Umbanda	
  and	
  Hybridity	
  	
  
Steven	
  Engler	
  
Department	
  of	
  Humanities,	
  Mount	
  Royal	
  College,	
  Calgary,	
  T3E	
  6K6,	
  Canada	
  
sengler@mtroyal.ca	
  

Abstract	
  
Scholars	
  of	
  religion	
  continue	
  to	
  talk	
  of	
  syncretism	
  where	
  their	
  colleagues	
  have	
  moved	
  on	
  to	
  talk	
  of	
  hybridity.	
  This	
  
paper	
   reviews	
   critiques	
   of	
   the	
   latter	
   concept	
   and	
   argues	
   that	
   ‘hybridity’	
   can	
   be	
   a	
   useful	
   concept,	
   but	
   only	
   if	
  
further	
   specified.	
   I	
   follow	
   Peter	
   Wade	
   in	
   distinguishing	
   between	
   hybridity	
   of	
   origin	
   (the	
   combination	
   of	
   pre-­‐
existing	
  forms),	
  and	
  hybridity	
  of	
  encounter	
  (the	
  result	
  of	
  diasporic	
  movements).	
  I	
  propose	
  a	
  third	
  type,	
  hybridity	
  
of	
  refraction,	
  in	
  order	
  to	
  highlight	
  the	
  manner	
  in	
  which	
  religious	
  or	
  cultural	
  phenomena	
  refract	
  social	
  tensions	
  
within	
   a	
   specific	
   nation	
   or	
   society,	
   resulting	
   in	
   a	
   spectrum	
   of	
   ritual,	
   doctrinal	
   and/or	
   religious	
   forms.	
   The	
  
typology	
   is	
   not	
   meant	
   to	
   be	
   complete	
   or	
   mutually	
   exclusive:	
   it	
   suggests	
   the	
   value	
   of	
   adopting	
   distinct,	
   potentially	
  
overlapping,	
   perspectives	
   on	
   hybridization.	
   I	
   illustrate	
   the	
   heuristic	
   value	
   of	
   this	
   approach	
   with	
   the	
   case	
   of	
  
Umbanda,	
  a	
  twentieth-­‐century	
  Brazilian	
  religion.	
  
Keywords	
  	
  
hybridity,	
  syncretism,	
  Umbanda,	
  Candomblé,	
  Kardecism,	
  Brazil	
  

The	
   interdisciplinarity	
   of	
   Religious	
   Studies	
   is	
   a	
   source	
   of	
   both	
   strengths	
   and	
   weaknesses.	
   The	
  
field	
  draws	
  fruitfully	
  on	
  concepts	
  and	
  theories	
  from	
  other	
  disciplines,	
  but	
  it	
  tends	
  to	
  do	
  so	
  late	
  
in	
  the	
  game	
  and	
  often	
  uncritically.	
  For	
  a	
  generation	
  now,	
  scholars	
  of	
  post-­‐colonial	
  and	
  cultural	
  
studies,	
   literary	
   criticism,	
   intellectual	
   history,	
   communications,	
   qualitative	
   sociology	
   and	
  
other	
   fields	
   have	
   used	
   and	
   critiqued	
   the	
   concept	
   of	
   “hybridity.”	
   Scholars	
   of	
   religion	
   have	
  
covered	
   some	
   of	
   the	
   same	
   ground	
   in	
   dealing	
   with	
   “syncretism,”	
   but	
   the	
   newer	
   term	
   offers	
  
distinct	
  advantages	
  and	
  raises	
  fresh	
  problems.	
  Given	
  that	
  “hybridity”	
  appears	
  poised	
  to	
  play	
  a	
  
more	
   prominent	
   role	
   in	
   the	
   study	
   of	
   religion—as	
   scholars	
   in	
   the	
   field	
   increasingly	
   research	
  
diasporic	
  religion,	
  draw	
  on	
  post-­‐colonial	
  theory,	
  and	
  so	
  forth—this	
  is	
  an	
  opportune	
  moment	
  
to	
  learn	
  what	
  lessons	
  we	
  can	
  from	
  the	
  hybridity	
  debates	
  in	
  other	
  fields.	
  “Hybridity”	
  is	
  valuable	
  
for	
   four	
   reasons:	
   it	
   reminds	
   us	
   that	
   analyses	
   of	
   religious	
   mixture	
   must	
   take	
   into	
   account	
  
broader	
   cultural	
   interactions,	
   not	
   just	
   relations	
   among	
   those	
   elements	
   considered	
   “religious”;	
  
it	
  usefully	
  highlights	
  the	
  prevalence,	
  creativity	
  and	
  dynamism	
  of	
  cultural	
  mixture,	
  especially	
  in	
  
our	
  current	
  global	
  context;	
  it	
  reminds	
  us	
  that	
  the	
  study	
  of	
  contemporary	
  religious	
  phenomena	
  
requires	
  attention	
  to	
  very	
  specific	
  historical,	
  regional,	
  and	
  social	
  contexts;	
  and	
  it	
  reminds	
  us	
  to	
  

be	
   wary	
   of	
   reifying	
   or	
   attaching	
   normative	
   weight	
   to	
   the	
   boundaries	
   that	
   are	
   crossed	
   and	
  
blurred	
  during	
  cultural	
  mixture.	
  

	
  

In	
   this	
   paper,	
   I	
   first	
   support	
   the	
   claim	
   that	
   scholars	
   of	
   religion	
   continue	
   to	
   talk	
   of	
  
syncretism	
  where	
  their	
  colleagues	
  have	
  moved	
  on	
  to	
  talk	
  of	
  hybridity.	
  I	
  then	
  review	
  a	
  number	
  
of	
   critiques	
   of	
   the	
   latter	
   concept.	
   This	
   results	
   in	
   some	
   practical	
   points	
   regarding	
   the	
   critical	
  
use	
   of	
   “hybridity.”	
   I	
   then	
   offer	
   a	
   brief	
   overview	
   of	
   a	
   specific	
   case,	
   Umbanda,	
   a	
   twentieth-­‐
century	
   Brazilian	
   religion.	
   In	
   analyzing	
   this	
   case	
   I	
   propose	
   a	
   threefold	
   distinction	
   between	
  
types	
   of	
   hybridity.	
   This	
   typology	
   is	
   proposed	
   for	
   its	
   heuristic	
   value,	
   not	
   because	
   it	
   is	
   complete	
  
or	
  mutually	
  exclusive.	
  I	
  first	
  draw	
  on	
  work	
  by	
  Peter	
  Wade	
  in	
  order	
  to	
  distinguish	
  two	
  types	
  of	
  
hybridity,	
  which	
  I	
  rename	
  hybridity	
  of	
  origin	
  (the	
  combination	
  of	
  two	
  pre-­‐existing	
  forms),	
  and	
  
hybridity	
   of	
   encounter	
   (the	
   result	
   of	
   diasporic	
   movements).	
   Based	
   on	
   my	
   discussion	
   of	
  
Umbanda,	
  I	
  argue	
  that	
  Wade’s	
  distinction	
  can	
  be	
  usefully	
  complemented	
  by	
  the	
  addition	
  of	
  a	
  
third	
   type,	
   hybridity	
   of	
   refraction.	
   In	
   suggest	
   the	
   latter	
   in	
   order	
   to	
   underline	
   that	
   certain	
   cases	
  
of	
   cultural	
   mixture	
   must	
   be	
   analyzed	
   in	
   terms	
   of	
   social	
   tensions	
   within	
   a	
   specific	
   nation	
   or	
  
culture.	
  	
  
My	
  conclusion	
  is	
  that	
  the	
  concept	
  of	
  hybridity	
  usefully	
  draws	
  our	
  attention	
  to	
  an	
  important	
  
set	
  of	
  issues,	
  but	
  that	
  this	
  contribution	
  of	
  the	
  concept	
  remains	
  at	
  a	
  very	
  general	
  level.	
  The	
  real	
  
work	
   of	
   analysis	
   comes	
   down,	
   as	
   always,	
   to	
   careful	
   work	
   with	
   cases.	
   In	
   this	
   work,	
   general	
  
concepts	
   like	
   syncretism	
   and	
   hybridity	
   are	
   of	
   little	
   value	
   except	
   as	
   flags	
   of	
   allegiance	
   to	
   a	
  
certain	
  approach.	
  
	
  
Beyond	
  “Syncretism”	
  
Scholars	
   of	
   religion	
   have	
   continued	
   to	
   talk	
   “syncretism”	
   for	
   an	
   entire	
   academic	
   generation	
  
during	
   which	
   most	
   of	
   their	
   colleagues,	
   in	
   other	
   areas	
   of	
   the	
   humanities	
   and	
   social	
   sciences,	
  
have	
   moved	
   on	
   to	
   talk	
   of	
   “hybridity.”	
   “Hybridity”	
   came	
   to	
   prominence	
   primarily	
   in	
   post-­‐
colonial	
   theory,	
   where,	
   by	
   a	
   decade	
   ago,	
   it	
   was	
   “one	
   of	
   the	
   most	
   widely	
   employed	
   and	
  
disputed	
  terms”	
  (Ashcroft,	
  Griffiths,	
  and	
  Tiffin	
  1998:	
  118).	
  The	
  concept	
  became	
  influential	
  in	
  
other	
   fields,	
   intersecting	
   with	
   a	
   wide	
   variety	
   of	
   related	
   terms,	
   both	
   old	
   and	
   new:	
   e.g.,	
  
acculturation,	
   articulation,	
   bricolage,	
   creolization,	
   fusion,	
   heterogeneity,	
   in-­‐betweenness,	
  
interstitiality,	
   mélange,	
   mestizaje,	
   multiple	
   identity,	
   pastiche,	
   polyphony,	
   subalternity,	
   third	
  
space,	
  transculturation,	
  etc.	
  For	
  better	
  or	
  for	
  worse,	
  Religious	
  Studies	
  has	
  been	
  curiously	
  slow	
  
to	
  jump	
  on	
  this	
  terminological	
  bandwagon.	
  

A	
  keyword	
  search	
  of	
  research	
  databases	
  offers	
  a	
  useful	
  measure	
  of	
  our	
  field’s	
  provincialism	
  
on	
   this	
   issue.1	
   Comparing	
   the	
   relative	
   number	
   of	
   occurrences	
   of	
   syncretism-­‐terms	
   with	
  
hybridity-­‐terms	
  makes	
  it	
  clear	
  that	
  scholars	
  of	
  religion	
  are	
  unusual	
  in	
  their	
  preference	
  for	
  the	
  
former.	
  In	
  the	
  ATLA	
  Religion	
  with	
  Serials	
  database,	
  heavily	
  weighted	
  toward	
  Religious	
  Studies	
  
publications,	
   5.5%	
   of	
   all	
   references	
   to	
   either	
   syncretism-­‐	
   or	
   hybridity-­‐terms	
   were	
   constituted	
  
by	
   the	
   latter.	
   That	
   is,	
   publications	
   in	
   the	
   database	
   referred	
   to	
   syncretism	
   seventeen	
   times	
  
more	
  often	
  than	
  they	
  did	
  to	
  hybridity.	
  In	
  the	
  Francis	
  database,	
  with	
  solid	
  but	
  proportionately	
  
less	
   Religious	
   Studies	
   representation,	
   the	
   numbers	
   of	
   occurrences	
   were	
   around	
   equal.	
   In	
  
SocINDEX,	
   representing	
   sociology	
   with	
   no	
   weighting	
   toward	
   Religious	
   Studies,	
   “hybridity”	
  
references	
  constituted	
  87%	
  of	
  the	
  total.	
  In	
  Academic	
  Search	
  Premier,	
  with	
  its	
  more	
  general	
  set	
  
of	
  sources,	
  99.1%	
  of	
  the	
  references	
  were	
  to	
  “hybridity.”	
  Clearly,	
  scholars	
  of	
  religion	
  prefer	
  the	
  
s-­‐word	
  where	
  their	
  colleagues	
  do	
  not.	
  
The	
  situation	
  in	
  Religious	
  Studies	
  is	
  beginning	
  to	
  change.	
  An	
  increasing	
  number	
  of	
  papers	
  
are	
  being	
  published	
  that	
  frame	
  issues	
  in	
  terms	
  of	
  hybridity.	
  However,	
  even	
  apparent	
  attempts	
  
to	
   privilege	
   the	
   newer	
   terminology	
   effectively	
   treat	
   “syncretism”	
   and	
   “hybridity”	
   as	
  
synonyms:	
   a	
   special	
   issue	
   of	
   Social	
   Compass	
   framed	
   by	
   the	
   editors	
   as	
   “Rethinking	
   Religious	
  
Hybridity”	
   (McGuire	
   and	
   Maduro	
   2005)	
   included	
   three	
   papers	
   on	
   “hybridity”	
   and	
   two	
   on	
  
“syncretism.”	
  
There	
   is,	
   of	
   course,	
   a	
   complex	
   debate	
   on	
   the	
   origin,	
   history,	
   allegiances,	
   and	
   value	
   of	
   the	
  
concept	
   of	
   syncretism	
   (see	
   Stewart	
   1999;	
   Leopold	
   and	
   Jensen	
   2004;	
   Martin	
   and	
   Leopold	
  
2004).	
   Many	
   of	
   the	
   strengths	
   and	
   weaknesses	
   of	
   “syncretism”	
   also	
   apply	
   in	
   the	
   case	
   of	
  
“hybridity”	
   (see	
   Kraidy	
   2002;	
   Hutnyk	
   2005).	
   The	
   most	
   obvious	
   criticism	
   is	
   that	
   “syncretism	
  
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
1

Table 1. Relative occurrence of hybridity terms (‘h’) and syncretism terms (‘s’) in keyword searches of four research
databases: ATLA Religion with Serials; Francis; SocINDEX; and Academic Search Premier. Search performed
18/04/07.
ATLA
Fran.
SocIN.
ASP
hybridity
19
137
525
762
hybrid
47
1143
1427
43154
hybridized
32
47
1867
Total ‘h’
66
1312
1999
45783
syncretism
1017
1084
178
288
syncretic
54
69
96
117
syncretistic
74
21
26
32
Total ‘s’
1145
1174
300
437
h/s
0.1
1.1
6.7
104.8
h/ h+s (%)
5.5
52.8
87.0
99.1
	
  

essentializes	
   too	
   much,	
   implying	
   that	
   there	
   were	
   once	
   well-­‐behaved	
   pure	
   breeds	
   before	
   the	
  
new	
   religious	
   mutts	
   gnawed	
   through	
   their	
   leashes”	
   (Johnson	
   2002b:	
   302).	
   However,	
   it	
   is	
  
possible	
   to	
   draw	
   a	
   broadly	
   useful	
   distinction	
   between	
   the	
   two	
   concepts:	
   syncretism	
   is	
   a	
  
mixture	
  of	
  religious	
  elements;	
  hybridity	
  is	
  a	
  broader	
  mixture	
  of	
  cultural	
  elements.	
  This	
  is	
  less	
  a	
  
rigorous	
  definition	
  than	
  a	
  pointer	
  in	
  the	
  direction	
  of	
  hybridity’s	
  value	
  as	
  a	
  more	
  appropriate	
  
concept	
  than	
  syncretism	
  for	
  studying	
  religion	
  in	
  an	
  increasingly	
  globalized	
  age.	
  
Syncretism	
   is	
   generally	
   presented	
   as	
   a	
   phenomenon	
   internal	
   to	
   religion.	
   Michael	
   Pye	
  
defined	
   it	
   as	
   “the	
   temporary	
   ambiguous	
   coexistence	
   of	
   elements	
   from	
   diverse	
   religious	
   and	
  
other	
   contexts	
   within	
   a	
   coherent	
   religious	
   pattern”	
   (1971:	
   93).	
   Charles	
   Stewart	
   offers	
   “the	
  
broadest	
   and	
   most	
   general	
   definition	
   of	
   syncretism:	
   the	
   combination	
   of	
   elements	
   from	
   two	
   or	
  
more	
  different	
  religious	
  traditions	
  within	
  a	
  specified	
  frame”	
  (2004:	
  282).	
  Similarly,	
  theorist	
  of	
  
hybridity	
  Nestor	
  Garcia	
  Canclini	
  defines	
  syncretism	
  as	
  “a	
  combination	
  of	
  traditional	
  religious	
  
practices”	
  (Canclini	
  2006[2001]:	
  xxviii).	
  	
  
Hybridity,	
   on	
   the	
   other	
   hand,	
   more	
   directly	
   acknowledges	
   the	
   complex	
   interactions	
  
between	
   religions	
   and	
   their	
   historical,	
   political,	
   social	
   and	
   cultural	
   contexts.	
   The	
   distinction	
   is	
  
not	
  sharp;	
  it	
  points	
  to	
  two	
  ends	
  of	
  a	
  spectrum.	
  Scholars	
  of	
  religion	
  highlight	
  religious	
  elements	
  
in	
  their	
  analyses	
  of	
  hybrid	
  forms,	
  but,	
  in	
  general,	
  a	
  conceptual	
  turn	
  to	
   “hybridity”	
  highlights	
  a	
  
broader	
  range	
  of	
  cultural	
  dimensions	
  of	
  religious	
  change.	
  In	
  addition	
  to	
  focusing	
  on	
  cultural	
  
mixture	
   more	
   broadly,	
   hybridity-­‐talk	
   emphasizes	
   the	
   normality,	
   creativity,	
   dynamism	
   and	
  
political	
   implications	
   of	
   such	
   mixture.	
   Ulf	
   Hannerz	
   for	
   example,	
   emphasizes	
   the	
   advantages	
   of	
  
“a	
  creolist	
  point	
  of	
  view”:	
  	
  
	
  
It	
   identifies	
   diversity	
   itself	
   as	
   a	
   source	
   of	
   cultural	
   vitality;	
   it	
   demands	
   of	
   us	
   that	
   we	
   see	
   complexity	
   and	
  
fluidity	
  as	
  an	
  intellectual	
  challenge	
  rather	
  than	
  as	
  something	
  to	
  escape	
  from.	
  It	
  should	
  point	
  us	
  to	
  ways	
  of	
  
looking	
  at	
  systems	
  of	
  meaning	
  which	
  do	
  not	
  hide	
  their	
  connections	
  with	
  the	
  facts	
  of	
  power	
  and	
  material	
  life.	
  
(1987:	
  556)	
  

	
  
A	
  turn	
  toward	
  “hybridity,”	
  then,	
  offers	
  two	
  immediate	
  advantages	
  for	
  scholars	
  of	
  religion:	
  it	
  
highlights	
  religions’	
  complex	
  relations	
  to	
  other	
  dimensions	
  of	
  their	
  cultures	
  and	
  societies;	
  and	
  
it	
   opens	
   doors	
   to	
   existing	
   cross-­‐disciplinary	
   discussions	
   of	
   these	
   broader	
   issues.	
   Of	
   course,	
  
“hybridity”	
  is	
  not	
  without	
  its	
  own	
  weaknesses.	
  
	
  

Evaluating	
  “Hybridity”	
  
The	
   shift	
   from	
  “mixture”	
   as	
   a	
   phenomenon	
   internal	
   to	
   religion	
   to	
   one	
   that	
   reflects	
   religions’	
  
complex	
  interactions	
  with	
  their	
  historical,	
  political,	
  social	
  and	
  cultural	
  contexts	
  is	
  a	
  valuable	
  
one.	
   However,	
   scholars	
   of	
   religion	
   need	
   not	
   reinvent	
   the	
   wheel.	
   Whether	
   or	
   not	
   the	
   actual	
  
term	
   “hybridity”	
   is	
   deemed	
   of	
   value,	
   scholars	
   of	
   religion	
   should	
   pay	
   attention	
   to	
   the	
   extensive	
  
debates	
  over	
  its	
  strengths	
  and	
  weaknesses	
  as	
  they	
  grapple	
  with	
  religions’	
  places	
  in	
  local	
  and	
  
global	
   contexts	
   of	
   cultural	
   mixture.	
   The	
   concept	
   of	
   hybridity	
   has	
   its	
   problems.	
   Several	
  
critiques	
  have	
  emerged	
  in	
  fields	
  outside	
  Religious	
  Studies.	
  	
