8/18/2019 Yoruba Names and Gender - Orie AL 2002
1/30
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/262637470
Yoruba names and gender marking
ARTICLE · JANUARY 2002
CITATION
1
READS
103
1 AUTHOR:
Olanikė Ola Orie
Tulane University
26 PUBLICATIONS 55 CITATIONS
SEE PROFILE
https://www.researchgate.net/profile/Olanike_Orie?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_7https://www.researchgate.net/profile/Olanike_Orie?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_7https://www.researchgate.net/profile/Olanike_Orie?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_7https://www.researchgate.net/profile/Olanike_Orie?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_7https://www.researchgate.net/profile/Olanike_Orie?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_7https://www.researchgate.net/institution/Tulane_University?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_6https://www.researchgate.net/profile/Olanike_Orie?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_5https://www.researchgate.net/profile/Olanike_Orie?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_4https://www.researchgate.net/?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_1https://www.researchgate.net/publication/262637470_Yoruba_names_and_gender_marking?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_3https://www.researchgate.net/publication/262637470_Yoruba_names_and_gender_marking?enrichId=rgreq-00eba93f-e995-4771-95f1-54b1d5661efa&enrichSource=Y292ZXJQYWdlOzI2MjYzNzQ3MDtBUzoxMDczNzY1MjcwMjQxMjhAMTQwMjYxMTk2ODIzNQ%3D%3D&el=1_x_2
8/18/2019 Yoruba Names and Gender - Orie AL 2002
2/30
Trustees of Indiana University
Anthropological Linguistics
Yoruba Names and Gender MarkingAuthor(s): Ọlanikẹ Ọla OrieReviewed work(s):Source: Anthropological Linguistics, Vol. 44, No. 2 (Summer, 2002), pp. 115-142Published by: The Trustees of Indiana University on behalf of Anthropological LinguisticsStable URL: http://www.jstor.org/stable/30028837 .
Accessed: 16/05/2012 10:50
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Trustees of Indiana University and Anthropological Linguistics are collaborating with JSTOR to digitize,
preserve and extend access to Anthropological Linguistics.
http://www.jstor.org/action/showPublisher?publisherCode=tiuhttp://www.jstor.org/action/showPublisher?publisherCode=anthlinghttp://www.jstor.org/stable/30028837?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/30028837?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=anthlinghttp://www.jstor.org/action/showPublisher?publisherCode=tiu
8/18/2019 Yoruba Names and Gender - Orie AL 2002
3/30
Yoruba Names and Gender
Marking
OLANIKEOLA
ORIE
Tulane
University
Abstract. Attributive names constitute the
principal
locus of
gender
dis-
tinction
in
Yoruba.
Masculine
names have the tone
pattern
LLH
and contain
two
monosyllabic
verbs
denoting
semantic themes such as
bravery
and inten-
tional
possession;
in
contrast,
feminine
names have LLH or LHH tone
patterns
and contain verbs
reflecting
themes
involving nurturing.
These
properties
are
analyzed
as
resulting
from the
interaction
of
phonology, morphology, syntax,
and semantics.
Furthermore,
differences in the
frequency
of masculine and
feminine names are
analyzed
as
following
from markedness.
Finally,
whereas
frequency,
femininity,
and aesthetics
play
some role in the selection of feminine
names,
semantics
plays
the dominant role.
1. Introduction. Gender is a term used to
classify
nouns
as
masculine,
feminine,
and
neuter.
Two
types
of
gender marking
occur
crosslinguistically:
grammatical gender
and natural
gender.
Grammatical
gender regulates gender
agreement
between
words,
whereas individual words
carry
natural
gender
information. French exhibits
grammatical
gender.
Hence,
cooccurring
articles
and
nouns must
agree
with
respect
to
gender.
On the other
hand,
English
does
not
impose gender agreement
on
cooccurring words,
but it exhibits natural
gender,
which is seen
only
in
third-person singular pronouns
and a few words
such
as
prince/princess
and
actor/actress.
Although
gender
is marked
in
many
languages,
some
languages
do not
classify
nouns or
pronouns
in terms
of
gender.
Yoruba
(of
the
Benue-Congo family, Nigeria)
is considered an
example
of such a
language;
it classifies
pronouns
in terms of
person
and number
(Bamgbose
1966;
Awobuluyi
1978),
but
not on the basis of
gender.
Furthermore,
there are
no
affixes
that contrast
nouns in terms of
gender.
While it is true that
gender
is not a
general property
of
Yoruba
nouns,
there
is evidence for
gender
marking
in
personal
attributive names.
According
to
Oyetade (1991),
two tonal
patterns
are
used
in
forming
attributive names:' low-
low-high
(LLH)
and
low-high-high
(LHH).2
The LLH
pattern
is
the
most
com-
mon and is
used for both masculine and feminine names.
This form is derived
by
prefixing
a low-toned
A
to
a
sequence
of two
monosyllabic
verbs,
as in table
1
below.
The LHH
pattern
is also derived
by attaching
a low-toned
A
to two
monosyllabic
verbs,
resulting
in forms that
are feminine
names,
such as those
in
table 2.3
115
8/18/2019 Yoruba Names and Gender - Orie AL 2002
4/30
116
ANTHROPOLOGICAL
INGUISTICS
44
NO. 2
Table 1. Masculine and
Feminine Names Derived with
the LIH Tone
Pattern
MASCULINE
FEMININE
A-ji-gbd A-l--ki
PREF-fight-carry
PREF-emerge.uniquely-pamper
A-lk-ni
A-we-ki
PREF-emerge.uniquely-possess PREF-bathe-pamper
A-kin-bt
A-be-ki
PREF-meet.intentionally-born PREF-beg-pamper
A-yin-de A-yin-lke
PREF-praise-arrive
PREF-praise-pamper
Table 2.
Feminine Names Derived with the LHH
Tone Pattern
A-gb-lki
PREF-carry-pamper
A-to-ke
PREF-nurture-pamper
A-j(-ki
PREF-wake.up-pamper
A-ni'-ki
PREF-possess-pamper
This article addresses
three issues related to the data in
tables
1
and 2. The
first
concerns the
characterization
of
the
gender
markers: which
properties sup-
ply
the
gender
distinction?
Oyetade
proposes
that
gender
marking
is
derived
from
the semantics of the
verbs contained in the name and from
the two tonal
patterns
LLH
and LHH. The article
argues
that
these two
properties
are neces-
sary
but
not sufficient to account for
gender marking.
The
existing
account
does
not
explain why
only
a
verb
phrase
with a serial verb
construction
is a valid
base. In
addition,
it
does
not
explain why
the
serial
verb must contain
exactly
two
monosyllabic
verbs. Since
Yoruba is
a
serializing language,
which allows
two
or more
verbs to occur in a
sequence
within
a sentence
(see,
e.g.,
Bamgbose
1974;
Oyelaran
1982;
Awoyale 1988),
why
is it
impossible
to
have
an attributive
name
with
more than two
monosyllabic
verbs? To
explain
these
restrictions,
it is
proposed
that the tonal
and semantic account
must be
supplemented
by syn-
tactic and prosodic requirements, which make a serial verb constituting a binary
foot the
optimal
base of
prefixation.
Analyzing
the base of
prefixation
as a con-
stituent
governed by
prosodic
and
syntactic
constraints
demonstrates,
contrary
to the
proposal
of Selkirk
(1986),
that
prosodic-based processes
cannot
always
be
defined in
purely prosodic
or
morphoprosodic
terms.
The second issue is
why
the
LLH
pattern
is
more common than
the LHH
pattern.
It is
argued
that
the masculine
pattern
is
the more common
pattern
8/18/2019 Yoruba Names and Gender - Orie AL 2002
5/30
2002
OLANIKELA
ORIE
117
because
it is
the unmarked form.
Hence,
both masculine and feminine attri-
butive names can be derivedfrom it. The
LHH
pattern
is less common
because
it
is the markedform,the form reservedexclusivelyforcreatingfeminine attribu-
tive names. That
is,
the use of the feminine tonal
pattern provides
more
specific
information
and rules
out the
possibility
of masculine
reference,
whereas
the use
of the masculine tonal
pattern
does not exclude the
possibility
of
feminine
reference
(Baker 1992).
