|
|
ASTRONOMY
OF AFRICA |
|
 |
Introduction The modern study of
sub-Saharan African (i.e. non-Egyptian) archaeoastronomy and ancient calendrical reckoning has a varied history. Although investigations have been made for well over a century for some sites, these have sometimes been hampered by ignorance of proper archaeological techniques, highly speculative conjectures about astronomical alignments, and even political policy opposed to the scientific evidence being brought to light (Great Zimbabwe Ruin, for example). Thus archaeoastronomy of sub-Saharan Africa can still be said to be in its infancy. We will therefore sample, in this article, a few cases of interest that can point the way for future directions of research. These will include an extensive look at the suggested archaeoastronomical site called "Namoratunga" in northwest Kenya along with the calendrical system of the Borana of southern Ethiopia that it has been suggested to represent. We will then mention preliminary work on the possibly Egyptian-like calendrical system of the ancient Kush of central Sudan followed by a sketch of the discovery of possible lunar symbols in certain caves of Tanzania as an indication of many sites that await investigation in this area. We then mention some of the initial archaeoastronomical work being done on the megalithic site of Great Zimbabwe and some of the difficulties involved. Finally, we survey the controversial evidence connecting the Dogon of Mali, in west Africa, with the binary companion star Sirius B. These investigations should serve as good examples of the kinds of research that are presently in progress on the astronomical ideas and methods of the ancient inhabitants of sub-Saharan Africa. |
|
Namoratunga and the
Borana Calendar
In 1973 A. Legesse first
documented the unique calendrical system of the Borana -- a Cushitic-speaking people of
southern Ethiopia and northwest Kenya (Legesse 1973). In this system calendrical time was
determined using only seven special stars in "conjunction" with various phases
of the moon (the sun, except indirectly via lunar phases, is ignored). In 1978
B.M. Lynch
and L.H. Robbins discovered, near Lake Turkana in Kenya, a site of nineteen stone pillars
called by the local people "Namoratunga" meaning "stone people", and
suggested that the stone pillars there were used for the alignments needed to derive the
Borana calendar (Lynch and Robbins 1978). Petroglyphs on these pillars were similar to
those on stones at a burial site (also called Namoratunga) 100 kilometers to the south
which was dated at 300 B.C. Thus an ancient age was indicated for the astronomical site
(called Namoratunga II) because of its association with the burial site (Namoratunga I).
The local Turkana people have names for each of these petroglyphs and indicate that they
are very old and relate to long-time family names. Later investigations found
the pillars to be magnetic (Soper 1982, Doyle and Wilcox 1986) and they consequently had
to be remeasured to see if the astronomical alignments persisted. The new measurement
positions are shown in Figure 1a. In addition,
the working of the Borana calendrical system they were supposed to represent was found to
be astronomically inconsistent, requiring a reinterpretation of the original description
of its working (Doyle 1986, reviewed in Thomsen 1984). Finally, the most recent field work
on the Borana calendrical system (Bassi 1988) indicates that the present working of this
calendar may differ from the historic one. Let us first examine how the Borana calendrical
system may have worked and then see how one might determine if the Namoratunga site was
indeed used as an astronomical calendrical marker for this system. |
|
As described by Legesse (1973) the
Borana calendar uses six star positions marked by seven stars or star groups. These seven
star groups are: Triangulum, Pleiades, Aldebaran, Bellatrix, Orion's Belt, Saiph, and
Sirius (for the Borana names of stars, days, and months, see Legesse 1973 or Doyle 1986).
