ARTICLES

9.

b y G . K.  Nukunya

The  Importance  of geomancy  in West Africa  Is well  known.
Its principal  exponents  include  the  Yoruba,  Dahomey and the  Ewe.
Although  the Yoruba are generally  credited with  Its origin,, some
liorin  Yoruba accounts quoted by  Clarke  suggest  that  It was  learned
from  the  Nupe.^  Similar  practices are also  found among the Anago
Yoruba  of  Brazil  and Cubans of  Yoruba  descent.  Whatever  its origin,
afa  divination,  as it  Is called among the  Ewe,  has become  Integrated
in diverse ways  into  the  religious  systems of all  the societies  which
now  practise  it.  The only  traces of  Yoruba  origin  In  the  non-Yoruba
areas now are  probably  the  unintelligible  Yoruba  language  employed
in  the  process and the verses the diviners  recite  which  are all  based
at  |fe,  the  traditional  ritual  centre  of  the  Yoruba.

Whiie  much has been written  about  the  Yorubo and the

Dahomean systems notably  by  Bascom and  Herskovfts  respectively,
very  little  has been done on the  Ewe side. 
In this paper an attempt
will  be made  to describe  the  technical  procedures and other  salient •
features  in  the  Ewe practice.

*  Dr.  G.K.  Nukunya  is a  Lecturer  in  the  Department  of  Sociology.

1 . 

2. 

This study  formed part of  the  1967 Summer  Research  School
organised  in  Ghana  jointly  by  the African  Studies Programme,
Northwestern  University  and the  Institute-of  African  Studies,
Legon.  Grateful  acknowledgement  Is made  to the  two
institutions* for  financial  support and other  forms of  cooperation
and to  the  following  students  for  their  assistance  in  fieldwork:
Joan  Bragginton and Margaret  Rosembloom of  Northwestern
and  Ernest  Dumor,  Brfgirre  Kumapley and Susan Gilbertson,
of  Legon.
Clarke,  1939,  p.238,

ARTICLES 

10.

The  home  of  the  Ewe-speaking  peoples  now  lies  in  south-

eastern  Ghana  and  the  southern  half  of  Togo.  But  they  have  not^
been  living  here  for  more  than  five  centuries.  They  have  a  history
of  migration  which  puts  their  origin  somewhere  in  Yoruba  country
around  fhe  present  Dahomey-Nigeria  border.

Before  coming  under  European  rule  at  the  end  of  the  last
century,  the  Ewe  never  lived  under  one  political  authority.  They
were  organised  info  a  number  of  chiefdoms  varying  in  size  from  a
few  villages  to  large  and  powerful  kingdoms  like  Anlo  and  Peki
comprising  several  towns  and  villages.  Despite  the  absence  of
political  unity  the  Ewe  share  among  others  a  common  myth  of  origin
and  history  of  migration;  a  common  Ewe  language  spoken  through-
out  Eweland;  and  finally  the  consciousness  of  being  one  people.

The  Ewe area  described  here  is  the  Kingdom  of  Anlo,

It  is a  low  country  whose  land-

which  lies  in  the  extreme  southeastern  corner  of  Ghana,  immediately
east  of  the  Volta  River,  with  an  area  of  883  square  miles  and  a
population  of  231,017  in  I960, 
scape  and  economy  are  dominated  principally  by  fhe  large  Kefa
Lagoon  (which  covers  about  a  third  of  the  area  and  separates  the
littoral  from  the  hinterland)  and also  by  the  numerous  streams  that
connect  the  Lagoon  with  the  Volta  River.  During  the  main  rainy
season  which  centres  around  June, large  areas  around  fhe  lagoon
are  flooded  for  months  while  in  the  dry  season,  November  to
Januaryf  fhe  floods  recede  and  the  greater  part  of  the  lagoon
itself  dries  up?  yielding  large  incrustations  of  salt  which  provide
an  important  article  of  trade.

Or.  the  sea  coast  and  in  the  villages  surrounding  the

lagoon^fishing  is  the  chief  occupation  but  Inland  to  the  north  the
people  cultivate  cassava,  corn  and  vegetables.  Other  occupations
include  cloth  weaving  on  hand  Iooms#  basket  making  and  poultry
farming.
i 
| 

. i n.  t i u

^ 

^ 

^ 

1 . 

1960  Population  Census  of  Ghana,  V o l.  1 .  The  Gazetteer,

• Census  O f f i c e,  Accra,  1962,  P,  X X I I I.

ARTICLES

1 1.

Like  the other  Ewe groups the Anlo are  parrilinea!.  There
are  fifteen  agamous,. dispersed,,  totemic  clans whose  local  segments
form exagamous  lineages.  Two of  the clans have  royal  dynasties
from whom the Awoamefia  or  King  is alternately  chosen.

