hotel facilities. 
      It is probable that the immediate vicinity of the point 
could not support a large population.  Further, the occurrence of 
a site with only one house indicated a close association with a 
larger settlement.  In other cultures such a site would reflect a 
temporary fishing camp or agricultural field house.  It might 
also reflect the permanent settlement of a single family 
(hamlet), who moved a short distance from the main settlement to 
relieve localized pressure on limited resources (especially 
available agricultural land).  Therefore, the probability that a 
site with only one house existed near Pitts Town Landing is 
enhanced by the discovery of a large settlement in the general 
vicinity. 
      About two kilometers east of Pitts Town Point we discovered 
the largest site of our Crooked Island survey.  This site is 
apparently the village visited by Columbus during his search for 
water.  He reports that the village was half a league from his 
anchoring point.  Morison interprets Columbus' along-shore league 
as equal to between one and 1.5 nautical miles.  Thus, the 
distance from Columbus' anchorage to the village would equal 
between 0.93 and 1.4 km.  Since Columbus was anchored between the 
point and the settlement his estimate adequately falls within the 
two kilometer distance between these locations. 
      This archaeological evidence supports the identification of 
the northwest point of Crooked Island as the area visited by 
Columbus.  The next stage of research should be controlled 
excavation to determine whether or not Spanish trade items occur 
on the site. 
      During his travels Columbus was motivated by a desire to 
discover gold and other valuables.  As such his reports are 
coloured by continuous references to the impending discovery of 
riches.  This consuming desire to discover riches is the reason 
that reaching Samaot became an obsession.  An obsession that 
Columbus finally forsakes with the realization that the Lucayans 
possessed little gold and few riches:  ". . . although I place 
little confidence in their assertions, both because I do not 
understand well and because I see that they are so poor in gold 
that any small quantity worn by this King would seem to be a 
great deal." 
      If we assume that Columbus' descriptions of Samaot were 
reasonably accurate, then certain conclusions may be drawn.  
First, Samaot must have been the village of a very influential 
chief.  Columbus interprets the assertations of Lucayans from San 
Salvador as indicating that the chief was "master of all these 
neighboring islands".  Second, no person or community of similar 
status were mentioned during his visits to the three islands to 
the north.  The only settlements of superior status were those 
reported for Haiti and Cuba.  Thus, while Columbus was correct in 
recognizing that the Lucayans possessed little gold, Samaot was 
described as relatively much wealthier than other Lucayan com- 
munities.  Finally, the association of the name Saomete with an 
area encompassing both Crooked Island and Acklins Island indi- 
cates a level of political complexity approaching that reported 
for the Taino Arawaks of the Greater Antilles. 
      Settlement evidence can be used to determine whether a site 
should be designated Samaot.  The relationship of settlements 
when measured along a dimension of size provides a good indica- 
tion of their relative wealth and power.  For example, if all 
settlements are of similar size their inhabitants usually pos- 
sessed similar access to resources ("egalitarian"); a hier- 
archical arrangement of settlements by size would indicate 
differential access to resources ("stratified"). 
      A special form of settlement is the "primate community", one 
whose population size is at least twice that of the second 
largest, and is often greater than the combined population of the 
next three largest communities.  The growth of primate com- 
munities is based on the importation of capital (both human 
labour and other resources), which requires political manipula- 
tion to ensure the flow of resources into the community.  Such 
differential access to resources through the manipulation of 
exchange relationships is a common pattern in the evolution of 
chiefdoms and states. 
      From Columbus' descriptions we would expect Samaot to 
approach the size recognized for primate communities.  Since 
population numbers cannot be determined it will be assumed that 
site size provides an adequate measure of relative population 
numbers. 
      During the survey of Acklins Island we found a series of 
fifteen sites extending from the cay connected by causeway to 
Delectable Bay Settlement to the point across from Jamaica Cay, a 
distance of six kilometers.  These sites range in size from two 
potsherds AC-11 to AC-23 which measures 200 m by 20 m.  The sites 
are of such close proximity that they should be considered 
household and hamlet components of a single large site.  they 
were assigned individual numbers to facilitate recording and 
surface collecting during the rapid survey; hereafter they are 
collectively referred to as the Delectable Bay complex. 
      The total area covered by the Delectable Bay complex is 
about 120,000 square meters, of which 21,000 square meters are 
concentrated midden.  Due to environmental conditions settlement 
was restricted to a narrow coastal dune about 30 meters wide.  
The dune is backed by a hard pan depression that appears to flood 
periodically.  Other large sites in the Bahamas archipelago seem 
to conform to a roughly circular pattern with a central plaza 
surrounded by houses (e.g., MC-6, MC-12, Pigeon Creek Site).  
Therefore, the surface areas of plaza sites and the linear 
Delectable Bay complex are not directly comparable. 
      The Delectable Bay complex has the longest linear dimension 
of any known site in the archipelago (6 km).  Archaeological 
surveys of the other islands in the southern Bahamas have failed 
to reveal settlements that approach even one-fifth of that 
length.  The extremely long linear dimension for the Delectable 
Bay complex, especially in comparison to other Lucayan sites in 
the area, is consistent with the prominence assigned to Samaot by 
the Lucayans who directed Columbus, and with Columbus' obsession 
to reach the settlement after viewing its smaller relatives to 
the north. 
      In comparison with other sites on Acklins Island and Crooked 
Island the concentrated midden area of the Delectable Bay complex 
covers at least twice the area of its two nearest competitors:  
CR-13 with 10,000 square meters, and AC-27 with 9000 square 
meters.  When the entire extent of the complex is considered it 
is several magnitudes larger than those sites. 
      A final possible piece of evidence is the discovery of what 
appears to be a prehistoric house foundation at the largest 
component of the complex (AC-23).  Of special interest is the 
alignment of stones to form tow chambers.  These chambers are 
roughly rectangular with the larger (2.4 m by at least 2 m) 
facing the ocean, and the smaller (1.5 m by 1 m) attached to the 
rear.  Similar stone alignments are known only from Middle 
Caicos, Turks and Caicos Islands.  Sullivan has argued that two- 
chambered structures at MC-6 are the house/temples of the 
caciques, with the anteroom used for the storage of ritual 
articles.  His argument is based on ethnohistoric evidence for 
the Taino Arawaks in the Greater Antilles. 
 
