GEOMETRY AT POVERTY POINT
F. Romain and Norman L. Davis
than three thousand years ago, in northeastern Louisiana, Native Americans
built an enormous earthwork complex that is unique in all the world (Figure 1).
At a minimum, the complex covers 162 hectares (400 acres) (Gibson 2001:4) and
includes the second largest mound north of Mexico (Kidder 2011: 97). The
earthwork builders built this complex using sophisticated geometric shapes.
Moreover, they aligned these shapes to celestial events. The earthwork complex
is known as Poverty Point.
Figure 1. Artist's representation of Poverty Point. Drawing
by Herb Roe.
notion that Poverty Point was constructed using geometric shapes is not new.
Early-on the site was described as a series of concentric circles or octagons
(e.g., Ford and Webb 1956:16). Based on more recent ground surveys the site is
better described as a series of concentric C-shaped ridges. If the trajectory
of these ridges were to be extended into the swampy area off to the east, they
would form a series of nested ovals or ellipses. Likewise, the idea that the
site is aligned to celestial events is not new. More than thirty years ago,
Brecher and Haag (1980, 1981, 1983; also see Haag 1993) suggested that the
northwest and southwest aisles that crosscut the concentric rings were oriented
to the solstices. Contrary to this, Robert Purrington (1983; also see
Purrington and Child 1989) argued that the aisles were not solstice-aligned.
LiDAR analysis shows that Purrington was correct - the summer solstice sunset
is not in alignment with the northwest aisle. And while Brecher and Haag's
posited winter solstice sightline through the southwest aisle is a better fit,
it is off by about 6.5 degrees, or the equivalent of 13 sun diameters.
Purrington offered alternative stellar alignments; but ultimately, neither
Brecher and Haag, nor Purrington's alignments were widely accepted - basically
because accurate maps were not available for rigorous assessments.
the situation rested for several years until newly acquired LiDAR data were
utilized by the authors to re-assess potential alignments. LiDAR is an acronym
for Light Detection and Ranging. The technology uses reflected near-infrared
laser beams aimed at the earth from an aircraft, to develop accurate images of
the ground topography (Romain and Burks 2008a, 2008b, 2008c). Figures 2 - 5
show images for Poverty Point using LiDAR data retrieved from the Louisiana
Statewide GIS Atlas (http://atlas.lsu.edu/). As Figure 2 shows, the Poverty
Point complex is comprised of at least five mounds (not including the Motley
Mound about 2 km to the north and Lower Jackson Mound about 3 km to the south),
six concentric man-made ridges with swales in-between and crosscut by two or
three aisles, a central plaza, and several borrow pits.
E is a flat-topped rectangular-shaped mound, about 4 meters in height.
Approximately 30 meters of the southern extent of the mound has been truncated
in modern times by road construction and land leveling (Diana M. Greenlee,
personal communication 2013; Kidder, et. al. 2004). In Figure 2, the dotted
black line shows the estimated original footprint of Mound E.
B is a conical mound about 6 meters in height and 60 meters in diameter
(Ortmann 2010:665). It was built in several stages (Gibson 2001:87; Kidder et.
A is an enormous, combination platform mound, conical mound, and ramp. It is
about 22 meters in height, 207 meters in width and 210 meters in length (Ortmann
and Kidder 2013:67). It is one of the largest mounds in North America and
appears to have been built very quickly - in a matter of months (Ortmann and
C is best-described as an "oval shaped conical mound" (Ortmann
2010:668). Today it is about 60 meters in length. Its original width is
uncertain, as it has been eroded by the Bayou Macon.
At present the mound is about 2.6 meters in height,
with about 2 meters of the mound above the present plaza surface and 0.5 - 0.6
meters below the surface (Diana M. Greenlee, personal communication 2013).
Mound C is comprised of at least 16 thin layers of prepared surfaces of
different colored and textured soils at its base, capped by a 1.5-meter layer
forming the conical shape of the mound. Post holes suggested to Gibson
(2001:88) that "post-in-ground" structures were built on several of
the floors. Artifact evidence including hematite, red ochre, crystalline
quartz, and other materials led Gibson (2001:89) to further conclude that Mound
C activities were of a "sacred and ceremonial character." Related are
Ortmann's (2010:669) comments that: "The Mound C stratigraphy indicates it
was not a mound during its initial construction history. The series of prepared
floors were purposeful constructions that did not appreciably raise the height
of the feature." Further,
"Although it may not have been distinguishable from the plaza as a raised
architectural feature, the unique construction methods and unique signature of
cultural debris and features suggest it was differentiated from the rest of the
site as a specialized activity area" (Ortmann 2010:669).