There  is a belief  in a  High  God,  Mawuy  who  like many
other such West African  gods disappeared after  creating  the  world
because of  the  frequent demands on him by his"people  while he
lived near  them.  There  is no organised worship for him:  no priest-
hoodf  shrine  nor church.  He  is invoked by  individuals  in  everyday
life  and  in  times of  crises, 
if  is according  to  the  Ewe themselves
a god of  the  thoughtful,  not of the multitude.  Next  to  Mawu are
the small gods or  trSwo  (singular, 
trS_) which are his  creations,
deriving  their  powers from him and residing  in nature objects  like
streams,  ponds,  lagoons and forests.  Unlike  the  High  God the
trSwo  have  organised worship with  shrines,  priests and churches.

There are also a number of  cults or  voduwo  owned by

certain  individuals.  Many Anlo voduwo are of  foreign  origin,
mainly  Dshomean and Togofese ond the  individual  desirous of
establishing  them travels  to  their  places of  origin  where  the
necessary  fees are  paid and the "rites performed.  Membership
of  both  trS  and  vody_ (singular)  cyts across  lineage and clan
lines,  though  children  usually  join  the gods served by  their
parents.

But by far  the most  important  supernatural  beings are

the ancestral  spirits  toabe^gljawo,  worshipped  in a'body  by
lineages and clans at  festivals and also by  individuals  who
normally  show their  reverence by giving  them  the first  morsel
of  food or the  first drops of  water or drinks  before satisfying  them
selves. 
ancestors through  the  leaders of  their descent groups.

Individuals  may also ask specific  favours from  their

Afa,  the deity  worshipped by afa  diviners^  bokowo

(s'n9>  k2k£_)  n as  a  somewhat ambiguous  position  in  Anfo

ARTICLES 

12.

It  is popularly  regarded as vodu_ rather  than trS_.

religious  system. 
It appears diviners  themselves also  regard  It as such at  least  by  <•
implication  as evidenced  in  their  greetings when  the state  of the
"vodu"  (meaning  the afa  deity)  is asked.  But afa  differs  from vodu
in some  important  respects.  Unlike vody_ it  is not a possession  cult,
and  its  intentions  and wishes are  revealed  through  sickness and
divination.  Besides  it  is through afa  that  the wishes of all  the other
gods and supernatural  beings are  revealed.  Also  each  practising boke_
can  both  find out and procure a cure  for sickness and misfortune  where-
as in  both JT5_  and  vodu these  functions are  the  prerogatives  of  the
priests.  Because of  these  reasons and the highly  specialised  nature  of
the divination  coupled with  the considerable  respect accorded them,
diviners  prefer  to keep afa  quite apart  from both vody_and trj5_.

All  the supernatural  beings  just enumerated  including  afa

are  capable  of sending death#  sickness and other  misfortunes on  their
worshippers  if  neglected or offended  in any way.  On  the other  hand
their  powers can be neutralized  by witches and sorcerers who may
harm anyone  regardless of  his devotion  to  Mawu,  tr5_#  vody_,  the
ancestors or afa.

The Anlo  themselves generally  attribute  death 

to these
supernatural  forces as well  as to witchcraft  and sorcery,  rarely  to
natural  causes.  Therefore  soon after  funeral  ceremonies are  completed
the  relations  of  the deceased approach a diviner  to  find out which  of
these  forces has caused death.  Almost  invariably  one or more of them
will  be mentioned,  which  is just  the answer the  relatives  themselves
expect,  . If  is only  in the  case of a very  old man that  it  is said Mawu  ,

1.  Life expectancy, in Anlo  is not great due  probably  to  malnutrition

and widespread alcoholism.  On a rough estimate  it  averages
about  forty-five  to  fifty  years for  those who do not  die  in  early
childhood whilst  almost  every  mother  before  the completion  of
her  reproductive  activities  experiences  either a still  birth  or the
death of an  infant.  Moreover  in  the  fishing areas,  drowning
and other accidents  with  this occupation are quite common.

ARTICLES

13.

has called him,  the equivalent-  of a natural  death.

Yet  fidning  the cause of death  Is only  one of  the bok$)Js_

many functions which  may be grouped under  the  following  headings;

(a)  To warn  individuals  and groups of  impending

misfortunes and to prescribe  remedies for  prevention.

(b)  To advise Individuals  and groups about  possible
actions which  may affect  prosperity  or health.

(c)  To give  explanations  for  misfortunes which  have

already  occurred,,  such as famine,  death  or sickness.

(d)  To find which ancestor  is re In coma ted  in «  child,

The following  is a sample of  cases witnessed one day  in a

diviner's  house:-

1.  A  woman was to undergo a surgfco'  operation and

wanted to know whether  it  would be successful.