New year starts when a new moon is seen "in conjunction" with Triangulum. Since
the lunar phase (synodic) cycle is 29.5 days long while the lunar position (sidereal)
cycle is 27.3 days long, the moon will arrive at the same position on the sky two days
before it has completed its phase cycle. The Borana have 27 day names (no weeks) and
continue counting days so that at this point the two day names are repeated, starting the
second month on a new day name about 29 or 30 days after the start of the first month
(depending on the observation). The second month starts when the new phase moon is seen
"in conjunction" with the Pleiades, and the following third through sixth months
start when the new moon is seen "in conjunction" with the
Aldebaran, Bellatrix,
Orion's Belt and Saiph taken together, and finally Sirius. The next six months are defined
by observations taken within the counting month. Month seven is defined as the month when
a full moon is seen "in conjunction" with Triangulum, while month eight is
defined when a three-quarter moon is "in conjunction" with Triangulum. The next
four months are defined when consequent less-waxed moons are seen "in
conjunction" with Triangulum and the year begins again when a new moon is finally
seen "in conjunction" with that position again. Allowing for a leap month every
three years, as observations dictate, the 354-day lunar (12-month) year can consequently
keep up with the 365-day solar cycle. |
|
However, what does "in
conjunction" mean? The lunar motion does not allow a "rising with"
interpretation of this term since Triangulum is, in general, too faint to be seen
next to a rising new moon (which is always seen in twilight). Also, the lunar motion does
not correspond to consecutive horizon risings with the given star positions at the start
of consequent months at any time within the date of existence of the calendrical system.
(Apparent star positions change with the precession of the pointing direction of the
Earth's rotation axis.) On the other hand, if "in conjunction" is interpreted as
meaning "rising at the same position on the horizon" then the 300 B.C. (but not
the present) positions of the Borana calendrical stars do indeed rise at the same horizon
rising positions as consequent new moons for the first six months of the calendar and the
given consequent phases of the moon rise at the horizon rising position of Triangulum for
the next six months, thus defining a workable calendrical system (Doyle 1986). The 300
B.C. positions of the Borana calendrical stars are shown in Figure
2a; the stars will rise almost vertically this close to the
equator. (Referring to Figure 2b, when
the Earth's equator is extended onto the sky, the distance a star lies above or below it
is called the star's "declination" measured from +90 to -90 degrees. The
distance around this celestial equator starting from the point where the Sun is on the
first day of spring (the vernal equinox) is the star's "right ascension"
measured from 0 to 24 hours around the sky or from 0 to 360 degrees as shown in
Figure 2c.) |
|
Additional investigation into the
Borana calendrical system (Bassi 1988) has indicated that the present ayantu (astronomers)
and historians disagree as to what constitute the Borana calendar stars. In general, the
historians call the original stars given by Legesse (1973) the Borana stars while the
present ayantu use a different overlapping set of stars (as they would have to do since
the first set of stars have changed their apparent positions and do not work as a
calendrical system as described any more). This present Borana system also has problems
since it is apparently based on a "rising with" interpretation of "in
conjunction". Most of the stars in this updated system are grouped together and as a
consequence the calendar cannot be used (due to interference from the Sun) for several
months out of the year. The system consequently seems a bit ad hoc, and Bassi (1988)
indicated that it might be some sort of remnant of the ancient system suggested by Doyle
(1986). It is clear, however, that some modification of a stellar-lunar
position-based calendrical system would have to have taken place if the origin of the
Borana calendrical reckoning extends back in time for more than a few hundred years. |
|
We next might ask, from the
interpretation of the ancient Borana calendrical system, if the 19 stone pillars at the
Namoratunga site could have been used to mark the 300 B.C. horizon-rising positions of the
seven Borana calendar stars. Remeasurement of pillar positions (Soper 1982, Doyle and
Frank -- 1983 Namoratunga expedition) produced 25 two-pillar alignments with the 300 B.C.
eastern rising positions of the Borana calendar stars. (Referring to Figure 1b, some examples of these 300 B.C.
stellar alignments with pillars found are: (Beta) Triangulum 1-8-10, Pleiades 1-2-12,
Aldebaran 1-3-4 also 19-8-12, Bellatrix 5-6-12, central Orion-Saiph 5-8-9 also 18-15, and
Sirius 18-13.) To decide if these
alignments are random or likely the result of intentional alignment with these seven
specific star positions on the sky, the following test was designed (Doyle and Wilcox
1986). Seven random star positions were generated and compared with the mapped pillar
positions at Namoratunga, giving the number of alignments found with these random star
positions on the eastern horizon. This process (generating seven random star positions and
comparing them with Namoratunga to determine how many alignments were thus made) was
repeated 10,000 times with the result that the number of times that 25 or more random
stellar alignments were found for the Namoratunga pillars was 41. This result would
indicate that there seems to be a less than 0.5% probability that the Namoratunga site
pillars could have randomly made as many as 25 or more alignments with the 300 B.C.