As to its age, Ortmann and Kidder
(2013:75) find that: “Mound C, or at least part of it, was built and used while
the ridges were being constructed” (also see Ortmann 2010:672). Both Gibson and
Ortmann's conclusions about the special nature of the mound, and Ortmann and
Kidder's findings about the age of the mound are important points – for as
discussed later, Mound C was an ideal location from which to observe the
solstices during the Poverty Point occupation. Indeed, the discovery of
specially prepared ground-level floors that ostensibly marked the backsight
location for such observations is consistent with this suggested use.
Mound D is a small
rectangular mound, about 2 meters in height. Radiocarbon-dated charcoal from
deep within the feature and diagnostic artifacts date to Poverty Point times
(Ortmann 2010:667-668). OSL dates and ceramics from higher levels, however,
indicate that significant mound construction occurred during the time of the
Coles Creek culture (ca. A.D. 700 - A.D. 1200) (Greenlee 2011) - long after the
Poverty Point florescence. Mound D is located on the elliptic trajectory of
Ridge 2. The ridge itself, however, at least in the southeast area of the site,
has been severely impacted by modern topographic changes. As a result, it is
not possible to visually determine whether Ridge 2 extended as far east as
Mound D. As a consequence, opinions have been divided on whether Mound D is a
Coles Creek mound built on top of a Poverty Point ridge; or alternatively, is
it a discrete Poverty Point mound that was built-up by the later Coles Creek
culture and not part of the ridge system? The most recent data and opinion
support the idea that Mound D was built on top of Ridge 2 (Greenlee 2011;
Ortmann 2010:667). If that is the case then due to its location on the bluff
edge, Mound D likely marked the southeast terminus for Ridge 2. If Mound D
started-out as a discrete Poverty Point mound, then its location at the edge of
Bayou Maçon is an interesting choice, perhaps related to the extensive view to
the east from that location. Either way, Mound D apparently held special
significance - including its role in a solstice alignment, discussed below.
As to the rings,
or ridges, the northern ridges are relatively undisturbed and between 1 – 2
meters in height. The southern ridges are lower - i.e., 20 - 30 cm in height
and have been subjected to plowing, although they may not have been as
topographically distinct as the northern ridges to begin with (Diana M.
Greenlee, personal communication 2013).
The ridges are separated by depressions or swales. Ridge 1 is labeled in
Figure 2 and was central to the overall design of the complex.
Figure 2. LiDAR image of Poverty
Point showing Ridge 1 and mound locations.
sequencing of mound construction at Poverty Point is the subject of continuing
research. Gibson (2001:96) suggests that most mound and ridge construction took
place between about cal. 1600 B.C. to 1300 B.C. (but see Connolly 2006; Ortmann
2010; and Ortmann and Kidder 2013, who suggest a wider time span). Whatever the
exact sequence and time span of mound building, there seems little doubt that
there was a "massive building boom at Poverty Point" that ended with
the very rapid construction of Mound A at ca. 3400 to 3200 cal. yr. B.P.
(Ortmann and Kidder 2013:75). Of course this does not really address the
question as to whether or not the end product that we see on the ground today
is the result of a master design plan envisioned by the founders of Poverty
Point; or, is the final form the result of a "self-organizing
phenomenon" that presents the illusion of a planned design "because
its parts articulate so well" (Kidder 2011:117)? The question is not
unique to Poverty Point. Indeed, almost every large-scale prehistoric
manifestation - whether Poverty Point, Cahokia, the Newark Earthworks Complex,
or Chaco Canyon is subject to the same question.
and Kidder (2013:78) propose that: “The complexity of the construction process
at Poverty Point demonstrates the venture was well planned and organized, and
was undertaken to transform the landscape into a preconceived and desired
form.” To this we would add that, based on the seamless integration of site
orientation, celestial alignments, bilateral symmetry of design points,
internal geometry, regularities in mensuration and relatively rapid monument
building, we believe construction likely proceeded - at least for the most part,
pursuant to a master plan or design template. The caveat is that such a plan
did not necessarily need to include every mound and ridge - as long as
additional constructions incorporated basic design principles understood by
successive mound builders and were consistent with the basic master design. The
end results would appear the same. In either case, the following observations
support the notion that Poverty Point was built according to a preconceived
master plan, or again, at the very least, a series of design phases that
integrated astronomical alignments, geometric shapes, and local topography.
