Endangered Language Alliance





Garifuna

an Afro-Indigenous Arawakan language, spoken in Honduras, Belize, Nicaragua, and Guatemala


Background

The Garifuna language is spoken primarily in Honduras and Belize, with smaller numbers of speakers in Nicaragua and Guatemala. A large population now lives in the United States as well, particularly New York City. The most reliable estimate for the number of speakers may be that of Grinevald (2007: 69, 71), who cites 22,000 for Honduras and 12,000 for Belize, although the Ethnologue reports higher numbers and there are several hundred thousand ethnic Garifuna. The most widely accepted account of the origins of the Garifuna people is that they are largely descended from West Africans who were transported to South America as slaves but escaped due to a fortuitous shipwreck off the island of St. Vincent, in the Lesser Antilles. Having arrived at St. Vincent, they intermarried with a local indigenous Arawak tribe, adopting many elements from their culture such as the cultivation of cassava and its related technology, as well as their singing styles and language.

Affiliation

Garifuna belongs to the Ta-Maipurean (or Caribbean) branch of the Arawakan (or Maipurean) language family, the largest indigenous language family in South America in terms of number of languages (59 according to the Ethnologue). Although increasingly threatened, Garifuna remains one of the most widely spoken Arawaknan language. It can be considered a “fusion language” which has an Arawakn base but also contains a signficant portion of African and European elements.

Endangerment

Regarding the Arawak family as a whole, Aikhenvald (1999:72) states:

“The overwhelming majority of Arawak languages are now endangered. Even in the few communities with over 1,000 speakers, a national language (Portuguese or Spanish) or a local lingua franca (Língua Geral Amazônica, Quechua or Tucano) is gradually gaining ground among younger people. A massive switch to Lingua Geral Amazônica took place around 1900 in the region of the Rio Negro, and resulted in the rapid loss of a number of languages.”

There is wide agreement that a great number of ethnic Garifuna have no command of the language and that the transmission of Garifuna to the younger generation in the majority of their areas has undergone rapid decline.

The downward trend of the last several decades is clear; while Garifuna was spoken in the coastal areas from as far as Nicaragua to Belize, it has now been almost completely replaced by Spanish and English in all of Nicaragua (Davidson 1980) and most areas in Belize (Bonner 2001) and Guatemala. There remain several strongholds in Honduras, where the Garifuna population is most dense, but even here a rapid shift towards Spanish is observed among the younger generations in the larger towns. The trend towards obsolescence in this area appears even more dire when one considers the planned development of Honduras’s Caribbean coast for tourism under the new government.

Academic Work

Aikhenvald (1999) notes that several decades later, the pioneering work of Taylor on the phylogenetic relation between Garifuna, Lokono, Guajiro (Wayuu) has not been surpassed. Further historical comparative work on the Caribbean subgroup has unfortunately not been forthcoming.

Garifuna represents a rare example of a language which shows no signs of creolization and or substrate effects (from African languages) despite having been originally acquired via peer-to-peer learning rather than by canonical intergenerational transmission. We can be relatively certain that peer-to-peer learning played a large role based on recent DNA studies which have shown surprisingly little genetic input from Amerindian populations to the Garifuna (Crawford et al 1981).

Also within the domain of contact phenomena, Garifuna is unique in having developed male and female registers based on two different languages, from the Carib and Arawak families, respectively (Aikhenvald 1999:74-75). Several hypothesis have been forwarded to account for this but the full range of alternations marked as men’s and women’s speech has not yet been documented. With a better understanding of comparative Carib and Arawakan morphology in recent years (see Gildea 1998, Aikhenvald 1999, Romero- Figeroa 2000, Courtz 2008), we are now in a better position to evaluate the Garifuna facts regarding the male and female registers, something which has not been explored beyond Taylor & Hoff 1980.

Language Structure

Phonology

The speech sounds of Garifuna are relatively simple, as seen in the phoneme inventory below. The only unusual sound is the nasalized palatal glide /j̃/, but this has been alternatively analyzed as a regular oral glide /j/ preceding /n/. The sound /p/ in the modern language is only found in borrowings. The /p/ of Island Carib has become /f/ in Garifuna, as can be seen in such pairs as…

Other changes include k → g (with some exceptions).

Labial Alveolar Palatal Velar Glottal
Nasals m n j̃ [ñ/ny]
Voiced stops b d g
Unvoiced stops (p) t k
Unvoiced fricatives f s h
Unvoiced affricate tʃ [ch]
Glides w j [y]
Liquids l r

Morphosyntax

The basic vocabulary and pronominal morphology of Garifuna clearly shows its connection with the Arawakan language family:

Features Garifuna Lokono
1sg n- da-
2sg b- by-
3sg.fem t- thy-
3sg.msc l- ly-
1pl wa- wa-
2pl h- hy
3pl ha- na-

Similarities in basic vocabulary and function words can be seen in the comparison below, where both words are clearly cognate. Interestingly, the order of determiner (“the”) and noun (“woman”) are opposite in Lokono.