2.  A  man brought a sick  daughtet  to  find our what was

troubling  her.  He said treatment  received  at  the
government  health  centre  felled  fo  cure  her,

3.  A  man who said he had been  having bad dreams

came to know  the cause.

4.  A  wornon whose prosperous business  collapsed
suddenly came  fo know why and fo  remedy  the
situation.

To this  list  may be added a  footnote  that  in Anio <a  large

section of  the  population  has still  not acquired any confidence  in
European trained medical  doctors and the hospitals. 
place  for  the sick  to approach  diviners  for diegnosis and cure. 

It  fs stiff  common-
In such

ARTICLES 

14.

an  environment-  the  importance  of  the  diviner  Is  considerable-

in  all  these  cases  the  bok_£_  uses a  complex  system

of  geomancy  of  which  there  are  two  methods  in  current  use,
In  one  a  rectangular  wooden  board  is  used.  After  a  light  greyish
powder  has  been  sprinkled  on  it,  the  diviner  shuffles  sixteen  palm
nuts  by  rapidly  pausing  them  from  one  hand  to  the  other,  his  aim
being  to  have  one  or  two  nuts  in  one  hand  and" the  rest  in  the  other
hand.  This  however  h  not  always  possible  after  every  shuffling.
In  any  case  If  two  nuts  ore  left  in  one  hand  he  marks  one  stroke,
if  one  nut  he  marks  two  strokes  in  the  powder  on  the  board, 
if
none  or  more  than  two  nuts  are  left  in  one  hand after  the  mani-
pulation  then  this  <s not  ominous  ana  the  process  is  repeated  t i l!
only  one  or  two  nuts  are  left  ot  a  time,  The  process  is  completed
when  eight  ominous  results  am  obtained,  marked  on  the  board  in
two  vertical  columns  of  four  groups  of  single  or  double  strokes  such
as shown  in  the  diagram  below.  This  combination  of  strokes  deter-
mines  the  client's  du,  rnc:ssage,or  kfjolj^  destiny,  which  are  used
interchangeably.  This  method  is  so complex  and  lengthy  that  it
lakes  several  minutes  *,o obtain  the  required  ominous  signs  or  strokes
and  is  consequently  not  employed  In  ordinary  cases  of  consultation.

in  the  less  compiex  one  the  diviner  uses  gjjm_acja:,  an

instrument-  made  of  eight  half-pods  of  the  rare  h_yku_  seed  strung
at  equal  intervals  on  a  cord  about  twenty  inches  in  length,  To
cast  the  gurrsacja  the  divine;"  holds  the  middle  of  the  cord  to  enable
the  seed  to  present  two  columns  of  four  groups  on  the  mat.  Each
of  the  half  pods  has  concave  and  convex  surfaces  so  that  when
^ne  SHfliSQS. 's  fhrown  on  the  mar  or  floor  the  pattern  presented  by
the  surfaces  determines  Hie  kjjolj  or  du-

Two  important  actions,  performed  simultaneously,  precede
both  processes  of  divination.  The  first  is a  prayer  in  somewhat  u n-
intelligible  Yoruba  and  Ewe  lasting  about  eight  minutes  in  which
the  names  of  some  famous  diviners  are  invoked  and an  appeal  made

ARTICLES

15.

to  the afa  deity  to assist  in  revealing  the  truth.  While  the  prayer  is being
recited  the diviner  shuffles  continuously  an  innumerable amount of  articles
which  he has emptied on the  floor  from his diving  bag.'  Before these
actions  the client  is expected  to whisper  his troubles  on a coin which is
then  placed on the mat,  for,proper  procedure  demands that  the  client's
intention  for  coming  to afa  be kept  secret  from the diviner  before  the

>oli  is revealed.

There are sixteen  possible  principal  kpoliwo  (pi.)  resulting  from

the diagram of strokes each of which  may be applicable  at a  time.  They
are arranged according  to seniority  as:

(1)  DZONGBE

(2)  YEKU-MEDZI

1 
1 
1 
1 

1
1
1
1

(5)  LOSO-MEDZI 

(6)

1 
1 
11 
11 

1
1
11
11

(9)  GUDA-MEDZI

(10)

1 
1 
•1 
11 

1
1
1
11

11 
11 
11 
11 

11
11
11
11

DI-MEDZI
11
11 
11 
11

1

1

SA-MEDZI
11
11 
1 
1
1
1 
1 
1

11 
1 
1 
11 

(3)  WOLI-MEDZI
11
1
1
11
ABLA-MEDZI
1
11
11
11

1 
11 
11 
11 

(7)

01)

KA-MEDZI
11
11 
1 
1
11
11 
11 
11

(4)  DI-MEDZI
1
11
11
1

1 
11 
11 
1 

(8)  AKLA-MEDZI

11  '  11
11
11 
11
11 
1 
1

(12)TRUKPE-MEDZI

11 
11 
1 
11 

11
11
1
11

\. 