  
First,	
  it	
  has	
  biased	
  roots.	
  As	
  Robert	
  Young	
  has	
  shown,	
  “hybridity”	
  is	
  rooted	
  in	
  the	
  racially	
  
loaded	
  discourse	
  of	
  nineteenth-­‐century	
  evolutionary	
  theory	
  (Young	
  1995;	
  Stross	
  1999).	
  This	
  
led	
  late	
  twentieth-­‐century	
  scholars	
  to	
  be	
  wary	
  of	
  metaphorical	
  language	
  that	
  draws	
  on	
  these	
  
roots:	
   e.g.,	
   species,	
   combination,	
   crossing	
   and	
   grafting.	
   This	
   politically	
   correct	
   reflexivity	
  
among	
  anthropologists	
  and	
  cultural	
  theorists	
  cast	
  a	
  shadow	
  on	
  an	
  entire	
  vocabulary,	
  given	
  the	
  
implicit	
  valorization	
  of	
  pure	
  parents	
  over	
  impure	
  offspring.	
  Paul	
  Gilroy	
  laments	
  “the	
  lack	
  of	
  a	
  
means	
   of	
   adequately	
   describing,	
   let	
   alone	
   theorizing,	
   intermixture,	
   fusion	
   and	
   syncretism	
  
without	
  suggesting	
  the	
  existence	
  of	
  anterior	
  ‘uncontaminated’	
  purities”	
  (Gilroy	
  2000:	
  250;	
  cf.	
  
McGuire	
  and	
  Maduro	
  2005:	
  411).	
  	
  
We	
  can	
  discount	
  this	
  first	
  problem	
  to	
  the	
  extent	
  that	
  we	
  question	
  that	
  concepts	
  with	
  dark	
  
pasts	
   necessarily	
   have	
   dark	
   futures.	
   The	
   issues	
   of	
   a	
   word’s	
   origin	
   may	
   or	
   may	
   not	
   be	
   relevant	
  
to	
  evaluating	
  its	
  current	
  uses	
  and	
  functions.	
  If	
  we	
  correct	
  for	
  biases	
  of	
  origin,	
  there	
  seems	
  no	
  
need	
  to	
  throw	
  out	
  the	
  concept.	
  (For	
  the	
  same	
  reason,	
  there	
  is	
  no	
  need	
  to	
  discard	
  the	
  concept	
  
of	
   “religion”	
   just	
   because	
   its	
   use	
   as	
   a	
   cross-­‐cultural	
   category	
   has,	
   in	
   part,	
   colonial	
   origins.)	
  
Reflexive	
  awareness	
  of	
  the	
  normative	
  dimensions	
  of	
  this	
  focus	
  on	
  “pure”	
  roots	
  mollifies	
  this	
  
first	
  critique.	
  
Scholars	
  of	
  religion	
  are	
  especially	
  well	
  positioned	
  to	
  adopt	
  this	
  reflexive	
  stance.	
  Writing	
  in	
  
the	
   context	
   of	
   Reformation	
   history,	
   Susan	
   R.	
   Boettcher	
   suggests	
   that,	
   because	
   scholars	
   of	
  
religion	
   “have	
   no	
   necessary	
   ethical	
   responsibility	
   to	
   take	
   confessional	
   sides,”	
   [they]	
   can	
   use	
  
the	
  concept	
  of	
  hybridity’s	
  ability	
  to	
  blur	
  the	
  observer’s	
  understanding	
  of	
  power	
  relationships	
  
“to	
  plumb	
  the	
  depths	
  of	
  the	
  frequent	
  ambiguities	
  of	
  religious,	
  cultural	
  and	
  political	
  power	
  at	
  
work”	
   (2005:	
   450).	
   On	
   the	
   one	
   hand,	
   we	
   should	
   not	
   be	
   too	
   quick	
   to	
   assume	
   that	
   we	
   are	
  
capable	
   of	
   some	
   sort	
   of	
   “pure”	
   objective,	
   outsider	
   stance.	
   On	
   the	
   other	
   hand,	
   the	
   study	
   of	
  
religion	
  has	
  long	
  had	
  a	
  very	
  healthy	
  debate	
  over	
  precisely	
  these	
  issues	
  of	
  reflexivity	
  regarding	
  

the	
  people	
  and	
  cultures	
  that	
  we	
  study.	
  
A	
  second	
  critique	
  focuses	
  on	
  the	
  descriptive	
  dimension	
  of	
  this	
  focus	
  on	
  “pure”	
  roots,	
  noting	
  
that	
   such	
   roots	
   tend	
   to	
   be	
   mixtures	
   themselves.	
   2	
   Hybridity	
   offers	
   little	
   analytical	
   purchase,	
  
because	
   it	
   is	
   hard	
   to	
   specify	
   what	
   is	
   not	
   hybrid:	
   “All	
   cultures	
   are	
   hybrid.	
   …	
   Culture	
   as	
   an	
  
analytic	
   concept	
   is	
   always	
   hybrid	
   …	
   since	
   it	
   can	
   be	
   understood	
   properly	
   only	
   as	
   the	
  
historically	
   negotiated	
   creation	
   of	
   more	
   or	
   less	
   coherent	
   symbolic	
   and	
   social	
   worlds”	
  
(Werbner	
  1997:	
  15).	
  	
  
Scholars	
  of	
  religion	
  are	
  especially	
  aware	
  of	
  this:	
  Anita	
  Leopold	
  reminds	
  us	
  that	
  “The	
  history	
  
of	
   religion	
   confirms	
   that	
   every	
   religion	
   is	
   in	
   ‘essence’	
   syncretistic—there	
   are	
   no	
   pristine	
  
origins	
   or	
   essences”	
   (Leopold	
   2004:	
   5).	
   This	
   again	
   is	
   only	
   a	
   problem	
   if	
   we	
   imagine	
   that	
   our	
  
concepts	
   must	
   be	
   absolute.	
   There	
   seems	
   to	
   be	
   little	
   difficulty	
   if	
   we	
   use	
   terms	
   such	
   as	
  
“syncretism”	
   or	
   “hybridity”	
   in	
   a	
   relative	
   sense,	
   marking	
   phenomena	
   whose	
   mixed	
   nature	
   is	
  
more	
   prominent	
   from	
   a	
   certain	
   perspective,	
   or	
   in	
   a	
   contextual	
   sense,	
   using	
   the	
   terms	
   as	
  
shorthand	
  to	
  highlight	
  selected	
  aspects	
  of	
  a	
  given	
  case.	
  As	
  Brian	
  Stross	
  puts	
  it,	
  	
  
	
  
One	
  might	
  say	
  that	
  there	
  are	
  no	
  truly	
  ‘pure’	
  forms,	
  …	
  completely	
  homogeneous	
  in	
  composition	
  (culturally)	
  
and	
   perhaps	
   never	
   have	
   been.	
   Thus	
   everything	
   is	
   a	
   ‘hybrid’	
   of	
   sorts.	
   Yet	
   the	
   term	
   has	
   both	
   utility	
   and	
  
meaning	
  for	
  most	
  of	
  us.	
  …	
  Pure	
  in	
  this	
  context	
  means	
  relatively	
  more	
  homogeneous	
  in	
  character	
  …,	
  having	
  
less	
   internal	
   variation.	
   Hybrid	
   …	
   is	
   of	
   course	
   more	
   heterogeneous	
   in	
   character,	
   having	
   more	
   internal	
  
variation.	
  (1999:	
  258)	
  

	
  
However,	
   these	
   uses	
   of	
   hybridity	
   terms	
   are	
   weak.	
   It	
   is	
   trivially	
   correct	
   but	
   hardly	
   helpful	
   to	
  
note	
  that	
  purity	
  and	
  hybridity	
  are	
  relative	
  terms.	
  This	
  offers	
  little	
  analytical	
  leverage	
  beyond	
  
highlighting	
   mixture	
   as	
   a	
   topic	
   of	
   interest,	
   leaving	
   the	
   important	
   work	
   to	
   a	
   closer	
  
consideration	
   of	
   what	
   is	
   mixed,	
   how,	
   to	
   what	
   degree,	
   under	
   what	
   circumstances,	
   and	
   with	
  
what	
  effects.	
  If,	
  as	
  scholars	
  of	
  religion	
  have	
  long	
  recognized,	
  hybridity	
  and	
  syncretism	
  are	
  the	
  
norm	
  rather	
  than	
  the	
  exception,	
  what	
  needs	
  explaining	
  is	
  why	
  so	
  much	
  importance	
  is	
  placed	
  
on	
   allegedly	
   pure	
   precedents	
   and	
   progenitors:	
   “what	
   is	
   problematic	
   is	
   not	
   hybridity	
   but	
   the	
  
fetishism	
  of	
  boundaries	
  that	
  has	
  marked	
  so	
  much	
  of	
  history”	
  (Nederveen	
  Pieterse	
  2001:	
  221);	
  
“Syncretism,	
  acculturation,	
  hybridity,	
  and	
  the	
  creole	
  are	
  no	
  longer	
  the	
  riddle	
  to	
  be	
  solved.	
  It	
  is	
  
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
2	
  This	
  section	
  draws	
  on	
  parts	
  of	
  Engler	
  2006.	
  

rather	
   zones	
   of	
   religious	
   purity	
   and	
   stability	
   that	
   now	
   seem	
   most	
   worthy	
   of	
   curiosity”	
  
(Johnson	
  2002b:	
  308).	
  
A	
   third	
   critique	
   of	
   hybridity	
   is	
   that	
   it	
   overemphasizes	
   diachronic	
   differences,	
   valuing	
  
historical	
   origins/roots	
   over	
   hybrid	
   actualities,	
   or	
   vice	
   versa.	
   This	
   distinction	
   has	
   much	
   in	
  
common	
   with	
   the	
   distinction	
   between	
   “real”	
   and	
   invented	
   traditions:	
   both	
   distinctions	
   are	
  
misleading	
   if	
   overly	
   sharp	
   and	
   especially	
   when	
   this	
   descriptive	
   distinction	
   is	
   given	
   a	
  
normative	
  dimension,	
  e.g.,	
  overemphasizing	
  the	
  static	
  nature	
  of	
  the	
  “old”	
  and	
  the	
  self-­‐serving	
  
tactical	
   innovations	
   of	
   the	
   “new”	
   (Engler	
   2005a;	
   2005b).	
   Ideological	
   appeals	
   to	
   invented	
  
traditions	
   can	
   be	
   smuggled	
   in	
   along	
   with	
   the	
   celebration	
   of	
   hybridity:	
   “in	
   signifying	
   the	
  
present,	
  something	
  comes	
  to	
  be	
  repeated,	
  relocated	
  and	
  translated	
  in	
  the	
  name	
  of	
  tradition,	
  in	
  
the	
   guise	
   of	
   a	
   pastness	
   that	
   is	
   not	
   necessarily	
   a	
   faithful	
   sign	
   of	
   historical	
   memory	
   but	
   a	
  
strategy	
  of	
  representing	
  authority	
  in	
  terms	
  of	
  the	
  artifice	
  of	
  the	
  archaic”	
  (Bhabha	
  1994:	
  35).	
  
Stuart	
   Hall	
   is	
   more	
   optimistic,	
   “hybrids	
   retain	
   strong	
   links	
   to	
   and	
   identifications	
   with	
   the	
  
traditions	
  and	
  places	
  of	
  their	
  ‘origin.’	
  But	
  they	
  are	
  without	
  the	
  illusions	
  of	
  any	
  return	
  to	
  the	
  
past”	
   (Hall	
   1993:	
   363).	
   Once	
   again,	
   the	
   warning	
   for	
   scholars	
   of	
   religion	
   tempted	
   to	
   work	
   with	
  
the	
  concept	
  of	
  hybridity	
  is	
  to	
  pay	
  explicit	
  attention	
  to	
  the	
  potential	
  problems	
  with	
  the	
  concept	
  
that	
  scholars	
  in	
  other	
  disciplines	
  have	
  already	
  flagged.	
  
Fourth,	
  “hybridity”	
  underemphasizes	
  synchronic	
  differences.	
  According	
  to	
  John	
  Hutnyk,	
  the	
  
concept	
   leads	
   to	
   a	
   “flattening	
   of	
   difference	
   [which]	
   is	
   secured	
   at	
   the	
   very	
   moment	
   that	
  
celebrates	
   difference	
   and	
   the	
   creative	
   productivity	
   of	
   new	
   mixings”	
   (Hutnyk	
   2005:	
   96).	
   On	
  
this	
  view,	
  “hybridity”	
  draws	
  attention	
  to	
  superficial	
  distinctions	
  while	
  erasing	
  more	
  important	
  
ones:	
   it	
   “is	
   inauthentic,	
   without	
   roots,	
   for	
   the	
   elite	
   only,	
   does	
   not	
   reflect	
   social	
   realities	
   on	
   the	
  
ground.	
  It	
  is	
  multiculturalism	
  lite,	
  highlights	
  superficial	
  confetti	
  culture	
  and	
  glosses	
  over	
  deep	
  
cleavages	
  that	
  exist	
  on	
  the	
  ground”;	
  above	
  all,	
  “hybridity”	
  assumes	
  equality,	
  hiding	
  issues	
  of	
  
power	
  (Nederveen	
  Pieterse	
  2001:	
  221,	
  224).	
  	
  
Fifth,	
   hybridity	
   has	
   become	
   too	
   glibly	
   associated	
   with	
   a	
   specific	
   political	
   agenda,	
  
sidestepping	
  the	
  detailed	
  analyses	
  needed	
  to	
  specify	
  this	
  relation	
  more	
  carefully.	
  Hybridity	
  is	
  
often	
  celebrated	
  precisely	
  because	
  hybridization	
  is	
  allegedly	
  a	
  politically	
  significant	
  process	
  of	
  
resistance	
   to,	
   for	
   example,	
   the	
   homogeneity	
   of	
   a	
   global	
   consumer	
   culture:	
   “Hybridity	
   has	
  
today	
   developed	
   into	
   a	
   code	
   word	
   associated	
   to	
   a	
   large	
   extent	
   with	
   hegemonic	
   politics”	
  
(Moreiras	
   1999:	
   388).	
   As	
   John	
   Hutnyk	
   notes,	
   it	
   is	
   often	
   the	
   case	
   that	
   “assertions	
   of	
   identity	
  
and	
   difference	
   are	
   celebrated	
   too	
   quickly	
   as	
   resistance,	
   in	
   either	
   the	
   nostalgic	
   form	
   of	
  

‘traditional	
  survivals’	
  or	
  mixed	
  in	
  a	
  ‘new	
  world	
  of	
  hybrid	
  forms’”	
  (Hutnyk	
  2005:	
  80).	
   Hutnyk’s	
  
ironic	
  conclusion	
  is	
  that	
  this	
  allegedly	
  political	
  attention	
  to	
  hybridity	
  fails	
  precisely	
  because	
  its	
  
conception	
  of	
  politics	
  is	
  overly	
  superficial.	
  Talking	
  “hybrids”	
  is	
  not	
  a	
  way	
  of	
  being	
  political	
  but	
  
rather	
  of	
  avoiding	
  doing	
  so;	
  it	
  offers	
  a	
  nod	
  and	
  a	
  wink	
  that	
  substitute	
  for	
  the	
  difficult	
  work	
  of	
  
getting	
  down	
  to	
  cases:	
  “syncretism	
  and	
  hybridity	
  are	
  academic	
  conceptual	
  tools	
  providing	
  an	
  
alibi	
   for	
   lack	
   of	
   attention	
   to	
   politics”	
  (Hutnyk	
   2005:	
   92).	
   These	
   last	
   two	
   critiques,	
   again,	
   are	
  
more	
  reminders	
  to	
  proceed	
  with	
  caution	
  than	
  reasons	
  to	
  abandon	
  the	
  concept.	
  
Gilroy’s	
   often-­‐cited	
   rant	
   against	
   “anterior	
   purities”	
   offers	
   a	
   useful	
   summary	
   of	
   these	
   issues,	
  
in	
  part	
  despite	
  its	
  explicit	
  thrust:	
  	
  
	
  
Which	
  culture	
  is	
  not	
  …	
  hybrid?	
  The	
  idea	
  of	
  ‘hybridity,’	
  of	
  intermixture,	
  presupposes	
  two	
  anterior	
  purities	
  ...	
  
[T]here	
  isn’t	
  any	
  purity;	
  there	
  isn’t	
  any	
  anterior	
  purity….	
  I	
  try	
  not	
  to	
  use	
  the	
  word	
  ‘hybrid’,	
  because	
  there	
  are	
  
degrees	
  of	
  it,	
  and	
  there	
  are	
  different	
  mixes...	
  Cultural	
  production	
  is	
  not	
  like	
  mixing	
  cocktails.	
  What	
  people	
  
call	
  ‘hybridity’,	
  I	
  used	
  to	
  call	
  ‘syncretism’...	
  I	
  would	
  prefer	
  to	
  stick	
  with	
  that—syncretism	
  is	
  the	
  norm,	
  but,	
  
that	
   dry	
   anthropological	
   word	
   does	
   not	
   have	
   any	
   poetic	
   charge	
   to	
   it.	
   There	
   isn’t	
   any	
   purity.	
   Who	
   the	
   fuck	
  
wants	
  purity?	
  Where	
  purity	
  is	
  called	
  for,	
  I	
  get	
  suspicious.	
  (Gilroy	
  1994:	
  54-­‐5)	
  

	
  
Ironically,	
  Gilroy’s	
  simile	
  of	
  the	
  cocktail	
  doesn’t	
  do	
  what	
  it	
  is	
  meant	
  to	
  do,	
  but	
  its	
  failure	
  sheds	
  
unintentional	
  light	
  on	
  several	
  dimensions	
  of	
  hybridity.	
  Gilroy	
  says,	
  “Cultural	
  production	
  is	
  not	
  
like	
   mixing	
   cocktails,”	
   but	
   this	
   simile	
   fails	
   spectacularly:	
   the	
   bottles	
   on	
   the	
   shelf	
   above	
   a	
  
cocktail	
   bar	
   themselves	
   contain	
   mixtures.	
   In	
   cocktail	
   mixing	
   as	
   well,	
   there	
   are	
   no	
   anterior	
  
purities:	
  in	
  that	
  sense	
  cultural	
  production	
  is	
  like	
  mixing	
  cocktails.	
  A	
  Manhattan,	
  for	
  example,	
  is	
  
made	
   from	
   Canadian	
   rye	
   whiskey,	
   Italian	
   or	
   French	
   sweet	
   vermouth,	
   and	
  
Trinidadian/Tobagan	
   or	
   Venezuelan	
   Angostura	
   bitters,	
   garnished	
   with	
   Maraschino	
   cherries.	
  
And,	
  of	
  course,	
  these	
  ingredients	
  themselves	
  are	
  mixtures:	
  for	
  example,	
  the	
  cherries	
  are	
  made	
  
by	
   soaking	
   them	
   in	
   Maraschino,	
   a	
   liqueur,	
   invented	
   by	
   sixteenth-­‐century	
   Dominican	
   monks	
   in	
  
Zadar,	
   Croatia,	
   which	
   is	
   fermented	
   from	
   Italian,	
   Croatian	
   or	
   Slovenian	
   Marasca	
   cherries,	
  
selected	
   herbs,	
   and	
   tropical	
   cane	
   syrup.	
   The	
   difference	
   is	
   not	
   one	
   of	
   origin	
   or	
   nature	
   but	
   of	
  
perceptions	
  or	
  framing.	
  Bottled	
  cocktail	
  ingredients	
  are	
  packaged,	
  branded,	
  and	
  marketed	
  as	
  
distinct	
  and	
  unitary	
  products;	
  cocktails	
  are	
  marketed	
  as	
  mixtures	
  of	
  these.	
  Both	
  commodities	
  
gain	
  in	
  value	
  due	
  to	
  the	
  perceived	
  naturalness	
  or	
  legitimacy	
  of	
  the	
  distinction	
  between	
  unitary	
  
originals	
   and	
   hybrid	
   product.	
   Where	
   everything	
   is	
   a	
   mixture,	
   the	
   question	
   becomes	
   when	
   and	
  

why	
  certain	
  mixtures	
  are	
  presented	
  or	
  perceived	
  as	
  pure.	
  	