Lokono
to
DEF.FEM
hiaro
woman
‘the woman’
Garifuna
hiñaro
woman
to
DEF.FEM
‘the woman’

The basic word order of a Garifuna transitive clause is Verb-Subject-Object. If one of the arguments of a verbal clause is a personal pronoun, as in the second example below, it shows up as agreement on the verb or auxiliary rather than in the position of regular noun phrase arguments.

Hou-tu
eat-3S.F.AOR
Maria
Maria
ereba
cassava_bread
‘Maria ate the cassava bread.’
Ariha-tina
see-1S.AOR
gasigamu
armadillo
‘I saw an armadillo.’

Another interesting feature of Garifuna are its prepositions which agree with their complements. In the first example below, the preposition -un ‘to’ takes the third person singular masculine prefix l- agreeing with the features of “Juan”. In the second example, the same preposition agrees with the first person singular. Just as with verbal predicates, agreement on prepositions does not co-occur with pronouns. Including the pronoun after the preposition in this example (i.e. *n-un nuguya 1SG-to 1SG) would thus be ungrammatical.

Ru-tu
give-3S.F.AOR
Maria
Maria
seinsu
money
l-un
3S.M-to
Juan
Juan
‘Maria gave money to Juan.’
Ru-tu
give-3S.F.AOR
Maria
Maria
seinsu
money
n-un
1S-to
‘Maria gave me money.’

Garifuna has a very complex system of auxiliaries, each one governing its own agreement pattern. Their use is determined by the aspect, tense, mood and transitivity of a clause. Some of these auxiliaries have cognates in related languages, such as Lokono, but few other Arawak languages seem to reflect a similar level of complexity in this area.

Lokono
Li
he
fary-fa
kill-FUT
aba
one
kabadaro.
jaguar
‘He will kill a jaguar.’ (Pet 2011:25)
Garifuna
L-afaru
3SM-kill
ba
FUT
aba
one
geigusi
jaguar
‘He will kill a jaguar.’

Notice in the following examples the relation between Garifuna tuma and Lokono oma. Unlike the Lokono, the Garifuna word that expresses “with” takes an agreement prefix that reflects the feminine singular features of the jaguar. Also note that Lokono “with” follows its complement (jaguar) while Garifuna “with” precedes it.

Lokono
Li
he
fara-fa
fight-FUT
to
the
kabadaro
jaguar
oma.
with
‘He will kill fight with the jaguar.’ (Pet 2011:27)
Garifuna
L-afaru
3SM-kill
ba
FUT
t-uma
3SF-with
geigusi
jaguar
to
3SF.DEF
‘He will fight with the jaguar.’

The following two sentences exemplify many of Garifuna’s fascinating grammatical properties.

Ru
give
b-umu-ti
2S-TRAN.AOR-3SM
l-un
 3SM-to
Juan
Juan
‘You gave it to Juan.’
Ka=ba
what=FUT
un
to
b-ichigu-ni?
2S-give-3SM
‘To whom will you give it?’

Note the following:

  • In the first sentence we see the transitive auxiliary umu inflected for a subject (second person singular, b-) and an object (third singular masculine -ti).
  • The recipient is expressed with the prepositional phrase lun Juan. In the second sentence, a question, we find numerous structural differences. First of all, “to whom” is expressed as ka un. The lun we found in the first sentence loses its inflectional prefix l- and is preceded by its complement ka rather than followed by it. This is pattern (referred to by linguists as pied-piping with inversion) is widespread in Meso-American languages but very rare outside this area (Kaufman, Campbell & Stark 1986). (The ungrammatical English equivalent of the pattern would be *Who to did you give the money?)
  • The auxiliary ba indicating the future tense intervenes between the two words in the interrogative phrase ka un “to who”. This auxiliary is a second-position clitic and follows the first word in the clause. The fact that it interrupts syntactic constituents represents an extremely rare phenomenon referred to by linguists as “phonological second-position”.
  • The words expressing “give” in the two sentences are unrelated to each other phonologically. In the first example we find ru and in the second case we find ichiga (which becomes ichigu in the future). There are a number of stems in Garifuna which are unprefixable. That means that in any context where a prefixed stem is required (future, progressive, questions, etc.), an unrelated suppletive form is employed, in this case, ichiga.
  • The object is expressed by -ti in the first example but by -ni in the second example. Object agreement suffixes that begin with -n are found in nominalizations as well as in questions. This exemplifies a pattern that is found in many other languages of South America: questions and clausal complements are made up of nominalizations.

ELA’s Work

The Garifuna are a maritime people–through their work on ships, they have been able to migrate to the United States in large numbers. Remarkably, it is thought that over a third of all Garifuna people now currently reside in New York.

ELA is currently undertaking work with the local Garifuna community to document forms of natural speech, as well as the more archaic language of traditional songs. In addition to compiling a substantial lexicon of the language and analyzing its highly complex morphosyntax, ELA is supporting efforts of Garifuna communities to revitalize the language in New York and St. Vincent, and community members and organizations have been featured at a number of ELA’s public events.

The following is a sample “mini-vocabulary” of Garifuna lexical items made by Daniel Kaufman in collaboration with Alex Kwabena Colón.