The third issue is
sociolinguistic
in nature. The
specific question
addressed
is the
following: given
that female names can be derived from the
male-based
LLH tonal
pattern
and the
exclusively
female-basedLHH tonal
pattern,
on what
basis do
parents
choose names
for
girls
from one set or the
other?4
n
addressing
this
question,
I show that the
flexibility
of name selection for
girls
results from a
range
of
factors,
including frequency, femininity,
aesthetics,
and
semantics.
First, frequency
is a
factor
because the
LLH tonal
pattern
is considered
more
common
and
popular
than the LHHtonal
pattern.
Second,
femininity
and aes-
thetics are
contributing
factors since the LHH
pattern
is viewed as more
feminine-sounding
and
attractive than the LLH
pattern.
Third,
semantics is a
factor
because some
parents
select names
just
to
express
the circumstancessur-
rounding
the birth of a
child,
to describe who the child
is,
or to
convey
their
wishes
for a child. For
such
parents,
the tonal
pattern may
be LLH or LHH. Of
all
these
factors,
however,
semantics is the most
important;
even when factors
such as
frequency,
emininity,
and
aesthetics
play
a role
in
the selection of
girls'
names,
meaning
is still crucial for all
parents.
The structure of the
article is as follows.
In
section
2,
sociolinguistic
back-
ground
is
briefly
reviewed.
Section
3
discusses traditional
naming among
the
Yoruba. In section 4, the strategies for name formation in Yoruba are presented
and the differences between
personal
names,
attributive
names,
and nicknames
are outlined in detail. Section
5
provides
an
account of
attributive
names show-
ing
the interaction of
morphosyntactic,
tonal, semantic,
and
prosodic properties.
In section
6,
the difference
in the
frequency
of LLH and
L;HH
names is
explained
as
following
from markedness factors. Section
7
examines
the
issue of name
selection for
girls,
and
section 8
gives
the conclusion.
Finally, sample
lists of
attributive names are
provided
in
appendices
1 and 2.
2.
Sociolinguistic background.
The
Yoruba of West
Africa are one of the
largest
ethnic
groups
south
of the Sahara
(Bascom 1969).
Their
language,
Yoruba,
although predominantly
spoken
in
Nigeria,5
is also
spoken
in Benin and
Togo
(see
map
1).
In
total,
there are over
twenty
million
speakers.6
Yoruba has more
than
twenty
distinct dialects.
Examples
are
Oyo, Ijesa,
Ife,
Igbomina, Ijebu,Egba,
Awori, Ondo, Ekiti,
Ilaje,
Ikale, Owo,
Ijo-Apoi,
Owe,
Ijumu,
Yagba,
Gbede,
Bunu, Shabe,
and Ketu
(see
map 1).7
Aside from
these
dialects,
there is a standard dialect
(Standard Yoruba),8
which
is
taught
in
Nigerian
schools and
used in
literary writing.
It is also the official
language
in
8/18/2019 Yoruba Names and Gender - Orie AL 2002
6/30
118 ANTHROPOLOGICALINGUISTICS
44
NO.2
southwestern
Nigeria,
and it is one of the
major languages
of the
media
(used
in
newspapers,
radio,
and television
broadcasting).
ISHA
SHABE
I/yrie
ANA
Atakpame
IDASBA
shabe
CRIdR.
OYO
OgunR.
OYO
IGBOMINA
Its
YAGBAI
AWORO
BUNU
TOGO
m o n o R
wemeR
KETU
ketu
Ibadan
Ebeokuta
owu
UESA
Ilesa
EKITI
Ife
Owo
BENIN
IFONYIN
EGBADO
EGBA
UISBU
Ijebu-ode
Ondo
ONDO
OWO
AWORI
Lagos
OSSER
NIGERIA
ILAJE
ITSEKIRI
Wari
NigerR.
0
20 40 60 80
100
scaleb mles
Map
1.
The location ofYorubadialects in
Nigeria, Benin,
and
Togo.
3. Traditional
naming among
the Yoruba.
Before
1840,
when Christian
names
and surnames
began
to be
used,
a Yoruba
person's
full
name had three or
four elements (Oduyoye 1972; Oyelaran 1976): ordko 'personal name',
ori'ki
zbiso
'attributive
name',9
oriki
Alije
'nickname','o
and oriki
or'le
'totemic
name'. Some
examples
are
given
in table
3.
Table 3. Traditional
Yoruba Names
PERSONALAMES
Ad6yemi
'crown
befits me'
Omddll6
'child
arrives home'
Akinola
'valor of
high
status'
Abiddin
'child born
during
a festival'
ATTRIBUTIVEAMES
Albi
'child who
emerges
singularly
to be
born'
Agbdke
'child to be carried
and
pampered'
Akiznjif
'childwho
brings
awakening'
Alik6
'child who
emerges
to be
pampered'
NICKNAMES
Eyinfinjowd
'white teeth'
Awele.gb
'tall and slim'
P6ldyejti
'pele
befits face'
Ayiluko
'plump
woman'
TOTEMICAMES
Qkin
'peacock'
Opo
'pillar'
Erin
'elephant'
Agbo
'ram'
NOTE: 6l6
denotes a
type
of
facial
marking.
8/18/2019 Yoruba Names and Gender - Orie AL 2002
7/30
2002
QLANIKELA
ORIE
119
In
general, people
are
universally
known
by
their
personal
names and are
known
familiarly by
their attributive names
(Johnson 1969:87).
Nicknames are
like attributive names because they are usually used by people who are familiar
with
the owner of
a name.
There are rules
regulating
the
use
of these
names.
For
example,
whereas
everyone may
address individuals
by
their
personal
names,
only
elders can address children
by
their attributive names when
they
want to
express
a
feeling
of endearment or affection for a child
(Johnson 1969:85).
In
contrast,
it is
considered
rude for a
younger person
to
address
an older
person by
his
or her
attributive name.
Unlike
personal
names, attributives,
and
nicknames,
which
belong
to indi-
viduals,
totemic names
belong
to families. In
addition,
they
have
accompanying
poems,
which encode information such as
family origin,
behavior and
character,
profession,
religion,
social
status,
and taboos
(Babalola
1967).
To
fully identify
a
person,
the
names
described
above are mentioned and connected to the names of
an individual's
parents,
as in
(1).
(1) Omo~ld AgbdkeAwe.lgbd Opd,
omo on Koldwpld,
mo
Ad&itutt
'Omodel6Agb6k6Awelegb6 0p6,
child of
K9lawole,
child of Adetutu'
Yoruba full
names,
like
fingerprints,
are
unique
to each
person.
In the words
of
Johnson,
When the
oriko (name),
the oriki and the
orile
(totem)
are
given,
the individual is
distinctive,
the
family
is
known,
and he can at
any
time be
traced
(1969:87).
In a North American
context,
the
equivalent
is a social se-
curity
number
(Oyelaran
1976).
As is well
known,
a social
security
number is a
distinctive number that
is
uniquely assigned
to one
person,
and all vital infor-
mation about that person, including birth, health, education, profession, re-
sidence,
tax
history,
vehicle
ownership,
and so
on,
is documented
using
the
assigned
number. Access to a social
security
number
provides
access to the life
of
an individual.
In
modern times,
Yoruba
naming
has
changed, especially among
the edu-
cated. For
example,
it is common
to find
people
with
only
three names-a first
name,
a middle
name,
and a surname.
Surnames
are names of children's
fathers;
first and middle names are
usually personal
and
attributive names.
Children of Christians and
Moslems are
also
given
Christian- and Moslem-
based first or middle
names,
such
as those in table
4.
Table 4. Names in
Contemporary
Times
FIRST
Adidayo.
Oldwindd
Filisia
Kirnmg
MIDDLE
Alsbt
Tem
Olidr6nki
Te'wogbade
SURNAME
Akinloldi
Fisold
Addwildl
Aydndald
8/18/2019 Yoruba Names and Gender - Orie AL 2002
8/30
120 ANTHROPOLOGICAL
INGUISTICS 44
NO. 2
In
the
following
section,
I address the issue of name selection and show that
personal
names,
attributive
names,
and nicknames are chosen based on factors
such as the circumstances surrounding the birth of a child, the type of family a
child
belongs
to,
and the
hopes
and
aspirations
of
parents.
Totemic names
are
not considered because
they
are inherited.
3.1. The choice
of
personal
names.
The
choice of
a name
is
a
solemn
under-
taking
for
parents
and
grandparents
because
the
Yoruba
believe that one's
name can have a
psychological impact
on one's behavior
(Oduyoye
1972:67).
