Lenca for

Linguists
(Sketch of an indigenous language of
Honduras)

Alan R. King
2017
Lenca for linguists
(Sketch of an indigenous language of Honduras)

Alan R. King
2017

© 2017 Alan King

Se reservan todos los derechos. Se puede distribuir libremente e imprimir para
fines de estudio y promoción de la lengua lenca. Queda terminantemente
prohibido cualquier actividad de reproducción, modificación y/o distribución con
ánimo lucrativo sin el permiso explícito por escrito de los propietarios del
copyright.

Este documento se podrá descargar gratuitamente de http://tushik.org/lenca-

kotik/.

In memory of Berta Cáceres (1971—2016)

2
PREFACE
Before commencing the current Lenca project, I had spent ten years endeavouring to
provide for the language needs of a nascent movement in El Salvador (Central
America) to recover the Nawat language, which still possesses speakers and has been
documented to a certain limited extent. That was a challenging task. However, the
challenges for Lenca put it into an entirely different category; that of an impossible
venture, according to many! The aim of this venture is nothing less than to bring back
to life, for an indigenous community, the language that was their ancestors’ until it
ceased entirely to be spoken a few decades ago.
The “bringing back to life” is in two senses: it is hoped that people will begin to learn it
and use it, but for that to happen the language needs to be described and codified to
make it teachable. Given that nobody has ever written a description of Honduran
Lenca before and nobody speaks Lenca today, it must therefore first be
“reconstructed” by linguists from the fragmentary information available before it can
begin to be recovered by the community.
And yet, the first on-line Lenca study groups have commenced this year, after months
of planning.1 Over one hundred individuals have so far enrolled in the virtual groups
using Facebook in conjunction with a supporting website2 as a platform. The
programme began to function in April 2017, one year after the assassination of the
influential Lenca political and ecological activist Berta Cáceres, after whom this
programme has been named.
Study of Lenca by those enrolling in the groups is at this stage self-study, given the lack
of conditions and resources for anything else, and we therefore refer to the framework
as “study groups” rather than “courses” or “classes”. The latter would be preferable
and it is to be hoped that the present programme can be instrumental in providing
preliminary training to a vanguard of Lencas or others with an adequate linguistic
competence out of which future Lenca speakers and teachers might eventually
emerge, and whence a more ambitious Lenca language programme may one day be
launched which, in the best of cases, might ensure a new future for the language.

1
Initially we tried to work through a programme supported by a Honduran governmental project linked
to the ministry of education, liaising with local administrators of the project, but there were some
serious failings in execution and it turned out to be a false start. Thanks to long experience with such
things and a reasonable amount of forethought, disaster did not ensue; rather, we quickly regrouped
and pressed on with Plan B, which takes advantage of the same research and materials but is not
dependent on any kind of official recognition or resourcing.
2
Tushik (meaning Our Navel in Nawat) is a portal for resources of many kinds related to the Nawat and
Lenca languages (http://tushik.org/). Most of the resources for Lenca that we have used or produced
are served from the website. There is a page listing and linking to the most important documentary
sources for Honduran Lenca at http://tushik.org/las-fuentes-lexicas-del-lenca-hondureno/. The main
materials we have developed targeting practical learners of Lenca are listed at http://tushik.org/lenca-
kotik/ from where they may be downloaded.

3
These study groups will offer a structured and loosely monitored programme of study,
articulated into a series of five-week modules (the higher modules are still under
development at this time). The first “course” in the beginning module was successfully
concluded last month, and a new group of beginners immediately got going, given the
high demand for enrolment. The graduates from the first group are now halfway
through study of the second module.
The “Berta Nap Laina” (Berta Is Here) Study Groups are each anchored to a specific-
purpose Facebook group through which materials, instructions, activities, progress
tests and communication are managed for the duration of one module. At the end of
the module, in which learners are evaluated on the basis of consistent participation
and self-reported progress checked by tests, all learners are removed and the group is
reset to begin again with a new group of learners. Learners who have achieved a
passing grade will be invited to re-enrol in the next module up; those failing to pass but
still interested in continuing are encouraged to re-enrol in the next edition of the
module they are in.
A sharp fall-off rate is fully expected in this procedure, with many in the first module in
particular failing to participate and dropping away almost immediately. This is
inevitable, and is best thought of as a self-selection process. Strategically the really
important thing is not how many would-be learners fail initially, but to harvest a few
who will persevere and make some real progress, and guide them forward. They are
the future hope for Lenca!
Central to the language study groups is a series of language materials being produced
in corresponding modules, titled Mol Pui Tam? (Can You Speak?).3 Modules consist of
five units; each unit provides for one week’s study. The units are made up of dialogues
(some of which have been recorded) with notes and exercises, grammar sections,
vocabulary lists and exercises. Additional, “ephimeral” support material, such as
illustrations, social-media-style “memes” and additional exercises are posted in the
study groups in an informal yet coordinated fashion. The groups are used to promote
interaction and maximum participation through games, puzzles, jokes and so on.
A more public Facebook group called “Iralapil” (We are sowing)4 has been engineered
to serve as a general interest group, an information hub, a channel to promote and
direct interest in learning Lenca and an initial recruitment platform and clearing house
for entry into the Berta Nap Laina programme. Although of fairly recent creation, other
Lenca interest groups now exist on Facebook, with which we interact and share in
order to publicise our work and ensure that others become aware of our work and
service.
The medium-term objective of the current study programme is to give people
exposure to some form of effective learning of the resuscitated Lenca language; in the
course of so doing, to generate a basic level of “use” of the language, thereby
constructing and enriching its corpus and collective competence; and eventually, it is
hoped, to lead a few of the best students forward to a significant degree of language
competence which will permit them in the future to occupy leadership roles to pursue
development and promotion of knowledge and recovery of Lenca. While aware of the

3
http://tushik.org/mol-pui-tam/.
4
https://www.facebook.com/groups/954746601258780/.

4
limited direct effect of our actions, we consider that we are planting seeds which may
give rise to eventual results provided there is interest and uptake.
This programme, however, presupposes the prior existence of a body of linguistic
knowledge and analysis which has had to be developed almost ex nihilo in the course
of the last two years, and the main purpose of the present sketch is to inform about
that work and some of its results. Chapters 2 and 3 summarise that process very
briefly, while Chapter 9 shows a sample of Lenca sentences as they appear in a source,
together with their linguistic analysis, in order to give an idea of the amount of analysis
that has been necessary and how it relates back to the input data. A bibliography at
the end of this document lists the sources and links to on-line copies of the texts used
are also provided. The other chapters focus on what is now known about the
Honduran Lenca language (with some passing references to Chilanga Lenca5 as well) in
linguistic terms.
Early in this process, we created a group6 on Facebook called “Masapeashpi” (We’re
Back, in Chilanga Lenca) to which a few linguists and specialists were invited, whose
purpose was to discuss the findings of my studies, first of Chilanga and later of
Honduran Lenca, which I shared by posting a series of documents in the group.
Subsequently some of these documents were also posted on the academia.edu site. A
third way to access my main linguistic writings about Lenca is to download documents
from the Tushik digital library, either from the library’s catalogue7 or via the individual
links such as those provided in the bibliography at the end of the present document.
Alan R. King, Zarautz, June 2017

5
Jan Morrow and I had worked together to support Nawat language recovery prior to turning to Lenca.
Nawat is spoken in El Salvador, which shares its northern border with Honduras. There are really two
Lenca languages, spoken in El Salvador and Honduras respectively. In the territory of modern-day El
Salvador prior to contact with Europeans, Lenca was the main language of the eastern part of the
territory of El Salvador, while Nawat was the most widespread language in western El Salvador. Extinct
since the middle of the twentieth century, Lenca was only documented at all properly in a single town in
El Salvador close to the Honduran border, Chilanga, but that documentation, including the only
contemporary language description of a form of Lenca, is sufficient to establish with certainty that
Chilanga and Honduran Lenca are not the same language. After having carried out some studies of the
Chilanga language, the focus of my attention shifted almost entirely to Honduran Lenca owing to greater
local interest, resulting in the project on which we are now embarked.
6
Membership is by invitation only; please contact Alan King or Jan Morrow.
7
The Tushik Digital Library is a collection of bibliographical items of interest to scholars researching
these and other languages of the area, access to which is by invitation; please contact me to request
access to the catalogue and documents.

5
Contents

Preface

1 What is Lenca? 7

2 Reconstructing Lenca 9

3 Dialects, variation and standardization 12

4 Overview of the language 15

5 Phonology 19

6 Morphology 21

7 Syntax 26

8 Lexicon 29

9 Sample source text 32

10 Sample teaching dialogue 34

Vocabulario Lenca-Español

Vocabulario Español-Lenca

Bibliography

6
 What is
1 Lenca?

Lenca is the name of an indigenous people of Central America who form part of the
ethnic and linguistic patchwork that was in place when contact with the Spanish
empire took place. The name “Lenca”, which was applied by the Spanish to the
historical Lencas and sometimes extended as a cover term for indigenous groups
of the region generally, is probably an exonym, of unknown origin and etymology.8
In the carving up of Central American territories in the postcolonial period which
gave rise to the current states of the region, the ethnic Lencas found themselves
divided between Honduras in the north and El Salvador in the south.
The historical Lencas spoke languages that all belonged to the same isolate family,
about whose early history nothing much is known. The last languages of this family
were no longer spoken well before the end of the twentieth century. A certain
amount of documentation of them had taken place in the second half of the
nineteenth century and the beginning of the twentieth in the locality of Chilanga in
El Salvador and in half a dozen places in an area in Honduras that extends from
Guajiquiro to La Esperanza, including the Marcala river valley. These are our only
primary sources of knowledge of the Lenca languages.
Analysing the available data, we find that the language of Chilanga is very different
from that of the Honduran Lencas; they could not have been mutually intelligible.
The Lenca from the Honduran side that is documented presents diversity,
presumably between dialects, which we can only imagine would have been greater
if we had information about the speech varieties of a larger number of territories.
But the differences we can observe are not great enough to require us to consider
that the known Honduran varieties do not all belong to a single language.
Hence there are two known Lenca languages, that of Honduras (with several
varieties documented) and that of El Salvador (with substantial documentation
only for one variety, that of Chilanga, but it is to be assumed that other districts
spoke different dialects, and this assumption is supported by local tradition). In
most of this sketch we are going to be talking about Honduran Lenca, to which the
term Lenca should be understood to refer in the present context unless otherwise
qualified.