The contents of a divining  bag  include  almost everything  the diviner  can
lay  hands on and may number nearly  a thousand articles. 
dre cowrie  shells,  coins,  palm nuts,  beads,  skulls of small animals all  of
which are used in the  process of vodzibabla  described  below.

Important  items

ARTICLES 

16.

03) TULA-MEDZI
1
11
1
1

1
11
i

i

(14)  LETE-MEDZI

(15)

1
1
11
1

1
1
11
1

TSYE-MEDZ!
...1-
11
1
11

1
11
1
11

(16)  FU-MEDZr1*

1

1

11 
1
11 
1

1
1
1
1

It

will  be

seen  that  all

but  one of

the

sixteen

principal

kpoliwo  have

the suffix  medzi  (which  means two  in  Yoruba)  because  in each  case  the number
and ord^i  of  the strokes  in both  columns are  identical. 
In other  words each
of  the sixteen  kpoliwo  is a combination  of  two  identical  columns.  No  clear 
explanation  has been given  for  the absence of  the suffix  in  the case of
Dzongbe. 
as the  chief  of all  the  kpoliwo  it  is not  necessary  to  repeat  its  identical
combination.

Informants  only  said  that  strictly  speaking  it  is also a medzi but

Each of  the sixteen  principal  kpoliwo  can be combined with  others

to  form various  permutations  of  ordinary  kpoliwo.  That  is,  one column  of
kpoli  Dzongbe  for  instance  can be combined with  one side  of any  of  the
other  fifteen  kpjiiwo  and vice  versa  to  form other  kpoliwo.  Thus we may
have  for  example  kpc*ii 
loso  -guda  which  is a combination  of  kpol?  loso
and kpoii  guda.  Such a kpuii  will  be  represented  in  terms of  strokes 
as:- 

, 

,

1 
11 

11
U 

It  will  be seen that  in  the  strokes  for  ioso-guda  the strokes  for  loso are on
the  right  and guda  on  the  left.  This  is because  in  every  combination  the
diviner  reads  from  right  to  !eft.  That  is,  if  loso were  on  the  left  and guda
on  the  right  we shall  have a different  kpoli  altogether  which  will  be known

1 .  This order of  the  figures  corresponds  exactly  to  Bascom's  for  the  Yoru-

ba but  differs  slightly  from both  Herskovits  for  the  Dahomey,  and
Fiawoo's  obtained  from  the same  Ewe area.  See Bascom  1966,  p.428,
Herskovits,  1938,  Vol. II,  p.210  and  Fiawoo,  1958,  p.69.

"

,  ,

w'

ARTICLES

17.

as guda  -  ioso  represented  in  strokes  as:~

I I

? •>

In  this  way  we  have  af  least  256  different Jcjjgjjwo  being  combinations
of  the  sixteen  majorkj3£Jjvyo_  with  each  other.  But  the  ordinary
kpoiiwo  have  not  fhe  suffix  medzi  because  their  columns  have  no
identical  number  or  order  of  strokes.

Each  principal  and  ordinary  kpy[i  has  proverbs,  verses,
pithy  sayings  and  a  number  of  anecdotes  cast  in  a  vague  way  to
explain  i t.  These  proverbs,  verses  e t c .,  cover  a  number  of  contin-
gencies  such  as death,  sickness,  wealth  or  marriage.  The  anecdotes
as  the  diviner  always  emphasises  ail  come  from  Yoruba  mythology
centred  at  lie  ife,  the  cradle  of  all  Yoruba  and  indeed  of  all  men,
according  to  the  Yoruba.  Thus  the  diviner  introduces  his  narration
by  the  phrase,  "Le  efe  mea 
There  are  also  a  number  of  taboos  usually  of  food  associated  with
the  articles  in  the  verses  and  anecdotes.

That  is,  "|n  Ife  of  old  . ..  "..

. . " 

The  determination  of  the  kgojj  and  its  anecdotes,  proverbs,

It  is

e t c .,  is  only  a  first  step  in  the  process  of  Afa  geomancy. 
further  necessary  to  know  whether  the _kpo|_[ carries  a good  or  bad
omen  for  the  client.  This  is determined  by  vodzibabla.  The  diviner
gives  a  nut  and  a  coin  or  cowrie  to  the  client  to  hold  in  his  hands
separately  and  at  random.  By  throwing  the  g_urr]aga_ two  times  he  is
able  to  determine  which  of  the  articles  is  ominous.  The  nut  carries
bad  omen  while  the  coin  and  the  cowrie  shell,a  good  omen.  A  bad
omen  means  that  disaster  is  likely  to  follow  if  sacrifice  is  not  made
to  offset  the  impending  danger  while  a  good  omen  on  fhe  other  hand
portends  good  fortune  but  only  if  validated  by  sacrifice.  Sacrifice
accompanies  every  completed  process  of  afa  divination  whether  the

ARTICLES

1 

• • ;.