Hence,
much care is taken in the selection of names.
Name selection is driven
by
a number of
considerations. For
instance,
a
name
may
reflect the circumstances
surrounding
the birth of a
child.
To illus-
trate,
there are
special
names for twins and for children born after
them.
The
first-born
twin,
who is
thought
to be
younger
because he or she was
sent
ahead
to
explore
the world
by
the second-born
twin,
is
called
Tiydwo
(from
to
aye
wo)
'taste or
explore
the world'. The second-born
twin,
who is
thought
to be older
because he or she
waited
patiently
for the first-born twin to
explore
the
world,
is
called
Kehinde
'last to arrive'. A child born
immediately
after twins is
given
the
name idwdti
'child
born
after twins'
and the
next child is Alibd 'child born after
after Idwdi'.
Names
may
be
given
based
on the
profession, religion,
or
status of a
child's
family.
For
example,
a child born into a
family
of artists is
likely
to have
a name
that
begins
with
ona
'art';
a child born into a
family
of diviners will
have a
name
beginning
with
ifi
(god
of
divination);
and a child born into a
royal family
will have a name that starts with adk
'crown'.
It
is considered an
oddity
to find a
child from a family of hunters (ode) with a name beginning with ade 'crown'.
Hence,
the
saying
Ild
la di
wo
ki
t66
somo idirdko
One
must
look at a
family
(its
status,
profession,
and
religion)
before
giving
a child a
name'.
Furthermore,
a name
may
reflect the
aspirations
of
parents
for
their child-
ren. For
example,
if a woman has several
children
in
succession who die at
childbirth,
such children are known as Abikd
'one
who
is
born to
die',
a child who
wishes to travel back and forth between heaven and earth
(Bascom
1969:74).
This child
is
likely
to be
given
a name like
Muilmod
'do not
go
again', Dir6jaye
'wait and
enjoy
life',
or Kiiti 'one
who cannot die'. These names
show
that
parents
desire that their ibikad children
would not leave them.
3.2. Choice
of
attributive names. Like
personal
names,
attributive names
may also depict the circumstances surrounding the birth of a child. For instance,
if a woman has several male children in
succession who died
at
childbirth,
a
surviving
male child born
after that
experience
is
likely
to be
given
the name
Ajhni'fight
to
possess',
whereas a female child
whom the
parents
waited for a
long
time to conceive
may
be
given
the name
Abebi
'child who was
begged
to be
born'.
8/18/2019 Yoruba Names and Gender - Orie AL 2002
9/30
2002
OLANIKE LA
ORIE 121
Attributive names
may express
what the child
is,
as
dictated
by
the child's
orn 'fortune'
(Oyelaran 1976).11
For
example, Alike
is a child that
emerges
singularly (out of all the possible children that her parents might have had at
the time she was
born)
to be
pampered,
Adisd is a child
who is believed to have
supernatural
powers,
which would make it
impossible
for
any
evil force to attack
him,
and
Adigtin
is a child born believed to be
perfect
in
every
respect.
In
addition,
an attributive name
may depict
what it is
hoped
that a
child
will
become
(Johnson 1969:85).
For
instance,
Adafer
s a male child whose
parents
hope
everyone
will
long
to love him and Adake. s a female child whose
parents
hope
everyone
will
long
to
pamper
her.
Unlike
personal
names,
attributivenames do not reflectthe
unique
features
of the
family.
That
is,
it is
impossible
to retrieve
information
such
as
family
profession,
status,
or
religion
from an attributive name.
3.3.
Choice
of
nicknames.
Nicknames are selected based
on
character,
physical appearance, profession,
or achievement of the individual. Nicknames
are often
used
by
women
in
addressing
children who were born before
they
were
married into the
family (Oyelaran
1976:228).12
Although
a woman
may
refer to a
child born after her
marriage by
name,
she
cannot address those born before
by
name because
they
deserve
respect
on account of existential
precedence.
To
avoid a violation of this cultural
norm,
a woman must invent a nickname for
each senior child. We have
already
seen some
examples
in table
3.
Other
examples
are
Eleyinjidege
'beautiful
eyes',
Ejffwimi 'gapped
teeth
please
me',
Sabre'-dowd
'turn
a needle into
money
(a tailor)', Athrf-oj6orunn-o-riMn
'head
does not allow the sun
to shine
(big
head)',
and
Opele'fge-subd-ltwo-bwo-o-
fi-o-subdi-lodd-odd-fAyaa slim personfalls on a plate, a plate is not broken,she
falls
on a mortar and it
breaks
(a
fat
girl)'.
4. The structural
properties
of
given
names.
Although given
names-
ordiko,
oriki
Abiso,
and orfki
Alfei--are functionally
similar in some
respects,
their structural characterizations
are different.
According
to
Oduyoye
(1972),
the structural
composition
of oriko
'name' may
be as follows: two
nouns,
as in
table
5;
a sentence
comprising
a noun and
a verb
phrase (comprising
a verb
plus
noun and sometimes
additionally,
a
prepositional phrase),
as in table
6;
or a
verb
phrase (comprising
a verb
plus
noun and
verb),
as in table
7.
Table 5. Noun
plus
Noun Names
NOUN
olh
ife
akin
wdtir
NOUN
oldiwa
oldwa
old
old
OUTPUT
Oldoldwa
Ifeioldiwa
Akinold
Wdarold
GLOSS
'the
high
estate
of God'
'the love of God'
'the valor of
high
status'
'gold
of honor'
8/18/2019 Yoruba Names and Gender - Orie AL 2002
10/30
122 ANTHROPOLOGICAL
INGUISTICS
44
NO.
2
Table 6. Sentential Names
NouN
ad4
omo
oldi
old
VERBPHRASE
wuokl~
dele
sina
jumoke
OUTPUT
Adiwpld
Omoddild
Oldisi.n.
Qldjusm 6.k
GLOSS
'the crown enters the house'
'child arrives
(at home)'
'the
Lord
opens
the
way'
'wealth accumulates to
pamper
a child'
Table
7.
Predicate
(Verb Phrase)
Names
VERBPHRASE
Mit~nmi
K6red4
Gbo.ldhin
Kdldwold
GLOSS
'donot deceiveme'
'arrive with
good things'
'exhibit honor'
'bring
honor
into
the
house'
The names in
tables
5-7
reflect
four
unique properties.
First,
none of the names
is derived
through
affixation.
They
are
mainly composed
from lexical items.
Second,
they
do not reflect
any
prosodic
restriction. In other
words,
they
do not
have to be of a
particular phonological shape
to be well-formed. All that is re-
quired
is that
they comply
with the rules of
phrasal
or sentential
composition.13
Third,
these names do not have a fixed
tonal
pattern.
Fourth,
most of these
names are
gender
neutral;
that
is,
they
can be used
by
males or females.14
Like
oriko,
oriki
lizje
'nicknames'
do
not
display prosodic
restrictions,
they
do not have
fixed tonal
patterns,
and
they
are
mostly gender
neutral. Unlike
oriko, however,
orizki
&lkjei
may
be derived
through prefixation (Oyelaran
1976:
244) and concatenation of lexical words. Examples are shown in table 8.
Table 8. Structure of Nicknames
.Eldyinjd;-ge
owner.of.eyeballs-delicate
'beautiful
eyes'
Sabr.d-dowd
turn-a.needle-into.money
'a
tailor'
Eji-wit-mi
gapped.teeth-please-me
'one with
gapped-teeth'
O.peigd-subd-lhwo- wo--f6d--6uibd-lodd-od6-f-ya
one-who.is.slim-falls-on.a.plate-a.plate-does.not-
break-she-falls-on.a.mortar-a.mortar-breaks
'a fat
girl'
(noun
phrase)
(verb
phrase)
(simple
sentence)
(coordinate
sentence)
8/18/2019 Yoruba Names and Gender - Orie AL 2002
11/30
2002
OLANIKE
OLA
ORIE
123
A-pon-bdpore.
one-who.is.red-and.friendly.with.palm.oil
'a fair skinned person'
O-pdldiige
one-who.is.slim
'a
slim
person'
(prefixation)
(prefixation)
Finally,
the
properties
of oriki
abizso
'attributive names' differ from the
properties
of
ortiko
and
onri
Alifje
in four
ways.
First,
they
are derived
through
prefixation only,
specifically
a low-toned
A
prefix.
