For most authors the term Pipil (Nawat) is used to refer to the language in Central America (excluding Mexico). However, the term (along with the synonymous Eastern Nahuatl) has also been used to refer to Nahuatl lects in the southern Veracruz, Tabasco, and Chiapas that like Pipil have reduced the earlier /tl/ sound to a /t/. The varieties in these three areas do share greater similarities with Nawat than the other Nahuatl varieties do (suggesting a closer connection); however, Campbell (1985) considers Nawat distinct enough to be considered a language separate from the Nahuatl complex, thus rejecting an Eastern Nahuatl subgrouping that includes Nawat.
For other authors the term Aztec is used to refer to all closely related languages in this region as a single language, not distinguishing Nawat from Nahuatl (and sometimes not even separating out Pochutec). Currently the widely accepted classifications by Lastra de Suárez (1986) and Canger (1988) see Pipil as a Nahuan dialect of the eastern periphery.
|Campbell (1985)||Lastra de Suárez (1986), Canger (1988)|
| || |
Pipil specialists (Campbell, Fidias Jiménez, Geoffroy Rivas, King, Lemus, and Schultze, inter alia) generally treat Pipil/Nawat as a separate language, at least in practice. It is certainly closely related to modern Nahuatl dialects and to the Classical Nahuatl language of the Aztecs, even if not descended directly from the latter.
Today Nawat is seldom used and only by a few elderly speakers in the Salvadoran departments of Sonsonate and Ahuachapán. Cuisnahuat and Santo Domingo de Guzmán have the highest concentration of speakers. Campbell's 1985 estimate (fieldwork 1970-1976) was 200 remaining speakers although as many as 2000 speakers have been recorded in official Mexican reports. Gordon (2005) reports only 20 speakers (from 1987). The exact number of speakers is difficult to determine because native speakers do not wish to be identified due to local conflict, such as the matanza ("massacre") of 1932 and laws that made speaking Nawat illegal. The varieties of Nawat in Guatemala, Honduras, and Panama are now extinct.
As an endangered language, Pipil is threatened with the possibility of extinction within the next few years unless effective steps are taken quickly to keep it alive. A few small-scale attempts or projects to revitalize Nawat in El Salvador were initiated in the course of the twentieth century. Recent relevant work includes projects by the Asociación Coordinadora de Comunidades Indígenas de El Salvador (ACCIES) and Universidad Don Bosco of San Salvador (both of which have produced some teaching materials), and also an on-line language course by Monica Ward. The Nawat Language Recovery Initiative is a grass-roots association currently engaged in several activities including an ongoing language documentation project, and has also produced a range of printed materials. Thus, as the number of native speakers continues to dwindle, there is growing interest in some quarters in keeping the language alive, yet non-existent official support from the government (cf. Various, 2002).
Localities where Pipil was reported by Campbell as spoken in the 1970s include the following:
In phonology, one of the most salient features of Pipil, the absence of tl and the occurrence of t where the former is found in Classical and some modern Nahuatl dialects, is not exclusively Pipil but shared by some Mexican forms of modern Nahuatl. Another salient characteristic, the Pipil pronunciation [u] rather than [o] as the predominant allophone of a single basic rounded vowel phoneme, is also shared with some modern Mexican varieties. Thus these features are characteristic but not diagnostic.
However, Pipil t corresponds not only to the two Classical phonemes /t/ and /tl/ but also to a word-final saltillo or glottal stop in nominal plural suffixes (e.g. Pipil -met : Classical -meh) and verbal plural endings (Pipil -t present plural, -ket past plural, etc.). This fact has been claimed by Campbell to be diagnostic for the position of Pipil in a genetic classification, on the assumption that this /t/ is more archaic than the classical Aztec reflex, where the direction of change has been t > saltillo.
One other characteristic phonological feature is the diachronic merger in Pipil of original /ll/ (kept in Classical Nahuatl, and itself resulting from earlier assimilations such as *l+tl and *l+y and internally reconstructible for the ancestor of Pipil) with single l.