  
Distilling	
   Gilroy’s	
   cocktail	
   simile	
   draws	
   attention	
   to	
   four	
   characteristics	
   of	
   hybridity	
   that	
  
serve	
  to	
  summarize	
  a	
  set	
  of	
  issues	
  that	
  scholars	
  of	
  religion	
  need	
  to	
  pay	
  attention	
  to:	
  	
  
	
  
•

The	
  contrast	
  between	
  unitary	
  originals	
  and	
  hybrid	
  product	
  is,	
  to	
  an	
  important	
  extent,	
  a	
  
construct.	
  The	
  distinction	
  between	
  pure	
  and	
  unadulterated	
  is	
  a	
  relative	
  one.	
  

•

Asserting	
   the	
   pure/impure	
   contrast	
   is	
   a	
   common	
   tactic,	
   but	
   only	
   one	
   of	
   many,	
   for	
  
projecting	
  normative	
  force	
  on	
  this	
  artificial	
  boundary.	
  

•

This	
   boundary	
   is	
   often	
   constructed	
   in	
   terms	
   of	
   a	
   diachronic	
   dimension,	
   with	
   further	
  
normative	
   force	
   drawing	
   on	
   the	
   distinction	
   between	
   tradition	
   (long-­‐established	
  
ingredients)	
  and	
  innovation	
  (new	
  mixture).	
  

•

Once	
   this	
   boundary	
   has	
   been	
   legitimized,	
   reified	
   or	
   naturalized,	
   eliding	
   it	
   can	
   have	
  
further	
  ideological	
  effects.	
  

	
  
This	
  forces	
  us	
  to	
  clarify	
  exactly	
  what	
  relative	
  and	
  contextualized	
  leverage	
  we	
  seek	
  to	
  gain	
  by	
  
using	
   “hybridity”	
   or	
   other	
   terms	
   to	
   point	
   to	
   mixtures.	
   Unless	
   we	
   problematize	
   the	
   concept	
  
adequately,	
  talking	
  about	
  hybrididy	
  is	
  just	
  as	
  vague	
  and	
  unhelpful	
  as	
  much	
  talk	
  of	
  syncretism	
  
has	
  been	
  in	
  the	
  field	
  of	
  Religious	
  Studies.	
  The	
  concept	
  is	
  only	
  useful	
  if	
  grounded:	
  	
  
	
  
Hybridity	
   becomes	
   a	
   floating	
   signifier	
   ripe	
   for	
   appropriation,	
   precisely	
   because	
   we	
   use	
   the	
   concept	
   without	
  
rigorous	
  theoretical	
  grounding….	
  [A]	
  nongrounded	
  use	
  of	
  hybridity	
  is	
  detrimental	
  to	
  theorizing	
  …	
  because	
  it	
  
encourages	
   superficial	
   uses	
   of	
   the	
   concept.	
   Such	
   uses	
   will	
   tend	
   to	
   be	
   descriptive	
   rather	
   than	
   analytical,	
  
utilitarian	
  rather	
  than	
  critical.	
  (Kraidy	
  2002:	
  323)	
  

	
  
Talk	
  of	
  hybridity	
  too	
  often	
  simply	
  points	
  at	
  difference;	
  two	
  further	
  steps	
  are	
  required.	
  First,	
  
we	
   need	
   to	
   pay	
   more	
   attention	
   to	
   a	
   close	
   analysis	
   of	
   specific	
   cases,	
   examining	
   the	
   specific	
  
social,	
   material	
   and	
   ideological	
   contexts	
   where	
   these	
   processes	
   work	
   themselves	
   out.	
   Second,	
  
we	
   must	
   go	
   beyond	
   the	
   basic	
   work	
   of	
   describing	
   hybrid	
   forms	
   to	
   ground	
   the	
   concept	
   more	
  
firmly	
   in	
   theory.	
   With	
   greater	
   attention	
   to	
   the	
   specific	
   details	
   of	
   what	
   is	
   mixed	
   under	
   what	
  
circumstances	
   and	
   with	
   what	
   recourse	
   to	
   discourses	
   of	
   purity,	
   the	
   concept	
   can	
   offer	
   useful	
  
analytical	
   leverage.	
   The	
   following	
   section	
   offers	
   a	
   brief	
   overview	
   of	
   Umbanda,	
   highlighting	
  

those	
  characteristics	
  that	
  will	
  be	
  drawn	
  upon	
  in	
  the	
  final	
  analytical	
  section.	
  
	
  

	
  

The	
  Spectrum	
  of	
  Umbandas	
  	
  
Umbanda	
   is	
   a	
   distinctively	
   Brazilian	
   religion	
   that	
   reflects	
   the	
   course	
   of	
   urbanization	
   in	
  
modern	
   Brazil	
   (Ortiz	
   1975:	
   89;	
   1999[1978]:	
   214).	
   It	
   formed	
   in	
   the	
   1920s	
   and	
   1930s	
   as	
   a	
   self-­‐
conscious	
   mixture	
   of	
   two	
   traditions	
   that	
   are	
   themselves	
   mixtures:	
   Candomblé,	
   an	
   Afro-­‐
Brazilian	
   tradition	
   that	
   evolved	
   as	
   various	
   West	
   African	
   beliefs	
   and	
   practices	
   encountered	
  
early-­‐modern	
   Iberian	
   Catholicism	
   in	
   the	
   context	
   of	
   a	
   colonial	
   slavery	
   system;	
   and	
   Kardecist	
  
Spiritism,	
  a	
  rationalized	
  French	
  version	
  of	
  American	
  Spiritualism,	
  with	
  strong	
  mesmerist	
  and	
  
Christian	
   influences.3	
   Two	
   other	
   religious	
   influences	
   were	
   also	
   important:	
   Catholicism,	
  
primarily	
   indirectly	
   through	
   both	
   Kardecism	
   and	
   Candomblé	
   but	
   also	
   with	
   some	
   direct	
  
influences;	
  and	
  indigenous	
  traditions,	
  not	
  directly	
  but	
  through	
  the	
  imagined	
  and	
  romanticized	
  
figure	
   of	
   the	
   caboclo	
   (Concone	
   2001),	
   but	
   also	
   (perhaps	
   more	
   directly)	
   in	
   possession	
   by	
   the	
  
spirits	
   of	
   animals	
   in	
   some	
   northeastern	
   groups	
   (Toop	
   1972:	
   73).	
   Arthur	
   Ramos,	
   in	
   a	
   classic	
  
study,	
  noted	
  that	
  “in	
  Brazil,	
  there	
  are	
  no	
  longer	
  pure	
  African	
  cults,	
  in	
  terms	
  of	
  their	
  origin”	
  and	
  
pointed	
   to	
   seven	
   distinct	
   degrees	
   of	
   mixture,	
   “in	
   order	
   of	
   increasing	
   syncretism,”	
   including	
  
Yoruba	
   and	
   Bantu	
   variants	
   as	
   well	
   as	
   Islamic,	
   “caboclo,”	
   Kardecist	
   and	
   Catholic	
   elements	
  
(2001[1934]:	
   138).	
   Umbanda,	
   the	
   most	
   recent	
   major	
   religious	
   innovation	
   to	
   draw	
   on	
   Afro-­‐
Brazilian	
   roots,	
   incorporates	
   a	
   broad	
   spectrum	
   of	
   these	
   influences.	
   In	
   this	
   context,	
   Ortiz	
  
argues	
   that	
   Umbanda	
   has	
   moved	
   past	
   syncretism	
   to	
   synthesis:	
   “If	
   ‘candomblé’	
   and	
   ‘macumba’	
  
are	
   African	
   religions,	
   the	
   spiritism	
   of	
   Umbanda	
   is,	
   on	
   the	
   contrary,	
   a—I	
   would	
   say	
   the—
national	
  religion	
  of	
  Brazil”	
  (1975:	
  96;	
  original	
  emphasis).	
  Further	
  clarification	
  of	
  Umbanda’s	
  
hybridity	
   is	
   clearly	
   in	
   order.	
   A	
   fruitful	
   place	
   to	
   begin	
   is	
   by	
   noting	
   that	
   it	
   is	
   a	
   hybrid	
   of	
   hybrids,	
  
and	
  one	
  that	
  reflects	
  the	
  social	
  and	
  historical	
  context	
  of	
  its	
  emergence:	
  “Umbanda	
  is	
  a	
  religion	
  
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
3	
  On	
  Candomblé	
  see	
  Carneiro	
  1977[1948],	
  Bastide	
  1960,	
  Prandi	
  1991,	
  Johnson	
  2002a,	
  Harding	
  2005	
  and	
  Silva	
  

2005[1994].	
   On	
   Kardecism	
   see	
   Kloppenburg	
   1964,	
   Bastide	
   1967,	
   Warren	
   1968,	
   Camargo	
   1973,	
   Aubrée	
   and	
  
Laplantine	
   1990,	
   Hess	
   1991,	
   and	
   Negrão	
   2005[1987].	
   On	
   Umbanda	
   see	
   Montero	
   1985,	
   Brown	
   and	
   Bick	
   1987,	
  
Brumana	
  and	
  Martinez	
  1989,	
  Brown	
  1994[1986],	
  Negrão	
  1996,	
  and	
  Ortiz	
  1999	
  [1978].	
  On	
  Neo-­‐Pentecostalism	
  
and	
   its	
   ritual	
   focus	
   (in	
   exorcism)	
   on	
   these	
   same	
   spirits	
   and	
   orixás	
   see	
   Birman	
   1997,	
   Campos	
   1999[1997],	
  
Mariano	
   1999,	
   and	
   Oro	
   2007.	
   I	
   use	
   the	
   word	
   “tradition”	
   to	
   point	
   to	
   the	
   dynamic	
   tension	
   between	
   strategies	
   of	
  
legitimation	
  and	
  authority	
  offered	
  by	
  ‘authentic’	
  and	
  ‘invented’	
  traditions	
  (Engler	
  2005a;	
  2005b).	
  All	
  translations	
  
from	
  Portuguese	
  and	
  French	
  are	
  mine.	
  

of	
  a	
  new	
  model	
  of	
  society,	
  as	
  Kardecism	
  was	
  previously”	
  (Prandi	
  1991:	
  62;	
  see	
  Fry	
  1982).	
  	
  
These	
   religions	
   are	
   relatively	
   small.	
   In	
   the	
   2000	
   census,	
   2.2	
   million	
   Brazilians	
   self-­‐
identified	
   as	
   Kardecists	
   and	
   397,000	
   as	
   Umbandists.	
   Candomblé,	
   the	
   largest	
   of	
   the	
   Afro-­‐
Brazilian	
   traditions,	
   is	
   much	
   smaller,	
   with	
   only	
   118,000	
   Brazilians	
   claiming	
   this	
   as	
   their	
  
primary	
   religious	
   affiliation	
   (Jacob	
   et	
   al.	
   2003:	
   101-­‐105).	
   More	
   nuanced	
   analyses	
   report	
  
slightly	
  higher	
  numbers	
  	
  (Pierucci	
  and	
  Prandi	
  2000).	
  Due	
  to	
  the	
  fact	
  that	
  Umbanda	
  is	
  seen	
  as	
  a	
  
provider	
   of	
   physical	
   and	
   spiritual	
   healing	
   services,	
   a	
   much	
   larger	
   number	
   of	
   Brazilians	
  
participate	
   regularly	
   in	
   the	
   rituals	
   of	
   Umbanda,	
   though	
   they	
   do	
   not	
   consider	
   themselves	
  
members	
  of	
  the	
  religion.4	
  
The	
  two	
  main	
  “roots”	
  of	
  Umbanda	
  are	
  quite	
  distinct	
  from	
  each	
  other.	
  Especially	
  significant	
  
in	
   the	
   emergence	
   of	
   Umbanda	
   are	
   factors	
   of	
   race	
   and	
   class	
   in	
   the	
   mixture	
   of	
   these	
   anterior	
  
impurities.	
   Candomblé,	
   one	
   of	
   a	
   wide	
   range	
   of	
   Afro-­‐Brazilian	
   religions,	
   places	
   fundamental	
  
emphasis	
  on	
  the	
  possession	
  of	
  initiated	
  members	
  by	
  orixás	
  (divinities	
  originating	
  primarily	
  in	
  
various	
   West	
   African	
   cultures	
   and,	
   at	
   times,	
   associated	
   with	
   Christian	
   saints).	
   Terreiros	
  
(grounds)	
   are	
   organized	
   as	
   a	
  familia-­‐de-­‐santo	
   under	
   the	
   leadership	
   of	
   the	
  pai-­‐de-­‐santo,	
   or	
   less	
  
commonly	
   mãe-­‐de-­‐santo	
   (saint	
   father/mother).	
   Key	
   rituals	
   include	
   the	
   roda-­‐de-­‐santo	
   (saint	
  
wheel)	
   in	
   which	
   initiated	
   members	
   dance	
   counter-­‐clockwise,	
   to	
   intensely	
   syncopated	
  
drumming,	
   until	
   they	
   enter	
   into	
   a	
   trance	
   state,	
   becoming	
   cavalos	
   (horses)	
   for	
   the	
   orixás,	
   as	
  
well	
  as	
  initiation	
  and	
  divination.	
  Candomblé	
  has	
  received	
  especially	
  intense	
  academic	
  scrutiny	
  
and,	
   arguably,	
   the	
   influential	
   studies	
   of	
   Edison	
   Carneiro	
   and	
   Roger	
   Bastide	
   went	
   beyond	
  
making	
   this	
   one	
   among	
   many	
   Afro-­‐Brazilian	
   religion	
   well	
   known:	
   “Bastide	
   did	
   not	
   limit	
  
himself	
  to	
  studying	
  Candomblé.	
  He	
  contributed	
  greatly	
  to	
  its	
  invention”	
  (Motta	
  1996:	
  32;	
  see	
  
Despland	
  2008).	
  	
  
There	
   are	
   a	
   number	
   of	
   differences	
   between	
   Candomblé	
   and	
   Umbanda	
   (see	
   Silva	
  
2005[1994]:	
   126-­‐127):	
   e.g.,	
   Umbanda	
   has	
   a	
   larger	
   and	
   more	
   doctrinally	
   elaborated	
   set	
   of	
  
supernatural	
   entities;	
   it	
   places	
   more	
   emphasis	
   on	
   mediumship	
   as	
   a	
   source	
   of	
   service	
   to	
  
clients;	
   it	
   places	
   less	
   emphasis	
   on	
   divination,	
   and	
   less	
   emphasis	
   on	
   the	
   pai-­‐de-­‐santo	
   as	
   central	
  
to	
   ritual	
   (often	
   foregoing	
   that	
   term	
   and	
   role	
   entirely);	
   it	
   places	
   less	
   or	
   no	
   emphasis	
   on	
  
initiation,	
   with	
   the	
   charismatic	
   authority	
   of	
   mediumship	
   playing	
   a	
   greater	
   role	
   than	
   the	
  
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
4	
   For	
   an	
   exemplary	
   study	
   of	
   the	
   appeal	
   of	
   Afro-­‐Brazilian	
   religions’	
   healing	
   functions,	
   resulting	
   in	
   multiple	
  

adherence	
  among	
  Catholics,	
  see	
  Oro	
  1989.	
  

ranking	
  of	
  initiation	
  and	
  period	
  of	
  study	
  in	
  its	
  institutional	
  hierarchy;	
  it	
  maintains	
  a	
  greater	
  
role	
  for	
  sorcery	
  (though	
  less	
  than	
  the	
  closely	
  related	
  religion	
  Quimbanda);	
  its	
  texts	
  and	
  hymns	
  
make	
  less	
  use	
  of	
  African	
  vocabulary	
  and,	
  at	
  the	
  “white”	
  end	
  of	
  the	
  spectrum	
  of	
  rituals	
  forms,	
  
uses	
   Christian	
   elements	
   (e.g.,	
   the	
   “Our	
   Father”).	
   These	
   differences	
   are	
   sufficient	
   that,	
   as	
  
Véronique	
   Boyer	
   suggests,	
   “Candomblé	
   and	
   Umbanda	
   form	
   poles,	
   tendencies	
   that	
   organize	
  
the	
  religious	
  universe	
  with	
  opposing	
  and	
  irreconcilable	
  currents”	
  (1996:	
  18).	
  
Kardecism	
   presents	
   itself	
   as	
   science,	
   philosophy	
   and	
   religion.	
   Its	
   beliefs	
   include	
   the	
  
possibility	
   of	
   communication	
   with	
   disembodied	
   spirits,	
   reincarnation,	
   karma,	
   the	
   universal	
  
spiritual	
   perfection	
   of	
   humankind,	
   “obsession”	
   caused	
   by	
   the	
   interference	
   of	
   non-­‐evolved	
  
spirits,	
   a	
   plurality	
   of	
   inhabited	
   worlds,	
   a	
   transcendent	
   God,	
   and	
   Jesus	
   Christ	
   as	
   an	
  
exceptionally	
   involved	
   spirit.	
   Key	
   rituals	
   include	
   consultation	
   with	
   or	
   reception	
   of	
   messages	
  
from	
  spirits	
  received	
  by	
  mediums,	
  the	
  passe	
  (a	
  form	
  of	
  blessing	
  similar	
  to	
  New	
  Age	
  cleansing	
  
of	
  the	
  aura)	
  and	
  study	
  sessions.	
  	
  
The	
   origin	
   of	
   Umbanda	
   and	
   its	
   ongoing	
   social	
   location	
   are	
   closely	
   tied	
   to	
   issues	
   of	
   race	
   and	
  
class.	
  Three	
  tendencies,	
  reflecting	
  the	
  Brazilian	
  “myth	
  of	
  three	
  races”	
  (indigenous,	
  black	
  and	
  
white	
   [DaMatta	
   1987:	
   58-­‐85])—were	
   present	
   in	
   the	
   formation	
   of	
   Umbanda.	
   First,	
   Spiritists	
  
looked	
  to	
  Afro-­‐Brazilian	
  traditions	
  for	
  a	
  more	
  intensely	
  emotional	
  and	
  corporeally	
  satisfying	
  
symbolism	
   and	
   ritual,	
   leading	
   to	
   the	
   empretecimento	
   (blackening)	
   of	
   Kardecism	
   (Ortiz	
  
1999[1978]:	
  40-­‐45):	
  	
  
	
  
[Umbanda’s	
   founders]	
   came	
   to	
   prefer	
   the	
   African	
   and	
   indigenous	
   spirits	
   and	
   divinities	
   present	
   in	
  
‘Macumba,’	
   considering	
   them	
   more	
   competent	
   than	
   the	
   highly	
   evolved	
   kardecist	
   spirits	
   in	
   terms	
   of	
   the	
   cure	
  
and	
  treatment	
  of	
  a	
  wide	
  range	
  of	
  diseases	
  and	
  other	
  problems.	
  They	
  found	
  the	
  rituals	
  of	
  ‘Macumba’	
  much	
  
more	
   stimulating	
   and	
   dramatic	
   than	
   those	
   of	
   Kardecism,	
   which	
   seemed	
   by	
   comparison	
   static	
   and	
   insipid.	
  