8
Suggested interpretations supposedly in terms of Lenca roots are highly fanciful.

7
The reason why “Lenca” is used to designate both of these languages is that we lack
reliable information about specific names for the languages, and “Lenca” has
commonly been applied indifferently to all the speech varieties of the ethnic group.
To the best of our knowledge, Lencas themselves, when they still spoke their
languages and wished to refer to them, used generic terms meaning “language”,
such as kotik in Honduran Lenca and putum in Salvadoran Lenca.
Based on a Hispanicization of the latter word, Lencas are sometimes referred to as
potones in El Salvador, and their language has on occasion been called lenca potón,
perhaps adapted from a Lenca expression Lenca putum ‘Lenca language’. Some
Spanish speakers reanalysed the meaning of potón as an ethnonym or a name of
the specific language rather than generically meaning ‘language’, while elsewhere
lenca potón was reinterpreted by people who only spoke Spanish as if potón were a
qualifier of lenca which designated one kind of Lenca as opposed to other kinds of
Lenca which, presumably, they thought of as not potón. A considerable amount of
confusion and inconsistent naming practices has thus arisen.
The areas inhabited by Lencas today, or in which many people self-identify as
Lencas, are frequently mountainous with numerous valleys of difficult access,
located in remote areas from the point of view of important population centres
such as Tegucigalpa or San Salvador. Contact between mainstream Spanish-
speaking society and the Lenca communities has been limited, and not very much
has ever been known to the rest of the world about Lenca society, customs and
languages.
Sadly the languages have been lost, but the sense of ethnic identity is still strong in
Lenca communities, and it is in this context that the interest recently shown by
some people in attempting a linguistic recovery is of importance. It is in response
to such expressions of interest that the work reported here has been undertaken.

8
 Reconstructing
2 Lenca

How do you get communities to learn, much less use, an ancestral language that
nobody speaks any longer, thereby performing language recovery from zero?
Furthermore, how can sufficient knowledge of a language be achieved in order to
be able to teach people to speak it when the language no longer has any speakers,
nobody knows the language at present and it was never fully or systematically
described: how can we achieve the “reconstruction” of a dead language?

RECORDED HONDURAN LENCA DIALECTS

Western
(Intibucá)

Central
(Sta Elena,
Chinacla)

LENCA
Eastern
(Guajiquiro, LENCA
Opatoro)

Southern
(Similaton) (Chilanga)

By calling the development of a comprehensive description of the language on the

basis of fragmentary and incomplete data language reconstruction, I do not, of
course, mean the same thing that the term reconstruction is taken to refer to in the
scholarly discipline of historical linguistics. In a new context a term may acquire a
different sense; in the present context, reconstruction refers to a process which
takes available fragmentary and unsystematic raw linguistic data as input and aims
to produce as its ouput a coherent practical, systematic description of that
language.
The data that can be taken as input for the development of a systematic grammar
of Honduran Lenca consist of word lists and some other materials contained in
publications dating from the late nineteenth and early twentieth centuries (see the

9
Bibliography). These materials list raw Lenca data collected in the named
localities, as shown in the following table. (Regarding the twentieth century dates
of some items, note that date of publication does not necessarily reflect when the
data of interest had been collected.)

Localities:
Guajiquiro Opatoro Similaton Sta. Elena Chinacla Intibuca
Doblado &
✓ ✓ ✓ ✓
Girard 1951

Hernández &
✓
Pinart 1897

Membreño
✓ ✓
1897

Moreno
✓
1924/1949

Squier 1858 ✓ ✓ ✓ ✓

This set of sources covers the practical totality of our primary data. The five
documents are, to the best of our knowledge and according to their own evidence,
mutually independent sources. That is no longer so when the list is extended to
include other documents, which not only hardly contribute any new information
(except when it is quite probably spurious, see below), but take to repeating (often
without attribution) each other’s data, proliferating inherited errors, introducing
new errata and generally confusing issues.
Works such as Campbell, Chapman & Dakin (1978), Herranz (1987) and Lehmann
(1920), which also contain Honduran Lenca material, were taken into account in
the project but are omitted from the table for specific reasons.
Lehmann’s (1920) account does not contain original work (unlike his section on
Chilanga Lenca); he limits himself to reproducing edited versions of the data
provided by his predecessors (already tabulated above).
Some modern enthusiasts set much store by Herranz (1987), for no good reason
since his data is too recent (obtained after the disappearance of the last people
who knew Lenca properly), containing a small amount of authentic primary
material mixed together in an insufficiently transparent manner with a larger
amount of material derived from the proper sources already at our disposal; its
contribution of new information is very limited, difficult to interpret and contains
abundant errors.
Campbell et al. (1978) is an interesting testimony of one of the last rememberers
but it was again produced at too late a date to be a good source of new information

10
about the language; the amount of data it supplies is very small and much of it is
obscure; the same applies to the tape recording which the brief article reports on.9

9
Other items sometimes taken into consideration are also superfluous and misleading in some
ways. Andrews (1970) contains tables collecting together a hand-picked selection of items copied
(usually accurately, sometimes not) from the primary sources for the purpose of proving an
extremely dubious hypothesis of genetic relationship between unrelated language families; its use
as a data source is pointless if the originals are to be had. Gavarrete (2004) has been discovered by
J. Morrow (and confessed to him regretfully by the author himself when pressed) to be the result of
a blatant plagiarism of a published work on Cacaopera (an unrelated language formerly spoken in a
part of present-day El Salvador very close to the Salvadoran Lenca area), wilfully mislabeled as if it
were a book on Lenca and published in the “author’s” native Honduras, thereby sowing serious
confusions with repercussions that persist today. Lardé y Larín (1951), which calls itself a
diccionario, reproduces a derivative compilation based on data from extant primary sources of both
Lenca languages, thrown together mechanically and alphabetized without any sort of analysis.

11
 Dialects,
3 variation,
standardization

From what is now known I estimate that the distance between Chilanga Lenca and
Honduran Lenca may be on an order of magnitude approaching that between Latin
and Greek of the classical period. Although clearly related in a general way, they
are markedly different in many details. Apart from some phonological contrasts,
considerable differences in lexicon coupled with significant morphological
contrasts would have reduced spontaneous mutual intelligibility to a percentage in
the single digits. Typologically they are syntactically similar, but we cannot include
complex sentence syntax in the comparison because hardly anything is known
about this aspect of Chilanga in the absence of any continuous text.

Some disclaimers are in order before

attempting any estimate regarding the Locality Source Words
mutual distance among the Honduran Guajiquiro Doblado 96
dialects. It is hard to be certain how widely Pinart 604
dialects might have varied in the language’s Membreño 432
heyday on the basis of the records that have Moreno 63
survived. We can compare the localities in Squier 93
the documented region, but the overall Opatoro Doblado 230
quantity of data is small and unevenly Squier 80
distributed. The adjoining table gives the
number of vocabulary items in word lists for Similaton Membreño 300
each dialect from the sources that we have Squier 36
used. Our knowledge of some of those few Sta. Elena Doblado 193
varieties is especially scanty and “ 75
incomplete; this will affect the accuracy of
(Chinacla Doblado 79)
any comparison that we can hope to
achieve. (Intibuca Squier 55)
Two of the localities mentioned cannot be described coherently from the data
available, Chinacla10 and Intibuca, on account of the insufficient quantity of data,

10
Doblado’s short Chinacla wordlist raises special issues; much of its content is scarcely
recognisable as Lenca and is difficult to explain or integrate into the picture provided by a
comparison of all other sources. It is therefore omitted from consideration henceforth. Doblado
provides two different word lists for Sta. Elena, whence the two word counts. It is the longer of the

12
which is just enough to confirm that Honduran Lenca was spoken but not enough
to characterise the dialects reliably. This leaves us with four actually describable
dialects: Guajiquiro, Opatoro, Similaton and Sta. Elena.
Sadly, the data we have cover a smaller area than the territory over which Lenca
was undoubtedly spoken in the past, and we have no way of judging the linguistic
relationship that must once have held between the area we can map linguistically
and those we can’t, other than to make the obvious assumption that Lenca in its
historical entirety would have been more heterogeneous than is evident from the
extant data, which come from a fairly compact region. Relying on this deficient
data corpus, I think that variation across the Lenca speaking area of Honduras
might have been about of the order of that amongst the North Germanic speech
varieties of Scandinavia: a respectable degree of surface variation within the
bounds of a common, coherent underlying system, with considerable mutual
intelligibility despite easily perceived local differences.
Guajiquiro is the best-known form of Lenca, so in the initial stage of my analysis I
focused on Guajiquiro Lenca to build a coherent picture of a single dialect before
expanding the net to include other varieties. A further advantage of making
Guajiquiro dialect the starting point for the analysis ensues from the fact that it is
the only variety for which a continuous text of any substance is found, This formed
the basis of my initial investigations into Lenca morphology and syntax. After
several months of painstaking analysis, I turned to the rest of the corpus and drew
up profiles identifying the salient characteristics of each dialects. On the strength
of this expanded database, I subsequently reappraised and reformulated the sum
of knowledge to date of the phonology, grammar and lexicon of Honduran Lenca.
We are aware that the Lencas have inherited a strong tradition of local identity and
autonomy, matching the mountainous terrain and remote population centres of
their historical territory. The notion of the former existence of a heterogeneous
network of differentiated speech varieties also appears to persist. In our
interactions with people interested in Lenca language and culture, we have been
told by some that Lenca doesn’t refer to a single language since each community
possessed its own language and that the Lencas were merely a confederation of
associated peoples. As a non-scientific appraisal and with much of the evidence
missing, I think we should be cautious to accept the linguistic part of this assertion
at face value.
Some members of present-day communities have expressed the wish to learn not
just “Lenca” but their Lenca, the old language of their specific town or valley. If
there were a real possibility of granting this request we would not be opposed to
the idea in principle. But for most of the areas where Lenca ethnic identity is still
celebrated there is no record of their local language of old.
Even if we had more complete knowledge of Lenca dialectology the question would
have to be asked whether it was practical to attempt to split up a barely nascent
wave of awakening interest in the Lencas’ ancestral language by recreating a
division into numerous dialects. As it is, there is not even a great deal of choice.
Therefore a pan-Honduran Lenca supra-dialectal norm is proposed which provides

two lists that has been found most useful; it gives us an enticing glimpse of a dialect with some
notable differences from the more southerly varieties.

13
a standard point of departure for learners, speakers, writers and teachers of the
language.
Lenca Unificado Moderno (LUM) is not intended to replace more complex historical
realities but to make the language approachable and learnable. Those wishing to
tackle matters requiring greater sophistication such as the study of regional
variation or analysis of historical texts can use LUM as a launching pad, while
others will probably be content to acknowledge LUM as a good enough
approximation to the Honduran Lenca language which captures the essence of the
language.
In the following discussion of Lenca, LUM forms are presented except where
otherwise noted, and are in bold, while any other forms cited, including dialectal
Honduran Lenca, Chilanga Lenca or other languages, are in italics.

