LIBRARY AND  MUSEUM REPORTS

88.

EXCAVATIONS  AT JAKPASERE  -  1968

by R.  Duncan Mathewson*

During the months of  March and April,  excavations were
carried out at the site known as Jakpasere situated about one mile
east of the village  of  Grunshi Zongo,  and four miles south of Salaga
on the Yeji-Tamale  road J  The excavation was directed on behalf of
the  Department of Archaeology  of the University  of Ghana and the
Ghana  National  Museum.  In addition,  a Training School was  conducted
at  the excavation  from 23rd March to 4th April.  Nine students participated:
three  from the  University  of  Ghana  Including a  Nigerian  graduate
student registered for the  M.A.  in African Archaeology,  three  from the
University  of Science and Technology,  two  from O'Reilly  Secondary
School,  and one from the  National  Museum.

Historically,  the site of Jakpasere  is of some interest,  as the
Kpembe oral  tradition  claims that  it was a  'palace'  constructed out of
a  mixture of shea-butter and honey which was built by  'Jakpa1,  the
legendary  hero of the  Gonja state. 
Preliminary  work  in August  1965
and May  1966 indicated that the site consisted of a  ruined rectangular
building  about seventy-five  feet  long'and thirty-five  feet  wide.**  Some
of the walls,  which were still standing,  were about six to ten feet high,
while  the highest wall  stood close to twelve  feet  high.  An  initial  test
of  the deposit within the structure  revealed  that there was over five  feet
of stratigraphy,  which  for the most part represented col lapsed walls.

FulUcale  mapping and excavation of the site Indicated that

it was far more extensive than previously  thought.  Several  large
settlement-mounds were discovered about one hundred yards to the east
of the main structure, associated with bilegas (bell-shaped water holes
cut  into the solid rock) all  of which  indicate  intensive occupation within
the  immediate vicinity  of the site. 
It became clear that the rectangular
building was not an isolated structure,  but part of a  larger complex of

*  Mr.  R.  Duncan Mathewson  is a  Research  Fellow  in the  Department

of Archaeology.

Jr" '•>

LIBRARY AND  MUSEUM REPORTS

89.

buildings enclosing an open compound.  There was a small building
at  the north-east  corner of the main structure which was roughly  per-
pendicular  to it.  This structure was about thirty-five  feet  long and
had four rooms,  two of which opened out on to the compound area.
Other  small buildings existed on the extreme eastern side of the site;^
these were  more or  less parallel  to the main building and formed the
other side of the enclosed compound.

The northern end of the main building  was completely

excavated and revealed a basement of small adjoining rectangular
rooms.  Small arched doorways of about  four to five  feet high
connected these rooms which were aligned on two  interior walls that
appeared to  run the entire  length of the building.  The only entrance
into the building  that was uncovered was a small passageway which
connected the basement  level  with some type of exterior  timber-roofed
structure.  The walls were built  of gravel-swish and contained  large
quantities of broken sherds, which to some extent must have helped to
bind the swish together.  They were built  in stages of horizontal swish
layers which were about three feet  high.5 
In the  lower courses, tfee
walls thickened,  producing a buttressing effect,  which helped to provide
the necessary stability  for the upper storey.  The building stood at  least
eighteen to twenty  feet high and had at  least one upper floor.  This was
indicated by joist holes in several of the walls.  Another row of foist
holes appear on the upper part of the'west wall and probably represent
the  flat  roof.6

The pottery showed a whole range of various wares,  including
several  types of painted pottery.  The most characteristic  painted design
consisted of  fine  red,  parallel  lines which occurred on several  different
fabrics.  Of  particular  interest was the occurrence of several snuff pots,
which appeared to be  identical  to the ones from Period V I,  at  Si lima
(NewBuipe).

The presence of a  tuyeVe fragment and a  considerable amount

of  Iron slag  indicates that  Iron was smelted in  large quantities on the
site.  The  Iron work was particularly  rich and a whole range of Iron
objects were recovered.  Of special  Interest was a  number of arrow-

LIBRARY AND  MUSEUM REPORTS 

90.

heads and  lanceheads,  as well  as part  of a horse's bit  and a curved
blade,  which  represents a short slashing sword of  probable Mande
t y p e/

The dating  evidence  for  Jakpasere  Is primarily  derived from

thirteen  tobacco-pipe  fragments which  indicate  that  the site was
occupied  during  the  middle  of  the seventeenth  century.  The occurrence
of  types  Y D - 1, 2,  and 3 of the  Yendi  Dabari  pipe typology suggests
that  the  period of occupation  may be very  provisionally  set from about
1620/30 to  1670/80.8  Closer  reliable  dating of  the site  is not  possible
at  this  time.  However,  there  does seem to be some verification  of  the
Kpembe oral  tradition  in  the sense that  the  provisional  date suggests
possible  correlation  with  " Jakpa1.

'Jakpa'  has been  identified  with Al-Lara,  the  fifth  ruling

king  of  Gonja.  He  is believed to  have reigned from  1622/3  to  1666/7,
when he abdicated and died six  years  later.^  On the  present  evidence,
it  therefore  seems quite  possible that  this  site  could very well  have been
associated with  'Jakpa1.  Why  it  was abandoned  is problematical,  but
it  seems reasonable  to suppose that  it  might  have been evacuated after
his abdication  from  power,  or shortly after  his death  in  1672/3.  However,
there  Is no doubt  that  Jakpasere  represents a period of  concentrated  Gonja
power and  indicates  consolidation  of the  Salaga area  from the  early
seventeenth  century.