Summary and Conclusions 
      Years of speculation have done little to resolve the par- 
ticulars of Christopher Columbus' voyage of discovery.  A variety 
of competing "theories" that appear to exhaust the range of 
possibilities have been presented.  Yet, these theories do not 
provide the criteria necessary to select between them.  To 
provide a final solution to these debates an integrated approach 
has been suggested. 
      Of all the available evidence, the remains of prehistoric 
settlements provides the best test for congruence between 
Columbus' descriptions and speculative reconstructions.  While a 
survey of every island identified as having been visited should 
be undertaken, this paper has focused on an island and village 
that were never visited.  Their significance obtains from their 
role in providing direction to Columbus' voyage through the 
Bahamas. 
      My research on Long Island, Crooked Island, Acklins Island, 
and in the Turks and Caicos Islands has led me to accept 
Morison's reconstruction.  It is supported, at least in its 
southern extent, by evidence presented in this paper.  The 
geographical orientation and physical characteristics of the 
islands and the location of prehistoric settlements all ade- 
quately correspond to the sailing data and descriptions in 
Columbus' journal. 
      The identification of the Delectable Bay complex as Samaot 
also is supported by a variety of evidence.  First, upon arrival 
at Crooked Island the Lucayans identified the island as Saomete.  
That Columbus believed Saomete to be a separate island suggests 
that the islands and the village bore the same names.  The 
variety of spellings suggests that the names derived from a 
single root word that probably was coincident with the Paramount 
Chief's name. 
      If that linguistic evidence is accurate it follows a pattern 
identified for the Taino Arawak.  Further, it is indicative of a 
level of political development expected of a chief who was 
"master of all these neighbouring islands".  While direct 
political "control" probably did not obtain, it is likely that 
the chief was capable of manipulating some relationships in the 
exchange network. 
      The location of Samaot in the southern Bahamas placed it in 
a superior exchange relationship with the developed chiefdoms of 
Hispaniola.  While direct exchange between Cuba and Long Island 
did occur, and may be responsible for the growth of large settle- 
ments on that island, settlements on Cuba did not achieve the 
complexity of those on Hispaniola and Puerto Rico.  The discovery 
of sites in the vicinity of Delectable Bay whose surface scatter 
had an unusually high percentage of imported pottery sherds (60 
percent; AC-27, AC-28; the average for the remainder of the 
Bahamas is 1 to 10 percent) indicates the close relationship that 
existed between this area and the Greater Antilles. 
      From these data we would expect Samaot to be the largest, 
wealthiest settlement in an interaction sphere that included San 
Salvador, Rum Cay, Long Island, Crooked Island and Acklins Island 
with connections to the Greater Antilles.  These expectations are 
fulfilled by the Delectable Bay complex with its fifteen house- 
hold to hamlet size components that cover an area of 120,000 
square meters with a length of six kilometers.  Presently the 
complex can be described as the primate community for the entire 
archipelago since it is at least three times the size of the 
second largest settlement. 
      The controlled excavation of the Delectable Bay complex 
should provide conclusive evidence of the settlement's age and an 
indication of its relative wealth and power.  Such excavations 
are being planned.  The evidence presented overwhelmingly sup- 
ports the identification of the Delectable Bay complex as Samaot, 
and demonstrates the utility of archaeological data for answering 
questions of historical significance. 
 
 
 
                                                           KEEGAN05.ART