14
 An
4 overview
of Lenca

Lenca11 varieties share a fairly simple basic phoneme inventory. In the stops three
places of articulation are distinguished: /p t k/. There are two fricative phonemes,
both of which are sibilants: /s ʃ/.12 The inventory is completed by two nasal
consonants, two liquids and two semivowels: /m n13 l r w j/.14 The phonetics of
this consonant system is complicated by the existence of allophones for the stops,
which may be voiceless or voiced non-distinctively: [pb], [td], [kɡ].15 There are
five cardinal vowel phonemes in Lenca (/a e i o u/), as well as falling diphthongs
(/aj ej uj aw ew ow/).16
The basic maximal shape of modern Lenca syllables is CVC. Certain types of
alternation and variation in word forms within and between dialects are best
explained by assuming a hypothetical proto-stage possibly with maximal CV
syllable projection, which has tended to evolve towards CV(C) structure via final
vowel deletion, leaving traces in the morphology and in variant forms, e.g. kar or

11
Although the overall focus in this description is on Honduran Lenca, statements in this chapter in
particular referring to “Lenca” that are not otherwise qualified are also applicable to the Chilanga Lenca
language of El Salvador.
12
There is evidence of non-sibilant fricatives but only as syllable-final allophones of stop phonemes;
these are dialect variants not reflected in the LUM standard: wak [wakwaɡwah] ‘foot’; map
[mapmabmahmob(a)mof] ‘woman’.
13
In syllable-final position before a pause or a heterosyllabic vowel, /n/ was realized as velar [ŋ], just as
in other languages of the area such as Nawat and local Spanish: kotan [kotaŋ] ‘forest’, kotan eta
[kotaŋeta] ‘a forest’. I assume there was probably resyllabification across “weak” morpheme boundaries
such as between a clitic and a host or in a verb-auxiliary sequence, which would have resulted in an
alveolar nasal in sequences such as kotan ap [kotanapkotanab] ‘in the forest’, lan una [lanuna] ‘I was’
(rather than *[kotaŋapb], *[laŋuna] etc., for which there is no evidence), since the /n/ is treated as
syllable-initial: /ko·ta·nap/, /la·nu·na/. Syllable-final /n/ followed by a consonant adopts the latter’s point
of articulation, e.g. lan pil [lam'bil] ‘we were’, lan tam [lan'tam] ‘you (sg.) were (dubitative)’, lan kil
[laŋ'ɡil] ‘you (pl.) were’, lan yem [laɳjem] ‘you (sg.) were’.
14
The Chilanga language augments this with the addition of five glottalized consonants corresponding to
the three stops and two fricatives; the glottalized sibilants are pronounced as affricates: /pɁ tɁ kɁ ͡tsɁ ͡tʃɁ/.
15
In Chilanga, also for the non-glottalized sibilants /s ʃ/, which may be either voiceless or voiced, and
either fricatives or affricates: [szt͡sd͡z], [ʃʒt͡ʃd͡ʒ].
16
In LUM orthography, written ai, au etc. except when a vowel follows in the same orthographic word,
e.g. wei ‘for, to’, layik ‘calling’, tau ‘house’, mawen ‘much’. The Guajiquiro dialect also has a rising
diphthong ia (ie in neighbouring Opatoro) which corresponds to some instances of e in other
dialects. In didactic materials ê is used to identify the known cross-dialectal correspondences, e.g.
kên (Guajiquiro [kiaŋ], elsewhere [keŋ]) ‘leg’.

15
kari ‘what’, -tam(i) ‘2s dubitative ending’, nap(a) ‘here’, lem(a) ‘bed’, -in(a) ‘3s
indicative ending’, mol ‘to speak (stem)’, mol-pil ‘we speak’, mol-ka ‘(variant
gerund form)’ but moli-k ‘speaking (usual gerund), moli-n ‘spoken’ (participle’),
moli-una ‘I speak’.
Stress in Honduran Lenca is poorly documented and open to conjecture. It is
thought that perhaps word-stress was weak or absent and phonological phrases
tended to be stressed on the final syllable.17
Full inflectional morphology only exists in verbs, and is of a suffixing kind. In finite
forms inflection consists of an optional tense marker and an obligatory subject-
indexing person-number marker: eni-una ‘I hear’, u-mi ‘I want to go’, u-p-una ‘I
shall go’. There is subject pro-drop: (Unan) eniuna ‘I hear’. Pronominal objects are
expressed by proclitics bound to the verb (am eniuna ‘I hear you’); lexical object
NPs are not obligatorily indexed by object clitics (e.g. Berta eniuna ‘I hear Berta’).
Nonfinite verb forms are obtained through suffixes to the invariable verb stem:
eni-k ‘hear (gerund)’, u-n ‘go (participle)’.
There are two verbs ‘to be’ more or less as in Spanish (ser, estar). The real copular
verb (corresponding to ser) has the anomaly that its stem (wa-) may be omitted in
many contexts; when this happens the subject marker remains, and becomes an
enclitic attaching to the preceding word, e.g. puki wa-ina ‘(he/she/it) is big’ 
puki ina (pronounced [puɡi'na]) ‘ditto’, cf. puki la-ina ‘(he/she/it) is big (now)’,
compare Sp. es grande versus está grande.
Most of the tense (TAM) structures in Honduran Lenca can be analysed as auxiliary
constructions. These follow the generic template: Verb - NF suffix + Auxiliary - SM
(NF = nonfinite, SM = subject marker), e.g. eni-k la-una ‘I am listening’. The past
tense construction eni-n una ‘I heard’ is formed with the copula wak ‘to be’ as
auxiliary, with the auxiliary stem wa- deleted in this case. The underlying presence
of an auxiliary verb is suggested by the nonfinite gerundial construction in which it
surfaces: eni-n wa-k ne ‘after hearing, having heard’. Auxiliary constructions have
such a strong presence in the Honduran Lenca verb system that it is tempting to
think that in origin all indicative verb forms originated as periphrastic
constructions which have reached different stages of morphological coalescence.
The non-indicative mood, which I call volitive, may go back to a deeper stratum of
inflection; this mood uses a different set of subject markers (e.g. indicative 1s -una,
volitive 1s -mi); the indicative subject markers are obviously related to the
personal pronouns (unan ‘I’...), to which the volitive markers, on the other hand,
are unrelated.18

17
In Chilanga Lenca, on the contrary, phrasal stress regularly falls on the penultimate syllable with
few exceptions. This stress assignment occurs after the incorporation of clitics into phrasal units,
e.g. Chilanga 'ishko ‘man’, ish'ko-na ‘the man’, sap ‘eye(s)’, 'u-sap ‘my eye(s)’. Clitic boundaries do not
limit the operation of allophony rules (hence ['sap] but ['uzap'ud͡ zap].
18
The Chilanga Lenca system of conjugation is not cognate to the Honduran system in a way that is
immediately evident, and the basic tenses in Chilanga are less obviously periphrastic on a synchronic
level. The Chilanga present tense form isakanpi [isa'kambi] ‘we sow, we are sowing’, for example, is
probably derived from the same pattern as the Honduran cognate ira-k la-pil [ira(ɡ)la'bil] ‘ditto’ (formed
like eni-k la-una ‘I am listening’), but cannot be described synchronically as periphrastic within
Chilanga. There exist periphrastic tense forms in Chilanga too, but their formation need not have a
common origin with the Honduran patterns. For example, one of the future tenses in Chilanga
transparently comprises a periphrasis using the present tense of ‘to go’ as auxiliary (in parallel with

16
Noun phrases have no case or number marking, and no gender or noun-class
system. Adjunct functions can be signalled by enclitic postpositions (kotan ap ‘in
the forest’, am man ‘with you’, api wei ‘for us’, Tegucigalpa nam ‘from
Tegucigalpa’). Topics can be signalled by the enclitic ne.19
Noun phrases have a proclitic and an enclitic position which may be filled by
various items. The proclitic position is available to possessive and demonstrative
determiners (u ‘my’, na ‘this’ etc.). The enclitic position can be used by the definite
article, the topic marker, a postposition or a clitic form of the copula. Most of these
enclitic categories are mutually exclusive so there is usually only one enclitic per
noun phrase, e.g. the definite article cannot co-occur with the topic marker, a
postposition or a short copula. There is also only one proclitic position per phrase
(*u na tau / *na u tau ‘this my house’, cf. Nawat ini nu-kal ‘this my-house’ which is
grammatical).
Sentence syntax is SOV, with many of the usual trappings of such languages.
Topicalization is the main way in which clauses may diverge from a strict SOV
order. Subordinate clauses may be finite or non-finite; subordinators take the form
of clause-final or post-verbal suffixes or particles. Adjunct clauses are free to
precede or follow the matrix clause.
Is Lenca an agglutinative language, as has occasionally been suggested? That
depends. A standard definition says that agglutination is “a type of morphological
structure in which words can be readily divided into a linear sequence of distinct
morphemes, each of which typically has a fairly consistent shape and a single
consistent meaning or function” (R.L. Trask, A dictionary of grammatical terms in
linguistics, Routledge, 1993, p. 12). But how are we to define “word” here?
Examples of agglutinative structures mentioned in non-specialist discussions
include Swahili a-li-ku-on-a ‘he/she saw you’ (cited by Trask), Turkish ev-ler-iniz-
den ‘from your houses’ (Wikipedia, sub “Agglutination”), Japanese tabe-sase-rare-
ru ‘can cause someone to eat’ (B. Comrie et al., The atlas of languages (rev. ed.),
2008, p. 53), and so on. These stock examples seem implicitly to assume that it is
an uncontrovertible fact that alikuona, evlerinizden and tabesaserareru are single
words — which they may well be, but what are the principles on which this
identification is made? Should we just ask a speaker of the language: Is this all one
word?20
Most of the data upon which we must base a reconstruction of the structure of
Honduran Lenca was gathered before the twentieth century by people who lacked
linguistic training, who made an effort to document the living language of their

Spanish); Honduran Lenca has its own periphrastic future tenses but does not use ‘to go’ in this way.
Thus some of the development of the tense systems of Honduran and Salvadoran Lenca may have
occurred independently, yet they nevertheless appear to be built out of a shared proto-system, in which
tense may have played a less significant role than aspect and inflectional morphology was perhaps limited
or even non-existent.
19
The ne topic construction is only attested in one Honduran dialect, Guajiquiro, but this is the only
dialect in which any substantial continuous text is extant, so it is not to be concluded that other dialects
lacked the construction. It is unlikely that ne existed (in that form, at least) in Chilanga as it would
probably have found its way into at least an occasional example if it had. It is possible that a non-cognate
Chilanga clitic form, pa, might have had a related function. (The only known morpheme in Honduran
Lenca which could be a possible cognate of Chilanga pa would be the locative postposition ap.)
20
Matters are complicated by the recognition of clitics as a distinct morpheme type which is neither
a “full word” nor a mere agglutinative affix.