(Brown	
  1985:	
  11)	
  	
  

	
  
A	
   second	
   tendency	
   was	
   the	
   late	
   nineteenth-­‐century	
   embranquecimento	
   (whitening)	
   of	
  
Candomblé	
   and	
   other	
   Afro-­‐Brazilian	
   traditions,	
   due	
   to	
   two	
   factors,	
   primarily	
   in	
   Rio	
   de	
   Janeiro	
  
and	
   São	
   Paulo:	
   the	
   presence	
   of	
   increasing	
   numbers	
   of	
   white	
   members,	
   often	
   new	
   immigrants;	
  
and	
  the	
  formation	
  of	
  a	
  “low	
  Spiritism”	
  among	
  the	
  lower	
  classes	
  (Camargo	
  1961:	
  34-­‐35;	
  Ortiz	
  
1999[1978]:	
  4-­‐40).	
  Edison	
  Carneiro’s	
  study	
  of	
  Afro-­‐Brazilian	
  religions	
  in	
  Bahia	
  in	
  the	
  1930s,	
  
for	
   example,	
   found	
   Spiritist	
   ideas	
   being	
   absorbed	
   into	
   candomblé	
   de	
   caboclo,	
   a	
   tradition	
  

already	
  incorporating	
  both	
  African	
  and	
  indigenous	
  elements:	
  “it	
  is	
  notable	
  that	
  some	
  of	
  these	
  
cults	
   have	
   reduced	
   themselves	
   to	
   the	
   so-­‐called	
   baixo	
   espiritismo	
   (‘low-­‐spiritualism’)”	
  
(Carneiro	
   1940:	
   276,	
   original	
   emphasis;	
   cf.	
   Leacock	
   1964a;	
   Ortiz	
   1999[1978]:	
   36).	
   The	
  
appropriation	
   of	
   kardecist	
   elements	
   was	
   an	
   important	
   factor	
   in	
   this	
   aspect	
   of	
   Umbanda’s	
  
emergence:	
  “’Cleaning	
  up’	
  the	
  new	
  religion	
  of	
  those	
  elements	
  most	
  compromised	
  by	
  a	
  secret	
  
and	
  sacrificial	
  initiation	
  tradition	
  was	
  to	
  take	
  Kardecism	
  as	
  a	
  model,	
  one	
  capable	
  of	
  expressing	
  
the	
  ideas	
  and	
  values	
  of	
  the	
  new	
  republican	
  society…”	
  (Prandi	
  1991:	
  49).	
  
The	
   third	
   factor	
   involved	
   the	
   other	
   of	
   Brazil’s	
   three	
   races.	
   In	
   the	
   1920s,	
   a	
   number	
   of	
  
kardecist	
   mediums	
   began	
   to	
   receive	
   the	
   spirits	
   of	
   Brazilian	
   Indians.	
   The	
   presence	
   of	
   these	
  
caboclos	
   was	
   rejected	
   by	
   mainstream	
   Spiritism	
   as	
   impure	
   and	
   incompatible	
   with	
   universal	
  
human	
   spiritual	
   progress.	
   (The	
   term	
   “caboclo,”	
   often	
   used	
   by	
   outsiders	
   to	
   characterize	
  
residents	
   of	
   Amazonia,	
   conveys	
   racial	
   mixture,	
   but	
   has	
   negative	
   connotations	
   of	
   rural	
  
backwardness	
   and	
   simplicity	
   [Pace	
   1997].)	
   In	
   the	
   years	
   since,	
   some	
   Kardecists	
   have	
   been	
  
open	
   to	
   rapprochement	
   with	
   Umbanda,	
   but	
   a	
   firm	
   rejection	
   has	
   been	
   more	
   prominent;	
   a	
  
statement	
  from	
  the	
  Kardecist	
  press	
  is	
  typical:	
  “Any	
  confusion	
  between	
  Spiritism	
  and	
  primitive	
  
forms	
  of	
  mediumship	
  [or]	
  manifestations	
  of	
  religious	
  syncretism	
  …	
  are	
  nothing	
  more	
  than	
  a	
  
miscomprehension	
   of	
   Spiritist	
   Doctrine	
   and	
   cannot	
   be	
   incorporated”	
   (cited	
   in	
   Kloppenburg	
  
1964:	
   55-­‐57).	
   The	
   presence	
   of	
   caboclos—who	
   function	
   as	
   spirits	
   of	
   nature,	
   in	
   structural	
  
opposition	
   to	
   the	
   domestic	
   spirits	
   of	
   pretos-­‐velhos	
   and	
   children—continues	
   to	
   be	
   a	
   central	
  
characteristic	
   of	
   Umbanda	
   (Concone	
   2001;	
   Motta	
   de	
   Oliveira	
   2007).	
   Their	
   absence	
   is	
   a	
  
defining	
   characteristic	
   of	
   Kardecism.	
   Caboclos	
   have	
   been	
   an	
   element	
   of	
   some	
   Afro-­‐Brazilian	
  
traditions	
  since	
  the	
  early	
  twentieth	
  century,	
  especially	
  Catimbó,	
  Jurema,	
  and	
  Batuque	
  (Boyer	
  
1992;	
  J.T.	
  Santos	
  1992;	
  Harding	
  2005:	
  122;	
  Prandi	
  2005:	
  121-­‐138).	
  
These	
  symmetrical	
  tendencies	
  can	
  be	
  interpreted	
  in	
  opposing	
  ways.	
  On	
  the	
  one	
  hand,	
  it	
  is	
  
possible	
  to	
  portray	
  the	
  formation	
  of	
  Umbanda	
  as	
  one	
  of	
  harmonious	
  mediation	
  of	
  tensions	
  in	
  
Brazilian	
   society.	
   A	
   recent	
   introduction	
   to	
   Afro-­‐Brazilian	
   religions	
   suggests	
   that	
   Umbanda’s	
  
“development	
   was	
   marked	
   by	
   the	
   search,	
   initiated	
   by	
   white	
   segments	
   of	
   the	
   urban	
   middle-­‐
class,	
  for	
  a	
  model	
  of	
  religion	
  that	
  could	
  legitimately	
  integrate	
  the	
  contributions	
  of	
  the	
  groups	
  
composing	
   the	
   national	
   society”	
   (Silva	
   2005[1994]:	
   15).	
   On	
   the	
   other	
   hand,	
   these	
  
developments,	
   especially	
   the	
   embranquecimento	
   of	
   Candomblé,	
   were	
   racist:	
   	
   “pioneering	
  
umbandists	
   were	
   anxious	
   to	
   situate	
   the	
   origins	
   of	
   Umbanda	
   within	
   the	
   respectability	
   of	
   the	
  
world’s	
   great	
   mystic	
   traditions,	
   and	
   they	
   envisioned	
   their	
   mission	
   to	
   be	
   that	
   of	
   saving	
  

Umbanda	
  from	
  the	
  negative	
  influences	
  associated	
  with	
  its	
  African	
  past,	
  and	
  of	
  purifying	
  it	
  of	
  
its	
  African	
  practices”	
  (Brown	
  1977:	
  33).	
   On	
  the	
  other	
  hand,	
  the	
  racism	
  that	
  was	
  a	
  dominant	
  
factor	
   in	
   the	
   emergence	
   of	
   Umbanda,	
   the	
   rejection	
   of	
   the	
   spirits	
   of	
   departed	
   black	
   and	
  
indigenous	
  people	
  as	
  unevolved,	
  was	
  inverted	
  to	
  some	
  extent	
  by	
  the	
  centrality	
  of	
  these	
  spirits	
  
in	
  Umbanda.	
  According	
  to	
  an	
  Umbanda	
  practitioner’s	
  guide,	
  	
  
	
  
The	
   pretos-­‐velhos	
   and	
   caboclos	
   …	
   were	
   rejected,	
   due	
   to	
   many	
   [Kardecist]	
   leaders’	
   lack	
   of	
   comprehension.	
  
Some	
  mediums	
  disagreed	
  with	
  this	
  discrimination,	
  because	
  the	
  disembodied	
  spirits	
  that	
  present	
  themselves	
  
as	
   ‘pretos-­‐velhos’	
   are,	
   for	
   the	
   most	
   part,	
   highly	
   evolved	
   spirits,	
   on	
   a	
   mission	
   of	
   charity.	
   (Pinto	
   and	
   Freitas	
  
1972:	
  29)	
  

	
  
The	
   fact	
   that	
   certain	
   racialized	
   doctrinal	
   and	
   ritual	
   tensions	
   led	
   to	
   the	
   formation	
   of	
   a	
   new	
  
religion	
   in	
   large	
   part	
   reflects	
   the	
   fact	
   that	
   the	
   social	
   sphere	
   where	
   Umbanda	
   originated	
  
straddled	
  racial	
  and	
  class	
  boundaries:	
  black/white;	
  and	
  lower-­‐class/middle-­‐class.	
  This	
  is	
  not	
  
to	
  suggest	
  that	
  there	
  exist	
  sharp	
  and	
  rigid	
  boundaries	
  between	
  these	
  groups	
  in	
  Brazil—racial	
  
and	
   social	
   distinctions	
   are	
   blurred—though	
   the	
   extremes	
   of	
   the	
   spectra	
   exhibit	
   dramatic	
  
differences	
   in	
   economic	
   and	
   political	
   power	
   as	
   well	
   as	
   in	
   cultural	
   status	
   and	
   capital.	
   Rather,	
  
the	
  diffusion	
  of	
  religious	
  ideas	
  and	
  the	
  adherence	
  of	
  new	
  types	
  of	
  members	
  led	
  to	
  a	
  broader	
  
than	
   usual	
   mix	
   of	
   co-­‐religionists.	
   It	
   is	
   important	
   to	
   keep	
   in	
   mind	
   that	
   race	
   in	
   Brazil	
   is	
   a	
  
complex	
   issue,	
   with	
   tensions	
   less	
   sharply	
   defined	
   than	
   in	
   other	
   areas	
   of	
   Latin	
   America	
   [Lovell	
  
and	
  Wood	
  1998;	
  Hoffman	
  and	
  Centeno	
  2003;	
  Fischer	
  2004].	
  As	
  a	
  result,	
  even	
  explicit	
  talk	
  of	
  
race	
  sometimes	
  masks,	
  e.g.,	
  talk	
  of	
  class,	
  and	
  vice	
  versa.	
  One	
  aspect	
  of	
  this	
  complexity,	
  and	
  of	
  
Umbanda’s	
  reflecting	
  social	
  tensions,	
  lies	
  in	
  the	
  emergence	
  of	
  a	
  new	
  critical	
  voice	
  among	
  the	
  
preto	
  velho	
  spirits	
  in	
  some	
  (but	
  certainly	
  not	
  all)	
  terreiros,	
  spirits	
  that	
  have	
  radically	
  shifted	
  
from	
   wise	
   house-­‐slaves	
   to	
   incisive	
   critics	
   of	
   racial,	
   gender	
   and	
   other	
   inequalities	
   in	
   Brazil	
  
(Hale	
  1997).	
  
Given	
  its	
  historical	
  development,	
  Umbanda	
  highlights	
  several	
  tensions	
  in	
  Brazilian	
  society.	
  
This	
  manifests	
  itself	
  in	
  tensions	
  internal	
  to	
  Umbanda	
  itself.	
  There	
  is	
  a	
  spectrum	
  between	
  the	
  
“white”	
   Umbanda,	
   closer	
   to	
   Kardecism,	
   and	
   the	
   popular	
   Umbanda	
   closer	
   to	
   Candomblé	
  
(Birman	
  1983:	
  80-­‐94).	
  The	
  particular	
  set	
  of	
  constitutive	
  elements	
  in	
  a	
  given	
  Umbanda	
  terreiro	
  
varies	
   along	
   a	
   spectrum	
   ranged	
   between	
   kardecist	
   and	
   Afro-­‐Brazilian	
   extremes:	
   “There	
   is	
   not	
  
one	
  Umbanda	
  but	
  many	
  Umbandas,	
  with	
  a	
  great	
  diversity	
  in	
  beliefs	
  and	
  rituals”	
  (Motta	
  2006	
  

[1999]:	
   25;	
   original	
   emphasis).	
   With	
   relatively	
   few	
   institutionally	
   imposed	
   or	
   maintained	
  
norms,	
   individual	
   Umbanda	
   terreiros	
   continue	
   to	
   manifest	
   this	
   spectrum	
   of	
   doctrinal	
   and	
  
ritual	
   characteristics,	
   from	
   Kardecist-­‐like	
   to	
   Candomblé-­‐like.	
   The	
   former	
   is,	
   to	
   a	
   greater	
  
extent,	
  a	
  middle-­‐class	
  phenomenon	
  and	
  the	
  latter	
  includes	
  a	
  greater	
  proportion	
  of	
  lower-­‐class	
  
members.	
   (Bastide’s	
   early	
   assertion	
   [1960;	
   1967],	
   later	
   qualified	
   [1974],	
   that	
   Umbanda	
   was	
  
primarily	
  and	
  uniformly	
  a	
  lower-­‐class	
  religion	
  has	
  been	
  soundly	
  rejected	
  [Ortiz	
   1999	
  {1978};	
  
Negrão	
  1979;	
  Brown	
  1994{1986}].)	
  
This	
  spectrum	
  is	
  also	
  correlated,	
  to	
  some	
  extent,	
  with	
  racial	
  variation:	
  Umbanda	
  branca	
  is	
  
“white”	
   not	
   only	
   because	
   it	
   places	
   more	
   explicit	
   emphasis	
   on	
   white	
   magic.	
   This	
   spectrum	
   is	
   in	
  
turn	
   correlated	
   with	
   different	
   manners	
   of	
   foregrounding	
   the	
   issue	
   of	
   origins.	
   White	
   Umbanda	
  
tends	
  to	
  downplay	
  Afro-­‐Brazilian	
  ritual	
  form,	
  though	
  it	
  preserves	
  the	
  pretos-­‐velhos	
  and	
  sees	
  
Umbanda’s	
  internal	
  fragmentation	
  as	
  degenerate.	
  The	
  point	
  is	
  not	
  that	
  Umbanda	
  is	
  race-­‐blind,	
  
though	
  this	
  is	
  asserted	
  by	
  umbandists:	
  “Umbanda	
  does	
  not	
  discriminate	
  against	
  blacks,	
  has	
  no	
  
prejudices,	
  neither	
  of	
  class	
  nor	
  colour”	
  (Matta	
  e	
  Silva	
  2004	
  [1969]:	
  33).	
  Rather,	
  it	
  manifests	
  a	
  
spectrum	
  of	
  beliefs	
  and	
  practices	
  that	
  reflects	
  the	
  social	
  spectra	
  of	
  race	
  and	
  class	
  in	
  Brazil.	
  	
  
It	
  is	
  worth	
  underlining	
  this	
  point	
  in	
  order	
  to	
  avoid	
  a	
  misreading.	
  People	
  of	
  all	
  classes	
  and	
  
races	
  participate	
  in	
  Candomble	
  and	
  Kardecism	
  as	
  well,	
  but	
  there	
  is	
  not	
  the	
  same	
  spectrum	
  of	
  
intra-­‐religious	
  phenomena	
  varying	
  in	
  correlation	
  with	
  racial	
  and	
  socio-­‐economic	
  factors.	
  It	
  is	
  
not	
   that	
   candomblecistas	
   are	
   poor	
   and	
   black	
   and	
   kardecistas	
   affluent	
   and	
   white,	
   with	
  
umbandistas	
   occyping	
   a	
   spectrum	
   of	
   demographically-­‐determined	
   position	
   in	
   between.	
  
Census	
   data	
   paint	
   a	
   much	
   more	
   nuanced	
   picture	
   (Jacob	
   et	
   al.	
   2003;	
   2006).	
   However,	
   the	
  
variables	
   of	
   (i)	
   doctrinal	
   elaboration,	
   ritual	
   form	
   and	
   institutional	
   structures	
   and	
   (ii)	
   race	
   and	
  
socio-­‐economic	
  status	
  track	
  each	
  other	
  to	
  a	
  much	
  greater	
  extent	
  in	
  the	
  case	
  of	
  Umbanda.	
  
The	
   spectrum	
   of	
   Umbanda	
   also	
   varies	
   in	
   terms	
   of	
   its	
   attitudes	
   toward	
   sexuality.	
   An	
  
important	
   similarity	
   between	
   Umbanda,	
   primarily	
   at	
   the	
   Afro	
   end	
   of	
   its	
   spectrum,	
   and	
  
Candomblé	
   is	
   their	
   offering	
   scope	
   for	
   the	
   performance	
   of	
   alternative	
   sexualities	
   in	
   a	
   society	
  
governed	
   by	
   very	
   conservative	
   heterosexual	
   gender	
   roles	
   (Landes	
   1947;	
   Fry	
   1982;	
   Birman	
  
1985;	
  1995;	
  Natividade	
  and	
  Oliveira	
  2007).	
  
Umbanda	
   has	
   always	
   manifested	
   strong	
   tensions	
   between	
   fragmenting	
   and	
   centralizing	
  
tendencies:	
   tensions	
   between	
   variation	
   of	
   doctrine	
   and	
   practice	
   depending	
   on	
   individual	
  
terreiros	
  and	
  the	
  development	
  of	
  associations	
  that	
  have	
  both	
  emphasized	
  doctrinal	
  regularity,	
  
in	
  order	
  to	
  make	
  public	
  claims	
  that	
  Umbanda	
  is	
  a	
  “religion,”	
  and	
  to	
  lobby	
  for	
  religious	
  freedom	
  

in	
   face	
   of	
   government	
   oppression	
   through	
   most	
   of	
   mid-­‐20th	
   century.	
   The	
   centralizing	
  
tendencies	
   attempted	
   to	
   impose	
   a	
   hierarchical	
   structure	
   and	
   tended	
   to	
   emphasize	
   the	
  
embranquecimento	
  of	
  Umbanda:	
  	
  
	
  
The	
   first	
   attempt	
   to	
   create	
   a	
   religious	
   hierarchy	
   for	
   the	
   various	
   umbandists	
   occurred	
   in	
   1937.	
   It	
   was	
  
accompanied	
  by	
  an	
  ideological	
  emphasis	
  on	
  the	
  embranquecimento	
  of	
  worship	
  forms	
  of	
  African	
  origin.	
  The	
  
Spiritist	
   Union	
   of	
   Umbanda	
   in	
   Brazil	
   …	
   proposed	
   a	
   religion	
   stripped	
   of	
   African	
   symbols	
   that,	
   at	
   the	
   same	
  
time,	
  placed	
  value	
  on	
  a	
  Gospel-­‐based	
  doctrinal	
  orientation.	
  (Birman	
  1983:	
  95)	
  

	
  
Umbanda	
  is	
  also	
  intermediary	
  in	
  terms	
  of	
  its	
  range	
  of	
  institutional	
  manifestations	
  of	
  internal	
  
divergence.	
   Kardecism	
   exhibits	
   a	
   high	
   degree	
   of	
   uniformity,	
   with	
   some	
   tensions	
   between	
  
“religious”	
   and	
   “scientific”	
   camps.	
   Doctrine	
   and	
   practices	
   are	
   relatively	
   explicit,	
   with	
   slight	
  
divergence	
   between	
   centres	
   associated	
   with	
   different	
   federations.	
   Distinctions	
   between	
   the	
  
various	
   Afro-­‐Brazilian	
   religions	
   reflect	
   historical	
   and	
   geographical	
   differences	
   (above	
   all,	
  
differences,	
   historical	
   and	
   constructed,	
   between	
   cultural	
   groups	
   of	
   slaves).	
   These	
   various	
  
traditions	
  manifest	
  complex	
  interrelations	
  in	
  terms	
  of	
  origins,	
  beliefs	
  and	
  rituals,	
  and	
  they	
  are	
  
generally	
   associated	
   with	
   distinct	
   regions:	
   e.g.,	
   Batuque	
   in	
   Rio	
   Grande	
   do	
   Sul;	
   Cabula	
  
(historically)	
  in	
  Espírito	
  Santo;	
   Candomblé	
   de	
   Caboclo	
   and	
   Jurê	
   in	
   Bahia;	
   Catimbó,	
   Cura	
   and	
  
Pajelança	
  from	
  Pernambuco	
  through	
  Amazônia;	
  Canjerê	
  in	
  Minas	
  Gerais;	
  Macumba	
  in	
  Rio	
  de	
  
Janeiro;	
   Toré	
   in	
   Sergipe;	
   Tambor	
   de	
   Mata	
   [or	
   Terecô]	
   in	
   Maranhão;	
   Tambor	
   de	
   Mina	
   in	
  
Maranhão	
   and	
   Pará;	
   Babassuê	
   in	
   Pará;	
   Xambá	
   in	
   Alagoas,	
   Pernambuco	
   and	
   Paraíba;	
   Xangô	
  
and	
   Jurema	
   in	
   Pernambuco;	
   and,	
   of	
   course,	
   Candomblé	
   in	
   Bahia,	
   later	
   spreading	
   to	
   Rio	
   de	
  
Janeiro,	
   São	
   Paulo	
   and	
   Rio	
   Grande	
   do	
   Sul.	
   (Umbanda’s	
   relation	
   to	
   Candomblé	
   reflects	
   its	
  
origins	
  in	
  Rio	
  de	
  Janeiro	
  and	
  São	
  Paulo.)	
  The	
  Afro-­‐Brazilian	
  religions	
  manifest	
  relatively	
  little	
  
institutionalization,	
  with	
  each	
  terreiro	
  being	
  largely	
  independent.	
  Umbanda	
  occupies	
  a	
  middle	
  
ground,	
  with	
  intellectuals	
  and	
  federations	
  arguing	
  sharp	
  lines	
  at	
  the	
  “white”	
  kardecist	
  end	
  of	
  
the	
  spectrum	
  and	
  with	
  Umbanda	
  blurring	
  into	
  Afro-­‐Brazilian	
  traditions,	
  with	
  an	
  emphasis	
  on	
  
charismatic	
   leadership	
   within	
   individual	
   terreiros,	
   at	
   the	
   other	
   end	
   of	
   the	
   spectrum.	
   This	
  
institutional	
   variation	
   is	
   also	
   reflected	
   in	
   increased	
   potential	
   for	
   internal	
   struggles	
   between	
  
those	
   who	
   emphasize	
   traditional	
   ritual	
   skills	
   and	
   those	
   who	
   manifest	
   organization	
   and	
  
intellectual	
  skills,	
  a	
  tension	
  analyzed	
  by	
  Yvonne	
  Maggie	
  as	
  one	
  between	
  “the	
  code	
  of	
  the	
  santo”	
  
and	
  “the	
  bureaucratic	
  code”	
  (2001[1977]).	
  