Second,
what
follows the
prefix
is
a
sequence
of verbs.
Third,
the names have fixed
tonal
patterns. Fourth, they
are
divided
along gender
lines.
Fifth,
verbs
denoting gender-based
semantic
themes
are
usually
chosen
in
deriving
these names.
For
instance,
the first verb
in a masculine name is a performative verb denoting notions such as bravery,
decisiveness,
unique
emergence,
and
praise;
the
second
verb,
which is resulta-
tive in
nature,
denotes
possession.
On
the other
hand,
verbs
denoting nurturing,
tenderness,
adulation, praise,
and
beauty
are chosen in
creating
feminine
names.
Examples
have been seen in
tables
1
and
2. Further
examples
of the two
tonal
patterns
are
given
in
tables
9
and 10.
Table 9.
Further
Examples
of Names with
LLH
Tone Pattern
MIASCULINE
A-ji-di
PREF-fight-restore
A-la-gbd
PREF-emerge-carry
A-kn-jif
PREF-meet-wake.up
FEMININE
A-w-ro6
PREF-bathe-adorn
beautifully)
A-bt-bi
PREF-beg-born
A-pe-ke
PREF-beckon-pamper
A-y -ki
PREF-rejoice-round.about
Table
10.
Further
Examples
of Names with LHH Tone Pattern
(Exclusively
Feminine)
A-bd-ki
PREF-join-pamper
A-tin-ke
PREF-repeat-pamper
A-rf-kie
PREF-see-pamper
A-ji-ke
PREF-wake.up-shine
8/18/2019 Yoruba Names and Gender - Orie AL 2002
12/30
124
ANTHROPOLOGICAL
INGUISTICS
44
NO.2
In
summary,
we see that attributive names are indeed
distinctive
with
respect
to their
phonological, syntactic,
and semantic characterization. These
unique properties converge n the verbsequencethat follows the prefixto differ-
entiate
gender.
In the next
section,
these
properties
are
explained
as
resulting
from an attributive
template
whose well-formedness is
regulated by morpho-
syntactic,
tonal, semantic,
and
prosodic
constraints.
5. Attributive names and
gender
differentiation, The
goal
of this
section
is to motivate the canonical
attributive
name
template.
As will
be
shown,
this
template comprises
a
prefix
nominalizer and a verb
phrase
that is
expressed
as
two
monosyllabic
serial verbs. Since it is
logically possible
to have
more than two
verbs in a serial verb
construction,
the verb
phrase
must
be
constrained to
exclude verb
sequences
involving
three
or more
syllables.
It is
proposed
that the
serial verb
phrase
must
obey
foot
binarity,
a
prosodic
constraint,
which
limits its
members to two
syllables
or moras
(McCarthy
and
Prince
1990).
In
addition,
given
the
gender-based
tonal and semantic
patterns
observed
in
section
4,
it is
proposed
that the verb
phrase
is also
subject
to tonal and
semantic constraints
(Oyetade
1991).
5.1.
Motivating
the
morphological
and
syntactic requirements.
Yoruba
is a
highly
prefixing
language (Oyelaran
1987;
Owolabi
1995).'5 Therefore,
new
words are derived from
existing
words
by attaching
a
prefix
to
roots, stems,
or a
given
syntactic
category.
For
example,
i
(here,
a concrete
nominalizer)
and
o/o516
(agentive nominalizer)
are
attached to roots that are verbs or
verb
phrases
containing
verbs and their
objects
or
adverbs.
(2a)
a-lo
PREF-go
'going'
(2b) A-t'egun
PREF-step-climb
'ladder'
(2c) A-rin-kiri
PREF-walk-about
'wandering'
(3a) o-kti
AG.NMZ-die
'corpse'
(3b)
o-le
AG.NMZ-lazy
'lazy person'
8/18/2019 Yoruba Names and Gender - Orie AL 2002
13/30
2002
OLANIKEOLA
RIE
125
(3c)
o-sere
o-se-ere
AG.NMZ-do-play
'performer'
(3d)
6-jogbon
o-je-ogbomn
AG.NMZ-eat-wisdom
'professor'
On
the
other
hand,
oni
(possessive
nominalizer)
is
always
attached to nouns or
noun
phrases,
as shown
in
(4a)-(4d).17
(4a)
oni'le
oni ile
POSSESSORhouse
'landlord'
(4b) onigbese
oni
igbese
POSSESSORebt
'debtor'
(4c)
elewon
on' ewon
POSSESSOR
prison
'prisoner'
(4d) ol6w6
oni ow6
POSSESSOR
money
'wealthy
person'
As mentioned
in section
4,
attributive
names are also derived
through
pre-
fixation. In order
to
derive
an attributive
name,
the
prefix
i is attached to a verb
phrase
composed
of a
sequence
of two
monosyllabic
verbs,
as
in
(5a)-(5d)
(drawn
from tables
9
and
10).
(5a) A-khn-jfi
PREF-meet-wake.up
'personwhom one meets and is awakenedby'
(5b) A-yo-k4
PREF-rejoice-round.about
'person
whom
one
rejoices
around'
8/18/2019 Yoruba Names and Gender - Orie AL 2002
14/30
126 ANTHROPOLOGICAL
INGUISTICS 44
NO.2
(5c) A-bd-ke.
PREF-join-pamper
'personwhom one rallies to pamper'
(5d) A-nr-ke
PREF-see-pamper
'person
that
one sees
and
pampers'
The
syntactic
restriction
that
the verb
phrase
of an attributive name must con-
sist of two verbs rules out verb
phrases
of other forms.
Thus,
verb
phrases
con-
sisting
of
verb-object
bases are not suitable for attributive
names,
as shown
by
(6a)-(6c),
nor are verb
phrases consisting
of verb-adverb
bases,
as
shown
by
(7).
(6a) *A-rdW
A-rd-iW
PREF-stand-house
'heir'
(6b)
*A-tenum6
a-te-enu-mo
PREF-press-mouth-attach
'emphasis'
(6c)
*A-cdbrire
A-de-il-bd-ire
PREF-arrive-house-meet-goodness
'meeting
fortune at home'
(7) *A-rin-kiri
PREF-walk-about
'wandering'
The
contrast between
permissible
and ill-formed
bases of
prefixation
is
explained
if we characterize the
morphosyntactic properties
of
attributive
names as in
figure
1.
NP
a
PREF
VP
vi Vj
Figure
1. Canonical
attributive name.
This
template
shows
that
the
nominalizing prefix
a&1s
elects a VP
that
consists
of a
serial verb
construction.19
Crucially,
there
must be
only
two verbs in this VP.
8/18/2019 Yoruba Names and Gender - Orie AL 2002
15/30
2002
OLANIKE
QLA
ORIE
127
By having
a
template
such as that in
figure
1,
forms such as
(5a)-(5d)
are
predicted
to be
well-formed,
whereas
cases such as
(6a)-(7)
are
ruled out.
5.2.
Motivating
the
tonal and semantic
requirements.
Although
the tem-
plate
in
figure
1
correctly
excludes forms
such as
(6a)-(7),
it
incorrectly
allows
cases such as those in table 11.
Table 11. Forms
Incorrectly
Predicted to
be Permissible
by Morphosyntactic
Template
DERIVED
NOUN
*A-t'e-g~n
PREF-step-climb
'ladder'
*A-f.-se
PREF-speak-come.to.pass
'incantation'
*A-pd-jo
PREF-complete-together
'reunion'
*A-n-yhn
PREF-own-select
'solicitude,
aspiration'
TONALATIERN
LLL
LLM
LHM
LHL
What the
examples
in
table
11
have
in
common s that the tonal
specifications
of
the verb phrase do not meet the required tonal specifications for the verb phrase
of an attributive name. That
is,
they
are
neither
LH
nor
HH.
Therefore,
these
forms are
disqualified.
Disqualifying
the
examples
in table 11 on the basis of tone
immediately
predicts
that forms such as
(8a)-(8d)
(with
LLH
pattern)
and
(9a)-(9d)
(with
LHH
pattern)
should be well-formed attributive
names,
but in
fact,
they
are not.