17
informants by eliciting and copying words and phrases. We may be fairly certain
that the collectors of these data did not know or understand any Lenca themselves,
hence they could only record (as best they knew how) the sounds they thought
they heard, but they were not in a position to analyse what they heard
morphologically or syntactically, and it seems that they rarely knew whether or
where to divide what they heard into discrete words. Hence Membreño’s text
corpus from Guajiquiro abounds with “words” such as carabelamga, ishteguantami,
and so on, whose actual meanings, disentangled for the first time by our linguistic
analysis, are ‘that you were going to die’ (in LUM spelling: kara pelam ka), ‘didn’t
you see?’ (LUM: ish te wan tami), etc.
The word lists sometimes contain phrases or even sentences, whether or not the
documenter knew it. Apart from items explicitly glossed by whole utterances, such
as ¿cabnam pulatami? ‘¿de dónde vienes? (source: Pinart) or animisheimomashiley
glossed ‘le está pegando a la mujer’ (Doblado), we also find other phrases such as
moltebuyina ‘mudo’ (Pinart) which really means ‘He/She cannot speak’. The LUM
spellings of these phrasal examples and accurate glosses of their meanings are,
respectively, Kap nam pu(k) latami? ‘Where are you coming from?’, Ana emeshi
i map mashi(k) lai(na) ‘that man is beating his wife’ and Mol tê pui ina ‘He/She
cannot speak’.
In the spelling system proposed for LUM, morphemes which can reasonably and
justifiably be treated as separate words (including clitics or auxiliaries) are so
treated. This does not result in a complete absence of inflections or simple
“agglutinative” structures. Examples from the sentences just cited are mashi-k
‘strike-GERUND’, pu-k ‘come-GERUND’, la-ina ‘estar-3s’, la-tam(i) ‘estar-
DUBITATIVE.2s’, pela-m ‘FUTURE-2s’ and so on.

In the raw source data

LUM spelling Source spelling
there are spurious word-
divisions and the same u wiran u-güiran my village
morpheme sequences u familia ú-familia my family
may be written without a u yêrta u-yarta give me
break or with one in u ish tê wan tami? ú ishteguantami didn’t you come to see me?
different instances, even u te tê ina utetena does not do [to] me
in adjacent items. In the ushak ne u-shagne upon arriving
short sample of source
text presented in Chapter 9, from which some of the above illustrations were
taken, we do not need to look far to see contradictions in the way word-divisions
are used: compare for example tinguishinuna ‘I waited’ with enin uná ‘I heard’ (in
LUM, tinkishin una and enin una).
It would be possible to respell Lenca to make it look more “agglutinative”, simply
by attaching all clitics and auxiliaries to the word they precede or follow, e.g.

LUM spelling “Agglutinative” respelling

U latan nan Carlos ina. Ulatannan Carlosina. My name is Carlos.
Nanan u seya ina Nanan useyaina. This is my younger brother.
Ishiuna kotik shuntê niwan yêm Ishiuna kotik shuntê I see that you have learnt a little
ka! niwanyêmka! Lenca.

18
 Phonology
5
The vowel inventory of Honduran Lenca is /a e i o u/ with cardinal phonetic
values.
Although rising diphthongs are not otherwise very common, data for Guajiquiro
dialect present the sequence “ia” in certain words, generally corresponding to /e/
in most other dialects. Following y and sometimes sh, the Guajiquiro variant may
present “a” rather than “ia”. However, some words with “(i)a” also have variants
with “e” in Guajiquiro, and no rule has been identified to predict when such
alternations occur. For LUM, I have decided to opt for e in the spelling of such
words, but for didactic purposes I propose the option of using ê to identify words
in which the e of other dialects is or may be replaced by “ia” or “a” in Guajiquiro,
e.g. kê ‘stone’ (Guaj. [ke] or [kia], other dialects [ke]), yêr- ‘give’ (Guaj. [jar-], other
dialects [jer-].

There are falling diphthongs, of which the most ei / êi ui eu ou

common are shown here. As a spelling convention, ai au
these forms are used except when followed by a vowel within the same word, in
which case y is used rather than i, and w rather than u, e.g. shêitam ‘do you
like/want?’ but eye ‘tortilla’, tau ‘house’ but shawa ‘tomorrow’. This rule does give
rise to some spelling alternations within paradigms: for example, the stem of ‘to
like, to want’, shêi- ([ʃaj] in Guajiquiro, see above), as in shêitam ‘do you
like/want?’, shêipil ‘we like/want’, etc., is written as shêy- in shêyina ‘he/she
likes/wants’, shêyiuna ‘I like/want’, the gerund shêyik or the participle shêyin.
In some words, a monosyllabic word form ending in a consonant has a variant in
which the final consonant is followed by a vowel, which is usually either a or i
(occasionally e), kar or kari ‘what’, kap or kapa ‘where’, am or ami ‘you’, umal or
umali ‘let’s go’, map or mapa ‘woman’ etc. In such words, these may be free
variants, and strict rules do not seem able to predict their occurence, except in the
case of verb stems, where there is also a pattern of alternation between i and zero
following a final consonant, but according to certain rules: thus e.g. mol- ‘speak’ in
mol-tam ‘do you speak?’, mol-pil ‘we speak’, mol-mal ‘let’s speak’, mol-ina
‘he/she speaks’ but moli-k (gerund), moli-n (participle), moli-una ‘I speak’. I
propose using the ad hoc term “fluid vowels” to refer to these phenomena.

19
The consonant phoneme inventory is shown here. The
phonetic realizations of most of these phonemes seems to p t k
be straightforward. /n/ had a velar realization before a
s sh
pause. Before stops it assimilates in place of articulation
to the following consonant. The main complications m n
concern the stop consonants /p t k/ themselves, which
each had two allophones, voiced and voiceless; /p/ and l r
/k/ also had fricative allophones in some dialects, it
y w
seems. The distribution of these allophones is mostly
determined by rules, but these are a little complex and
varied somewhat between dialects.
All three stop phonemes may be voiced after /n/, but voicing does not always
happen. For example, the imperative form tan-ta ‘sit down!’ is attested as tanda,
but en-ta ‘listen!’ is not given in the data as enda which is what would be expected.
There may have been free variation. For /p/ and /k/ voicing is usual after /n/.
The pronunciation of /k/ is voiced between vowels within a word (e.g. shoko
‘white’ [ʃoɡo]) and generally in word-internal positions unless adjacent to a
voiceless consonant (e.g. tepka ‘seven’ [tepka] or the dimunitive suffixes -ska [ska]
etc., and even then, sometimes /k/ is voiced, e.g. mokta ‘all’ [moɡta]). Word-
initially, it is voiceless, e.g. kê ‘stone’ [ke], [kia]. The pronunciation of /p/ is
voiceless between vowels within a word (e.g. napir ‘pig’ [napir]).
The contrast between the behaviour of these two stops is seen in the proclitics api
(first person plural) and aki (second person plural): api is pronounced [api],
whereas aki is [aɡi] (or [aj]).
Both /p/ and /k/ are usually voiced in initial position in grammatical morphemes
that are closely bound to a preceding word (whether as a suffix or an enclitic), as is
the case of the morphemes pil ‘we’, kil ‘you (pl.)’, pela- (future auxiliary), ka
(complementizer), -kin (verbal noun suffix). Exceptions occur sometimes when
the stop is adjacent to a voiceless consonant, e.g. kos-ki- ‘to remove’, tish-ki- ‘to
teach’, u-p-tam ‘will you go?’
For the most part these patterns are consistent across the data for different
dialects, but less consistency is observed when /p/ or /k/ is in word-final position,
where each may display at least three allophones: voiced stop, voiceless stop or
voiceless fricative; a fourth option is for the final stop to become silent. /p/ is
voiced if followed by a “fluid vowel”, e.g. nap ‘here’ may be [nap] or [nab], but also
[naba].
The source data contain scanty and sometimes contradictory evidence making it
hard to draw firm conclusions about the nature and placement of stress. Probably
most verb forms were stressed on the last syllable; it is to be borne in mind that
given Lenca’s SOV syntax, the verb typically stands last in its clause, so this would
actually be the last syllable of the clause. It is plausible that stress was assigned to
syntactic groups rather than to individual words. It is assumed that fluid vowels
were unstressed, e.g. nap ['naba], umal [u'mali].

20
 Morphology
6
The syntax of the noun phrase is clearly structured. It can involve both proclitic
and enclitic elements which may surround a lexical nucleus. In the noun phrase,
proclitics are either demonstrative (na tau ‘this house’) or possessive (u tau ‘my
house’); enclitics comprise a determiner (tau nan ‘the house’), postpositions (tau
ap ‘in/at/to the house’) etc. There is a postposed topic particle ne which may
follow NPs.

The proclitics are shown in this

Demonstrative Possessive Object
table. The demonstrative na this u my u me
proclitics always precede a noun. ina that (near) am your (sg.) am you (sg.)
The possessive and object ana that (far) i his, her, its i him, her, it
proclitics are identical in form; api our api us
the former precede nouns, the aki your (pl.) aki you (pl.)
latter verbs. These proclitics also al their al them
precede postpositions (u man
‘with me’) and the topic marker ne (u ne ‘I TOPIC’).
The possessive proclitics express number of the possesser since number (singular
or plural) is incorporated into the person system, as shown. Quantifiers may also
encode number lexically in obvious ways (eta ‘one, a(n)’ is singular, other
numerals are plural...). Apart from these, there is no grammatical expression of
number at all in the noun phrase.21
Nouns cannot be preceded by both a demonstrative and a possessive; verbs cannot
be preceded by two object markers.

21
Glosses are conventionally cited as singular for ease of exposition, but are really neutral for
number (except as mentioned) and gender, e.g. na is not only ‘this’ but also ‘these’; i is ‘his, her, its’.

21
The enclitics that can Def. article Postposition Topic mkr. Copula
follow a nominal nucleus una I am
containing any of (Noun) tu am I?
(Adjective) (Quantifier) yêm you are (sg.)
are shown in this table. ap in, at, to tam are you (sg.)?
The postpositions (some nan the man with (ne) ina is
listed below) are counted etc. ti is?
pil we are
as enclitics on
kil you are (pl.)
distributional grounds
lana they are
(tau nan ‘the house’, tau
ap ‘in/to the house’, *tau nan ap), as are the short forms of the copula (tau ina ‘it’s
the house’, *tau nan ina).
The position of the topic marker in the table is less clear: it does not co-occur with
the article nan (tau ne ‘the house TOPIC’, *tau nan ne) but it does not preclude
postpositions (tau ap ne ‘in the house TOPIC); however, ne is also itself a
postposition meaning ‘about’ (kayu ne molmal ‘let’s talk about the horse’), in
which case it does not co-occur with another postposition or with topic ne.
In phrases where it is syntactically compatible (e.g. tau nan), the article nan is less
obligatory than English the in definite noun phrases. Enclitics including the article
may co-occur with any of the proclitics (na tau nan ‘this house’, u tau nan ‘my
house’, u tau ap ‘in/to my house’, u tau ne ‘my house TOPIC’, u tau ina ‘it is my
house’).
Quantifiers may co-occur with enclitics (tau mawen ap ‘in many houses’). This
applies to eta ‘one’ which is also used as indefinite article (tau eta’p ‘in/to a/one
house’, tau eta ina ‘it is a house’ etc.).