Pipil lacks some grammatical features present in Classical Nahuatl, such as the past prefix o- in verbs. It distributes others differently: for example, 'subtractive' past formation, which is very common in the classical language, exists in Pipil but is much rarer. On the other hand, reduplication to form plural nouns, of more limited distribution in the language of the Aztecs, is greatly generalised in Pipil. Still other grammatical features that were productive in Classical Nahuatl have only left fossilised traces in Pipil: for example, synchronically Pipil has no postpositions, although a few lexical forms derive etymologically from older postpositional forms, e.g. apan 'river' < *'in/on the water', kujtan 'uncultivated land, forest' < *'under the trees'; these are synchronically unanalyzable in modern Pipil.
|plural marking||limited in Classical||generalized||taj-tamal 'tortillas'|
sej-selek 'tender, fresh (pl.)'
|plural formation||mostly suffixes||mostly redup.|
|absolute -tli (Pipil -ti)||generally kept||often absent||mistun 'cat (abs.)'|
|construct /C_||-wi or zero||always zero||nu-uj 'my path'|
|inalienability||nouns generally have absolutes||many inalienables||*mey-ti, *nan-ti...|
|possessive prefixes||lose o before vowel||retain vowel (u)||nu-ikaw 'my brother'|
|articles||no generalized articles in Classical||definite ne, indefinite se||ne/se takat 'the/a man'|
|post/prepositions||postpositions||no post-, only prepositions||tik ne apan 'in the river'|
Many nouns are invariable for state, since -ti (cf. Classical -tli, the absolute suffix after consonants) is rarely added to polysyllabic noun stems, while the Classical postconsonantal construct suffix, -wi, is altogether unknown in Pipil: thus sin-ti 'maize' : nu-sin 'my maize', uj-ti 'way' : nu-uj 'my way', mistun 'cat' : nu-mistun 'my cat'.
An important number of nouns lack absolute forms and only occur inalienably possessed, e.g. nu-mey 'my hand' (but not *mey or *mey-ti), nu-nan 'my mother' (but not *nan or *nan-ti), thus further reducing the number of absolute-construct oppositions and the incidence of absolute -ti in comparison to Classical Nahuatl.
|inflection||more complex||less complex; analytic substitutes||kuchi nemi katka 'used to stay and sleep'|
|past prefix o-||found in Classical + some dialects||no||ki-neki-k 'he wanted it'|
ni-kuch-ki 'I slept'
|subtractive past formation||common in Classical + some dialects||limited|
|past in -ki||no||yes|
|perfect in -tuk||no||yes||ni-kuch-tuk 'I have slept'|
|imperfect tense||-ya||-tuya (stative)||ni-weli-tuya 'I could'|
|-skia, -tuskia conditionals||no||yes||ni-takwika-(tu)-skia 'I would sing/I would have sung'|
|initial prefixes /_V||lose i||mostly retain i||niajsi 'I arrive',|
kielkawa 'he forgets it'
To form the past tense, most Pipil verbs add -k (after vowels) or -ki (after consonants, following loss of the final vowel of the present stem), e.g. ki-neki 'he wants it' : ki-neki-k 'he wanted it', ki-mati 'he knows it' : ki-mat-ki 'he knew it'. The mechanism of simply removing the present stem vowel to form past stems, so common in Classical Nahuatl, is limited in Pipil to polysyllabic verb stems such as ki-talia 'he puts it' → ki-tali(j) 'he put it', mu-talua 'he runs' → mu-talu(j) 'he ran', and a handful of other verbs, e.g. ki-tajtani 'he asks him' → ki-tajtan 'he asked him'.
Pipil has a perfect tense in -tuk (synchronically unanalyzable), plural -tiwit. Another tense suffix, -tuya, functions both as a pluperfect (k-itz-tuya ne takat 'he had seen the man') and as an imperfect of stative verbs (inte weli-tuya 'he couldn't'), in the latter case having supplanted the -ya imperfect found in Mexican dialects.
Pipil has two conditional tenses, one in -skia expressing possible conditions and possible results, and one in -tuskia for impossible ones, although the distinction is sometimes blurred in practice. A future tense in -s (plural -sket) is attested but rarely used, a periphrastic future being preferred, e.g. yawi witz (or yu-witz) 'he will come'.
In serial constructions, the present tense (really the unmarked tense) is generally found except in the first verb, regardless of the tense of the latter, e.g. kineki / kinekik / kinekiskia kikwa 'he wants / wanted / would like to eat it'.
There are also some differences regarding how prefixes are attached to verb-initial stems; principally, that in Pipil the prefixes ni-, ti-, shi- and ki- when word-initial retain their i in most cases, e.g. ni-ajsi 'I arrive', ki-elkawa 'he forgets it'.