The  fype  of  socnFice  to  be  performed  in  each

omen  is good  or  riot 
case  i's determined  by  H".e  kps*'?.  *iv;r  nature  of  the  omen,  the  relevant
verses  and  the  inferprcrnricris  <~f !hy  diviner.  Should  a  client  fail  to
perform  sacrifice  hecoasff  he. (e,:h  'he  omens  are  good  he  will  face
misfortune  instead.  K o m ^ c,  or  divination  without  sacrificing  and  its
consequences  are  some  oi  the  :.;x>sons which  keep  some  people  away
from  diviners,  for  sacrifice:,  o> " i ll  be  shown  later  are  sometimes  very
expensive  and  one  w i ll  not  know  before  consulting  a  diviner  whether
or  not  an  expensive  sonrii•>.-"•:  -j'iW be  demanded.

it  hai, been  F.bcv<n  ;hor  the  anecdotes  are  cast  in  vague  terms

which  reveal  or  for^c^ir  vory  l i t t l e.  Precision  and  detail  may  be
obtained  by  further  threw inc!  of  the jgurnaga_or  manipulation  of  the  nuts
which  may  also  involve  ft.-rrher  vocizibabja.  Leading  questions  are
askft-J  and  the  client  7K.y  ;;yen  be  asked  to  say  something  about  his
life  history  a."!--,; exp^W-vnces.  For  example, 
!o
client  o? chUd
is  no  she  may  •.-•t
jf  the  <.V:vA.'er  h
;. ;;f:n;  ••-ill  •;•.'.o,i  r.oMCiive. 
p;; k,nn  :;i  :o  co;;firn;  the  safe  delivery  the  kgp[i_portents  . 
b oj  .:•:.  j..';,c,••,a,i
r^y wili  bring  misfortune  unless  sacrifice  for  Fumedzi
h  pfcHoii-vicJ.,  A  rrvcse  clier.i-  on  F:umedzi  aiso  faces  similar  questions
abouf  his  wife  or  fiirnafe  relatives.

thir  is  ?evecied  the  first  question  put  to  a  female
a;?rg  oo&  'mf
If  the  answer
nsked  v./heiher  any  of  her  close  relatives  is  pregnant.
cujin  in  tb.3 negative,  the  diviner  w i ll  then  say  the

!f  *ho  omens  are  good  sacrifice  should  be
If  they  are

k£»)j£  Fumedzi  is  related

  MAre  you  pregnant?" 

As  ftu  r-js  ih's  sfavjs  oF the  procedure  is  concerned  there

oppacrs  ;o  be  some  difference  between  the  Anio  and  other  exponents
of  vhs  -.yjffitn.  Whilt  amonq  tho  Yccuba  if  is  the  client  who  selects
the  s(,'itable  ve^?  'Trsona  the  Usi- of  vsrsps  narrcted  by  the  diviner,  '
the  [we  boko 
•A'har  [if:  consi'jr'rs  !o  be  fhc-  rno?' appropriate  for  his  client.  He  is

.?tlfn  iiCirrcMitu!  lii'?'- stories  of  the  client's  kpoli  selects

bascorn,

ARTICLES

19.

guided  in  his choice  by such consideraHons as the  complexion  of
his client^  his weight, build  and other  physical  traits as well as
other  facts  he  is able  to  solicit  from  him. 
practice  conforms substantially  to  that  of  the  Dahomey where
"complete  frankness" between  the diviner  and his client  is
demanded,^ 
not  reveai  the nature  of  his  problem  to  the  b_oko_  before  the
casting  of  the gumaga and the determination  of  the Jkgoflas  is the
case  in  Dahomey,

the only  difference  being  that  the Anlo  client  does

In  this  respect  the  Anlo

Afa  is both a deity  and a system of  divination.  Actually

it  is the deity  of  the system of  divination  that  bears  its name.  How-
ever  not all  its servants are diviners.  Although  both  male arid
female  may serve  it,  only  male servants may be diviners and some
of  the  latter  too do not  practise  it  either  because their  economic
activities  do not allow  them or because they  fack  the skill  for  it.