This table shows some of the most basic Locative ap, ’p ‘in, at, to’
postpositions. The full form ap occurs after Ablative nam ‘from’
consonants or diphthongs (kin ap ‘on the road’, Comitative man ‘with’
Instrumental lan ‘with, by’
tau ap ‘in/to the house’, but wara’p ‘in/to the
Dative wei ‘to, for...’
river’, ili’p ‘in/to the tree’). (A)p may be omitted Theme ne ‘about’
in complements of certain verbs (wiran umal
‘let’s go to town’).

Self-standing pronominal forms ending Dem. pronouns Indep. pers. pronouns

in -nan exist for the demonstratives and na(nan) this unan I
personal pronouns (see above, ina(nan) that (near) amnan you (sg.)
proclitics). In the demonstrative ana(nan) that (far) inan he, her, it
pronouns -nan is optional (na or nanan apinan we
‘this’). In the independent personal akinan you (pl.)
pronouns -nan is obligatory (unan ‘I’, alnan they
*u). The personal proclitics, not the independent personal pronouns, are used with
postpositions or topic-marking ne (u man ‘with me’, u ne ‘I TOPIC’, *unan man,
unan ne).
Object proclitics primarily indicate a patient (u ishyêm ‘you see me’) but with
some verbs may also express a recipient (u yêryêm ‘you give me’, u molyêm ‘you

22
tell me’ etc.). Indirect objects can also be expressed as postpositional phrases with
wei ‘to, for’ (u wei yeryêm = u yeryêm).
Verbs have finite and nonfinite forms. Three basic nonfinite forms are essential to
periphrastic conjugation: the stem form (no inflection), the gerund (-k) and the
participle (-n).
Verb stems are sorted into two classes for convenience when describing their
inflection. Those whose stem form ends in a consonant insert i between the stem
and some endings including the -k of the gerund and the -n of the participle (mol,
molik, molin ‘speak’); these constitute the first class. All others, which do not
insert a vowel, make up the second class (liwa, liwak, liwan ‘buy’). I have decided
to use the gerund as a citation form for verbs, hence molik ‘to speak’, liwak ‘to
buy’.
All gerunds in which -k is preceded by a vowel other than i (like liwak ‘buy’, kok
‘enter’, uk ‘go’, puk ‘come’, stems liwa-, ko-, u-, pu-) belong to the second class.
Most ending in -ik belong to the first class (like molik, stem mol-), but in some the
i is a permanent part of the stem so they are grouped in the second class (e.g.
molshik ‘read’, tipik ‘go up’, with invariable stems molshi-, tipi-). These are
predictable from phonological shape: such verbs either have a stop or a consonant
cluster preceding the i.

There are two main synthetic finite forms. Indicative: Volitive:

Both are unmarked for tense, but contrast in Affirmative Dubitative
mood: indicative (called unmarked tense) moli -una mol -tu mol -mi
and optative-cohortative-imperative-jussive mol -yêm mol -tam mol -ta
(called volitive). Both the indicative and the mol -ina mol -ti mol -iu
volitive are formed from the stem by adding mol -pil mol -mal
distinct sets of person (subject) endings. In mol -kil mol -tal
the indicative there is a second set of subject mol -lana mol -iu
markers in the singular which are typically
used in questions (called dubitative; the
non-dubitative indicative endings may be Indicative: Volitive:
called affirmative). The upper table shows Affirmative Dubitative
the paradigm of a Class I verb (molik liwa -una liwa -tu liwa -mi
‘speak’) in these synthetic finite forms; the liwa -yêm liwa -tam liwa -ta
lower table shows that of a Class II verb liwa -ina liwa -ti liwa -yu
(liwak ‘buy’). liwa -pil liwa -mal
liwa -kil liwa -tal
The indicative synthetic tense is unmarked liwa -lana liwa -yu
for tense and may refer to present or past
time.22 It is used with stative verbs generally to express the present (launa ‘I am,
estoy’, tishiuna ‘I know’), and with dynamic verbs to express either habitual or
past (moliuna ‘I speak’ or ‘I spoke’). The verb wak ‘to be, ser’ has a full form that is
regular (X waina ‘is X’) and a short form where the stem wa is omitted and the
ending alone is used as an enclitic (X ina ‘is X’).

22
In Chilanga Lenca, the apparent cognates of the Honduran unmarked tense, which lacks a segmental
tense marker, provides a past tense, except for the irregular and defective verb yan- ‘estar’ which
responds to Honduran lak, where it gives a present tense: Chilanga ya[ŋ]u, yanmi, yanpa, yanpi..., LUM
launa, layêm, laina, lapil... ‘I am, you are, he is, we are...’.

23
The volitive23 has a variety of uses, some of which depend on the person; uses are
different in main and subordinate clauses. In main clauses, the 1s volitive may
expresses volition or intention (molmi ‘I want to speak, I shall speak’); in the 1p,
‘let’s’ (molmal ‘let’s speak’), in the second person it is imperative (molta, moltal
‘speak!’ sg./pl.), in the third person it is jussive (moliu ‘let (him/them) speak’).

Main verb Auxiliary Other finite forms are generally

Imperfective: gerund + lak periphrastic constructions, consisting of
Perfective: participle + wak a non-finite form of the main verb
Prospective: stem + pelak followed by an auxiliary. The main
patterns are shown here. The auxiliaries
conjugated in the unmarked indicative tense are understood as present tense.

In the imperfective Present Past Future

construction, the Affirmative
gerund (the verb’s molik launa molin una mol pelauna or molpuna
citation form, ending “ layêm “ yêm “ pelayêm molpêm
in -k) is followed by “ laina “ ina “ pelaina molpena
the verb lak ‘estar’, “ lapil “ pil “ pelapil molpepil
e.g. molik laina ‘is “ lakil “ kil “ pelakil molpekil
speaking’. In the “ lalana “ lana “ pelalana molpelana
perfective Dubitative
construction, the molik latu molin tu molpetu
participle (in -n) is “ latam “ tam molpetam
followed by the short “ lati “ ti molpeti
form of the copula (molin ina); this has the meaning of a past or perfect, ‘spoke,
has spoken’. In the future or prospective construction, the plain verb stem is
followed by a form of pelak (mol pelaina ‘will speak’). There is an alternative
future tense which is perhaps a contraction of the “long” future: molpena.
More complex periphrastic tenses may be obtained by using the auxiliary in a
different tense form, including periphrastic tenses (molik lan ina ‘was speaking’,
mol pelan ina ‘was going to speak’).

Negation of verbs involves the particle tê, which

is placed after the verb’s stem and is followed by a Unmarked Past
form of the copula wak, e.g. mol tê ina ‘does not Negative Negative
speak’, past mol tê wan ina ‘did not speak, has mol tê una mol tê wan una
not spoken’ (wan ina is the past of wak). “ “ yêm “ “ “ yêm
“ “ ina “ “ “ ina
In the domain of the noun phrase, there is some “ “ pil “ “ “ pil
evidence for variation in both the form and the “ “ kil “ “ “ kil
distribution of the definite article enclitic nan. “ “ lana “ “ “ lana
Variant forms include anan, ana, na. It’s use in the Neg. dub. Neg. dub.
Guajiquiro texts is frequent and seems unmarked, mol tê tu mol tê wan tu
yet it is completely unattested in some dialects. “ “ tam “ “ “ tam
Phonologically based variation in conjugation “ “ ti “ “ “ ti
morphology includes the optional addition or truncation of final vowels in many

23
The Chilanga cognate of the Honduran volitive paradigm is its imperative system. It is not known
whether these forms had the other functions of the Honduran volitive given the scarcity of syntactic data.

24
forms (shêitami = shêitam ‘you like (dub.)’, lin = laina ‘está’, umali = umal ‘let’s
go’) and alternations involving diphthongs (leina = laina ‘está’).
Besides -una, there is evidence for an alternative 1s indicative affirmative subject
marker, -on. It is natural to attempt to account for this alternation fully in terms of
the phonological phenomena just mentioned, but some difficulties remain.24 This
phenomenon is not to be confused with a different one of suffixing -on to other
finite verb forms (not replacing the subject marker but in addition to it), e.g.
lainon = laina ‘está’, shêitamon = shêitam ‘you like (dub.)’, umalon = umal ‘let’s
go’. It is thought that this added a stylistic nuance and my guess is that it may have
signalled familiarity between interlocutors.

Chilanga Lenca, from El Salvador, has two 1s endings, -on and -u, the latter corresponding to
24

Honduran -una.

25
 Syntax
7
The syntax of noun phrases was discussed in Chapter 6 in connection with clitics.
On a macro-level the pattern would be: (PROCLITIC) + NUCLEUS + (ENCLITIC).
Note that there is only one proclitic and one enclitic slot. Thus while different types
of proclitic and of enclitic are eligible to be used in these slots, the choice of on
proclitic (e.g. a possessive) excludes the possibility of co-occurrence of another
(e.g. a demonstrative), and similarly, only one enclitic per noun phrase is possible.
The Nucleus may include an adjective and/or a quantifier, in that order, after a
noun. Thus a fuller statement of noun phrase structure is:

(PROCLITIC) + NUCLEUS + (ENCLITIC)

Def. article
Demonstrative Postposition
Noun (Adjective) (Quantifier)
Possessive Topic marker
Short copula

N-N genitive constructions are scarcely attested in the corpus, and we are left to
assume, on the basis of Lenca’s typological profile, data from Chilanga Lenca where
it is documented and the fact that such a pattern is commonplace in indigenous
languages of the area generally, that the possessor precedes the possessed noun
which takes the 3s possessive i: map i wewe nan ‘the woman’s child’ (woman 3s
child DEFINITE, i.e. ‘woman her child’).
Turning now to the syntax of the simple clause, unmarked order is verb-final, i.e.
SOV. There is pro-drop, so the S (and O) need not be represented by a clause
constituent if adequately specified (in a finite clause where the verb is inflected for
person of subject), e.g. Kotik moliuna ‘I speak Lenca’ (Lenca speak.1s), or if overt
specification is considered unnecessary (in a nonfinite clause), e.g. Ushak ne am
mol pelauna ‘When [I] arrive [here], I will tell you’ (arrive.GERUND TOPIC 2s speak
FUTURE.1s). Subject-indices are obligatory on finite verbs; object marking is not,
and is only used if a nominal O is not expressed, e.g. U ishina ‘He/She/It sees me’, I
ishina ‘He/She/It sees him/her/it’ but Map nan ishina ‘He/She/It sees the
woman’, not *map nan i ishina (or if this is possible, it is at least not required).
Notice that if the object is a personal pronoun it is represented by a proclitic
satellite of the verb from which it is inseparable (Map nan u ishina, not *U map
nan ishina, except when the latter means ‘My wife sees’ or ‘He/She sees my wife’).