celestial north pole is the point in the heavens around which the stars and
constellations appear to rotate. For many cultures, celestial north is a center
place, both geographically and cosmologically. Accordingly, many cultures seek
to anchor or link their monumental architecture to this center point. One way
this is done is by orienting a site to true north. This works because, if an imaginary
line is drawn straight down from the celestial north pole, the resulting
direction on the ground will be true north. Since true north is a geographic
expression of celestial north, sites that are oriented to north are by
definition, astronomically aligned. (The same thing applies if the sun's
meridian transit is used to determine north.) With reference to Figure 3,
several observations suggest that Poverty Point was intentionally oriented to
1) The sightline between mounds
E-A-B (and the Lower Jackson Mound not shown here) extends north-south (Gibson
2) Ridge 1 follows the shape of an
ellipse, or oval. The major axis of this ellipse is aligned north-south and is
therefore parallel to the sightline between mounds E-A-B. Thus the north-south
axis for the site is expressed in two different, but mutually supportive ways.
3) Mound C is situated on the
north-south major axis of the inner oval.
Figure 3. LiDAR image showing Poverty Point design features
Design Points 1 and 2
locations appear to have been of special importance in the design of Poverty
Point. For convenience we refer to these locations as Design Point 1 (DP1) and
Design Point 2 (DP2). DP1 is situated on the east side of Bayou Maçon; DP2 is
situated on Ridge 1, directly opposite DP1. Both DP1 and DP2 are located where
the minor axis of the Ridge 1 ellipse intersects with the perimeter of that
DP2 is at a location that could have served as a ground-based observation point
as well as a design point on a master plan, the situation is not so clear with
regard to DP1. The uncertainty derives from the fact that the Poverty Point
site is situated on the edge of an elevated topographic feature known as Macon
Ridge. DP1 is located east of this ridge, in the Mississippi River floodplain,
at an elevation about 20 feet lower than the main site. As a result, in order
to see the sunsets over mounds E and B from DP1, DP1 would had to have been
either a 20-foot high mound, or a built-up wooden platform.
complicating the matter is that although DP1 is, today, situated in a wooded
swamp, Gibson (2001:7) suggests that at some point in the past, water levels in
that area were higher than they are today. If he is correct, the implication is
that DP1 was under water. Of course timing is everything; and even if water
levels were higher in the past, what we do not know is when that might have
been the case.
the moment, however, let us assume that DP1 was under water when Poverty Point
was built. Given that scenario, it may be that, rather than being an
observation point, DP1 was a focal point for design purposes. In other words,
if Poverty Point was built according to a scaled master plan drawn on a flat
surface, using predetermined solstice azimuths, then, DP1 was simply a design
point used to establish geometric and astronomic relationships between mounds.
In this case, the elevation of DP1 was a moot issue since the point was simply
one component of a design plan superimposed onto the landscape. For whatever
reasons, the main Poverty Point site was required to be open to the east and
situated on the edge of the Maçon Ridge. The result was that DP1 was going to
be located off to the east - in swamp or water, no matter what. That said, the
following observations support the notion that DP1 was a design focal point:
1) DP1 is situated where the
east-west axis and perimeter of the Ridge 1 ellipse intersect.
2) The major axis of Mound A points
3) The trajectory of two aisles
visible in the LiDAR imagery converge at DP1.
4) The sightline from Mound B
through Mound C points to DP1.
An east-west line drawn from a point
halfway between the southern edge of Mound B and the northern edge of Mound E
intersects the Ridge 1 ellipse at DP1. In Figure 3, the edges of mounds B and E
used to establish the halfway point are shown by dotted lines.
following solstice azimuths (Figure 4) are calculated for 1700 B.C., (although
observation dates +/- 500 years will not result in azimuth differences
discernable to the naked eye), using an estimated western horizon elevation of
0°.75 degrees, corrected for refraction and lower limb tangency (where the
bottom edge of the sun is tangent to the horizon). Since the horizon elevation
for Poverty Point thousands of years ago is unknown and depends on tree height
and the extent of area clearing, the following azimuths are likely accurate to
plus or minus one degree.
1) Line DP1 to Mound B is aligned
to the summer solstice sunset.
2) Line DP1 to Mound E is aligned
to the winter solstice sunset.
3) Viewed from Mound C, the summer
solstice sun will set over Mound B.
4) Viewed from Mound C, the winter
solstice sun will appear to set not over, but rather, into the side of Mound A.
The placement of Mound C near Bayou Maçon allowed for a long sightline to Mound
A, but also resulted in the location for Mound A in a place that seems
not-symmetrical with the overall site plan.