Kardecism	
   draws	
   sharper	
   boundaries	
   than	
   Afro-­‐Brazilian	
   traditions,	
   with	
   Umbanda	
   in	
  
between	
   with	
   respect	
   to	
   this	
   characteristic.	
   At	
   the	
   white	
   end	
   of	
   the	
   umbandist	
   spectrum,	
  
Umbanda	
  has	
  appropriated	
  elements	
  of	
  Kardecism,	
  but	
  the	
  reverse	
  is	
  not	
  the	
  case.	
  At	
  the	
  Afro	
  
end	
   of	
   the	
   spectrum,	
   there	
   is	
   mutual	
   admixture	
   between	
   Umbanda	
   and	
   Afro-­‐Brazilian	
  
traditions,	
   e.g.,	
   in	
   “umbandized”	
   Xangô	
   and	
   the	
   strong	
   presence	
   of	
   elements	
   of	
   Jurema	
   in	
  
umbandist	
   terreiros	
   in	
   the	
   interior	
   of	
   Brazil’s	
   northeast	
   (Motta	
   2006	
   [1999]:	
   27-­‐30;	
   Assunção	
  
2001).	
  The	
  blurring	
  of	
  the	
  Afro	
  extreme	
  of	
  the	
  Umbanda	
  spectrum	
  into	
  Candomblé	
  and	
  other	
  
Afro-­‐Brazilian	
   traditions	
   is	
   illustrated,	
   for	
   example,	
   by	
   Leacock’s	
   fieldwork	
   in	
   the	
   1960s,	
  
which	
   discovered	
   quite	
   flexible	
   insider	
   labels:	
   “Members	
   do	
   not	
   call	
   the	
   cult	
   ‘Batuque,’	
   but	
  
refer	
  to	
  it	
  as	
  either	
  ‘Nagô,’	
  ‘Mina,’	
  or	
  ‘Umbanda,’	
  depending	
  on	
  minor	
  variations	
  in	
  belief	
  and	
  
ritual”	
   (1964b:	
   354	
   n.2).	
   Similarly,	
   many	
   senior	
   practitioners	
   of	
   Afro-­‐Brazilian	
   traditions	
   in	
  
Minas	
  Gerais	
  distinguish	
  between	
  Umbanda	
  and	
  Canjerê	
  while	
  emphasizing	
  their	
  fundamental	
  
continuity	
   (Tavares	
   and	
   Floriano	
   2003:	
   167-­‐168).5	
   The	
   greater	
   fluidity	
   of	
   boundaries	
   at	
   the	
  
Afro	
   end	
   of	
   the	
   Umbanda	
   spectrum	
   is	
   also	
   illustrated	
   by	
   the	
   extent	
   to	
   which	
   that	
   sub-­‐set	
   of	
  
Umbandas	
   has	
   been	
   “re-­‐africanized”	
   by	
   Candomblé	
   since	
   the	
   latter’s	
   growth	
   in	
   the	
   urban	
  
centres	
  of	
  southeastern	
  Brazil	
  since	
  the	
  1960s	
  (Prandi	
  1991:	
  74;	
  2000:	
  644).	
  	
  
Umbanda	
   is	
   also	
   intermediary	
   in	
   terms	
   of	
   the	
   extent	
   to	
   which	
   it	
   is	
   drawn	
   upon	
   by	
   New	
  
Religious	
   Movements,	
   which	
   are	
   largely	
   associated	
   with	
   middle-­‐	
   and	
   upper-­‐class	
   urban	
  
religiosity.	
   Kardecism	
   has	
   an	
   historical	
   relation	
   to	
   Mesmerism,	
   and	
   it	
   manifested	
   significant	
  
tensions	
   in	
   the	
   late	
   nineteenth	
   century	
   between	
   scientific	
   and	
   esoteric	
   tendencies	
   (Monroe	
  
2008).	
   In	
   Brazil,	
   these	
   characteristics	
   inform	
   its	
   marked	
   tendency	
   to	
   serve	
   as	
   an	
   important	
  
element	
   in	
   a	
   range	
   of	
   NRMs:	
   e.g.,	
   Círculo	
   Esotérico	
   da	
   Comunhão	
   do	
   Pensamento	
   (1909),	
  
Ordem	
   Mística	
   Espiritualista	
   Agla-­‐Avid	
   (1959),	
   Ordem	
   Espiritualista	
   Cristã/Vale	
   do	
  
Amanhecer	
   (1969)	
   and	
   others	
   (often	
   with	
   esoteric	
   or	
   Masonic	
   influences).	
   The	
   “white”	
   end	
   of	
  
the	
   Umbanda	
   spectrum	
   expresses	
   this	
   same	
   tendency,	
   though	
   to	
   a	
   much	
   lesser	
   extent:	
   it	
  
informs	
   Umbandaime	
   (an	
   emerging	
   current	
   within	
   Santo	
   Daimé);	
   and	
   several	
   groups	
   have	
  
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
5	
  An	
  example	
  from	
  my	
  own	
  fieldwork	
  illustrates	
  another	
  dimension	
  of	
  these	
  fluid	
  boundaries.	
  Informants	
  took	
  

me	
   to	
   what	
   they	
   called	
   a	
   “Candomblé”	
   in	
   a	
   small	
   city	
   in	
   Minas	
   Gerais.	
   It	
   was,	
   in	
   fact,	
   a	
   terreiro	
   of	
   Umbanda	
   at	
   the	
  
Afro	
  end	
  of	
  its	
  spectrum.	
  This	
  reflects	
  the	
  prominence	
  of	
  white	
  Umbanda	
  in	
  their	
  own	
  experience	
  of	
  that	
  religion	
  
and	
  their	
  primary	
  concern	
  with	
  therapeutic	
  services	
  rather	
  than	
  insider	
  or	
  academic	
  categories.	
  Of	
  course,	
  this	
  
experience	
  leads	
  me	
  to	
  interrogate	
  the	
  “in	
  fact”	
  of	
  my	
  own	
  categorizations.	
  	
  

adopted	
  the	
  label	
  of	
  Esoteric	
  Umbanda	
  (Guerriero	
  2006).	
  Afro-­‐Brazilian	
  traditions	
  have	
  little	
  
presence	
  in	
  Brazilian	
  NRMs.	
  
The	
   spectrum	
   of	
   Umbanda’s	
   ritual	
   variants	
   also	
   reflects	
   divergent	
   processes	
   of	
  
secularization	
   and	
   rationalization	
   in	
   Brazilian	
   society.	
   The	
   spectrum	
   between	
   kardecist	
   and	
  
Afro-­‐Brazilian	
   extremes	
   of	
   Umbanda	
   represents	
   varying	
   degrees	
   of	
   rationalization	
   of	
   Afro-­‐
Brazilian	
   traditions,	
   or,	
   according	
   to	
   Renato	
   Ortiz,	
   “levels	
   of	
   secularization”	
   that	
   allow	
   us	
   to	
  
study	
  how	
  “traditional	
  magico-­‐religious	
  practices	
  …	
  cross	
  class	
  boundaries,	
  penetrating	
  both	
  
lower-­‐	
   and	
   middle-­‐classes”	
   (1999[1978]:	
   214;	
   see	
   Motta	
   2006[1999]:	
   24).	
   A	
   prevalence	
   of	
  
scientific	
   metaphors	
   (especially	
   electro-­‐magnetic	
   terms)	
   at	
   the	
   white	
   end	
   of	
   the	
   spectrum	
  
reflects	
  the	
  admixture	
  of	
  “scientific”	
  and	
  “philosophical”	
  kardecist	
  doctrines	
  (reflected	
  also	
  in	
  
the	
   themes	
   of	
   spiritual	
   evolution/perfection	
   and	
   moralization	
   of	
   worldly	
   activity)	
   (Camargo	
  
1961:	
  115-­‐117;	
  Ortiz	
  1999[1978]:	
  168-­‐173).	
  	
  
These	
   various	
   characteristics	
   of	
   Umbanda	
   all	
   stand	
   as	
   variables	
   along	
   which	
   the	
   religion	
  
manifests	
   a	
   spectrum	
   of	
   religious,	
   especially	
   ritual,	
   forms	
   that	
   are	
   correlated	
   with	
   broader	
  
tensions	
  in	
  Brazilian	
  society.	
  In	
  a	
  classic	
  study,	
  Candido	
  Procópio	
  Ferreira	
  de	
  Camargo	
  argued	
  
that	
  Umbanda	
  and	
  Kardecism	
  form	
  extremes	
  of	
  a	
  continuum	
  of	
  Brazilian	
  religious	
  practices	
  of	
  
mediumship	
   (1961).	
   This	
   is	
   misleading,	
   as	
   Umbanda	
   varies	
   widely	
   from	
   ritual	
   forms	
   like	
  
Kardecism	
   to	
   those	
   like	
   Candomblé,	
   where	
   Kardecism	
   itself	
   varies	
   relatively	
   little.	
   Renato	
  
Ortiz	
  has	
  suggested	
  the	
  more	
  defensible	
  idea	
  of	
  “a	
  religious	
  gradient	
  between	
  two	
  poles:	
  the	
  
more	
  westernized	
  and	
  the	
  less	
  westernized”	
  (1999:	
  97).	
  Ortiz	
  superposes	
  two	
  distinctions	
  in	
  
suggesting	
   “westernization”	
   as	
   the	
   criterion	
   of	
   the	
   spectrum	
   of	
   ritual	
   Brazilian	
   spirit-­‐
possession	
   religions:	
   traditional/modern	
   and	
   African/European.	
   This	
   is	
   helpful,	
   but	
   it	
   both	
  
begs	
   the	
   important	
   question	
   of	
   invented	
   traditions	
   and	
   does	
   not	
   go	
   far	
   enough	
   in	
   clarifying	
  
the	
  complex	
  negotiations	
  of	
  race	
  and	
  culture	
  in	
  the	
  Brazilian	
  landscape.	
  Moreover,	
  it	
  fails	
  to	
  
capture	
   a	
   crucial	
   element:	
   the	
   relation	
   between	
   these	
   religious	
   dimensions	
   and	
   the	
   social	
  
context.	
  	
  
	
  
Three	
  Types	
  of	
  Hybridity	
  
Discussions	
  of	
  hybridity	
  usefully	
  highlight	
  issues	
  of	
  race,	
  class,	
  and	
  power,	
  though	
  the	
  above	
  
problems	
  warn	
  us	
  to	
  be	
  critical	
  in	
  our	
  use	
  of	
  the	
  concept.	
  An	
  effective	
  typology	
  of	
  hybridities	
  
should	
   take	
   account	
   of	
   different	
   sorts	
   of	
   crossings	
   of	
   different	
   boundaries	
   in	
   different	
  
contexts,	
  and	
  it	
  should	
  distinguish	
  scholarly	
  from	
  insider	
  perceptions	
  of	
  hybridity.	
  

In	
  this	
  light,	
  Peter	
  Wade	
  proposes	
  a	
  useful	
  distinction	
  between	
  two	
  types	
  of	
  hybridity:	
  	
  
	
  
The	
  first,	
  which	
  as	
  a	
  shorthand	
  I	
  will	
  call	
  roots-­‐hybridity,	
  depends	
  on	
  a	
  simple	
  syncretism	
  of	
  two	
  anterior	
  
wholes	
   to	
   make	
   a	
   third	
   new	
   whole.	
   In	
   this	
   teleological	
   mode,	
   roots	
   and	
   belonging	
   are	
   paramount	
   and	
  
exclusive	
  essentialisms	
  can	
  easily	
  be	
  reproduced.	
  The	
  second,	
  which	
  I	
  will	
  label	
  routes-­‐hybridity,	
  depends	
  
on	
   unpredictable	
   diasporic	
   movements,	
   creating	
   unstable	
   complex	
   networks,	
   not	
   reducible	
   to	
   teleological	
  
progressions,	
  but	
  moving	
  to	
  and	
  fro	
  erratically	
  in	
  time	
  and	
  space.	
  In	
  this	
  mode,	
  routes	
  and	
  movement	
  are	
  
paramount	
   and	
   exclusivism	
   gives	
   way	
   to	
   more	
   inclusive	
   identities	
   based,	
   for	
   example,	
   on	
   perception	
   of	
  
common	
  interests	
  and	
  goals,	
  rather	
  than	
  common	
  origins.	
  (Wade	
  2005:	
  256-­‐257)	
  

	
  
On	
  the	
  one	
  hand,	
  roots-­‐hybridity	
  is	
  the	
  outdated	
  essentialist	
  view	
  of	
  syncretism,	
  where	
  recent	
  
theorists	
  of	
  hybridity	
  see	
  routes-­‐hybridity	
  “in	
  some	
  sense	
  as	
  a	
  progression	
  from	
  or	
  challenge	
  
to	
  the	
  former,	
  if	
  not	
  as	
  its	
  simple	
  opposite”	
  (2005:	
  257).	
  On	
  the	
  other	
  hand,	
  Wade	
  makes	
  two	
  
points	
   that	
   suggest	
   the	
   continued	
   value	
   of	
   this	
   distinction:	
   “thinking	
   in	
   terms	
   of	
   roots	
   and	
  
origins	
  is	
  not	
  necessarily	
  as	
  essential	
  and	
  exclusivist	
  as	
  it	
  might	
  first	
  seem”;	
  and	
  “the	
  routes	
  
form	
  of	
  hybridity	
  cannot	
  escape	
  from	
  the	
  roots	
  form.	
  The	
  two	
  are	
  mutually	
  implicated	
  and	
  co-­‐
dependent”	
   (2005:	
   257).	
   That	
   is,	
   in	
   addition	
   to	
   being	
   cautious	
   in	
   our	
   scholarly	
   use	
   of	
   these	
  
concepts,	
  we	
  need	
  to	
  recognize	
  that	
  the	
  distinction	
  points	
  to	
  something	
  significant	
  in	
  insider	
  
perceptions	
   of	
   hybridity.	
   Because	
   Wade’s	
   visually	
   catchy	
   labels,	
   “roots”	
   and	
   “routes,”	
   are	
  
homophones,	
  I	
  propose	
  alternative	
  terms:	
  hybridities	
  of	
  “origin”	
  and	
  “encounter.”	
  
Several	
  elements	
  of	
  analytic	
  caution	
  are	
  crucial.	
  First,	
  the	
  distinction	
  between	
  hybridities	
  of	
  
origin	
   and	
   encounter	
   does	
   not	
   presume	
   that	
   the	
   roots	
   of	
   the	
   former	
   are	
   pure	
   nor	
   that	
   the	
  
parties	
  that	
  encounter	
  in	
  the	
  latter	
  are	
  hybrid.	
  Second,	
  the	
  distinction	
  is	
  not	
  that	
  between	
  past	
  
and	
   present,	
   tradition	
   and	
   innovation:	
   origins	
   can	
   be	
   current	
   developments	
   and	
   encounter	
  
historical	
   ones;	
   both	
   scholarly	
   “facts”	
   and	
   insider	
   “inventions”	
   are	
   relevant	
   to	
   both.	
   (In	
   this	
  
light,	
   Wade’s	
   emphasis	
   on	
   teleology	
   is	
   misleading,	
   as	
   it	
   characterizes	
   “roots-­‐hybridity”	
   from	
  
the	
  critical	
  perspective	
  that	
  he	
  tries	
  to	
  move	
  past.)	
  	
  
The	
   distinction	
   frames	
   distinct	
   perspectives,	
   issues	
   and	
   sets	
   of	
   questions	
   not	
   distinct	
  
hybrid	
  realities.	
  “Hybridity	
  of	
  origin”	
  marks	
  (i)	
  insider	
  perceptions	
  of	
  the	
  origin	
  and	
  character	
  
of	
   distinct	
   cultural	
   forms,	
   whether	
   seen	
   in	
   essential	
   and	
   exclusivist	
   terms	
   or	
   not,	
   and	
   (ii)	
  
limited	
   scholarly	
   attention	
   to	
   the	
   “internal	
   history”	
   of	
   doctrine,	
   practice	
   and	
   institutional	
  
forms	
  (including	
  that	
  of	
  current	
  developments).	
  The	
  scholar	
  or	
  insider’s	
  choice	
  to	
  invoke	
  this	
  

type	
  of	
  hybridity	
  highlights	
  the	
  general	
  issues	
  of	
  origins	
  but	
  leaves	
  many	
  complex	
  questions	
  
open.	
  For	
  example,	
  focusing	
  on	
  insider	
  accounts	
  of	
  roots,	
  Candomblé	
  is	
  an	
  African	
  religion,	
  but	
  
the	
  extent	
  to	
  which	
  this	
  origin	
  is	
  constructed,	
  invented	
  or	
  imagined	
  remains	
  debated	
  among	
  
scholars.	
   Kardecism	
   presents	
   itself	
   as	
   largely	
   independent	
   of	
   place,	
   though	
   its	
   European	
   roots	
  
are	
  often	
  associated	
  with	
  past	
  and	
  present	
  status	
  claims,	
  and	
  some	
  works	
  argue	
  that	
  its	
  origin	
  
is	
  (mythically)	
  Brazilian.	
  Umbandist	
  texts,	
  again	
  manifesting	
  a	
  spectrum,	
  sometimes	
  point	
  to	
  
roots	
   in	
   Africa,	
   sometimes	
   to	
   India,	
   Brazil,	
   Atlantis,	
   or	
   other	
   planets,	
   and	
   sometimes	
   claim	
   a	
  
universality	
  free	
  of	
  geographic	
  roots.	
  