(8a)
*A-lk-yd
PREF-split-understand
'explanation'
(8b)
*A-kA-yd
PREF-read-understand
'comprehension'
(8c) *A-yhn-mo
PREF-select-attach
'destiny'
8/18/2019 Yoruba Names and Gender - Orie AL 2002
16/30
128
ANTHROPOLOGICAL
INGUISTICS 44
NO.2
(8d)
*A-fo-m6
PREF-jump-attach
'mistletoe, parasite'
(9a)
*A-nr-fin
PREF-see-scrutinize
'an insult'
(9b) *A-s-ri
PREF-open-see
'secret'
(9c)
*A-bi-ku
PREF-born-die
'one
who
dies
again
and
again'
(9d)
*A-wf-gbo
PREF-speak-listen
'obedience'
The
major
problem
with
(8a)-(9d)
is that the semantics of the
cooccurring
verbs
is
inappropriate
for an attributive name. As
indicated
in
section
4,
the verbs of
attributive names are
semantically
restricted. For
example,
the first
verb
in
a
masculine
name is
usually
an
action verb
denoting
semantic themes
such as
bravery
(jiA
fight')
or decisiveness
(kAn
meet
intentionally
or
purposefully',yAn
'choose');
the second verb is a
resultative
verb
implying possession
(ni
'to
pos-
sess',
gbe 'carry',
mu
'take',
bi'give
birth').
On the other
hand,
feminine names
have verbs reflecting semantic themes involving nurturing (ke 'pamper', be
'beg', we 'bathe').
Combining
the tonal
and semantic
restrictions, then,
we see that forms such
as those in table 11 and
(8a)-(9d)
cannot be
attributive names in
Yoruba. These
restrictions mean that the
following specifications
must be
added to the tem-
plate
as
given
in
figure
1:
*
A
masculine
name
VP has a
low-high
tonal
pattern;
the VP
begins
with
an
action verb
denoting
themes such as
bravery, decisiveness,
unique
emer-
gence,
and
praise;
it ends with a
resultative
verb
expressing possession.
*
A feminine
name VP
has
a
low-high
or
high-high
tonal
pattern;20
the VP
must contain verbs
reflecting
semantic themes
involving nurturing.
5.3.
Motivating
the
prosodic requirement.
So
far,
it has
been
shown that
attributive names are
subject
to various
restrictions-morphosyntactic,
tonal,
and semantic
constraints.
Recognizing
the
important
role of these
requirements
enables us to
understand
why only
certain verbs and tones are
licensed
in
the
verb
phrase
of
the base of
prefixation.
This section
shows that there is
yet
8/18/2019 Yoruba Names and Gender - Orie AL 2002
17/30
2002
OLANIKELA
ORIE
129
another
requirement,
a
prosodic
constraint,
which is needed to characterize
attributive names
properly.
As illustrated by the attributive names in (10a)-(10d), the base of pre-
fixationhas a verb
phrase
with two
monosyllabic
verbs.
(10a)
A-kitn-ji
PREF-meet-wake.up
'person
whom one meets and is awakened
by'
(10b) A-kin-fe
PREF-meet-love
'person
whom one meets and loves
(10c)
A-we-r6
PREF-bathe-adorn
'person
who is adorned
(beautifully)
after
being
bathed'
(10d)
A-du-ke
PREF-scramble-pamper
'person
whom one scrambles to
pamper'
Requiring
the VP of an attributive
name
template
to be a serial verb reflects the
strong serializing
tendency
of Yoruba.
As shown
by
(11a)-(lc),
it is
possible
to
have a
sequence
of two or more verbs that
are not
connected
by
an
overt
conjunction
within
a
single
Yoruba clause
(Bamgbose 1974;
Oyelaran
1982;
Awoyale
1988).
(11a) Ade ra isu je.
ade
buy yam
eat
'Ade
bought
(some)
yams
and ate them.'
(11b)
Ade
gb omo
sd
19.
ade
carry
child run
go
'Ade carried the child and ran
away.'
(11c)
Ade rd
aso
jf gbW w'o pdei oba.
ade saw
garment
steal
carry
wear meet
king
'Ade saw a
garment,
stole
it,
carried and wore it and met the
king
in it.'
Given the
existence of serial verb
expressions
such
as
these,
one would
expect
a
sequence
of
more
than two
verbs to be
possible
in an attributive
name.
However,
as shown
in
(12a)-(12d),
forms with three verbs are
unacceptable.
(12a)
*A)-kn-ji-fef
PREF-meet-wake.up-love
'person
whom one
meets,
whom one
is
awakened
by
and
whom one loves'
8/18/2019 Yoruba Names and Gender - Orie AL 2002
18/30
130
ANTHROPOLOGICAL
INGUISTICS 44
NO.2
(12b)
*Ai-kn-bi-fe.
PREF-meet-give.birth-love
'personwhom one meets, gives birth to and loves'
(12c)
*A-we-r-Te
PREF-bathe-adorn-pamper
'person
who
is
adorned
(beautifully)
after
being
bathed and is
pampered'
(12d)
*t--dri-ff.-ke.
PREF-scramble-love-pamper
'person
whom one scrambles to love and
pamper'
Moreover,
not all two-verb
sequences
are
acceptable
in
attributive names.
Examples
in
(13a)-(13d)
show that forms
containing
a
monosyllabic
verb
and a
disyllabic
verb
are
ungrammatical.
(13a) *A-pide-ii
PREF-meet-wake
p
'person
whom
one
meets and is awakened
by'
(13b)
*A-we.-ddro
PREF-bathe-stand
'person
who
stands
(beautifully)
after
being
bathed'
(13c) *A-jijhdd-ke
PREF-scramble-pamper
'person
whom one
scrambles
to
pamper'
(13d) *A-kin-fer.fn
PREF-meet-love
'person
whom
one
meets
and
is
awakened
by'
Interestingly, examples
(10a)-(10d)
and the
unacceptable (13a)-(13d)
are simi-
lar in several
respects. Semantically,
they
are identical.
Morphosyntactically,
they
are also
alike;
each name is derived
by prefixing
i
to
a
base
formed of two
verbs. With the
exception
of
(13d),
the
basic
LLH
tonal
melody
is satisfied
by
the
two sets of data.2' As can
be
seen, however,
(10a)-(10d)
are
well-formed but
(13a)-(13d)
are
unacceptable
as attributive
names.
What the forms in
(10a)-
(10d)
have in common is
that
each
verb in the
verb
phrase
is
monosyllabic.
In
contrast,
in
(13a)-(13d),
there is at
least one verb in the
sequence
of
verbs
that
has more than one syllable. That is, the VP has two syllables in (10a)-(10d),
whereas in
(12a)-(13d),
it
has more than two
syllables.
In
order to account for this
contrast,
I
propose
the constraints in
(15)
and
(16),
whereby
the VP of an attributive
name must be
exactly
two
syllables long
and the VP must be a
binary
foot
(see,
e.g., McCarthy
and Prince
1990, 1993;
Hewitt
1994).
8/18/2019 Yoruba Names and Gender - Orie AL 2002
19/30
2002
OLANIKEOLA
ORIE
131
(15)
Prosodic constraint
on Yoruba attributive name bases:
The base of
prefixation
for
deriving
an attributive name must be
a
binary
foot.
(16)
Foot
binarity:
A foot is
binary
at the
syllabic
or moraic level.
Given these
constraints,
forms such
as
those
in
(12a)-(13d)
are ill-formed
because foot
binarity
is not
respected
in the VP. In
*A-werdke, given
in
(12c),
and
*A-pde.-ji,
in
(13a),
the foot has three
syllables.
In
*A-jijhdu.-ke,
shown
in
(13c),
there are
four
syllables
within the
foot;
although
one could
group
these
four
syllables
into two feet in
conformity
with the demand
of
foot
binarity,
such
a form is still
unacceptable
because the base
of an attributive name
requires
one
foot,
not two feet.
Diminutive
reduplication provides
evidence that the
VP of an attributive
name
is a
binary
foot. As demonstrated in table
12,
diminutives
are formed
by
reduplicating a name,
which is shortened
to a
binary
foot
(Orie 1997:146-47).
Table
12. Foot-based
Process: Diminutive
Reduplication
FULLNAME DISYLLABIC
ORM
DIMINUTIVE UNATTESTED
AkinolA
Akin or
Old
Akzinakin r
OldolA *AkinoldakinolA
WarAohild
Wdrbor
Old
Wdrdiwurbr
Olioll *Wdridlidwuraoll
Addewole
Ade or
Wold
Acdad& r
Wdlwwol
*Adwoladewole
Oltidsin
Old or
SinA
Oldolu or
SindsinA *hisintdolusinh
One can observe
that,
in addition to
reduplication,
the
tonal
pattern
of the base
is
displaced
by
a
high-high-mid-low
(HHML)
pattern,
the tone
of the diminutive.