To become a diviner  one has to  undergo an apprentice-
ship  lasting at  least  three  months after  initiation.  This period may
be extended  to nearly  three  years  if  the novice  is slow  in  mastering
the  techniques.  During  the  period of apprenticeship  the  novice
remains  in  his master's novitiate  till  he  is rirually  outdoored.  The'
apprenticeship  which  may cost about  £100  in  cash and goods  involves
memorizing all  the  figures and their  verses,  pithy  sayings^  proverbs
and anecdotes as well  as their  accompanying  sacrifices and herbs.
The  technique  of  making a gymacja  is also  learned/  for  every
practising  boka_  must make his own

Training  in afa  technique  is compulsory  for  every  male
servant  regardless of  his prospective  occupation.  Although  very
successful  diviners  live  solely  on  income  from divination,  it  is
usually  not a  full  time  business and may be combined with  other
economic  activities.

1.  Herskovitsf  op.cit.  p.216.

ARTICLES 

20.

Afa  diviners  are organised  into  Four sects:  anago, dzisa,
mayehoe  or  zev> arid ajsycike...  The anago differ  from  the others'
in  certain  respects.  While  they are  forbidden  to  indulge  in any
form of sorcery  or magic,  the  others are  c!osely  connected  with
these  practices.  And  in  ihe  casting  of  the gjymag£ the anago  throw
it  with  the  middle  handle  falling  towards  the  client  whilst  the
others  throw  it  in  the  opposite  diiecvion,  tha* is with  the  handle
failing  towards  the diviner.  The reading  in  both  cases  is however
the  same,  from  top  to bottom and  right  to  left.  But despite  their
differences  the  four sacts cooperate  in  many  ways such as dancing
and drumming and holding  of  common  initiation  activities.

The  principal  officers'  for  the  seers  are:-

i

^.QaJi? 

Dj?-isa_(Mayehoe,  Arsyake)

]
2.
3.
4.
5.

BoKa  (Oludwo)
D?. op! bone
A^.aba
Selikpawo
Ordinary  dlvlnera

1.
2.
3.
4.
5.

Oluawo
Dzogbana
A4,aba
Seiikpawo
Ordinary  diviners

Ihe  iViba  or  Oluawo  is the- figure  head but  the  Dzogbana is

it  is  like  having a society  in which  the

the  rno-it  important  leader, 
secretary's  powers carry  greater  weight  than  the  president's. 
this  case the secretary  will  be the  Dzogbana  and the  president,  the
Baba or Oluawo.  The A^aba assists the  Dzogbana while  the  Seii-
kpawo serves as a messenger,  a position  which  carries great  influence
among  diviners.

In

1 ,  Some of  these  titles are also  found among the Yoruba.  Clarke

gives  the  following  hierarchy  of  babalawo  (diviners) without
giving  their  functions:

] .  Oluawo
2,  Akoda
3,  Aseda
4,  Arabo
5,  Ordinary  babalawo.

ARTICLES

21 .

It  has already  been  mentioned  how afa  has become an

Anlo  institution,  and  integrated  into  the  traditional  religious
system.  Today very  few Anlo apart  from the bokowo know or
claim  that  it  originated  from  Yorubaiand.  Among  that  section
of  the  population  which  consults  the diviners,  belief  in  the
efficacy  of  the system  is almost  complete.  The ordinary  Anlo
traditional  believer  has nothing  to  complain  about  it  even when
a  number of  the  predictions  and sacrifices  fail  to  materialise.
According  to  their  belief,  the supernatural  forces of afa  control
all  the elements of  the  procedure  which  others may ascribe  to
chance.  The prayers  preceding  the divination,  the  boko  claims
and  the  client  believes,  help  *o solicit  supernatural  participation
in  the  process. 
It  is further  believed by  both diviner  and  client
that  the  former's  initiation  into  the afa  cult  and rhe  training he
has acquired all  help  to  put  him  in a special  ritual  condition, a
relationship  with  the sacred,  which  helps him divine  with  super-
natural  sanction.  Another  source of  confidence  for  the  client
derives  from  the anecdotes,  verses,  proverbs and  the sayings
which  explain  the  figures.  Without  exception  they  refer  to
legendary  characters who approached Afa  with  similar  problems
as the client,  sacrificed  the  things  the diviner  now demands  for
sacrifice,  and prospered,  the belief  being  that  since  that  legendary
character  succeeded by sacrificing  the  very  things now  mentioned,
the  present  client  will  also succeed by  following  suit.  This
corresponds very  closely  to  Frazer's  Law of similarity  which,
despite  Bascom's statements on the  Yoruba  to  the  contrary,  finds
expression  in  many afa  sacrifices  in Anio.  Two specific  examples
will  suffice.  A  barren  woman on  Fu-medzi  was asked to make,
among others,  an  image of a baby,  carry  it  on  her  back 
in  the way
babies are  carried by  mothers and then  leave  it  under any  tree  in a
flowering  state.  Also a client  who was standing  for an  election was
asked to trample  his opponent's  mud image  under  his feet.  The
implication  for  the  law  of  similarity  in  these actions  is clear.  By

ARTICLES

22.

imitating  a  mother  and  coming  into  contact  with  a  fertile  tree  the
client  w i ll  also  achieve  fertility  and  eventually  become  a  mothe/.
The  second  client  also,  according  to  this  belief,  w l l!  defeat  his
opponent  after  bearing  him  in  rhe  sacrifice.