“Hybridity	
  of	
  encounter”	
  marks	
  the	
  social	
  context	
  of	
  cultural	
  interaction,	
  the	
  strategies	
  and	
  
tactics	
   of	
   mutual	
   influence,	
   the	
   agency	
   of	
   participants.	
   Both	
   concepts	
   are	
   appropriate	
   for	
  
talking	
   of	
   historical	
   or	
   contemporary	
   developments.	
   Both	
   are	
   useful	
   for	
   analyzing	
   diasporic	
  
religions	
  as	
  well	
  as	
  the	
  more	
  constrained	
  interactions	
  of	
  long-­‐term	
  coexistence	
  	
  within	
  a	
  given	
  
cultural	
   context.	
   (For	
   example,	
   Umbanda’s	
   emergence	
   is	
   not	
   a	
   diasporic	
   mixing	
   but	
   a	
  
development	
  internal	
  to	
  a	
  well-­‐established,	
  albeit	
  eminently	
  hybrid	
  and	
  post-­‐colonial,	
  society:	
  
the	
   encounter	
   here	
   is	
   not	
   that	
   of	
   diaspora	
   but	
   of	
   urbanization.)	
   As	
   analytical	
   tools,	
   the	
   two	
  
concepts	
  are	
  complementary.	
  Hybridity	
  of	
  origin	
  is	
  not	
  the	
  self-­‐conscious	
  construct	
  of	
  novelty,	
  
a	
  movement	
  toward	
  an	
  end,	
  but	
  a	
  reaction	
  to	
  specific	
  historical,	
  religious,	
  and	
  often	
  political	
  
circumstances.	
   It	
   reflects	
   the	
   present	
   as	
   much	
   as	
   the	
   past.	
   Similarly,	
   hybridity	
   of	
   encounter	
  
necessarily	
   draws	
   on	
   its	
   roots.	
   It	
   reflects	
   the	
   past	
   as	
   much	
   as	
   the	
   present.	
   Discussions	
   of	
  
hybridity	
   are	
   useful	
   when	
   they	
   foreground	
   the	
   struggles	
   that	
   draw,	
   label,	
   prioritize,	
  
naturalize,	
   and	
   sacralize	
   boundaries;	
   they	
   are	
   misleading	
   when	
   they	
   take	
   these	
   boundaries	
  
and	
  the	
  significance	
  of	
  their	
  crossing	
  or	
  blurring	
  for	
  granted.	
  
Analyzing	
   Umbanda	
   in	
   terms	
   of	
   Wade’s	
   two	
   concepts	
   of	
   hybridity	
   would	
   miss	
   one	
   of	
   the	
  
religion’s	
  defining	
  characteristics.	
  To	
  draw	
  this	
  out,	
  I	
  propose	
  a	
  third	
  type	
  of	
  hybridity:	
  that	
  of	
  
“refraction.”	
  Umbanda	
  consists	
  in	
  a	
  spectrum	
  of	
  individual	
  groups	
  that	
  span	
  the	
  same	
  racial	
  
and	
   class	
   divisions	
   that	
   sparked	
   its	
   emergence	
   in	
   the	
   early	
   twentieth	
   century.	
   It	
   is	
   not	
   a	
  
diasporic	
   religion	
   (hybridity	
   of	
   encounter).	
   It	
   did	
   indeed	
   originate	
   in	
   the	
   mixture	
   of	
   distinct	
  
religious	
   roots,	
   themselves	
   hybrids	
   (hybridity	
   of	
   origin).	
   But	
   to	
   stop	
   here	
   would	
   leave	
   out	
   a	
  
crucial	
  dimension	
  of	
  the	
  religion’s	
  hybridity:	
  the	
  way	
  that	
  it	
  continues	
  to	
  manifest	
  internally	
  a	
  
series	
   of	
   tensions	
   that	
   were	
   implicated	
   in	
   its	
   hybrid	
   roots	
   and	
   that	
   continue	
   to	
   be	
   constitutive	
  
of	
   Brazilian	
   society.	
   The	
   concept	
   of	
   hybridity	
   of	
   refraction	
   refers	
   to	
   this	
   way	
   in	
   which	
   the	
  
social	
  boundaries	
  that	
  are	
  symbolically	
  elided,	
  inverted,	
  or	
  echoed	
  within	
  a	
  system	
  of	
  religious	
  

beliefs	
  and	
  practices	
  reflect	
  or	
  refract	
  homologous	
  boundaries	
  present	
  in	
  a	
  given	
  society.	
  
Umbanda	
   is	
   a	
   modern	
   religion	
   that	
   spans,	
   symbolically	
   elides,	
   yet	
   ultimately	
   reinforces	
  
important	
   social	
   boundaries	
   in	
   Brazilian	
   society.	
   It	
   is	
   an	
   especially	
   important	
   case	
   of	
   the	
  
hybridity	
   of	
   refraction	
   because	
   its	
   origin,	
   trajectory,	
   and	
   status	
   are	
   so	
   intimately	
   tied	
   with	
  
issues	
   of	
   race	
   and	
   class	
   in	
   Brazil	
   and	
   because	
   it	
   reflects	
   these	
   tensions	
   in	
   its	
   doctrinal	
  
elaboration,	
   ritual	
   form,	
   and	
   institutionalization.	
   Various	
   scholars	
   have	
   noted	
   the	
   marked	
  
extent	
   to	
   which	
   Umbanda	
   reflects	
   Brazilian	
   society.	
   Peter	
   Fry	
   argues	
   that	
   Umbanda	
   reflects	
  
the	
  social	
  and	
  political	
  structures	
  of	
  Brazilian	
  society	
  (1982).	
  Concone	
  notes	
  that	
  the	
  religion’s	
  
various	
   spirits	
   “are	
   obviously	
   drawn	
   from	
   the	
   national	
   reality.	
   …This	
   is	
   precisely	
   the	
   most	
  
interesting	
   aspect	
   of	
   the	
   umbandist	
   religion:	
   the	
   fact	
   that	
   it	
   dives	
   so	
   deeply	
   into	
   Brazilian	
  
reality,	
  …	
  transforming	
  popular	
  figures	
  into	
  symbols	
  …”	
  (2001:	
  282).	
  (The	
  symbolic	
  work	
  of	
  
Umbanda	
   is	
   a	
   particularly	
   dynamic	
   aspect	
   of	
   its	
   flexibility	
   and	
   mutability	
   [Malandrino	
   2006].)	
  
Brumana	
  and	
  Martinez,	
  in	
  their	
  invaluable	
  study,	
  analyze	
  Umbanda	
  as	
  a	
  “subaltern	
  cult”	
  that	
  
“elaborates	
   symbolically	
   the	
   social	
   condition	
   of	
   the	
   client”	
   (1989:	
   45).	
   Ortiz	
   argues	
   that	
  
“umbandist	
   ideology	
   preserves	
   and	
   transforms	
   Afro-­‐Brazilian	
   cultural	
   elements	
   within	
   a	
  
modern	
   society,	
   [while,	
   at	
   the	
   same	
   time]	
   manifesting	
   rupture,	
   forgetting,	
   and	
  
reinterpretation	
   of	
   older,	
   traditional	
   values”	
   (1999[1978]:	
   212).	
   Patricia	
   Birman	
   underlines	
  
Umbanda’s	
  symbolic	
  and	
  ritual	
  engagement	
  with	
  Brazilian	
  social	
  reality:	
  
	
  
Possession	
   in	
   Candomblé	
   involves	
   the	
   state	
   and	
   audience	
   in	
   scenes	
   that	
   are	
   more	
   perfect	
   the	
   more	
   they	
  
involve	
   criteria	
   irreducible	
   to	
   the	
   civilized	
   world—a	
   world	
   of	
   alterity	
   is	
   recognized	
   by	
   this	
   criterion,	
  
valorizing	
  the	
  Africanness	
  that	
  it	
  presents.	
  On	
  the	
  other	
  hand,	
  umbandist	
  possession	
  …	
  is	
  worthy	
  of	
  credit	
  to	
  
the	
   extent	
   that	
   it	
   contextually	
   invokes	
   its	
   relation	
   with	
   the	
   world	
   as	
   experienced	
   by	
   its	
   audience.	
   (1995:	
   44-­‐
45)	
  

	
  
What	
   I	
   add	
   to	
   this	
   frequent	
   recognition	
   than	
   Umbanda	
   is	
   especially	
   responsive	
   to	
   the	
  
structures	
   and	
   tensions	
   of	
   Brazilian	
   society	
   is	
   a	
   more	
   precise	
   characterization	
   of	
   this	
  
responsiveness:	
   this	
   is	
   not	
   a	
   relation	
   between	
   a	
   uniform	
   or	
   generic	
   type	
   of	
   Umbanda	
   and	
  
Brazilian	
   society;	
   it	
   is	
   a	
   relation	
   between	
   distinct	
   variants	
   of	
   Umbanda	
   and	
   specific	
   social	
  
tensions.	
  That	
  is,	
  Umbanda	
  manifests	
  the	
  hybridity	
  of	
  refraction.	
  
Umbandist	
   doctrine	
   and	
   ritual	
   manifest	
   both	
   the	
   positive	
   and	
   negative	
   aspects	
   of	
   Brazil’s	
  
myth	
   of	
   harmonious	
   mestiçagem	
   between	
   three	
   races:	
   it	
   celebrates	
   a	
   certain	
   form	
   of	
   racial	
  

inclusiveness,	
  yet	
  without	
  challenging	
  racism’s	
  material	
  manifestations;	
  and	
  it	
  does	
  so	
  within	
  
traditional	
   hierarchical	
   and	
   largely	
   patriarchal	
   social	
   forms.	
   It	
   also	
   reflects	
   a	
   range	
   of	
   class	
  
positions,	
   marked	
   by	
   geographical	
   location	
   of	
   the	
   terreiros	
   within	
   communities,	
   by	
   middle-­‐
class	
   participation,	
   and	
   by	
   varying	
   degrees	
   of	
   intellectualization,	
   emphasis	
   on	
   texts,	
   and	
   the	
  
prominence	
  of	
  semi-­‐conscious	
  rather	
  than	
  unconscious	
  trance	
  states.	
  The	
  spectrum	
  of	
  types	
  
of	
   Umbanda,	
   from	
   kardecist	
   to	
   Afro-­‐Brazilian,	
   with	
   their	
   different	
   stances	
   regarding	
   the	
  
religion’s	
   origin	
   and	
   fragmented	
   nature,	
   manifests	
   this	
   same	
   ambiguity	
   both	
   eliding	
   and	
  
reflecting	
  social	
  boundaries.	
  	
  
Umbanda	
   is	
   not	
   unique	
   in	
   manifesting	
   the	
   hybridity	
   of	
   refraction.	
   Kardecism	
   manifests	
  
more	
  elite	
  and	
  popular	
  variants.	
  Bastide	
  distinguished	
  between	
  upper,	
  middle	
  and	
  lower-­‐class	
  
Spiritism;	
   significantly,	
   he	
   defined	
   the	
   latter	
   as	
   “another	
   type	
   of	
   spiritism,	
   the	
   spiritism	
   of	
  
Umbanda,”	
   noting	
   that	
   “one	
   finds	
   between	
   Kardecism	
   and	
   Umbanda	
   a	
   whole	
   series	
   of	
  
transitions”	
   (1967:	
   9,	
   11).	
   Candomblé	
   also	
   manifests	
   the	
   hybridity	
   of	
   refraction,	
   albeit	
   to	
   a	
  
lesser	
  degree	
  than	
  Umbanda.	
  In	
  the	
  late	
  twentieth	
  century,	
  Candomblé	
  underwent	
  what	
  Paul	
  
Christopher	
  Johnson	
  calls	
  a	
  “social	
  extension”,	
  moving	
  from	
  “traditional”	
  to	
  “public”	
  forms,	
  in	
  
part	
   as	
   a	
   result	
   of	
   the	
   increasing	
   prominence	
   of	
   Afro-­‐Brazilian	
   elements	
   in	
   popular	
   culture:	
  
“the	
   religion	
   that	
   was	
   ethnically	
   specific	
   is	
   presented	
   as	
   universally	
   available”	
   (2002b:	
   313;	
  
see	
  2002a).	
  Inseparable	
  from	
  this	
  development	
  are	
  a	
  series	
  of	
  recent	
  movements	
  within	
  Afro-­‐
Brazilian	
  traditions	
  that	
  aim	
  at	
  reclaiming	
  the	
  purity	
  of	
  perceived	
  tradition	
  through	
  processes	
  
of	
   “re-­‐africanization,”	
   “de-­‐syncretization,”	
   “de-­‐catholicization”	
   etc.	
   (Caroso	
   and	
   Bacelar	
   1999).	
  
If	
   this	
   were	
   a	
   recent	
   development,	
   this	
   would	
   be	
   the	
   hybridity	
   of	
   encounter,	
   not	
   of	
   refraction.	
  
However,	
  a	
  tension	
  between	
  traditional/African	
  terreiros	
  and	
  those	
  more	
  open	
  to	
  a	
  broader	
  
social	
  spectrum	
  has	
  been	
  prominent	
  in	
  Candomblé	
  since	
  at	
  least	
  the	
  late	
  nineteenth	
  century	
  
(Harding	
   2000;	
   Parés	
   2007:	
   132-­‐138).	
   This	
   tension	
   was	
   amplified	
   in	
   the	
   late	
   twentieth	
  
century	
   by	
   immigration	
   from	
   the	
   northeast	
   to	
   the	
   large	
   urban	
   enters	
   in	
   the	
   south	
   (Prandi	
  
1991;	
  2005).	
  To	
  a	
  limited	
  extent,	
  then,	
  Candomblé—less	
  than	
  Umbanda	
  but	
  much	
  more	
  than	
  
other	
   Afro-­‐Brazilian	
   traditions—spans,	
   in	
   its	
   limited	
   ritual,	
   doctrinal,	
   and	
   institutional	
  
variants,	
   important	
   social	
   boundaries	
   in	
   Brazilian	
   society:	
   i.e.,	
   ethnic	
   and	
   racial	
   distinctions	
  
with	
  some	
  correlated	
  variation	
  in	
  class	
  adherence.	
  
The	
   three	
   types	
   of	
   hybridity	
   that	
   I	
   have	
   distinguished—those	
   of	
   origin,	
   encounter,	
   and	
  
refraction—are	
  not	
  mutually	
  exclusive	
  but	
  represent	
  different	
  perspectives	
  or	
  emphases,	
  as	
  is	
  
illustrated	
   by	
   the	
   parallel	
   between	
   Candomblé	
   and	
   Umbanda.	
   Both	
   religions	
   began	
   in	
   a	
  

context	
   of	
   religious	
   mixture	
   (hybridity	
   of	
   origin);	
   both	
   resulted	
   from	
   and	
   reacted	
   to	
   the	
  
interaction	
   of	
   different	
   religious	
   and	
   cultural	
   currents,	
   primarily	
   diaspora	
   and	
   urbanization	
  
respectively	
   (hybridity	
   of	
   encounter);	
   and	
   both	
   have	
   internal	
   variations	
   that	
   reflect	
  
constitutive	
   social	
   tensions	
   in	
   their	
   national	
   context	
   (hybridity	
   of	
   refraction).	
   My	
   claims	
   are	
  
that	
  Umbanda	
  is	
  distinct	
  in	
  terms	
  of	
  the	
  degree	
  of	
  importance	
  of	
  the	
  latter	
  factor,	
  and	
  that	
  the	
  
concept	
   of	
   hybridity	
   of	
   refraction	
   helps	
   to	
   highlight	
   some	
   of	
   the	
   most	
   important	
   features	
   of	
  
this	
  new	
  Brazilian	
  religion.	
  
A	
   fuller	
   analysis	
   would	
   take	
   account	
   of	
   three	
   important	
   dimensions	
   of	
   Umbanda	
   that	
  
manifest	
   not	
   a	
   spectrum	
   of	
   positions	
   between	
   Afro-­‐Brazilian	
   traditions	
   and	
   Kardecism	
   but	
  
distinct	
  alternatives.	
  First,	
  Umbanda	
  creates	
  a	
  space	
  where	
  ambiguous	
  moral	
  agency	
  is	
  prized:	
  
it	
   demonstrates	
   “the	
   legitimacy	
   of	
   the	
   rogue,	
   the	
   underhanded	
   and	
   the	
   personal	
   favour	
   [do	
  
malandro,	
  da	
  sacanagem	
  e	
  do	
  favor]”	
  (Fry	
  1982:	
  13;	
  see	
  Concone	
  2001:	
  284-­‐286).	
  The	
  power	
  
that	
  mediumship	
  gives	
  in	
  Umbanda	
  is	
  more	
  ambivalent,	
  capable	
  of	
  being	
  used	
  for	
  good	
  or	
  bad	
  
ends:	
   it	
   is	
   little	
   constrained	
   by	
   an	
   explicit	
   moral	
   system,	
   as	
   in	
   Kardecism,	
   or	
   by	
   the	
  
subsumption	
   of	
   individual	
   agency	
   through	
   identification	
   with	
   the	
   possessing	
   supernatural	
  
entity,	
   as	
   in	
   Candomblé	
   (Brumana	
   and	
   Martinez	
   1989:	
   40-­‐42).	
   Second,	
   Umbanda	
   “positions	
  
itself	
  as	
  a	
  religion	
  that	
  encourages	
  social	
  mobility,	
  …	
  [and]	
  this	
  mobility	
  is	
  open	
  to	
  all,	
  without	
  
exception”	
   (Prandi	
   1991:	
   58).6	
   Third,	
   Umbanda	
   reflects,	
   more	
   explicitly	
   than	
   most	
  
manifestations	
  of	
  Brazilian	
  religiosity,	
  the	
  centrality	
  of	
  patriarchal	
  patron-­‐client	
  relations.	
  The	
  
mediums	
  are	
  possessed	
  by	
  helpful	
  spirits	
  who	
  act	
  as	
  patrons	
  to	
  their	
  clients,	
  many	
  of	
  whom	
  
return	
   week	
   after	
   week	
   to	
   speak	
   to	
   the	
   same	
   caboclo	
   or	
   preto-­‐velho.	
   The	
   pai-­‐de-­‐santo	
   in	
   an	
  
Umbanda	
   terreiro	
   is	
   “the	
   center	
   of	
   a	
   network	
   of	
   distribution	
   where	
   magical	
   services	
   are	
  
exchanged	
  for	
  money	
  with	
  wealthy	
  clients,	
  celebrations	
  are	
  exchanged	
  for	
  recognition	
  by	
  the	
  
general	
  public	
  and	
  the	
  filhos-­‐de-­‐santo	
  and	
  money	
  invested	
  in	
  the	
  terreiro	
  become	
  symbols	
  of	
  
success”	
  (Fry	
  1982:	
  75).	
  	
  
	
  
Conclusion	
  
This	
   paper	
   has	
   argued	
   that	
   Umbanda	
   manifests	
   a	
   spectrum	
   of	
   beliefs,	
   rituals	
   and	
   institutional	
  
forms	
   that	
   is	
   correlated	
   with	
   broader	
   tensions	
   in	
   Brazilian	
   society,	
   above	
   all	
   race	
   and	
   class.	
  
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
6	
  I	
  argue	
  elsewhere	
  that	
  specific	
  characteristics	
  of	
  umbandist	
  ritual	
  function	
  to	
  orient	
  the	
  agency	
  of	
  participants	
  

in	
  a	
  manner	
  consistent	
  with	
  this	
  social	
  mobility	
  (Engler	
  2007;	
  2008;	
  2009).	
  

Umbanda	
   is	
   not	
   unique	
   in	
   this	
   sense	
   but	
   (i)	
   it	
   is	
   characterized	
   by	
   a	
   particularly	
   significant	
  
degree	
   of	
   this	
   intra-­‐religious	
   variation,	
   (ii)	
   it	
   is	
   intermediate	
   between	
   Kardecism	
   and	
   Afro-­‐
Brazilian	
  traditions	
  in	
  an	
  especially	
  wide	
  variety	
  of	
  senses,	
  and	
  (iii)	
  that	
  these	
  characteristics	
  
are	
   uniquely	
   implicated	
   in	
   Umbanda’s	
   origin.	
   Again,	
   it	
   is	
   not	
   that	
   Kardecism	
   is	
   white	
   and	
  
upper-­‐class,	
   Candomblé	
   black	
   and	
   lower-­‐class,	
   and	
   Umbanda	
   in	
   the	
   middle.	
   All	
   races	
   and	
  
classes	
   are	
   involved	
   in	
   all	
   three	
   religions.	
   Rather,	
   variations	
   in	
   these	
   social	
   tensions	
   are	
  
mapped	
  onto	
  variations	
  in	
  belief,	
  practice,	
  and	
  institutionalization	
  to	
  a	
  much	
  greater	
  extent	
  in	
  
the	
   case	
   of	
   Umbanda:	
   the	
   many	
   Umbandas	
   of	
   Brazil	
   are	
   an	
   especially	
   clear	
   case	
   of	
   the	
  
hybridity	
  of	
  refraction.	
  