Many of  the architectural  features of  the main building suggest

a very  close similarity  to  the structural  evidence  from the  previously
excavated mound sites. 
reflects  the widespread occurrence of  traditional  Sudanlc  architecture
throughout  northern  Ghana.

This resemblance  Is not  fortuitous and

The excavation  was primarily  conducted on a  horizontal basis,

in other words,  a complete area was stripped with an emphasis on the
plan of  the site  rather  than on vertical  sections  through  I t . ''  This
method results  In a  far  more complete  picture  of a  living area,  and there-
fore  provides a considerable amount of  Information  as to  how people  lived
and not  just when  they  lived.  The association  of different  types of  pottery

LIBRARY AND  MUSEUM REPORTS

91.

and small  finds (iron objects,  bone,  spindle worls,  etc.)  on the basement
floor of the  Individual  rooms, as well  as,  In the collapsed wall  material
will  provide some indication as to the normal activities within the
various parts of the building  just prior to  Its Initial  collapse. 
It  there-
fore will  become possible to say much more about some of the economic
and social aspects of the  Gon|a  community and of the  Late  Iron Age  in
the Volta  Basin.  With  this in mind,  it  is hoped that a second season at
Jakpasere can be planned which will  complete  the excavation of the
main building and its smaller associated structures.

Through the facilities  of the  United States  Information  Service
arrangements were made to film the students talcing part  In the training
programme.  For the most part,  the filming was done at  the excavation
and during the evening  lecture-work  sessions.  Some shots were also
taken on an excursion to the excavated mound site at  New  Bulpe and
to the  Ktntampo rockshelters. 
It was thought that a  film of this type
might be of some Interest  to the general  public,  as well  as  being a
valuable aid  In the teaching of Archaeology by visually  portraying the
various methods and techniques employed In an excavation.

During the course of the excavation,  the site was visited by

a  number of  Interested local  people,  as well  as, by twenty-five  teachers
from the Salaga  region, and over fifty  students from three primary and
three middle schools.  All  showed a genuine  interest  in the site and it
is hoped that  in the very near future Jakpasere will  be scheduled as a
national  monument.

The success of this excavation and training programme was
largely  due to the generous assistance and co-operation which was
In particular we wish to thank the
received  from a  number of people. 
United African Company for providing funds which helped to cover the
students1 expenses;. the Director of the  United States Information
Service,  Mr.  Gordon WInkler,  and the cameraman,  Mr.  George  Bracher,
for making  possible the  filming of the training  programme; the  District
Administrative  Officer at  Salaga,  Mr.  S.K.  Dosavl, the  Public Works

LIBRARY AND  MUSEUM REPORTS

92.

Department  Engineer,  Mr.  A.K.  Dagadu, and the  Medical  Officer,
Dr.  A.W.  Swiecicki  who all  helped to provide a most pleasant stay
in  Salaga;  the  Kpembewura and his elders for the entertaining  Durbar
and for their  hospitality and  interest  in the excavation;  and the people
of  Kpembe,  Salaga,  and Grunshl  Zongo,  for their  friendliness and co-
operation  throughout.

!

,

- * •*

LIBRARY  A ND  MUSEUM  REPORTS

93.

Footnotes

1. 

2. 

3. 

4. 

5. 

6. 

7. 

8. 

9. 

10. 

11. 

I wish to thank  Jack  Goody  for first bringing  this site to my
attention.

The site  is very briefly  mentioned by both  E.  Meyerowitz  in
The Akcn Traditions of Origin  1952:  58  n.2  and D.H.  Jones
in  "Jakpa and the  Foundation of  Gonja",  Trans.  Hist.  Soc.
Ghana,  Vol.VI,  1962:  23.

Mathewson,  R.D.  in  Research Review,  I.A.S.,  Vol.2,  N o . l,
1965:  46 and  in Archaeology  in the Volta  Basin  1963-66,
V.B.R.P.  and  Department or Archaeology,  Legon,  1967:  21-22.

As time was short,  this area  wen only quickly tested.

The walls were  previously described as being  constructed of
'larerite-blocks*.  This should be disregarded.

From the depth of the deposit  it would seem that  there was only
one upper storey.

See Freeman,  R.A.,  Travels and  Life  in Ashanti and Jotnan,
1898:  417.

See  "Excavation at  Yendi  Dabari",  Shinnie,  P.L.  and Ozanne,
P.C.,  Trans.  Hist.  Soc.  Ghana,  Vol.VI,  1962:  94-106.

Wilks,  I .,  "TheChronology  of the  Gonja  Kings",  Ghana  Notes
and Queries,  No.8,  1966:  26-28.

Mathewson,  R.D.,  "Some Notes on the  Settlement-Mounds of
central  Gonja",  Research Review, 
1968:  108-114.

!.A.S.,  Vol.4,  No.2,

I should like  to thank  the students and other  members of the
Department  in assisting  in the excavation.

V"

.'•¥