Thus,
foot
reduplication
and the
HHML tonal
specification
are distinctive
pro-
perties of the diminutive.
As
the
examples
in table
13 demonstrate,
attributive names
can also
be
turned
into diminutives
by reduplicating
the last two
syllables.22
Table 13.
Foot-based Process:
Diminutive
Reduplication
of Attributive
Names
BASE DIMINUTIVE
ORM
UNATTESTED
A-kin-jif A-kin-jif-kan-ji *A-kin-ji-akan-ji
A-y-ki A-y-kd-yo-ka *A-y6-kdi-ayokA
A-bd-ki A-bd-kd-ba-ke
*A-bd-kd--abd-ke
A-r6-ke
A-r-ke-ri-ke.
*A-r6-ke-ari-ke.
The
reduplication
pattern
in
table
13
provides strong
evidence
that the VP of an
attributive name is a binary foot. As we can see, the reduplicants of unaccept-
able
forms
have morethan two
syllables
in violation
of
foot
binarity
and are thus
predicted
to be
ungrammatical.
Incorporating
the
prosodic requirement
into
the set
of constraints
already
established,
the
final
set of
requirements,
which
derive
attributivenames are
as
follows:
8/18/2019 Yoruba Names and Gender - Orie AL 2002
20/30
132
ANTHROPOLOGICAL
INGUISTICS
44
NO. 2
*
The
morphosyntactic
structure
is
[NPa[VPViVj]]
(cf.
figure
1).
*
The VP
is a serial verb and a
binary
foot.
*
The VP of a masculine name
has a
low-high
tonal
pattern.
The first verb
in
the VP is an action verb
denoting
themes
such
as
bravery,
decisiveness,
unique
emergence,
and
praise;
the
second
verb,
a resultative
verb,
denotes
possession.
*
The VP of a feminine name has a
low-high
or
high-high
tonal
pattern
and
has verbs
reflecting
semantic themes
involving nurturing
(pampering,
pleading, bathing).
The
characteristics of this
template
show that the VP
base
is the
crucial con-
stituent
responsible
for
gender marking.
As shown
in
sections
5.1-5.3,
there are
other VPs that resemble the ones in the
template
in
some
ways,
but
are
ungram-
matical as attributive
names because
they
do
not
exhibit all the
necessary
pro-
perties.
Finally,
the
analysis presented
here
necessitates a
rethinking
of
one funda-
mental notion within
prosodic
theory,
that the domain for
prosodic processes
should be
prosodic
or
morphoprosodic
ather than
morphosyntactic
e.g.,
Selkirk
1986;
Inkelas
1990).
While it
is true that
many prosodic
processes apply
within
prosodic
and
morphoprosodic
domains
in
Yoruba
(Orie 1997),
I have shown that
attributive names at least demonstrate
that a
morphosyntactic
constituent
may
define the domain of a
prosodic process.
For
instance,
in a
purely prosodic
or
morphoprosodic
account,
the
two-syllable
size limit on
the base
of
prefixation
would be
explained
as
resulting
from foot
binarity.
However,
foot
binarity
allows
any
sequence
of
two
syllables
to be a valid
base.
It
cannot
explain why only
verb-
verb sequences are possible base forms and why verb-object sequences are ex-
cluded. As
shown,
an
adequate
account
of attributive names
must
make refer-
ence to both
morphosyntactic
and
prosodic
domains.
5.4.
Comparison
with a
previous
analysis.
Having
established that a
canonical attributive name
template
must include
morphosyntactic, semantic,
tonal,
and
prosodic information,
I
consider
an alternative
proposal.
In
Oyetade's
(1991) account,
attributive
names
are
derived
through prefixation
of
A
to two
verbs,
and the
observed
gender
differences result from the tonal
patterns
and
from semantic
restrictions
on verb selection.
There are
two
major problems
with
this
proposal. First,
in
accounting
for the
morphological
derivation of
attributive
names,
Oyetade's
account
simply stipu-
lates that the prefix h is normally attached to two monosyllabic verbs. No moti-
vation or
explanation
is
provided
for this
stipulation.
As shown in
section
5.1,
it
is
possible
to
derive nouns
by attaching
this
prefix
to verb
phrases
of various
different forms:
verb-verb,
verb-object,
verb-object-verb, erb-adverb,
and so on.
Since
attributive names
cannot select all of these
plausible
bases,
a
principled
account must
explain why
the attested base is
well-formed and
why
unattested
8/18/2019 Yoruba Names and Gender - Orie AL 2002
21/30
2002
QLANIKE LA
ORIE
133
bases are
unacceptable.
The
existing
account
does
not
mention these
possibili-
ties,
and it does not
provide any principled proposal
for
distinguishing
attested
and unattested cases.
Second,
Oyetade's
account
gives
no
explanation
for
why
forms such as
*A-p~ie-ji
and
*A-jijjdui-k,
which meet the tonal and
semantic
specifications
of attributive
names,
are ill-formed. One
might
try
to
argue
that these forms are
unacceptable
because
they
do not contain two
monosyllabic
verbs,
but this
argu-
ment is
flawed,
since it
provides
no means of
accounting
for
why
two mono-
syllabic
forms
are
necessary.
6.
Asymmetries
in
the
function of
gender
tonal
patterns.
In this
section,
a final
unique property
of attributive names is considered. As noted in
Oyetade
(1991),
the
LLH
melody may
be used to create both masculine and feminine
names,
whereas the LHH
melody
is used
exclusively
to derive feminine names.
Thus, in terms of frequency, there are more LLH-based names than L;HH-based
names.
Interestingly,
there are
ILH-based
names that
may
be
used
by
males or
females,
as shown in table
14.23
Table 14. LLH-based Names: Masculine or Feminine
MASCULINE FEMININE
MASCULINE
RFEMININE
A-kiAn-mi
A-li~--ki
A-d-b
PREF-meet-take
PREF-emerge-pamper
PREF-scramble-love
A-mo-b
A-mo-pi A-kin-ke
PREF-know.before-born
PREF-know-be.complete PREF-meet-pamper
A-kAn-bz
A-bb-kk
A-ko-ki
PREF-meet-born
PREF-beg-pamper
PREF-meet-greet
A-yin-dd
A-yin-ke
A-m -ri
PREF-praise-arrive PREF-praise-pamper
PREF-know.before-see
A-jt-nt
A-we-r6
A-y o-fe
PREF-fight-possess
PREF-bathe-adorn
PREF-rejoice-love
A-di-grin A-bg-b
A-beg-fe
PREF-wrap-be.perfect PREF-beg-born
PREF-beg-love
Table 15.
LHH-based Names
(Exclusively Feminine)
A-gbe-ke
PREF-carry-pamper
A-to--ki
PREF-nurture-pamper
A-ji-ke
PREF-wake.up-pamper
A-nf-ke
PREF-possess-pamper
8/18/2019 Yoruba Names and Gender - Orie AL 2002
22/30
134
ANTHROPOLOGICAL
INGUISTICS 44
NO. 2
One
may ask,
is this
asymmetry
in
frequency
an accidental
gap,
or does
it
follow from
any
principle
of
grammar?
Let us
first
of all
clarify
that
it would
not
be reasonable to dismiss the observed pattern as an accidental gap. As noted by
Oyetade,
one of the tonal
patterns
discovered
to be dominant
in
oriki
Abiso
[attributive
names]
is the
LLH
pattern.
It is
by
far the
commonest
. . .
the
pattern
is used for both male and female
(1991:58).
He observes
further,
the
second tonal
pattern
observed is the
LHH
... names with this
pattern
are
fewer
than those of
LLH
pattern.
This
pattern
is for
female
only.
I
have not found a
single
male oriki Abiso with the LHH
pattern
(1991:59).
As can
be deduced
from
these
observations,
there is a
systematic
robust
pattern
here.
Dismissing
it
as an accident amounts to
claiming
that the
pattern
is
insignificant,
which
is
clearly
not desirable.
Alternatively,
the
asymmetry may
be
explained
as
following
from
marked-
ness. In
languages
such as
English
and
French,
masculine
pronouns
are some-
times used as universal
signifiers
that
may
have
masculine or feminine refer-
ents. For
example,
in
French,
elles
'they'
is
used
when
all
referents are
feminine.