Reviewing	
   the	
   literature	
   underlines	
   the	
   fact	
   that	
   the	
   processes	
   of	
   mixing	
   that	
   concepts	
   like	
  
“syncretism”	
   and	
   “hybridity”	
   point	
   to	
   are	
   complex:	
   varying,	
   for	
   example,	
   from	
   internal	
   elision	
  
of	
   ideal-­‐typical	
   boundaries	
   between	
   religions	
   to	
   complex	
   influences	
   among	
   diverse	
   cultural	
  
forms	
  in	
  pluralistic,	
  diasporic	
  communities.	
  Discussions	
  of	
  different	
  forms	
  of	
  mixture	
  rightly	
  
draw	
   attention	
   to	
   the	
   reorganization	
   of	
   social	
   spaces	
   in	
   the	
   face	
   of	
   modernization,	
  
globalization	
  and	
  diaspora,	
  issues	
  that	
  the	
  study	
  of	
  religion	
  must	
  address.	
  But	
  these	
  concepts	
  
tend	
   to	
   take	
   boundaries	
   too	
   much	
   for	
   granted	
   in	
   the	
   attempt	
   to	
   theorize	
   their	
   crossing	
   and	
  
elision.	
   They	
   are	
   useful	
   when	
   they	
   foreground	
   the	
   ideological	
   forces	
   that	
   draw,	
   label,	
  
prioritize,	
   naturalize,	
   and	
   sacralize	
   boundaries;	
   they	
   are	
   misleading	
   when	
   they	
   reify	
   those	
  
boundaries.	
  	
  
The	
   typology	
   of	
   hybridities	
   proposed	
   here	
   offers	
   one	
   tool	
   to	
   help	
   focus	
   on	
   specifics.	
   The	
  
distinction	
   between	
   hybridity	
   of	
   origin	
   and	
   hybridity	
   of	
   encounter	
   draws	
   our	
   attention	
   to	
  
distinct	
   modes	
   of	
   analysis:	
   the	
   former	
   concept	
   highlights	
   characteristics	
   of	
   hybrid	
   forms	
   as	
  
permutations	
  and	
  combinations	
  of	
  other	
  forms,	
  and	
  the	
  second	
  underlines	
  the	
  social	
  context	
  
of	
  the	
  mixing	
  process.	
  The	
  third	
  type	
  that	
  I	
  propose	
  here,	
  hybridity	
  of	
  refraction,	
  highlights	
  one	
  
relation	
   between	
   these	
   two	
   dimensions	
   of	
   analysis:	
   the	
   extent	
   to	
   which	
   variations	
   among	
  
religious	
  or	
  cultural	
  phenomena	
  reflect	
  social	
  tensions	
  within	
  a	
  specific	
  nation	
  or	
  culture.	
  
Brazilian	
   culture	
   offers	
   a	
   useful	
   case	
   for	
   rethinking	
   religion’s	
   relation	
   to	
   race,	
   class,	
  
syncretism	
  and	
  hybridity,	
  given	
  its	
  rich	
  religious	
  landscape	
  and	
  complex	
  history	
  of	
  racial	
  and	
  
cultural	
   mixing.	
   More	
   specifically,	
   Umbanda	
   contains	
   within	
   itself	
   a	
   spectrum	
   of	
   beliefs	
   and	
  
practices	
  that	
  refract	
  tensions	
  of	
  race	
  and	
  class	
  in	
  Brazil,	
  and	
  this	
  is	
  not	
  the	
  case,	
  to	
  anything	
  
like	
   the	
   same	
   extent,	
   with	
   Afro-­‐Brazilian	
   traditions	
   or	
   Kardecism.	
   Umbanda’s	
   variation	
  
reflects	
   social	
   tensions	
   in	
   Brazilian	
   society,	
   and	
   it	
   is	
   this	
   that	
   justifies	
   the	
   term	
   “hybridity	
   of	
  

refraction.”	
  
In	
   its	
   origins	
   and	
   beliefs,	
   Umbanda	
   incorporates	
   key	
   racial	
   and	
   class	
   tensions	
   prominent	
   in	
  
Brazil,	
  levelling	
  or	
  inverting	
  elements	
  that	
  are	
  hierarchically	
  arranged	
  in	
  the	
  broader	
  society.	
  
DaMatta	
   suggests	
   that	
   the	
   mestiçagem	
   present	
   in	
   Umbanda	
   reveals	
   yet	
   displaces	
   the	
  
hierarchical	
   relations	
   present	
   in	
   Brazilian	
   society:	
   “Umbanda	
   and	
   Carnaval	
   …	
   ,	
   along	
   with	
  
their	
   cousin,	
   futebol,	
   foster	
   powerful	
   ties	
   of	
   brotherhood,	
   uniting	
   the	
   powerless	
   by	
   virtue	
   of	
  
their	
  magical	
  and	
  mystical	
  powers”	
  (1983[1987]:	
  137).	
  Umbanda	
  reframes	
  social	
  tensions	
  in	
  
part	
  by	
  offering	
  a	
  symbolic	
  resolution	
  of	
  their	
  tensions	
  in	
  a	
  manner	
  distanced	
  from	
  material	
  
effects.	
   However,	
   this	
   is	
   not	
   an	
   illusory	
   or	
   merely	
   compensatory	
   relation.	
   Umbanda	
   allow	
  
millions	
   of	
   Brazilians	
   to	
   rehearse	
   modes	
   of	
   ritual	
   agency	
   that	
   both	
   reflect	
   and	
   reframe	
   the	
  
constraints	
  that	
  they	
  experience	
  in	
  their	
  society	
  (Engler	
  2007;	
  2008;	
  2009).	
  Its	
  effectiveness	
  
in	
   doing	
   so	
   is,	
   in	
   part,	
   a	
   function	
   of	
   the	
   fact	
   that	
   its	
   spectrum	
   of	
   ritual	
   forms	
   reflects	
   the	
  
broader	
  set	
  of	
  constraints	
  that	
  impact	
  practitioners’	
  experience	
  as	
  social	
  agents.	
  	
  
	
  

	
  

Acknowledgments	
  
An	
  earlier	
  version	
  of	
  this	
  paper	
  was	
  presented	
  at	
  the	
  annual	
  conference	
  of	
  the	
  North	
  American	
  
Association	
   for	
   the	
   Study	
   of	
   Religion	
   (NAASR)	
   in	
   Washington,	
   D.C.	
   in	
   2006.	
   I	
   am	
   grateful	
   for	
  
questions	
   from	
   several	
   people	
   who	
   attended	
   that	
   session	
   and	
   to	
   comments	
   by	
   Gustavo	
  
Benavides	
   and	
   Ipsita	
   Chatterjea	
   on	
  previous	
   drafts.	
   I	
   am	
   especially	
   indebted	
   to	
   my	
   colleagues	
  
in	
   the	
   Programa	
   de	
   Estudos	
   Pós-­‐Graduados	
   em	
   Ciências	
   da	
   Religião	
   at	
   the	
   Pontifícia	
  
Universidade	
  Católica	
  de	
  São	
  Paulo	
  for	
  their	
  hospitality	
  and	
  generosity	
  during	
  the	
  two	
  and	
  a	
  
half	
  years	
  that	
  I	
  spent	
  there	
  as	
  a	
  Visiting	
  Research	
  Professor	
  (2005-­‐2007).	
  Research	
  for	
  this	
  
paper	
  was	
  partially	
  funded	
  by	
  fellowships	
  from	
  the	
  Fundação de Amparo à Pesquisa do Estado
de São Paulo (FAPESP) and the	
  Coordenação	
  de	
  Aperfeiçoamento	
  de	
  Pessoal	
  de	
  Nível	
  Superior	
  
(CAPES)	
  and	
  by	
  Mount	
  Royal	
  College’s	
  Research	
  Reserve	
  Fund.	
  
References
Ashcroft,	
  Bill,	
  Gareth	
  Griffiths,	
  &	
  Helen	
  Tiffin.	
  1998.	
  Key	
  Concepts	
  in	
  Post-­‐Colonial	
  Studies.	
  New	
  York:	
  Routledge.	
  
Assunção,	
   Luiz.	
   2001.	
   “Os	
   mestres	
   da	
   Jurema:	
   Culto	
   da	
   Jurema	
   em	
   terreiros	
   de	
   Umbanda	
   no	
   interior	
   do	
  
nordeste.”	
   In	
   Reginaldo	
   Prandi	
   (ed.),	
   Encantaria	
   brasileira:	
   o	
   livro	
   dos	
   mestres,	
   caboclos	
   e	
   encantados,	
   Rio	
   de	
  
Janeiro:	
  Pallas,	
  182-­‐215.	
  
Aubrée,	
   Marion	
   &	
   François	
   Laplantine.	
   1990.	
   La	
   table,	
   le	
   livre	
   et	
   les	
   Esprits:	
   Naissance,	
   evolution	
   et	
   actualité	
   du	
  
movement	
  social	
  spirite	
  entre	
  France	
  et	
  Brésil.	
  Paris:	
  Éditions	
  Jean-­‐Claude	
  Lattès.	
  

Bastide,	
   Roger.	
   1960.	
   Les	
   religions	
   africaines	
   au	
   Brésil:	
   Vers	
   une	
   sociologie	
   des	
   interpretations	
   des	
   civilisations.	
  
Paris:	
  Presses	
  Universitaires	
  de	
  France.	
  
———.	
  1967.	
  “Le	
  spiritisme	
  au	
  Brésil.”	
  Archives	
  de	
  Sciences	
  Sociales	
  des	
  Religions	
  24(1):	
  3-­‐16.	
  
———.	
  1974.	
  “Ultima	
  scripta.	
  Textes	
  recueillis	
  et	
  commentés	
  par	
  Henri	
  Desroche.”	
  Archives	
  de	
  Sciences	
  Sociales	
  
des	
  Religions	
  38(1):	
  3-­‐47.	
  
Bhabha,	
  Homi	
  K.	
  1994.	
  The	
  Location	
  of	
  Culture.	
  London:	
  Routledge.	
  
Birman,	
  Patrícia.	
  1983.	
  O	
  que	
  é	
  Umbanda?	
  São	
  Paulo:	
  Editora	
  Brasiliense.	
  
———.	
  1985.	
  “Identidade	
  social	
  e	
  homosexualismo	
  no	
  Candomblé.”	
  Religião	
  e	
  Sociedade	
  12(1):	
  2-­‐21.	
  
———.	
  1995.	
  Fazer	
  estilo	
  criando	
  gêneros.	
  Possessão	
  e	
  diferenças	
  de	
  gênero	
  em	
  terreiros	
  de	
  Umbanda	
  e	
  Candomblé	
  
no	
  Rio	
  de	
  Janeiro.	
  Rio	
  de	
  Janeiro:	
  Ed.	
  UERJ/Relume	
  Dumará.	
  
———.	
   1997.	
   “Males	
   e	
   malefícios	
   no	
   discurso	
   Neopentecostal.”	
   In	
   Patrícia	
   Birman,	
   Regina	
   Novaes	
   &	
   Samira	
  
Crespo	
  (eds.),	
  O	
  mal	
  à	
  brasileira,	
  Rio	
  de	
  Janeiro:	
  EdUERJ,	
  62-­‐80.	
  
Boettcher,	
   Susan	
   R.	
   2005.	
   “Post-­‐Colonial	
   Reformation?	
   Hybridity	
   in	
   16th-­‐Century	
   Christianity.”	
   Social	
   Compass	
  
52(4):	
  443-­‐452.	
  
Boyer,	
  Véronique.	
  1992.	
  “De	
  la	
  campagne	
  à	
  la	
  ville:	
  la	
  migration	
  du	
  caboclo.”	
  Cahiers	
  d’Études	
  africaines	
  32/125:	
  
109-­‐127.	
  
———.	
  1996.	
  “Le	
  don	
  et	
  l'initiation.	
  De	
  l'impact	
  de	
  la	
  littérature	
  sur	
  les	
  cultes	
  de	
  possession	
  au	
  Brésil.”	
  L'Homme	
  
36(138):	
  7-­‐24.	
  
Brown,	
  Diana	
  Degroat.	
  1977.	
  “O	
  papel	
  histórico	
  da	
  classes	
  média	
  na	
  Umbanda.”	
  Religião	
  e	
  Sociedade	
  1:	
  31-­‐42.	
  
———.1985	
  .	
  “Uma	
  história	
  da	
  umbanda	
  no	
  Rio.”	
  In	
  Umbanda	
  e	
  Política,	
  Rio	
  de	
  Janeiro:	
  Marco	
  Zero.	
  
———.1994	
  [1986].	
  Umbanda:	
  Religion	
  and	
  Politics	
  in	
  Urban	
  Brazil.	
  New	
  York:	
  Columbia	
  University	
  Press.	
  
Brown,	
   Diana	
   Degroat	
   &	
   Mario	
   Bick.	
   1987.	
   “Religion,	
   Class,	
   and	
   Context:	
   Continuities	
   and	
   Discontinuities	
   in	
  
Brazilian	
  Umbanda.”	
  American	
  Ethnologist	
  14(1):	
  73-­‐93.	
  
Brumana,	
  Fernando	
  Giobellina	
  &	
  Elda	
  Gonzales	
  Martinez.	
  1989.	
  Spirits	
  from	
  the	
  Margins:	
  Umbanda	
  in	
  São	
  Paulo.	
  
Uppsala:	
  Uppsala	
  University	
  Press.	
  
Camargo,	
   Cândido	
   Procópio	
   Ferreira	
   de.	
   1961.	
   Kardecismo	
   e	
   Umbanda:	
   uma	
   interpretação	
   sociológica.	
   São	
   Paulo:	
  
Pioneira.	
  
———.	
  ed.	
  1973.	
  Católicos,	
  Protestantes,	
  Espíritas.	
  Petrópolis:	
  Editora	
  Vozes.	
  
Campos,	
   Leonildo	
   Silveira.	
   1999	
   [1997].	
   Teatro,	
   templo	
   e	
   mercado:	
   organização	
   e	
   marketing	
   de	
   um	
  
empreendimento	
  neopentecostal.	
  2nd	
  ed.	
  São	
  Paulo:	
  Vozes.	
  
Canclini,	
   Nestor	
   Garcia.	
   2006	
   [2001].	
   “Introdução.”	
   In	
   Culturas	
   Híbridas:	
   Estratégias	
   para	
   Entrar	
   e	
   Sair	
   da	
  
Modernidade,	
  Trans.	
  Gênese	
  Andrade,	
  São	
  Paulo:	
  EdUSP,	
  xvii-­‐xliii.	
  
Carneiro,	
   Edison.	
   1940.	
   “The	
   Structure	
   of	
   African	
   Cults	
   in	
   Bahia.”	
   The	
   Journal	
   of	
   American	
   Folklore	
   53(210):	
   271-­‐
278.	
  
———.	
  1977	
  [1948]	
  Candomblés	
  da	
  Bahia.	
  5th	
  ed.	
  Rio	
  de	
  Janeiro:	
  Editora	
  Civilização	
  Brasileira.	
  
Caroso,	
   Carlos	
   &	
   Jeferson	
   Bacelar	
   (eds.)	
   1999.	
   Faces	
   da	
   Tradição	
   Afro-­‐Brasileira.	
   Rio	
   de	
   Janeiro/Salvador:	
  
Pallas/CEAO.	
  
Concone,	
   Maria	
   Helena	
   Villas	
   Boas.	
   2001.	
   “Caboclos	
   e	
   pretos-­‐velhos	
   da	
   Umbanda.”	
   In	
   Reginaldo	
   Prandi	
   (ed.),	
  

Encantaria	
  brasileira:	
  o	
  livro	
  dos	
  mestres,	
  caboclos	
  e	
  encantados,	
  Rio	
  de	
  Janeiro:	
  Pallas,	
  281-­‐303.	
  
DaMatta,	
   Roberto.	
   1983	
   [1978].	
   Carnavais,	
   malandros	
   e	
   heróis:	
   para	
   uma	
   sociologia	
   do	
   dilema	
   brasileiro,	
   5th	
   ed.	
  
Rio	
  de	
  Janeiro:	
  Zahar.	
  [Carnivals,	
  Rogues,	
  and	
  Heroes:	
  An	
  Interpretation	
  of	
  the	
  Brazilian	
  Dilemma,	
  Trans.	
  John	
  
Drury,	
  Notre	
  Dame,	
  IN:	
  University	
  of	
  Notre	
  Dame	
  Press,	
  1991.]	
  
———.	
  1987.	
  Relativizando:	
  Uma	
  introdução	
  à	
  antropologia	
  social.	
  Rio	
  de	
  Janeiro:	
  Rocco.	
  
Despland,	
  Michel.	
  2008.	
  Bastide	
  on	
  Religion:	
  The	
  Invention	
  of	
  Candomblé.	
  London:	
  Equinox.	
  
Engler,	
  Steven.	
  2005a.	
  “Tradition’s	
  Legacy.”	
  In	
  Steven	
  Engler	
  &	
  Gregory	
  P.	
  Grieve	
  (eds.),	
  Historicizing	
  Tradition	
  in	
  
the	
  Study	
  of	
  Religion,	
  Berlin	
  and	
  New	
  York:	
  DeGruyter,	
  357-­‐378.	
  
———.	
  2005b.	
  “Tradition.”	
  In	
  Kocku	
  von	
  Stuckrad	
  (ed.),	
  The	
  Brill	
  Dictionary	
  of	
  Religion,	
  4	
  vols.,	
  Leiden:	
  Brill,	
  4:	
  
1907-­‐1911.	
  
———.	
   2006.	
   “Rethinking	
   ‘Hybridity’:	
   Anterior	
   Purities	
   and	
   the	
   Invention	
   of	
   Tradition.”	
  Religião	
   e	
   Cultura	
   10:	
   9-­‐
20.	
  	
  
———.	
  2007.	
  “Patronage	
  and	
  Distributed	
  Agency	
  in	
  Brazilian	
  Spirit-­‐possession.”	
  Paper	
  presented	
  at	
  the	
  annual	
  
conference	
  of	
  the	
  North American Association for the Study of Religion, San Diego.	
  
———.	
  2008. “Reflexivity,	
  Virtuality	
  and	
  Social	
  Relations	
  in	
  Brazilian	
  Umbanda.”	
  Paper	
  presented	
  at	
  the	
  annual	
  
conference	
  of	
  the	
  American	
  Academy	
  of	
  Religion,	
  Chicago.	
  
———.	
   2009.	
   “Brazilian	
   Spirit	
   Possession	
   in	
   the	
   Light	
   of	
   Recent	
   Theory	
   of	
   Ritual.”	
   Keynote	
   address	
   for	
   the	
  
conference	
  “Performing	
  Self	
  and	
  Community:	
  New	
  Perspectives	
  on	
  Ritual	
  Practice”	
  organized	
  by	
  The	
  Centre	
  
for	
  Research	
  on	
  Religion	
  /	
  Centre	
  de	
  Recherche	
  sur	
  la	
  Religion	
  (CREOR),	
  McGill	
  University,	
  forthcoming	
  in	
  
ARC:	
  The	
  Journal	
  of	
  the	
  Faculty	
  of	
  Religious	
  Studies,	
  McGill	
  University.	
  
Fischer,	
   Brodwyn.	
   2004.	
   “Quase	
   pretos	
   de	
   tão	
   pobres?	
   Race	
   and	
   Social	
   Discrimination	
   in	
   Rio	
   de	
   Janeiro’s	
  
Twentieth-­‐century	
  Criminal	
  Courts.”	
  Latin	
  American	
  Research	
  Review	
  39(1):	
  31–59.	
  