26
NB. In periphrastic TAM constructions, it is the lexical verb of which the
pronominal object is an argument that is preceded by the object proclitic, e.g. u
ishik laina ‘is seeing me’ (1s see.GERUND estar.3s), not *ishik u laina. This does not
necessarily apply to other kinds of embedding construction, e.g. puk u ishin ina
‘he/she came and saw me’ (see.GERUND 1s see.PARTICIPLE be.3s).
Among noun phrases or other constituents preceding the verb in the unmarked
order, the subject usually precedes others (including nominal objects), hence the
default order is SOV (Map nan shui nan ishin ina ‘The woman saw the dog’, Map
nan wiran ap liwak laina ‘The woman is buying in the town’).
For various reasons this default order may be disrupted: topicalization is frequent.
A constituent treated as a topic is placed in one of the topic positions, of which a
main clause has two: the beginning of the clause and the end of the clause. Thus
there is a “nuclear clause” consisting of (S) (O) V which may be preceded in its
entirety and/or followed in its entirety by topics, hence (T1) (S) (O) V (T2).
T1 and T2 may be differentiated pragmatically as foreground-topic and
background-topic. Any nominal constituent can optionally be topicalized;
sometimes topicalization is obligatory, when an interrogative constituent pre-
empts the position immediately preceding the verb, e.g. Map nan tau ap laina
‘The woman is in the house’, but Kunan laina tau ap? ‘Who is in the house?’ with
tau ap displaced to T2 position. Adjuncts, whether phrasal or clausal, are very
frequently topicalized, e.g. Shawa mokta wiran ap umal ‘Let’s all go to town
tomorrow’ (tomorrow [T1] all [S] town to go.VOLITIVE.1p).
Topics (in either T1 or T2 position) are frequently (but optionally) followed by the
topic marker ne, e.g. Kunan laina tau ap (ne)?, Shawa (ne) mokta wiran ap
umal). A personal pronoun subject can be topicalized with ne, using the proclitic
form of the pronoun, as before postpositions, e.g. U ne wiran ap umi ‘I (T1) want
to go to town’ (I TOPIC town to go.VOLITIVE.1s), cf. Unan wiran ap umi ‘I want to go
to town’ (I town to go.VOLITIVE.1s), without marked topicalization of the subject
pronoun.
To judge from our texts, subordination was widely used in Lenca discourse and its
various constructions demand careful study. Not so coordination, for which limited
resources are in evidence. Not only is there a dearth of coordinating connectors
like but, or and so on, there is no single equivalent to and either. Clause
constituents may be coordinated or listed in several ways, depending on the
context.
One such is the use of a postposition such as man ‘with’ with the second of two
constituents, e.g. eye shinak man ‘tortillas and beans’ (tortilla bean with). Another
postpositon, li, occurs when human referents are coordinated (u pap u li ‘my
father and I’). The first item in this coordination construction is often followed by
ne, the topic marker (u pap ne u li). Use of this type of construction is limited by
the fact that it cannot be governed by a postposition to indicate the function of the
compound NP within the clause because the enclitic postposition slot is already
occupied by man or li, but it is available for constituents like subjects and objects
which do not require such a postposition, e.g. Eye shinak man kormi ‘I want to
eat tortillas and beans’, U pap ne u li ta nan irak lapil ‘My father and I are sowing
the cornfield’.

27
Another option for conjoining any number of nominal constituents is a “listing
construction” in which every nouns in the list is followed by a, e.g. Miguel a, Berta
a pupelana ‘Miguel and Berta will come’ (M. a B. a come.FUTURE.3p); Tau a, ta a
muk lalana ‘They have houses and cornfields’; Ama a, mira a, shinak a korpil
‘We eat maize, plantains and beans’.
Neither of these devices can conjoin clauses. If two clauses sharing the same same
subject are to be coordinated, this may be achieved by using the gerund of the verb
in the first clause and only fully conjugating that of the last clause, e.g. U seya nan
puk u ishin ina ‘My younger brother came and saw me’ (1s younger.brother
DEFINITE come.GERUND 1s see.PARTICIPLE COPULA.3s).

A morpheme which may be used to coordinate both clause constituents and

clauses is ikwa. Placed between two nouns it coordinates them; the second item
may still be followed by man, e.g. Kashlan eria ikwa shui man am yêrmi ‘I’ll give
you four chickens and a dog’. When ikwa is used to conjoin clauses it is placed
after the gerund of the verb of the preceding clause, and its meaning is ‘as soon as’,
e.g. Ushak ikwa am mol pelauna ‘As soon as [I] arrive [here], I will tell you’
(arrive.GERUND ikwa 2s speak FUTURE.1s).
The gerund is also used to form various kinds of nonfinite subordinate clause.
When followed by ne (the topic marker), the gerundial clause has an adjunct
function, often but not always a time clause, e.g. Ushak ne am mol pelauna ‘When
[I] arrive [here], I will tell you’ (arrive.GERUND TOPIC 2s speak FUTURE.1s).
Other types of subordinate clause, such as complement clauses and some reason
clauses, may have a finite verb, which stands at the end of the clause and is
followed by a subordinator, e.g. Enin una kara pelam ka ‘I heard that you are
going to die’ (i.e. you are dying) (hear.PARTICIPLE COPULA.1s die FUTURE.2s ka),
Unwayêm kali pun una ‘I came because you are ill’ (be.ill.2s kali come.PARTICIPLE
COPULA.1s). Subordinate clauses such as these are free to be placed before or after
the matrix clause as one wishes.
A subordinate volitive clause may express purpose, e.g. Kayu eta u yêrta u wiran
umi ‘Give me a horse so that I may go to my village’ (horse one 1s give-VOLITIVE.2s
1s town go-VOLITIVE.1s). Volitives may also be found in conditional and concessive
clauses. In subordination we encounter “tensed volitives”, that is a periphrastic
TAM construction of the same kind already seen but with the last auxiliary in the
volitive mood, e.g. kara wan ta ne ‘if you had died’ (die COPULA-PARTICIPLE
COPULA.VOLITIVE.2s TOPIC).

28
 Lexicon
8
The documents that provide our corpus are mostly made up of word lists,
generally not very long ones, but between them they contain five or six hundred
distinct vocabulary items, some of which are clear and are attested more than
once, while others are presented in such a way that they need to be reconstructed
and some are only attested once.

From this vocabulary it can be inferred that Monosyllabic (CVC) Bisyllabic (CVCV)
Lenca words are either simple or compound. kê ‘stone’ ama ‘maize’
Primary lexical stems for the most part kên ‘arm’ ili ‘tree’
conform to a monosyllabic, or at the most a kin ‘path’ kashi ‘sun’
(C)VCV phonetic shape, with a few longer lem ‘bed’ kata ‘pot’
mon ‘rabbit’ mina ‘mother’
ones (e.g. CVCVC or CVCCV). Probably the pap ‘father’ mira ‘banana’
monosyllables historically derive from shapes rak ‘flesh’ pala ‘hill’
of the CVCV kind in which the final vowel has sin ‘cup’ sela ‘hammock’
tended to be lost in many but not all cases, ta ‘cornfield’ toro ‘head’
whence the “fluid vowel” phenomenon. tau ‘house’ wala ‘hand’
wak ‘foot’ wara ‘river’

(Class I) (Class II) The range of phonological shapes

Monosyllabic -C Monosyllabic -V Bisyllabic -Vof verbal roots generally follows a
ay- ‘say’ ko- ‘enter’ ira- similar pattern to that of nominal
‘sow’
en- ‘hear’ la- ‘estar’ kara- ‘fall, die’
roots, with the relationship
ish- ‘see’ mu- ‘have’ kolko- ‘open’
between monosyllabic and
kor- ‘eat’ pu- ‘come’ kopi- ‘close’
kos- ‘go out’ te- ‘make’ liwa- bisyllabic roots made more evident
‘buy’
mol- ‘speak’ u- ‘go’ niwa- by the fact that an i is added (or
‘learn’
tan- ‘sit’ wa- ‘be’ restored) to consonantal (Class I)
shapa- ‘find’
stems in certain inflected forms (mol ‘speak’, gerund molik, participle molin, 1s
moliuna; cf. liwa ‘buy’, gerund liwak, participle liwan, 1s liwauna). Verb stems
with more complex phonological structures than these are usually compound,
derived or borrowed.

Lenca has its share of loan words. Borrowing in Borrowings < Sp. Spanish
different periods is evidenced by the presence of kashlan ‘hen’ *castellana
words which can be identified as loans from another kayu ‘horse’ caballo
indigenous language or from Spanish. Loans from kushtal ‘sack’ costal
indigenous languages in Lenca sometimes undergo leshe ‘milk’ leche
patush ‘duck’ patos
native derivation, e.g. tet ‘work (n.)’ > Nawat te[ɣ]it → shupu ‘liquor’ *chupo
tet-wa-k ‘to work’ formed with Lenca affixes. (The wakash ‘cow’ vacas
earliest form of the loan in Lenca was probably *teiti,

29
since tetwak has a variant teitiwak; in accordance with Lenca historical
phonology, the loss of the interconsonantal i (> *teitwa-) would have led to
obligatory levelling of the diphthong to avoid a closed syllable containing a
diphthong (> tetwa-).) A pronunciation of tetwak is attested with voicing of the
internal /t/, a sporadic Lenca phonological development: te[d]wak.
In nominal compounds the qualifying component precedes, as in ketau ‘cave’ (cf.
kê ‘stone’, tau ‘house’).
Compound noun ← Components
amasarin ‘grain of maize’ ama ‘maize’ sarin ‘seed’
inshuk ‘beard’ in ‘mouth’ shuk ‘body hair’
ketau ‘cave’ kê ‘stone’ tau ‘house’
kuishoror ‘hurricane’ kui ‘rain’ shoror ‘wind’
waktik ‘sandal’ wak ‘foot’ tik ‘face’
walalasel ‘finger’ wala ‘hand’ lasel ‘digit’
washkata ‘pitcher’ wash ‘water’ kata ‘pot’
wemap ‘girl’ we(we) ‘child’ map ‘woman’
yukanta ‘pine grove’ yukan ‘pine’ ta ‘(corn)field’

Compound verbs are less common. The verbs kolkok ‘open’ and kopik ‘close’ may
take in- before them: inkolkok, inkopik; this is the noun in ‘mouth’ (which also
compounds with nouns, cf. inshuk above), and these are exactly parallel with the
Nawat verbs (ten)tapua and (ten)tzakwa, where ten ‘mouth’ denotes an edge or opening.
However, such formations, which are commonplace in Nawat, seem to be rare in Lenca.
Diminutive nouns are formed by adding one of the endings -(i)skan, -(i)shkan or
-(i)nka (weskan ‘small child’, cf. we(we) ‘child’). The same occur with other words
(e.g. adjectives, quantifiers) except that -(i)s(h)ka does not end in n (shuntiska or
poriska ‘little’, cf. shun ‘a little’, pori ‘small’; sheulinka ‘reddish, orange, yellow’, cf.
sheula ‘red’).