However,
if one of the referents
is
masculine,
ils
'they'
is
adopted.
Since
the
masculine form can be used
for reference to
both
sexes,
it is
considered to be
the
unmarked version
(Baker 1992).
On
the other
hand,
the feminine
form
can
only
have a feminine reference and
is thus treated as
the marked form.
By
analogous reasoning,
the
asymmetry
in
attributive
name tonal
frequency
can
be
explained
as
resulting
from
markedness distinctions. The
observation
that
the
LL;H
pattern
is more common than
the LHH
pattern
because
both
male
and female
names are formed from it is
explained
if the
LLH
pattern
is
the
unmarked form-the
universal
generic
form
that
may
have both
masculine and
feminine referents. The LHH pattern is less common because it is the marked
form,
the form
reserved
exclusively
for
creating
feminine
praise
names.
Follow-
ing
Baker
(1992),
this shows that
the use of the feminine tonal
pattern provides
more
specific
information
and rules
out
the
possibility
of
masculine
reference,
whereas the use of
the masculine tonal
pattern
does not
exclude the
possibility
of feminine
reference.
7. Factors
governing
the selection of
feminine names.
In
closing,
I
address a
sociolinguistic question.
Given that female names can
be
derived from
the
male-based
LLH
tonal
pattern
and the
exclusively
female-based LHH
tonal
pattern,
on what
basis do
parents
choose names for
girls
from
one set or
the
other? In
addressing
this
question,
I
interviewed
ten Yoruba
adults who
are
parents
of
girls.24
The
result
of
my
research
reveals that three
major
factors are
crucial for the
selection
of
female
attributive
names--frequency
and
popularity,
femininity
and
aesthetics,
and semantics. Of
all these
factors,
semantics
is
the
most
prominent
because
consultants
consider it
crucial
even
when
other
factors
seem
to
play
a
role. Each factor
is considered in
turn below.
8/18/2019 Yoruba Names and Gender - Orie AL 2002
23/30
2002
OLANIIKELA
ORIE
135
7.1.
Frequency
and
popularity.
Of all the ten consultants
interviewed,
only
one considers
frequency
and
popularity
to be a crucial factor in
selecting girls'
names. As pointed out by this consultant, who has two daughters with LLH
names,
the attributive names of these
girls
(Ab'en
and
Awer6)
have
LLH
tones
because that is the more common and
popular
tonal
pattern.
This view is in
harmony
with the observation noted in sections
4-6
that LLH names have a
higher frequency
of
occurrence
than LHH names because both males and fe-
males can use
LLH
names.
Intriguingly,
this consultant
equates frequency
with
popularity
and notes that
popular
names are more
appealing
than
unpopular
ones.
However,
he observes that
meaning
is
also essential: his first
daughter's
name is Abeni 'child
whom one
pleaded
or
begged
to
possess'
because
they
had
several
miscarriages
before she was
born;
his second
daughter's
name is
Aw.ero
'child who is bathed and adorned
beautifully'
because she is beautiful.
7.2.
Femininity
and
aesthetics.
The second
factor,
which was
suggested
by
another Yoruba
speaker
who has two
daughters
with LHH
names,
is
femininity
and aesthetics.
According
to this
speaker,
the names
Agbeke.
and
Abdke.
were
chosen for her
daughters
because
they
sound
more feminine
and attractive than
names such as
Aduke.
or
Aweke'.
In
essence,
this
speaker
views LHH names as
displaying femininity
and
aesthetics,
factors that
may explain
why
LHH names
are
exclusively
feminine.
Again,
the semantics of attributive
names was considered
important:
Agbeke
was chosen because
her desire is that the child be
carried and nurtured
by
all
and
sundry; Abike.
was selected because she felt
that her
daughter belongs
to
the
community
and her
nurturing
should
be
community-based.
7.3. Semantics.
The third and final factor
responsible
for the selection of
female names
is semantics. As
pointed
out
by eight
consultants,
the
meaning
of
an attributive
name
is more
important
than the
tonal
patterns.
These
speakers
note that a feminine name
may
be
LLH or LHH as
long
as it
explains
the
cir-
cumstances
surrounding
the birth of a child
or
conveys
the wishes of the
giver
of
the name. For these
consultants,
factors
such as
frequency,
femininity,
and
aesthetics are not crucial.
7.4.
Summary.
On
the basis
of these
findings,
we see
that the semantics of
attributive
names is central to the selection
of
girls'
names. Even
when other
factors are
involved,
semantics
is
important.
Table 16 summarizes
the identified
patterns based on the three basic factors observed--frequency and popularity,
femininity
and
aesthetics,
and semantics.
Recall that even those
parents
who
mentioned factors other
than
semantics as
important
also rated semantics
as
essential
(sections 7.1-7.2).
8/18/2019 Yoruba Names and Gender - Orie AL 2002
24/30
136
ANTHROPOLOGICAL
INGUISTICS
44
NO. 2
Table
16.
Summary
of
Factors
Responsible
for the Selection of
Girls'
Attributive Names
FACTOR
Frequency
and
popularity
Femininity
and aesthetics
Semantics
PERCENTAGEFPARENTSWHO
RATED HAT ACTOR S
CRUCIAL
10%
10%
100%
Finally,
I
should
point
out that I discovered two
additional
groups
of
parents
recently
whose
daughters'
attributive names were selected
by
their
parents
or
grandparents.
Whereas the
parents
in the first
group
do not know
why
a
parti-
cular name
was
chosen,
the
parents
in the second
group
do;
their
daughters
in-
herited
either
their mother or
grandmother's
names.
Consequently,
for
these
parents,
none of
the
factors
in
sections
7.1-7.3
is
relevant. The
response
of
these
two groups is an indicator that the factors described in this section are not the
only
ones that
affect name choice. Further research
may
show
that still other
factors
play
a
role
in
the
selection
of
female names.
8. Conclusion. In
summary,
it has been shown that
gender-based
names
(oriki
abiso)
are
unique
in
several
respects.
Morphologically, they
are formed
through
prefixation
to a
verb
phrase. Syntactically,
the verb
phrase
must
consist
of a serial verb
construction.
Semantically,
the first verb in
a
masculine name
illustrates themes such as
bravery, decisiveness,
unique emergence,
and
so
on;
the
second verb denotes
possession.
Feminine
names, too,
have
semantic
restric-
tions on
verbs--only
verbs
reflecting
nurturing
(pampering, pleading, bathing,
and so
forth)
are selected.
Phonologically,
two restrictions
apply. First, the tonal
melody
of
a
masculine name is
LILH,
whereas feminine names
may
have LLH
or
LHH tonal
melodies.
Second,
the verbal base of
prefixation
must be a
binary
foot,
expressed
as two
syllables.
To
account for these
properties,
I
proposed
a
canonical
attributive name
template,
which
incorporates
the
morphosyntactic,
semantic,
tonal and
prosodic
requirements
mentioned
above. The
advantage
of
this
analysis
over the
alternative account is that it
explains
the attested
pat-
terns and
rules
out unattested forms.
As
regards
the
asymmetry
in
the
use of masculine and
feminine tonal
pat-
terns,
it is
suggested
that this results
from
markedness
distinctions.
In
particu-
lar,
the
proposal
is that the
masculine tonal
pattern
is
the universal
generic
unmarked
form; hence,
it
may
be used
of both
male and female
referents. In
contrast, the feminine tonal pattern is reserved exclusively for creating female
attributive
names because it is the
marked form.
Finally,
the
flexibility
of
name
selection for
girls
is shown to
result from a
range
of
contributing factors,
including
frequency,
femininity, aesthetics,
and
semantics.
Frequency
is a
factor,
since some LLH names
are selected
based
on
the
fact that the LLH
tonal
pattern
has a
higher frequency
than the LHH
tonal
pattern.
Femininity
and
aesthetics are
factors,
because
some LHH
names
are
8/18/2019 Yoruba Names and Gender - Orie AL 2002
25/30
2002
OLANIKE
OLA
ORIE
137
viewed
as more
feminine-sounding
and attractive than the
LLH
pattern.
Se-
mantics is a crucial
factor,
because
meaning,
not tonal
pattern,
is
crucial
to
some
parents in the selection of girls' names. Research shows that semantics is the
most
important
factor
because
the
meaning
of an
attributive name is considered
crucial
even
when other factors
are
involved.