horizon rising positions of the seven original Borana calendrical stars. |
|
However, several archaeologists
conclude that the stones are instead soddu (burial) stones (Stiles 1983). But few question
that they are man-made, as the basalt pillars have a square cross-section while basaltic
material, in general, naturally cleaves in a hexagonal cross-section. Interpretation of
the petroglyphs on the pillars also remain controversial as does the connection with
Namoratunga I to the south. Additional dating techniques (desert varnish, for example) may
be applicable to constraining the date of the Namoratunga II site, and
additional such
sites should be locatable if such pillars were in general use for calendrical reckoning.
Finally, identification of the petroglyphs with other familial symbols in east Africa
(symbols found on the artifacts from the Kushitic pyramids in the Sudan, for example)
might prove to be a uniquely African historic tracer supplementing linguistic data on
migration patterns and cultural history. In conclusion, the Namoratunga II pillars may
have been the site of an ancient calendrical observatory, but much yet remains to be done
for this to be convincingly concluded. Perhaps the best possible investigation would be
the discovery of another such megalithic site where similar analyses as have been outlined
here might be performed. |
|
From Kushitic Pyramids
to Tanzanian Cave Symbols
Examples of locations of
calendrical interest in east Africa that have undergone only preliminary astronomical
investigation range from astronomically interesting alignments of the pyramids of Kush to
possible lunar symbols found in certain caves in Tanzania. These two cases are presented
here as examples of the types of investigations that remain to be done into the
astronomical ideas of the varied peoples of this region of the world. The Kingdom of
Kush, in
central Sudan, lasted from about 1000 B.C. to A.D. 200 around the fifth cataract of the
Nile (the capitol being first at Napata then relocated to Meroe). While archaeological
investigation has revealed an extensive civilization (with a perplexing language and
writing) the sites of numerous Kushitic pyramids have yet to be thoroughly investigated
for astronomical alignments similar to those found in Egyptian pyramids to the north (see
Arkell 1973, Millet and Kelley 1977, Shinnie and Bradley 1980, for reviews and
references.) However, from preliminary investigations of maps of the area around
Meroe, it
is plain that most of the pyramid entrances face very close to the direction of the
eastern rising of the star Sirius indicating, perhaps, an Egyptian calendrical influence
brought back from the Kushitic Pharaohs (Kush conquered Egypt about 800 B.C.) Any
contrasts between calendrical systems of the Kush and the ancient Egyptians could
certainly help illuminate some of the cosmology of a still little know civilization that
flourished in sub-Saharan Africa almost two millennia ago. |
|
As another example of such research,
among ancient cave drawings in Tanzania, M. Leaky (1983) has discovered a number of
precise circles, drawn with an accuracy that indicated to her that they might have been
used for some sort of counting process. She dubbed them "Suns, for want of a better
name". They consist of concentric circles carefully spaced apart and ranging from
just one circle up to 29 or 30 circles depending on the location. It is interesting that
the number of days in the lunar synodic cycle, 29-30, comes up again. Perhaps they could
be called "moons" more accurately. The number of concentric circles at a
particular site may represent, for example, a meeting or gathering cycle since the travel
time between cave sites could be several days. The investigation of these and other
symbols to see if they do correlate in a calendrical way could indeed open up new insights
into the timekeeping as well as possibly the cosmological ideas of the early inhabitants
of these regions . |
|
Great Zimbabwe Ruin and
the Pre-Shona Calendar
While it has been
estimated that over 1000 megalithic sites can be found in Zimbabwe and surrounding
countries, the most famous of these is Great Zimbabwe Ruin itself in southeastern
Zimbabwe. It consists principally of the Great Enclosure with an adjacent area on a nearby
hill called the Hill Complex. The pre-Shona people (sometimes called the Karanga) started
building here around 400 A.D. and finished the present structure seen today around the
middle of the fourteenth century. Since an initial survey around the turn of the century
(Bent and Swan 1969 republished edition) there have been a number of various astronomical