The participle (ending in -n) and the Base word Derived verb
gerund (in -k) of verbs may sometimes kosik ‘go out’ koskik ‘remove’
tishik ‘know’ tishkik ‘teach’
be used as derived nouns. The lak ‘be, estar’ lankik ‘remain’
participle-derived noun, which is more mam ‘feces’ mankik ‘defecate’
common, designates a participant in the arik ‘burn (intr.)’ arshik ‘burn (tr.)’
“action”, sometimes objective (a patient kulik ‘copulate’ kulshik ‘plough’
or result) and other times subjective (an lumak ‘shake (intr.)’ lumshik ‘shake (tr.)’
agent), and occasionally an instrument, karak ‘fall, die’ kashik ‘kill’ (?)
e.g. washan ‘urine’, cf. washak ‘to lip ‘lightning’ lipshik ‘flash’
shê ‘good’ sheshik ‘get better’
urinate’; tishin ‘doctor’, cf. tishik ‘to
pashik ‘finish (intr.)’ pashak ‘destroy’
know’; wakin ‘grindstone’, cf. wakik talik ‘drink’ talak ‘get drunk (?)’
‘to grind’. A gerund is sometimes wash ‘water’ washak ‘urinate’
nominalized as a designation of an kui ‘rain (n.)’ kuiwak ‘rain (v.)’
action or the outcome or objectivication teli ‘evening’ telwak ‘be/get late’
of the action (sherak ‘rent’). tet ‘work (n.)’ tetwak ‘work (v.)’

Another deverbal noun-forming suffix is -(k)in, which can be used to derive nouns
bearing a variety of relations to the meaning of the source verb, usually objects or
concepts, less frequently people (korkin ‘food’, cf. korik ‘to eat’; purkin ‘broom’, cf.
purik ‘to sweep’; talkin ‘drunkard’, cf. talik ‘to drink’). These may be termed verbal
nouns.

30
A number of verb-forming suffixes or “extensions” are observed in the Lenca lexicon.
The syllables -ki- and -shi- added to the root of primary verb stems sometimes increase
their valency (producing a causative: koskik ‘remove’, cf. kosik ‘go out’; tishkik
‘teach’, cf. tishik ‘know’; arshik ‘burn (tr.)’, cf. arik ‘burn (intr.)’; kulshik ‘plough’,
cf. kulik ‘*roll over (intr.)’ > ‘copulate’; lumshik ‘shake (tr.)’, cf. lumak ‘shake
(intr.)’; probably also kashik ‘kill’, cf. karak ‘die’), though other times they seem to
modify the meaning in an almost random way without necessarily affecting valency
(lankik ‘remain’, cf. lak ‘estar’). These suffixes can also derive verbs from nouns
(lipshik ‘flash’, cf. lip ‘lightning’; mankik ‘defecate’, cf. mam ‘feces’; sheshik ‘get
better’, cf. shê ‘good’).
For valency reduction, there is some evidence to suggest the existence of a reciprocal-
reflexive-mediopassive verbal derivative suffix -ay-. There is also some evidence to
suggest that an alternation between i and a in the root-final vowel can also signal a
transitivity alternation (pashik ‘finish (intr.)’  pashak ‘destroy’) or a semantic
modification (koskik ‘remove’, see above  koskak ‘hunt’; talik ‘drink’  talak ‘get
drunk’). This area requires further study.
Perhaps -a- is also a denominative verb-deriving suffix in washak ‘urinate’, cf. wash
‘water’. But the most productive denominal suffix may be -wa- (kuiwak ‘rain (v.), cf.
kui ‘rain (n.)’; telwak ‘be/get late’, cf. teli ‘evening’). The homonymy between the
suffix -wa- and the copula wa- results in some doubts: is cashiguaina ‘ya amaneció’ to
be parsed as kashi waina (day COPULA.3s) or kashiwaina (day.DERIV.3s)? In addition
to deriving verbs from Lenca nouns, -wa- can derive Lenca verbs from borrowed words,
as in tetwak ‘to work’ (from a Nawat loanword) and lamarwak ‘to embrace’ if this
comes from Sp. amar, as is probable.
However, the most usual and productive way to turn a Spanish verb into a Lenca one is
by combining the Spanish infinitive with the verb tek ‘make, do’ which functions as a
light verb and may be fully conjugated, e.g. creer am teuna ‘I believe you’ (believe 2s
make.1s).

31
 Sample
9 source
text

In this chapter I will take a sample of continuous conversational text in Honduran

Lenca as noted in the late nineteenth century. Proceeding line by line, I shall first
present the sentence in its original spelling in the source together with the
translation provided in the source (in Spanish). On a second line, I rewrite the
Lenca sentence as I reconstruct it following the conventions proposed for Modern
Unified Lenca (LUM), accompanied by an idiomatic translation (in English) of the
reconstructed meaning (which may differ slightly from that in the source
document. On a third line, I provide a morpheme-by-morpheme analysis, followed
by a literal morphological gloss.
The source of this text is Alberto Membreño, Hondureñismos. Vocabulario de los
provincialismos de Honduras, Tegucigalpa, 1897.

(Line)
[2] U-shagne enin uná amnan unguac Cuando vine supe que Ud. estuvo de muerte
carabelamga. de su enfermedad.
Ushak ne enin una amnan unwak When I arrived I heard that you were ill
kara pelam ka. and about to die.
usha-k ne eni-n una am-nan un-wa-k kara pela-m ka
arrive.here-GERUND TOPIC hear-PARTICIPLE COPULA.1s 2s-PRONOUN illness-DERIV-GERUND die
PROSPECTIVE-2s COMPLEMENTZR

[3] Quintegh pug ú ishteguantami am Y cómo no fuiste a verme: yo te esperé

tinguishinuna magüen. mucho.
[a] Kin tek puk u ish tê wan tami? How come you didn’t come and see me?
Kin te-k pu-k u ish tê wa-n tam?
which do-GERUND come-GERUND 1s see NEG COPULA-PARTICIPLE COPULA.DUB.2s
[b] Am tinkishin una mawen! I waited for you a long time!
Am tinki-shi-n una mawen
2s wait?-DERIV-PARTICIPLE COPULA.1s much

32
[6] Magüen ú tuniamá ú-familia, neuly Siento mucho, lo mismo como mi familia,
am yun ab isteguanbilga inian no haberte visto en tus necesidades; pero
pulauná cariari am ofrecer-tiba hoy vine á ofrecerte algún servicio.
ishta.
[a] Mawen u tunina u familia ne u li My family and I are very sorry that we
ami un ap ish tê wan pil ka, didn’t see you in your illness,
mawen u tun-ina u familia ne u li am un ap ish tê wa-n pil ka
much 1s hurt-3s 1s family TOPIC 1s by 2s illness in/to see NEG COPULA-PARTICIPLE
COPULA.1p COMPLEMENTZR
[b] inê puk launa kariari am ofrecer te therefore I am coming for you to see
para ishta. whether anything may [be] offer[ed] you.
inê pu-k la-una kariari am ofrecer te para ish-ta
so come-GERUND estar-1s anything 2s offer do PROSPECTIVE.SUBORDNTR see-VOLITIVE.2s

[7] Norane ofrecer utetena, olonguayu Hasta hoy no se me ofrece nada, porque
toni mulaunà; sheshliara-nan-ñe estoy provisto aún de remedios: sólo que
cayu eta u-yarta u-güiran humí. en mi restablecimiento necesito una bestia
para conducirme a mi pueblo.
[a] Nora[p] ne ofrecer u te tê ina, I don’t need anything now,
nora-p ne ofrecer u te tê ina
this.time-in/to topic offer 1s do neg 3s
[b] olonwayu toni muk launa; I have enough medicine;
olonwayu toni mu-k la-una
medicine enough have-gerund estar-1s
[c] sheshiara anap ne kayu eta u yêrta [but] when [I] get better lend me a horse so
u wiran umi. I can go to my village.
shê-shi-ara anap ne kayu eta u yêr-ta u wiran u-mi
good-deriv-subord anaph topic horse one 1s give-volitive.2s 1s town go-volitive.1s

33
 Sample
10 teaching
dialogue

The following is an excerpt from a dialogue written as study material as part of a

coursebook for learners of Lenca.

Ah, Diana, pun yêm! Ah, Diana, you have come!

ah Diana pu-n yêm ah Diana come-PARTICIPLE COPULA.2s

Kap lan tam? Where have you been?

kap la-n tam where estar-PARTICIPLE COPULA.2s

Telwaina! It’s late!

tel-wa-ina late-DERIV-3s

Inê, tishiuna telwaina ka. Yes, I know that it’s late.

inê tishi-una tel-wa-ina ka so/yes know-1s late-DERIV-3s COMPLEMTZR

Kunan ti na? Who is this?

kunan ti na who COPULA.DUB.3s this

Nanan u seya ina. This is my younger brother.

na-nan u seya ina this-PRONOUN 1s younger.brother COPULA.3s

Shê latam? How are you? (lit. Are you well?)

shê la-tam good estar-DUB.2s

Shê launa. I am well.

shê la-una good estar-1s

Amnan kotik niwaptam Are you going to learn Lenca also?

anaman?
am-nan kotik niwa-p-tam ana-man 2s-PRONOUN language learn-FUTURE-DUB.2s that-with

34
Niwami! I want to learn!
niwa-mi learn-VOLITIVE.1s

U latan nan Carlos ina. My name is Carlos.

u latan nan Carlos ina 1s name DEF Carlos COPULA.3s

Shê laina, Carlos. Okay, Carlos.

shê la-ina, Carlos good estar-3s, Carlos

Ishiuna kotik shuntê niwan yêm I see that you have [already] learnt a little Lenca!
ka!
ishi-una kotik shun-tê niwa-n yêm ka see-1s language little-APPROX learn-PARTICIPLE COPULA.2s
COMPLEMENTZR

35
 Elementary Vocabulary
Lenca — Español
inê sí, así
A
irak sembrar
a y (posp.)
isis diez
am tu, te
ishaipepil nos veremos
ama maíz
ishik ver
amnan tú
ishkik mostrar
ana aquel, -llo
anaman también K
anap allá ka que
ap en, a kalapa nueve
api nuestro kali porque
apinan nosotros kanashti canasta
apishaya despacio kap dónde
arshik quemar kar qué
aulak llegar (allá) karak caer, morir
awinki venado kariai por qué
ayak bajar karman cuándo
ayik decir kashi día
kashlan gallina
B
kasi largo
bus autobús
kasip lejos
C kata olla
cafe café kau cacao
castellano castellano, español kaurak caminar
clase clase kayu caballo
E kê piedra
enik oir, escuchar kelik asar, tostar
eria cuatro kelkin comal
eshara trapo kên pierna
eta un, -a kenin brazo
eye tortilla kepan peña
ketau cueva
G kil sois
galan galán kin camino
H kin’ hora’p ¿a qué hora?
hora hora kina ¿cuál?
hora mokta’p siempre kinê cómo
kok entrar
I kolkok abrir
ili árbol kopik cerrar
ilik correr kopikopi mariposa
in tê no hay korik comer
ina1 ese, eso korkin comida
ina2 es (ver wak)
inan él, ella
inap ahí