Fry,	
  Peter.	
  1982.	
  Para	
  inglês	
  ver.	
  Rio	
  de	
  Janeiro:	
  Zahar.	
  
Gilroy,	
  Paul.	
  1994.	
  “Black	
  Cultural	
  Politics:	
  An	
  Interview	
  with	
  Paul	
  Gilroy	
  by	
  Timmy	
  Lott.”	
  Found	
  Object	
  4:	
  46-­‐81.	
  
———.	
  2000.	
  Between	
  Camps.	
  London:	
  Penguin.	
  
Guerriero,	
  Silas.	
  2006.	
  Novos	
  movimentos	
  religiosos:	
  o	
  quadro	
  brasileiro.	
  São	
  Paulo:	
  Paulinas.	
  
Hale,	
  Lindway	
  Lauren.	
  1997.	
  “Preto	
  Velho:	
  Resistance,	
  Redemption,	
  and	
  Engendered	
  Representations	
  of	
  Slavery	
  
in	
  a	
  Brazilian	
  Possession-­‐Trance	
  Religion.”	
  American	
  Ethnologist	
  24(2):	
  392-­‐414.	
  
Hall,	
  Stuart.	
  1993.	
  “Culture,	
  Community,	
  Nation.”	
  Cultural	
  Studies	
  7:	
  349-­‐363.	
  
Hannerz,	
  Ulf.	
  1987.	
  “The	
  World	
  in	
  Creolisation.”	
  Africa:	
  Journal	
  of	
  the	
  International	
  African	
  Institute	
  57(4):	
  546-­‐
559.	
  
Harding,	
   Rachel	
   E.	
   2000.	
   A	
   Refuge	
   in	
   Thunder:	
   Candomblé	
   and	
   Alternative	
   Spaces	
   of	
   Blackness.	
   Bloomington:	
  
Indiana	
  University	
  Press.	
  
———.	
   2005.	
   “Afro-­‐Brazilian	
   Religions.”	
   In	
   Lindsay	
   Jones	
   (ed.),	
   Encyclopedia	
   of	
   Religion,	
   2nd	
   ed.,	
   New	
   York:	
  
Macmillan	
  Reference,	
  119-­‐125.	
  
Hess,	
  David	
  J.	
  1991.	
  Spirits	
  and	
  Scientists:	
  Ideology,	
  Spiritism,	
  and	
  Brazilian	
  Culture.	
  University	
  Park:	
  Pennsylvania	
  
State	
  University	
  Press.	
  
Hoffman	
  Kelly	
  and	
  Miguel	
  Angel	
  Centeno.	
  2003.	
  “The	
  Lopsided	
  Continent:	
  Inequality	
  in	
  Latin	
  America.”	
  Annual	
  

Review	
  of	
  Sociology	
  29:	
  363–90.	
  
Hutnyk,	
  John.	
  2005.	
  “Hybridity.”	
  Ethnic	
  and	
  Racial	
  Studies	
  28(1):	
  79-­‐102.	
  
Jacob,	
   Cesar	
   Romero,	
   Dora	
   Rodrigues	
   Hees,	
   Philippe	
   Waniez	
   &	
   Violette	
   Brustlein.	
   2003.	
  Atlas	
   da	
   filiação	
   religiosa	
  
e	
  indicadores	
  sociais	
  no	
  Brasil.	
  São	
  Paulo:	
  Edições	
  Loyola.	
  
———.	
  2006.	
  Religiao	
  e	
  sociedade	
  em	
  capitais	
  brasileiras.	
  Rio	
  de	
  Janeiro	
  and	
  São	
  Paulo:	
  PUC-­‐Rio/Edições	
  Loyola.	
  
Johnson,	
  Paul	
  Christopher.	
  2002a.	
  Secrets,	
  Gossip,	
  and	
  Gods:	
  The	
  Transformation	
  of	
  Brazilian	
  Candomblé.	
  Oxford	
  
and	
  New	
  York:	
  Oxford	
  University	
  Press.	
  
———.	
   2002b.	
   “Migrating	
   Bodies,	
   Circulating	
   Signs:	
   Brazilian	
   Candomblé,	
   the	
   Garifuna	
   of	
   the	
   Caribbean,	
   and	
   the	
  
Category	
  of	
  Indigenous	
  Religions.”	
  History	
  of	
  Religions	
  41(4):	
  301-­‐327.	
  
Kloppenburg,	
  Boaventura.	
  1964.	
  O	
  espiritismo	
  no	
  Brasil.	
  Petrópolis:	
  Editora	
  Vozes.	
  
Kraidy,	
  Marwan	
  M.	
  2002.	
  “Hybridity	
  in	
  Cultural	
  Globalization.”	
  Communication	
  Theory	
  12(3):	
  316-­‐339.	
  
Landes,	
  Ruth.	
  1947.	
  The	
  City	
  of	
  Women.	
  New	
  York:	
  MacMillan.	
  
Leacock,	
  Seth.	
  1964a.	
  “Fun-­‐Loving	
  Deities	
  in	
  an	
  Afro-­‐Brazilian	
  Cult.”	
  Anthropological	
  Quarterly	
  37(3):	
  94-­‐109.	
  
———.	
  1964b.	
  “Ceremonial	
  Drinking	
  in	
  an	
  Afro-­‐Brazilian	
  Cult.”	
  American	
  Anthropologist	
  (n.s.)	
  66(2):	
  344-­‐354.	
  
Leopold,	
  Anita.	
  2004.	
  “General	
  Introduction.”	
  In	
  Anita	
  M.	
  Leopold,	
  &	
  Jeppe	
  S.	
  Jensen	
  (eds.),	
  Syncretism	
  in	
  Religion:	
  
A	
  Reader,	
  London:	
  Equinox,	
  2-­‐12.	
  
Leopold,	
  Anita	
  M.	
  &	
  Jeppe	
  S.	
  Jensen,	
  eds.	
  2004.	
  Syncretism	
  in	
  Religion:	
  A	
  Reader.	
  London:	
  Equinox.	
  
Lovell,	
   Peggy	
   A.	
   &	
   Charles	
   Wood.	
   1998.	
   “Skin	
   Color,	
   Racial	
   Identity,	
   Chances	
   in	
   Brazil.”	
   Latin	
   American	
  
Perspectives	
  25(3):	
  90–109.	
  
Maggie,	
  Yvonne.	
  2001[1977].	
  Guerra	
  de	
  Orixá:	
  um	
  estudo	
  de	
  ritual	
  e	
  conflito.	
  Rio	
  de	
  Janeiro:	
  Jorge	
  Zahar.	
  
Malandrino,	
  Brígida	
  Carla.	
  2006.	
  Umbanda:	
  mudanças	
  e	
  permanências.	
  São	
  Paulo:	
  Editora	
  PUC-­‐SP.	
  
Mariano,	
  Ricardo.	
  1999.	
  Neopentecostais:	
  sociologia	
  do	
  novo	
  pentecostalismo	
  no	
  Brasil.	
  São	
  Paulo:	
  Loyola.	
  
Martin,	
   Luther	
   H.	
   and	
   Anita	
   M.	
   Leopold.	
   2004.	
   “New	
   Approaches	
   to	
   the	
   Study	
   of	
   Syncretism.”	
   In	
   Peter	
   Antes,	
  
Armin	
   W.	
   Geertz,	
   and	
   Randi	
   R.	
   Warne	
   (eds.),	
   New	
   Approaches	
   to	
   the	
   Study	
   of	
   Religion.	
   Volume	
   2:	
   Textual,	
  
Comparative,	
  Sociological	
  and	
  Cognitive	
  Approaches,	
  Berlin	
  and	
  New	
  York:	
  De	
  Gruyter,	
  93-­‐107.	
  
Matta	
  e	
  Silva,	
  W.W.	
  da.	
  2004	
  [1969].	
  Umbanda	
  do	
  Brasil.	
  3rd	
  ed.	
  São	
  Paulo:	
  Ícone	
  Editora.	
  
McGuire,	
  Meredith	
  B.	
  &	
  Otto	
  Maduro.	
  2005.	
  “Rethinking	
  Religious	
  Hybridity.”	
  Social	
  Compass	
  52(4):	
  411–415.	
  
Monroe,	
   John	
   Warne.	
   2008.	
   Laboratories	
   of	
   Faith:	
   Mesmerism,	
   Spiritism	
   and	
   Occultism	
   in	
   Modern	
   France.	
   Ithaca	
  
and	
  London,	
  Cornell	
  University	
  Press.	
  
Montero,	
  Paula.	
  1985.	
  Da	
  doença	
  à	
  desordem:	
  a	
  magia	
  na	
  Umbanda.	
  Rio	
  de	
  Janeiro:	
  Graal.	
  
Moreiras,	
  Alberto.	
  1999.	
  “Hybridity	
  and	
  Double	
  Consciousness.”	
  Cultural	
  Studies	
  13(3):	
  373-­‐407.	
  
Motta,	
  Roberto.	
  1996.	
  “A	
  invenção	
  da	
  África:	
  Roger	
  Bastide,	
  Edison	
  Carneiro	
  e	
  os	
  conceitos	
  de	
  memória	
  coletiva	
  e	
  
pureza	
  Nagô.”	
  In	
  Tânia	
  Lima	
  (ed.),	
  Sincretismo	
  religioso:	
  o	
  ritual	
  afro.	
  Anais	
  do	
  IV	
  congresso	
  afro-­‐brasileiro,	
  
Recife,	
  maio,	
  1994,	
  Recife:	
  Fundação	
  Joaquim	
  Nabuco/Editora	
  Massangana,	
  24-­‐32.	
  
———.	
   2006	
   [1999].	
   “Religiões	
   afro-­‐recifenses:	
   ensaio	
   de	
   classificação.”	
   In	
   Carlos	
   Caroso	
   &	
   Jeferson	
   Bacelar	
  
(eds.),	
  Faces	
  da	
  Tradição	
  Afro-­‐Brasileira,	
  Rio	
  de	
  Janeiro/Salvador:	
  Pallas/CEAO,	
  17-­‐35.	
  
Motta	
  de	
  Oliveira,	
  José	
  Henrique.	
  2007.	
  “Eis	
  que	
  o	
  caboclo	
  veio	
  à	
  Terra	
  ‘anunciar’	
  a	
  Umbanda.”	
  História,	
  imagem	
  e	
  
narrativas	
  4(2):	
  176-­‐188.	
  

Natividade,	
   Marcelo	
   &	
   Leandro	
   de	
   Oliveira.	
   2007.	
   “Religião	
   e	
   intolerância	
   à	
   homosexualidade:	
   tendências	
  
contemporâneas	
   no	
   Brasil.”	
   In	
   Vagner	
   Gonçalves	
   da	
   Silva	
   (ed.),	
   Intolerância	
   religiosa:	
   impactos	
   do	
  
neopentecostalismo	
  no	
  campo	
  religioso	
  afro-­‐brasileiro,	
  São	
  Paulo:	
  EdUSP,	
  261-­‐302.	
  
Nederveen	
  Pieterse,	
  Jan.	
  2001.	
  “Hybridity,	
  so	
  what?	
  The	
  anti-­‐hybridity	
  backlash	
  and	
  the	
  riddles	
  of	
  recognition.”	
  
Theory,	
  Culture	
  &	
  Society	
  18(2-­‐3):	
  219-­‐245.	
  
Negrão,	
   Lisias Nogueira. 1979.	
   “A	
   Umbanda	
   como	
   expressão	
   da	
   religiosidade	
   popular.”	
   Religião	
   e	
   Sociedade	
   4:	
  
171-­‐191.	
  
———.	
  1996.	
  Entre	
  a	
  Cruz	
  e	
  a	
  Encruzilhada:	
  formação	
  do	
  campo	
  umbandista	
  em	
  São	
  Paulo.	
  São	
  Paulo:	
  EdUSP.	
  
———.	
  2005	
  [1987].	
  “Kardecism.”	
  In	
  Lindsay	
  Jones	
  (ed.),	
  Encyclopedia	
  of	
  Religion,	
  2nd	
  ed.,	
  New	
  York:	
  Macmillan	
  
Reference,	
  5089-­‐5091.	
  
Oro,	
   Ari	
   Pedro.	
   1989.	
   “Immigrants	
   européens	
   et	
   religions	
   afro-­‐brésiliennes	
   dans	
   le	
   sud	
   du	
   Brésil.”	
   Archives	
   de	
  
Sciences	
  Sociales	
  des	
  Religions	
  68(1):	
  125-­‐140.	
  
———.	
  2007.	
  “Intolerância	
  religiosa	
  Iurdiana	
  e	
  Reações	
  Afro	
  no	
  Rio	
  Grande	
  do	
  Sul.”	
  In	
  Vagner	
  Gonçalves	
  da	
  Silva	
  
(ed.),	
   Intolerância	
   religiosa:	
   impactos	
   do	
   neopentecostalismo	
   no	
   campo	
   religioso	
   afro-­‐brasileiro,	
   São	
   Paulo:	
  
EdUSP,	
  29-­‐69.	
  
Ortiz,	
   Renato.	
   1975.	
   “Du	
   syncrétisme	
   à	
   la	
   synthèse.	
   Umbanda,	
   une	
   religion	
   brésilienne.”	
   Archives	
   de	
   Sciences	
  
Sociales	
  des	
  Religions	
  40(1):	
  83-­‐97.	
  
———.	
   1999	
   [1978].	
   A	
   morte	
   branca	
   do	
   feiticeiro	
   negro:	
   Umbanda	
   e	
   sociedade	
   brasileira,	
   2nd	
   ed.	
   Petrópolis:	
  
Vozes.	
  
Pace,	
  Richard.	
  1997.	
  “The	
  Amazon	
  Caboclo:	
  What’s	
  in	
  a	
  Name?”	
  Luso-­‐Brazilian	
  Review	
  34(2):	
  81-­‐89.	
  
Parés,	
  Luis	
  Nicolau.	
  2007.	
  A	
  formação	
  do	
  Candomblé:	
  história	
  e	
  ritual	
  da	
  nação	
  jeje	
  na	
  Bahia.	
  Campinas:	
  Editora	
  
da	
  UNICAMP.	
  
Pierucci,	
  Antônio	
  Flávio	
  &	
  Reginaldo	
  Prandi.	
  2000.	
  “Religious	
  Diversity	
  in	
  Brazil:	
  Numbers	
  and	
  Perspectives	
  in	
  a	
  
Sociological	
  Evaluation.”	
  International	
  Sociology	
  15(4):	
  629–639.	
  
Pinto,	
  Tancredo	
  da	
  Silva	
  &	
  Byron	
  Tôrres	
  de	
  Freitas.	
  1972.	
  Umbanda:	
  guia	
  e	
  ritual	
  para	
  organização	
  de	
  terreiros.	
  
7th	
  ed.	
  Rio	
  de	
  Janeiro:	
  Editora	
  Eco.	
  
Prandi,	
  

Reginaldo.	
  

1991.	
  

Os	
  

candomblés	
  

de	
  

São	
  

Paulo.	
  

São	
  

Paulo:	
  

Hucitec/EdUSP.	
  

<http://www.fflch.usp.br/sociologia/prandi/csp-­‐down.htm>	
  	
  
———.	
   2000.	
   “African	
   Gods	
   in	
   Contemporary	
   Brazil:	
   A	
   Sociological	
   Introduction	
   to	
   Candomblé	
   Today.”	
  
International	
  Sociology	
  15(4):	
  641-­‐663.	
  
———.	
  2005.	
  Segredos	
  guardados.	
  Orixás	
  na	
  alma	
  brasileira.	
  São	
  Paulo:	
  Companhia	
  das	
  Letras.	
  
Pye,	
  Michael.	
  1971.	
  “Syncretism	
  and	
  Ambiguity.”	
  Numen	
  18(2):	
  83–93.	
  
Ramos,	
   Arthur.	
   2001[1934].	
   O	
   negro	
   brasileiro.	
   Vol.	
   1.	
   Etnografia	
   religiosa.	
   Rio	
   de	
   Janeiro:	
   Graphia	
   Editorial.	
   [The	
  
Negro	
  in	
  Brazil,	
  Trans.	
  Richard	
  Pattee,	
  Washington,	
  D.C.:	
  Associated	
  Publishers,	
  1951.]	
  
Santos,	
   Jocélio	
   Teles	
   dos.	
   1992.	
   “La	
   divinité	
   caboclo	
   dans	
   le	
   candomblé	
   de	
   Bahia.”	
   Cahiers	
   d’Études	
   africaines	
  
32(125):	
  83-­‐107.	
  
Silva,	
   Vagner	
   Gonçalves	
   da.	
   2005	
   [1994].	
   Candomblé	
   e	
   Umbanda:	
   Caminhos	
   da	
   devoção	
   brasileira,	
   2nd	
   ed.	
   São	
  
Paulo:	
  Selo	
  Negro	
  Edições.	
  

Stewart,	
  Charles.	
  1999.	
  “Syncretism	
  and	
  its	
  Synonyms:	
  Reflections	
  on	
  Cultural	
  Mixture.”	
  Diacritics	
  29(3):	
  40-­‐62.	
  
———.	
  2004.	
  “Relocating	
  Syncretism	
  in	
  Social	
  Science	
  Discourse.”	
  In	
  Anita	
  M.	
  Leopold	
  &	
  Jeppe	
  S.	
  Jensen	
  (eds.),	
  
Syncretism	
  in	
  Religion:	
  A	
  Reader,	
  London:	
  Equinox,	
  264-­‐285.	
  
Stross,	
   Brian.	
   1999.	
   “The	
   Hybrid	
   Metaphor:	
   From	
   Biology	
   to	
   Culture.”	
   The	
   Journal	
   of	
   American	
   Folklore	
   112:	
   254-­‐
267.	
  
Tavares,	
   Fátima	
   Regina	
   Gomes	
   &	
   Maria	
   da	
   Graça	
   Floriano.	
   2003.	
   “Do	
   Canjerê	
   ao	
   Candomblé:	
   notas	
   sobre	
   a	
  
tradição	
   afro-­‐brasileiro	
   em	
   Juiz	
   de	
   Fora.”	
   In	
   Fátima	
   Regina	
   Gomes	
   Tavares	
   &	
   Marcelo	
   Ayres	
   Camurça	
   (eds.),	
  
Minas	
  das	
  devoções:	
  diversidade	
  religiosa	
  em	
  Juiz	
  de	
  Fora,	
  Juiz	
  de	
  Fora:	
  Universidade	
  Federal	
  de	
  Juiz	
  de	
  Fora,	
  
165-­‐177.	
  
Toop,	
   Walter	
   R.	
   1972.	
   “Organized	
   Religious	
   Groups	
   in	
   a	
   Village	
   of	
   Northeastern	
   Brazil.”	
   Luso-­‐Brazilian	
   Review	
  
9(2):	
  58-­‐77.	
  
Wade,	
   Peter.	
   2005.	
   “Rethinking	
   Mestizaje:	
   Ideology	
   and	
   Lived	
   Experience.”	
   Journal	
   of	
   Latin	
   American	
   Studies	
  
37(2):	
  239-­‐257.	
  
Warren	
  Jr.,	
  Donald.	
  1968.	
  “Spiritism	
  in	
  Brazil.”	
  Journal	
  of	
  Inter-­‐American	
  Studies	
  10(3):	
  393-­‐405.	
  
Werbner,	
   Pnina.	
   1997.	
   “Introduction:	
   The	
   Dialectics	
   of	
   Cultural	
   Hybridity.”	
   In	
   Pnina	
   Werbner	
   (ed.),	
   Debating	
  
Cultural	
  Hybridity:	
  Multi-­‐cultural	
  Identities	
  and	
  the	
  Politics	
  of	
  Anti-­‐Racism,	
  London:	
  Zed	
  Books,	
  1-­‐26.	
  
Young,	
  R.J.C.	
  1995.	
  Colonial	
  Desire:	
  Hybridity	
  in	
  Theory,	
  Culture	
  and	
  Race.	
  London:	
  Routledge.