Appendix
1: Additional
Examples
of LLH Attributive Names
MASCULINE
A-lk-de
PREF-emerge-arrive
A-m-bi
PREF-know-born
A-kin-bi
PREF-meet-born
A-yin-dd
PREF-praise-arrive
A-ja-ni
PREF-fight-restore
A-jh-nz
PREF-fight-possess
A-dd-fe.
PREF-scramble-love
A-la-mu
PREF-split-take
A-kan-mu
PREF-meet-take
A-yAn-fer
PREF-choose-love
A-lA-nt
PREF-emerge-possess
A-kin-nz
PREF-meet-own
A-si-ma
PREF-select-take
A-kin-dde
PREF-meet-arrive
A-ttn-d&i
PREF-lure-create
A-yhn-dd
PREF-choose-create
A-1a-b
PREF-emerge-born
A-di-gdin
PREF-wrap-be.perfect
FEMININE
A-la-ke
PREF-emerge-pamper
A-mo-pi
PREF-know-be.complete
A-be-ki
PREF-beg-pamper
A-yin-kse
PREF-praise-pamper
A-we-ro
PREF-bathe-adorn
A-pe-ki
PREF-beckon-pamper
A-be-bt
PREF-beg-born
A-b~-gbd
PREF-beg-carry
A-sun-kle
PREF-sleep-pamper
A-sh-bt
PREF-select-born
A-mo-kc
PREF-know-pamper
A-w~-ke
PREF-bathe-pamper
A-yo-nt
PREF-rejoice-own
A-y-ni
PREF-rejoice-pamper
A-be-nr
PREF-beg-own
A-si-nz
PREF-select-own
A-shA-k
PREF-select-pamper
A-b~-je
PREF-beg-answer
MASCULINE R
FEMININE
A-di-bi
PREF-scramble-born
A-kan-kle
PREF-meet-pamper
A-ko--ki
PREF-meet-greet
A-m-nri
PREF-know-see
A-yo-fe
PREF-rejoice-love
A-b~-fe
PREF-beg-love
A-pE-fe.
PREF-beckon-love
A-mofe
PREF-know-love
A-sPF-fee
PREF-select-love
8/18/2019 Yoruba Names and Gender - Orie AL 2002
26/30
138
ANTHROPOLOGICAL
INGUISTICS
44
NO.2
A-khn-fe.
PREF-meet-love
A-yin-ld
PREF-praise-lick
A-m-6*
PREF-know-fall.(in.
we)
A-kin-6*
PREF-meet-fall.(in.awe)
A-ji-6*
PREF-fight-fall.(in.awe)
A-yhn-o*
PREF-choose-fall.(in.awe)
A-du-k?.
PREF-scramble-love
A-din-ni
PREF-sweet-possess
A-begf*
PREF-beg-fall.(in. awe)
NOTE: The
meaning
of verbs with
deleted
consonants
is unclear.
Oyetade
proposes
that
the deleted consonant in
&mid
and &beo
s
w. If this
assumption
is
true,
then,
the
compositionofAmoois PREF-know-fallndthe compositionof Ab~o s PREF-beg-fall.
Appendix
2: Additional
Examples
of LHH Attributive Names
(Exclusively
Feminine)
A-bt-ke'
PREF-born-pamper
A-nr-kle
PREF-see-pamper
A-nt-kei
PREF-own-pamper
A-ji-ke
PREF-wake.up-pamper
A-pin-~ke
PREF-take.turns-pamper
A-tin-k4
PREF-repeat-pamper
A-tuin-n'
PREF-repeat-pamper
A-fi-n
PREF-wake.up-own
A-jf-le~
PREF-wake.up-shine
A-gbe-kz
PREF-carry-pamper
A-pe-k~
PREF-gather-pamper
A-bd-lk1
PREF-join-pamper
A-t6-ki
PREF-nurture-pamper
A-t6-ni
PREF-nurture-own
A-t6-l1
PREF-nurture-shine
Notes
Acknowledgments.
I
would like to thank Akin
Akinlabi,
Victoria
Bricker,
Laura
Downing,
and two
anonymous
reviewers
for
valuable
comments,
which
improved
both
the
substance and the
presentation
of this article. Thanks also
to
James Welch for
help
with
the
map. Special
thanks to
Olasope
Oyelaran,
who first showed me that
my
own
language
is
interesting
and whose work on
orinti
and Yoruba
morphology
s
a rich re-
source. I am
grateful
to him for
many
fruitful discussions about Yoruba and about this
topic
in
particular.
8/18/2019 Yoruba Names and Gender - Orie AL 2002
27/30
2002
OLANIKEOLA
ORIE
139
Abbreviations.The
following
abbreviationsare used:
AG.NMZ
agentive nominalizer,
NP
=
noun
phrase;
VP
=
verb
phrase;
V
=
verb;
PREF
=
prefix.
Transcription.
The
examples
in
this article are
given
in Standard Yoruba ortho-
graphy.
In Yoruba
orthography,
e
=
[e];
o
=
[o];
Vn
=
nasalized
vowel;
s
=
[U];
p
=
[kp];
an
acute
accent
[']
=
high
tone
(H);
a
grave
accent
[']
= low
tone
(L);
no accent
=
mid
tone
(M);
a
wedge
[']
=
rising
tone;
a tone-marked nasal
=
syllabic
nasal.
1.
Oyetade
(1991)
characterized attributive names
as
personal praise
names. Given
that
these
praise
names do not
always praise
an individual
(Oyelaran
1976),
I use the
terminology
attributivenames
instead
(Johnson 1969).
2.
Oyetade
(1991:59)
observes that there is one attributive
name,
Ajpk4,
which has a
LMH
pattern.
3.
Masculine names with the
LHH
pattern
are
extremely
rare,
but do exist as
exceptions
to the rule on feminine names. For
example,
while I was
conducting
this
research on this
topic,
some Yoruba
speakers pointed
out the
exception
Aijiz7i
one whom
one wakes
up
to salute.'
They
note,
however,
that this form is
exclusively
used for the
Creator,
who is sometimes referred to as 'mother' in the
expression
ablyamp,
aboji
gb~or gbo.ro 'mother, one who has a cloth sash, which is indefinitely long'. Qlasope
Oyelaran
(p.c.
2001)
also notes another
counterexample-a
nickname with a LHH
pattern,
Arisi 'fearful
person',
which was used in
addressing
a
man
in
a
poem.
4.
Many
thanks to Victoria Bricker for
raising
this
question.
5.
The Yoruba
of
Nigeria
are found
in southwestern
Nigeria,
especially
in the follow-
ing
states:
.yo, Ogun,
Ondo,
Osun,
Ekiti, Kwara,
Lagos,
the western local
government
of
Kogi,
and
Edo.
6.
According
to Johnstone
(1993),
there are
18,850,000 speakers
in
Nigeria
and
465,000
speakers
in Benin. Yoruba
speakers
in
Togo
are estimated at about
250,000.
In
addition,
there are several thousand Yoruba n
diaspora
in
Asia,
Europe,
the
Americas,
and Australia.
7.
Crozier and
Blench
(1992:111),
based
on classifications motivated in
Akinkugbe
(1976)
and
Capo
(1989),
group
these dialects as follows: Central Yoruba
(Ife,
Ijesa,
Ekiti),
Northwest
(Oyo,
Qsun,
Egba),
Northeast
(Yagba,
Gbede,
Ijumu),
Southwest
(Shabe and Ketu spoken in Benin and adjacent border areas of Kwara and Ogun
States),
Southeast
(Ondo, Qwo, Ikale,
Ilaje).
8. Because there
are several
resemblances between the
.y9
dialect and Standard
Yoruba,
it is
commonly
assumed that the latter
originated
from that
dialect. But
as
Bamgbose
(1986)
demonstrates,
there is evidence
that Standard Yoruba has
properties
found in other
dialects,
such
as
Ijesa
and
Ekiti.
Bamgbose's
view is in
harmony
with
Fagborun's
(1994)
account of the
development
of Standard Yoruba.
According
to
Fagborun,
Standard Yoruba
developed among
speakers
of the
Oyo
dialect,
who
mi-
grated
from the northwest
region
to the south under
pressure
of Fulani
expansion
in the
early
nineteenth
century.
These
immigrants
settled in
la