5) A line from DP1 through the
central plaza of the site marks the azimuth of the equinox sunset along the
northern edge of Mound A. One of the authors (Norman Davis) has witnessed this
phenomenon from the central plaza of Poverty Point on several occasions. On
these occasions the sun appeared to roll down along the northern edge of Mound
A before sinking into the western horizon.
the redundancy of the posited alignments, observations from DP1 were not
necessary if one wanted to observe these celestial events as linked to the
earthworks. Indeed, as a practical matter and given the location of DP1 off in
the swamp, Mound C may well have been the preferred location for actual
Figure 4. LiDAR image showing posited sunset alignments. SSS
= summer solstice set; WSS = winter solstice set.
addition to the alignments just noted, a near-perfect equilateral solstice
triangle is formed by Mound B - DP1 - Mound E, and a second such triangle
extends between Mound C - Mound B - Mound A (south edge). The same kind of triangle
(albeit to different sizes) is found at several other Archaic sites in the
Southeast (Clark 2004; Sassaman and Heckenberger 2004; Sassaman 2005) and is
sometimes referred to as a Clark-Sassaman-Heckenberger Triangle. An equilateral
triangle has three internal angles each equal to 60 degrees. A solstice
triangle has two sides that are congruent with solstice azimuths when one side
is oriented due north (Davis 2012). As shown, these conditions are met for the
Poverty Point Clark-Sassaman-Heckenberger triangle to within 2 degrees of arc.
Thus in addition to ellipses at Poverty Point, design triangles are in
a center place, Poverty Point was also place of balance in the sense that, in
addition to the sunset alignments just presented, conceptually opposite,
sunrise alignments are also found. These alignments, shown in Figure 5, are
again calculated for 1700 B.C., but this time using an estimated eastern
horizon elevation of 0°.1 degree, corrected for refraction and lower limb
tangency. As was the case for sunset alignments, azimuths are likely accurate
to plus or minus one degree due to uncertainties in tree height and clearing.
1) Viewed from DP2, the summer
solstice sun will rise over Mound C.
2) Viewed from DP2, the winter
solstice sun will rise over Mound D. (If in fact Mound D was constructed more
than two thousand years after the Poverty Point florescence, then, the
implication is that people of the Coles Creek culture understood, incorporated,
and further expanded upon the Poverty Point design for their own purposes.)
3) Viewed from DP2, the equinox sun
will rise in alignment with DP1.
one could raise various objections to the above – especially since neither DP1
nor DP2 have known marker features on the ground. On the other hand, DP1 and
DP2 could have been marked by poles that have yet to be identified by
postmolds. Or again, posited alignments may have simply been design components
for the overall layout of the site - not all of which would necessarily be
marked for viewing.
Figure 5. LiDAR image showing sunrise alignments. SSR =
summer solstice rise; WSR = winter solstice rise.
are many more interesting observations that could be added to this discussion.
There are, for example, additional findings relevant to the geometry and
mensuration of the site that have not been discussed. For now, however, the
foregoing introduces what we believe to be the basic astronomical alignments
and geometry of the site.
We cannot close
without noting that Brecher and Haag (1980) were right in their assessment more
than thirty years ago - i.e., Poverty Point does incorporate solstice
alignments. Brecher and Haag's proposed alignments are not the same as what has
been posited here. Nevertheless, as they suggested, Poverty Point may indeed be
the world's largest solstice marker.
Of course the
question that begs to be answered is: Why? Why was Poverty Point designed in
such a way that it connects geometric earthen forms to celestial bodies and
events at such a massive scale? Most likely there are multiple answers to this
question. And certainly we claim no direct connection between the builders of
Poverty Point and the Osage Indians. But there is an interesting narrative
given by an Osage elder to Francis La Flesche (1930:577) that might give a
No-ho-zhi-ga when formulating the tribal rites persistently held
up before the people the fundamental principle that in all their
activities as an organized body, a tribe, they musthave a unity of
purpose and a unity of action.|
They gave iterative emphasis to this
fundamental principle for the reason that
during their long years of contemplation of the great cosmic
bodies that move through the heavens in orderly precision they had
discerned the strength and power of this principle.
Perhaps by linking their great
earthwork to the heavens the builders of Poverty Point sought integrate the
strength, power, and order they witnessed in the heavens with their own earthly
existence. Perhaps in this they found meaning and purpose.
The authors wish to thank Chip
McGimsey, Dennis L. Jones, Diana M. Greenlee, and Kenneth E. Sassaman for
useful comments on earlier versions of this paper. The authors are solely
responsible for the content of this paper.