and mathematical features claimed for the Great Enclosure which consists of a large oblong
wall about 10 meters in height, 3 meters thick, and about 100 meters in length, with
various brick "altar" structures, interior walls, pillars, and stone monoliths
inside. At one end of the enclosure,
for example, are two brick towers of about 15 meters in height which were said in early
surveys to be related in that the radius of the larger was the circumference of the
smaller (that is, the ratio of their circumferences was the mathematical constant, pi =
3.14159...). Half a decade later some investigators also claimed that the circumference of
the whole enclosure was related to certain astronomical distances. Recent investigations
(for example, Garlake 1985, Huffman 1987, Doyle et. al. -- 1989 Great Zimbabwe
Expedition), found no such mathematical relationships inherent in either the internal
structures or the surrounding wall structure. Also, by comparing the oldest maps and
photographs it was found that the original ruins have been significantly tampered with --
certain smaller towers or pillars having been removed with one , at least, having been
added. In addition, some of the internal monoliths have been reseated recently. In spite
of this, however, preliminary investigations do reveal that the native African peoples
that built Great Zimbabwe were aware of the sky and may indeed have marked important
astronomical seasonal events. |
| For example, in a preliminary survey,
a "chevron" pattern on the southeast corner of the large outer wall is bisected
by the rising position of the Sun on the summer solstice from inside the enclosure, and
aligns with what has been called the "altar" as well as an original pillar
inside the enclosure. As this large patterning does not appear at any other place on the
outer wall it would appear to be a conspicuous candidate for a summer solstice marker
built into the Great Enclosure. In addition, a large passageway within the Great Enclosure
-- about 2 meters in width, 30 or so meters in (curving) length, with 10 meter high brick
walls on either side, would allow a limited view of the sky with an angular extent and
curvature matching the position and angular extent of the Milky Way overhead on the summer
solstice. While the Milky Way was a very important calendrical marker for the Karanga
people of this area (Sicard 1969, McKosh 1979) this observation too must be confirmed with
further research. Finally, from a cleared platform at the top of the Hill Complex, two
large stones (approximately 5 meters in height) in close proximity to each other can be
seen to form a slit directed precisely east which could have served as a solar marker for
the equinoxes. These and other observations are, however, preliminary and a better
understanding of the calendrical systems of the early inhabitants of this region would
substantially improve further investigations into any astronomical features that may have
been built into the ruins at Great Zimbabwe. |
|
|
The Dogon of Mali and
the Star Sirius B
Perhaps the most
controversial investigations into the ancient astronomy of sub-Saharan African have come
from the historical accounts of the Dogon of Mali, in western Africa (Temple 1987, but see
also Krupp 1991 and references therein for an astronomical viewpoint). These
investigations involve the star Sirius (A) which is the brightest star in the sky -- after
the Sun -- and has played an important historical role in, for example, the Egyptian
calendar (the helical rising of Sirius always signaled the beginning of the flooding of
the Nile in ancient times). However, the fact that Sirius is a double star was not
discovered by modern astronomers until 1844 when Friedrich Bessel deduced its existence by
the slight wobbling motion observed in Sirius A itself. Named Sirius B, this small
companion had to await the invention of larger telescope in order to be seen, and was
first spotted by Alvan Clark in 1862 using one of the largest telescopes at the time (an
18 1/2 inch aperture refracting telescope). Soon after this the period of the mutual orbit
of this binary star was determined to be about 50 years which also enabled a precise
determination of the two stars' masses and radii. The mass of Sirius A was found to be
about twice that of the Sun, which one would expect from a large, blue star. However, the
mass of the much smaller companion was found to be a little larger than the Sun. This came
as a surprise since it was only about the size of the Earth. This meant that the density
of this star would have to be about one million times that of an average star.