To  return  to  the  question  of  confidence  in  diviners,  their
general  behaviour  may  also  be  mentioned..  Both  as  individuals  and
as a  group  their  deportment  is  expected  to  be  above  suspicion  and
as  they  themselves  o!ways  emphasize,their  professional  white  attire
which  distinguishes  them  from  others  is  meant  to  reflect  their  moral
purity,  There  are  also  a  number  of  taboos  which  govern  their
behaviour  such  as  those  forbidding  them  to  quarrel  in  public  and  to
commit  adultery.  Moreover  the  knowledge  displayed  in  memorizing
hundreds  of  verses,  anecdotes  and  proverbs  commends  them  very
highly  to  the  ordinary  Anio  as  men  of  high  intellectual  standing.
It  has  been  suggested  tor  instance  that  the  amount  of  energy,  time
and  knowledge  that  gnv.:,  fn  qualify  a  person  cs  diviner  may  be  .
equivalent  to  if  not  greater  titan  what  one  requires  fora  Ph.D.
Whatever  the  merits  <v' ihh  claim,  the  respect  for  these  men  in  a
folk  society  rnusi  be 

r-irisidercblf!.

it  must  bej rne;:-iioneri  once  again  that  though  their
predictions  do  not  oiwjys  <-ome  Hue,  as  found  in  many  other
systems  of  divination,  fuik'res  do  noi  result  in  any  doubt  about
the  compettrics  o\  officifincv  of  either  the  diviner  or  the  system
as such,  Explanations  ond  alibi*,  including  incorrect  procedures
and  the  unhallowed  condition  of  the  client,  are  always  provided
to  cover  all  short-comings.  Besides,  a  client  who  feels  he  followed
all  the  instructions  of  the  diviner  and  stili  failed  to  get  the  desired
result  cannot  be  sure  that  nothing  has  been  achieved  or  that  the
advice  of  the  diviner  was  wrong. 
conditions  he  cannot  be  certain  that  if  he  had  refused  to  take  the

In  the  absence  of  controlled

1 .  Ottenberg  and  Ottenberg,  p.65.

ARTICLES

23,

diviner's  advice,  things  would not  have been worse.

These  than ore  some of  the  considerations which  enthuse

conscience  in  clients.  To Hie outside  observer and  fhe  non-believer
however  Hie patterns  presented by  the gunrciga and the number of nuts
remaining  in one hand at- o time  during  divination  are due  to mere
chance,  "How  could  the  throwing  of nuts reveal-the  unknown?"
To these  people  therefore  belief  in such a system sterns from either
supersirition  or  the  uncritical  mind'; of  clients  and the  members of
the  society  ar  large.  Today  if  is also  the widespread belief among
rrian-/,,especio!ly  ChristionSjthar  diviners are  frauds who use all  sorts
of  trickery  to  cheat  their  clients. 
frequent!1/  mode to  the many  failures  diviner?  encounter,  while'
obvious successes are explained  in  terms of  chance and  coincidence.
It  is quite  true  lhat  despiie  their  professional  codes of  conduct,,  un-
scrupulous  dividers  sometimes exist  whose behaviour  is very  suspicious
and who can  iherefore  fend themselves  very  easily  to charges of  fraud.
Eci'joMy  true  is the  fact  ihot  even among  ihe usually  respectable
diviners  doubts sometimes  urist  cibou* the  interpretation  of  the  figures
as well  os the  predictions  and  the  prescribed  remedies.  These weak-
nesses however  do not  affect  the  majority  of  them who are  in general
sincere  individuals  having  fervent  belief  in  their  procedures and pro-
fession 

Few  if  any will  always go out  to  cheat  their  clients.

In  this  regard  reference  is

Yet  there  can  be very  little  cheating  unless the  procedures

of  the  system  itself  makes this  possible.  Unfortunately  a fa  is not
altogether  proof  cgaingf  fraud*  Bascom mentions a number of  instances
when this  con be done among the Yoruba'  some of which apply  to the
Anlo  too.  Let's  take  firsf  the  throwing  of  the gumaga.  Perhaps this  is
the cnly  stage where  cheating  Is not  easy,  The casting of  the chain  is
done  in such a way  that  the diviner  cannot  influence  its  foil  and the
figures. 

It  hce; also been noticed among  the Anlo  (as Bascom did among

1 ,

Dp,  cit.

p. 50.