Brecher, Kenneth, and William G.
1980 The Poverty Point Octagon:
World's Largest Prehistoric Solstice Marker?
Bulletin of the American
Astronomical Society 12(4):886-887. (Abstract reprinted in 1981 in
1983 Astronomical Alignments at
Poverty Point. American Antiquity
Clark, John E.
2004 Surrounding the Sacred:
Geometry and Design of Early Mound Groups as Meaning and Function. In Signs
of Power: The Rise of Cultural Complexity in the Southeast, edited by Jon
L. Gibson and Phillip J. Carr, pp. 214-233. University of Alabama Press,
Connolly, Robert P.
2006 An Assessment of Radiocarbon
Age Results from the Poverty Point Site. Louisiana Archaeology 27:1-14.
Davis, Norman L.
2012 Solar Alignments at the Watson
Ford, James A., and Clarence H.
1956 Poverty Point, a Late Archaic
Site in Louisiana.
Anthropological Papers vol.
46, part 1. American Museum of Natural History, New York.
Gibson, Jon L.
1987 Poverty Point Earthworks
1999 Poverty Point: A Terminal
Archaic Culture in the Lower Mississippi Valley. 2nd ed. Anthropological Study
7. Department of Culture, Recreation and Tourism, Louisiana Archaeological
Survey and Antiquities Commission, Baton Rouge.
2001 The Ancient Mounds of
Poverty Point: Place of Rings.
University Press of Florida,
Greenlee, Diana M.
2011 Annual Report of the Station Archaeology Program
at Poverty Point Historic Site. Poverty Point Station Archaeology Program, Department of
Atmospheric Science, Earth Science & Physics, University of Louisiana at
Haag, William G.
1993 Archaeoastronomy in the Southeast.
In Archaeology of Eastern North America: Papers in Honor of Stephen Williams,
edited by James B. Stoltman, pp. 103-110. Mississippi Department of Archives
and History, Archaeological Report No. 25. Jackson, Mississippi.
Kidder, Tristram R.
2002 Mapping Poverty Point.
2011 Transforming Hunter-Gatherer
History at Poverty Point. In Hunter-Gatherer Archaeology as Historical
Process, edited by Kenneth E. Sassaman and Donald H. Holly Jr., pp. 95-119.
University of Arizona Press, Tucson.
Kidder, Tristram R., Anthony
Ortmann and Thurman Allen
2004 Testing Mounds B and E at
Poverty Point. Southeastern Archaeology 23:98-113.
La Flesche, Francis
1930 The Osage Tribe: Rite of the
Wa-xo-be. In Forty-fifth Annual Report of the Bureau of American Ethnology,
Smithsonian Institution 1927 - 1928, pp. 523 - 833. United States
Government Printing Office, Washington D.C.
Ortmann, Anthony L.
2010 Placing the Poverty Point
Mounds in Their Temporal Context. American Antiquity 75:657-678.
Ortmann, Anthony L., and Tristram
2013 Building Mound A at Poverty
Point, Louisiana: Monumental Public Architecture, Ritual Practice, and
Implications for Hunter-Gatherer Complexity.
Geoarchaeology: An International
Purrington, Robert D.
1983 Supposed Solar Alignments at
Purrington, Robert D., and C. A.
1989 Poverty Point Revisited:
Further Consideration of Astronomical Alignments. Archaeoastronomy
(Supplement to the Journal for the History of Astronomy) 13:S49-S60.
Romain, William F. and Jarrod
2008a LiDAR Imaging of the Great Hopewell Road. In Current
Research in Ohio Archaeology 2008. Electronic document,
http://www.ohioarchaeology.org/joomla/index.php, accessed March 13, 2013.
2008b LiDAR Assessment of the Newark Earthworks. In
Current Research in Ohio Archaeology 2008. Electronic document,
http://www.ohioarchaeology.org/joomla/index.php?option=com, accessed March 13,
2008c LiDAR Analyses of Prehistoric Earthworks in Ross
County, Ohio. In Current Research in Ohio Archaeology 2008. Electronic
document, http://www.ohioarchaeology.org/joomla/index.php. accessed March 13,
Sassaman, Kenneth E.
2005 Poverty Point as Structure,
Journal of Archaeological Method
and Theory 12(4):335-364.
Sassaman, Kenneth E., and Michael J. Heckenberger
2004 Crossing the Symbolic Rubicon
in the Southeast. In Signs of Power: The Rise of Cultural Complexity in the
Southeast, edited by J. Gibson and P. Carr, pp. 214-233. University of
Alabama Press, Tuscaloosa.