However, it was not until 1931 that the astrophysicist S. Chandrasekhar published the
first explanation of the structure of Sirius B and several other of these very dense
stars, which were to be called "white dwarfs". |
|
Also in 1931 the anthropologist and
ethnologist Marcel Griaule began collecting the lore of the Dogon of Mali, West Africa. He
found that the Dogon regard Sirius (which they call sigi tolo) as very important, but that
its invisible companion po tolo (meaning "deep beginning") is far more
important. Po tolo (Sirius B) was said to be closely associated with the fonio grain grown
in this region (it is sometimes called the "star of the fonio"). They said that
because the fonio grain is very small and white, po tolo is also very small and white.
They also said that po tolo is very heavy - the heaviest of all stars. They said that it
is at the center of the sky and its influence makes them stay in place. (The node of the
Earth's precession plane is very close to Sirius so that it will not move significantly
within several millennia.) Finally they say that po tolo circles the star Sirius every
fifty years, which is also correct, but they apparently have a special ceremony every
sixty years to celebrate the completion of one cycle associated with po tolo. Records of
the masks used in this ceremony apparently indicate that this ceremony - or one like it -
has indeed been held since around the 13th century. Dogon myths have also indicated,
however, that a third star exists in the Sirius system, -- a star that has not been seen
and whose mass influence should perhaps have shown up by now in perturbations of the
motions of the other two stars. |
|
|
| References Arkell, A.J., (1973), A History of the Sudan - From Earliest Times to 1821, Greenwood Press, Westport, CT. Bassi, M. (1988), " On the Borana Calendrical System: A Preliminary Field Report", Current Anthropology 29, 619-624. Bent, J.T. and R.M.W. Swan, (1969), The Ruined Cities of Mashonaland, Books of Rodesia, Bulawayo. Doyle, L.R., (1986), "The Borana Calendar Reinterpreted", Current Anthropology 27, 286-287. Doyle, L.R. and T.J. Wilcox, (1986), "Statistical Analysis of Namoratunga: An Archaeoastronomical Site in Sub-Saharan Africa?", Azania: Journal of the British Institute of East Africa 21, 125-129. Garlake, P., (1985), Great Zimbabwe Described and Explained, Zimbabwe Publishing House, Harare. Huffman, T.N., (1987), Symbols in Stone, Witwatersrand University Press, Johannesburg. Krupp, E.C., (1991), Beyond the Blue Horizon, Harper Collins, New York. Leakey, M. (1983), Africa's Vanishing Art: The Rock Paintings of Tanzania, Doubleday and Co., Garden City, N.Y. Legesse, A. , (1973), Gada: Three Approaches to the Study of African Society, Macmillan Pub. Co., New York. Lynch, B.M., and L.H. Robbins, (1978), "Namoratunga: The First Archaeoastronomical Evidence In Sub-Saharan Africa", Science 200, 766-768. McCosh, F.W.J., (1979), "The African Sky", NADA 12, 30-44. Millet, N.B. and A.L. Kelley, (1977), "Meroitic Studies - Proceedings of the 3rd International Meroitic Conference, Toronto", Akademie-Verlag, Berlin. Shinnie, P.L. and R.J. Bradley, (1980), The Capital of Kush, Akademie-Verlag, Berlin. Sicard, H.v., ( 1969), "Karanga Stars", NADA 2, 42-64. Soper, R., (1982), "Archaeo-astronomical Cushites: Some Comments, Azania: Journal of the British Institute of East Africa 17, 145-162. Stiles, D., (1983), The Azanian Civilization and Megalithic Cushites Revisited", Kenya Past and Present 16, 20-27. Temple, R.K., (1987), The Sirius Mystery, Inner Traditions International, New York. Thomsen, D.E., (1984), "What Mean These African Stones?", Science News 126, 168-169, 174. Laurance R. Doyle SETI Institute Edward W. Frank Tusker Trail and Safari Company |
|
|