ARTICLES 

24.

the  seeds are  joined by  rather

the  V'oruba)  lhat  in  some gjJfngfia 
inflexible  cords which  enable  certain  figures  to show more  frequently
than others.  This  is no cheating  as such.  The weakness here  derives
simply  from  the diviner's  inability  to make a good.ajjmaga  for  himself.
Moreover  as an  informant  claimed,  even  in an  inflexible  cord afa
will  influence  the  full1.

It  is  easier  to  control  the  Figures derived  from  shuffling

the  palm nuts since  oniy  the diviner  can determine  when  it  is lime
fo stop and check  the number of nuts remaining  in each hand.  Thus
a diviner  intent  on cheating  could  continue  shuffling  until  the
desired number  of  nuts  remains  in  one hand,  repeating  this  trick
through  the whoie  process.  Yet  it  has been  found  from  experience
and ciose observation  that  only  a very  clever  divine?' can  dishonestly
manipulate  the  nuts without  incurring  the suspicion  of  his  client.
In any  case  this method  is never  used in ordinary  divination  but  only
on special  occasions  when  cheating  will  not  benefit  the diviner  in
any  way.

!t  is when  reciting  the verses  that  the diviner  can  easily
falsify  by simply  inventing  verses  to suit  the  client's  circumstances
if  he surmises  them-  Perhaps  the oniy  safeguard against  this  is the
fear  that  the suspicious  client  may ask another  diviner  to  re-
interpret  the  figures  for  him.

Sacrifice  is another  vulnerable  aspect.  Although  the

sacrifice  for  the various  kp.oiiwo_is  determined by  the  applicable
verse,  there  is no fixed  number  or quantity  of  sacrificial  objects
in  each case.  Undoubtedly  this gives  the  best opportunity  to  the
diviner  to  cheat  if  he wishes  to.  According  to  informants  afa
sacrifice  is flexible  enough to  meet  the  pockets of every  client.
It  is claimed  that  if  rhe kgojj  demands sheep for sacrifice  the
diviner  can ask  a poor  client  tc  buy a piece  of  mutton  from a

I. 

Ibid.

ARTICLES

butcher  instead of  slaughtering  a whole  sheep.  Yet  few diviners are
considerate  enough to a Slow this.  This  is understandable  in  viewpf
the  fact  that  the  destination  of all  sacrificial  objects  after  necessary
rites are  performed,  is the  diviner's  pof  and kitchen.

\n his study  among the  Yoruba,Clarke  suggests that  the

diviner  uses telepathy  through which  hs knows consciously  or
unconsciously  what  the enquirer  has whispered  on the  coin.
Since  this  fieldwork  was  limited  to  the  Ewe area  there  is no direct
evidence  to  controvert  the  position  of  someone who studied among
the  Yoruba,  As far as the  Ewe are  concerned,  telepathy as an
integral  part  of a fa divination  is fotoliy  ruled out,  and  for what  is
known of  the  Yoruba system both  from the  literature  and from
observation  of  Yoruba  bgbgjawo  in  Ghana,  it  can only  be said  that
most of  its  procedures do nor differ  significantly  from the  Ewe system-
A  more  persuasive  explanation  may be empathy  the acceptance  of
which,  however,  tike  telepathy-,  would  render all  the  complex
processes  involved  unnecessary.,  for  it  would  mean that  the  figures
and the verses are only  used to  bamboozle  the  clieni,  a  contention
no one who has closely  studied  the  systems  in action  will  accept.
On  the  other  hand only  a belief  in  the efficacy  of  the afa  deity  can
change what  Is overtly  a chance operation  into a veritable  method
of  knowing  the unknown.

REFERENCES

Bascom, W.R.  1941 

"The Sanctions  of  I fa  Divination",  Journal
2l  'hp  Royal Anthropological  Institute,
Vol.7].

1 . 

Op.  cit.  p.,251..

ARTICLES

26.

Clarke,  J . D. 

1939, 

"Afa  Divination",  Journal  of  the  Royal
Anthropological  Institute,  Vol.69.

Fiawoo,  D.K. 

1958, 

"The  Influence  of  Contemporary  Social
Changes  on  the  Magico-Religious  Concepts
and  Organization  of  the  Southern  Ewe-
speakincj  People  of  Gh'ina".  Unpublished
pi'.D..  Thesis,  7dinbum  University.

f •'••".-rsknvjir.,  M.  J.  '\93n.  Da horn'.iy^  a 1_A! IS-i'-1'"••!_ We sr  African  Kingdo

(2'VolsV)  N  ;w~York  ."

• ^ . - ^ - v / i i l i a r : *,  P.  Bascom,  W.R.  and  Mcl.ellancJ,  E.M,  1966

"Two  Studies  of  Ifa  Divinafion",  Afnca_,

OHenberg  or\(i  O'renb  »rg, 1 960  -  Culture;  and  Societies  of  Africa .