36
korok vestirse mirin verdad
koron camisa, ropa misak robar
kosa bueno misi gato
kosik salir mishu hijo
koskak cazar miti frío
kotan monte mokta todo, -s
kotik lengua, lenca molik hablar
ku-eta cuarenta molshik leer
ku-pe ochenta mon conejo
kulshik arar muk tener
kumshik llenar mukuk llevar
kunan quién
N
kupik atar
na este, esto
kushik poner
nam desde, de (posp.)
kushtal saco, costal
nan el, la, los, las
kutek cortar (fruta)
nap aquí
kuto corto
napir cerdo
L nawe chile
lak estar ne acerca de, marcador de
lan con (instrumento) tópico
(posp.) niwak aprender
lankik quedarse niwan estudiante
lashayik acordarse norap ahora
latan nombre noraptê ahorita
latik vacío
O
lawa tres
olon medicina
layik llamar
lem cama, tapesco P
lepa tigre pala cerro
leshe leche palan pecho, mama
li y (posp.) pap padre
libro libro parik quitar
liwak comprar pashak destruir
lok jolote pashik acabarse
loron sombrero patush pato
lum tierra paulak cansarse
lumak menearse pe dos
pelak auxiliar futuro
M
pero pero
mailep antes, primero
pesha hija
mal ¡vamos a...! (= umal)
piak acostarse
man con (posp.)
pik vender
manipê cuesta
polko redondo
manishik soñar
pori pequeño
mashik golpear
poro hoja(s)
mashti machete
puk venir
mawen mucho
puki grande
merik romper
pulranka tarro
mes mes
purik barrer; volver
mesa mesa
purkin escoba
mina madre
mira plátano

37
 shololo alto
R
shua coyote
rak carne
shui perro
rapa dulce
shulu pollito
ripak brincar
shun poco
rok levantarse
shuna flor
ruwa duro
shuntê un poco (de)
S shuntiska pequeño
sai cinco shupak chupar (alcohol)
sarik dormir; tapiscar shupu guaro, chicha
sawik lavar shupuk rápido
sêk pagar shupuk-shupuk recio, fuerte
sela hamaca shur ave
semana semana shurishur ardilla
sepe sal shusha ratón
seya hermano menor
T
sin taza
ta milpa
sinka azul
tak1 ¿cuánto(s)?
sira abeja
tak2 plantar
sisi hormiga
talik beber
siwa caliente
tan suelo
siwan fruta
tanik sentarse
soi rana
taposhi cangrejo
sulik soltar
tashik botar, tirar
sunak despertarse
tashu anciano
SH tau casa
shak leña tayik cortar
shakanik platicar tê no
shala1 cuello teipan iglesia, templo
shala2 lago tek hacer
shalala culebra teka valle, llano
shali huevo telwak tardar, ser tarde
shapa barranco telwan ayer
shapak encontrar tepa ancho
shapon jabón tepere bajo
shawa1 mañana tepka ocho
shawa2 seco têrak olvidar
shê bueno, bien tesh tortuga
shêlinka bonito, lindo teta abuela
sheula rojo tetwak trabajar
sheulinka amarillo têyik pedir
shêyik querer tinkishik esperar
shika verde tipik subir
shina papel tishik saber, conocer
shinak frijoles tishkik enseñar
shiri negro tishkin profesor, -a
shirishir mosca toko arena
shishi agrio tolo abuelo
shit año toro cabeza
shiwik escribir toto bebé
shoko blanco towi malo, defectuoso

38
tuak bañarse wash agua
tukik meter, enterrar washkata cántaro
tumin dinero wayik llorar
tunik doler wei a, para, de (posp.)
tupan metate weri barriga
wesuk centro
U
wewe niño, -a
u mi
wie seis
uk ir
wira bravo
ulik bailar
wiran pueblo
umshik pastorear
wirik reñir, pelear
unan yo
wiska siete
unwak estar enfermo
wishik soplar
ushak llegar (allá)
uwa tabaco Y
yaku armadillo
W
yaru mono
wak1 ser
yawa hacha
wak2 pie
yek cuerpo
wakash vaca, ganado
yêrik dar
wakik moler
yeshonak pensar, contemplar
wala mano
yolik alegrarse, reirse
wamasta veinte
Yosh n’am sêyu gracias
wanko banco
yuk espalda
wanto después, luego
yuka fuego
wara río

39
 Elementary Vocabulary
Español — Lenca
bebé toto
A
beber talik
a wei, ap (posp.)
bien shê
abeja sira
blanco shoko
abrir kolkok
bonito shêlinka
abuela teta
botar tashik
abuelo tolo
bravo wira
acabarse pashik
brazo kenin
acerca de ne (posp.)
brincar ripak
acordarse lashayik
bueno shê, kosa
acostarse piak
bus bus
agua wash
ahí inap C
ahora norap caballo kayu
alegrarse yolik cabeza toro
alto shololo cacao kau
allá anap caer karak
amarillo sheulinka café cafe
anciano tashu caliente siwa
ancho tepa cama lem
antes mailep caminar kaurak
año shit camino kin
aprender niwak camisa koron
aquel, -llo ana canasta kanashti
aquí nap cangrejo taposhi
árbol ili cansarse paulak
ardilla shurishur cántaro washkata
arena toko carne rak
armadillo yaku casa tau
arar kulshik castellano castellano
asar kelik cazar koskak
así inê centro wesuk
atar kupik cerdo napir
autobús bus cerrar kopik
ave shur cerro pala
ayer telwan clase clase
azul sinka comal kelkin
comer korik
B
comida korkin
bailar ulik
cómo kinê
bajar ayak
comprar liwak
bajo tepere
con man, lan (posp.)
banco wanko
conejo mon
bañarse tuak
conocer tishik
barranco shapa
contemplar yeshonak
barrer purik
correr ilik
barriga weri

40
cortar tayik este, esto na
cortar (fruta) kutek estudiante niwan
corto kuto estudiar niwak
coyote shua
F
cuál kina, kinanan
flor shuna
cuándo karman
frijoles shinak
cuánto(s) tak
frío miti
cuarenta ku-eta
fruta siwan
cuatro eria
fuego yuka
cuello shala
fuerte shupuk-shupuk
cuerpo yek
cuesta manipê G
cueva ketau galán galan
culebra shalala gallina kashlan
chile nawe ganado wakash
chupar (alcohol) shupak gato misi
golpear mashik
D
gracias Yosh n’am sêyu
dar yêrik
grande puki
de nam (posp.), i, wei
guaro shupu
decir ayik
defectuoso towi H
desde nam hablar molik
despacio apishaya hacer tek
despertarse sunak hacha yawa
después wanto hamaca sela
destruir pashak hija pesha
día kashi hijo mishu
diez isis hoja(s) poro
dinero tumin hormiga sisi
doler tunik huevo shali
dónde kap I
dormir sarik iglesia teipan
dos pe ir uk
dulce rapa
J
E jolote lok
él, ella inan
el, la, los, las nan L
ellos, -as alnan lago shala
en ap largo kasi
encontrar shapak lavar sawik
enfermo, estar unwak leche leshe
enseñar tishkik leer molshik
enterrar tukik lejos kasip
entrar kok lenca (lengua) Kotik
escoba purkin lengua kotik
escuchar enik leña shak
ese, eso ina libro libro
espalda yuk lindo shêlinka
español castellano llamar layik
esperar tinkishik llano teka
estar lak llegar aulak, ushak

41
llenar kumshik para wei (posp.)
llevar mukuk papel shina
llorar wayik pastorear umshik
lo, la i pato patush
los, las, les al pecho palan
luego wanto pedir têyik
pelear wirik
M
pensar yeshonak
machete mashti
peña kepan
madre mina
pequeño pori, shuntiska
maíz ama
pero pero
malo towi
perro shui
mano wala
pie wak
mañana shawa
piedra kê
mariposa kopikopi
pierna kên
me u
plantar tak
medicina olon
plátano mira
menearse lumak
poco shun
mes mes
pollito shulu
mesa mesa
pollo kashlan
metate tupan
poner kushik
meter tukik
por qué kariai
mi u
porque kali
milpa ta
primero mailep (adv.)
moler wakik
profesor, -a tishkin
mono yaru
pueblo wiran
monte kotan
morir karak Q
mosca shirishir que ka
mostrar ishkik qué kar
mucho mawen quedarse lankik
quemar arshik
N
querer shêyik
negro shiri
quién kunan
niño, -a wewe
quitar parik
no tê
no hay in tê R
nombre latan rana soi
nos api rápido shupuk
nos veremos ishaipepil ratón shusha
nosotros apinan recio shupuk-shupuk
nueve kalapa redondo polko
nuestro api reirse yolik
reñir wirik
O
río wara
ochenta ku-pe
robar misak
ocho tepka
rojo sheula
oir, escuchar enik
romper merik
olla kata
ropa koron
olvidar têrak
os aki S
saber tishik
P
saco kushtal
padre pap

42
sal sepe tortuga tesh
salir kosik tostar kelik
seis wie trabajar tetwak
semana semana trapo eshara
sembrar irak tres lawa
sentarse tanik tu am
ser wak tú amnan
sí inê
U
siempre hora mokta’p
un, -a eta
siete wiska
usted amnan
soltar sulik
ustedes akinan
sombrero loron
soñar manishik V
soplar wishik vaca wakash
subir tipik vacío latik
suelo tan valle teka
¡vamos a...! umal, mal
T
veinte wamasta
tabaco uwa
venado awinki
también anaman
vender pik
tapesco lem
venir puk
tardar telwak
ver ishik
tarro pulranka
verdad mirin
taza sin
verde shika
te am
vestirse korok
templo teipan
volver purik
tener muk
vosotros akinan
tierra lum
vuestro aki
tigre lepa
tirar tashik Y
todo, -s mokta y a, man, li (posp.)
tortilla eye yo unan

43
 Bibliography

Click on the arrow symbols to download scanned copies of the items listed.

⇩ Andrews, E.W. “Correspondencias fonológicas entre el lenca y una lengua

mayance.” UNAM, 1970.
⇩ Campbell, Lyle, Anne Chapman & Karen Dakin, “Honduran Lenca”, 1978, in
International Journal of American Linguistics 44, p. 330 ff.
⇩ Doblado, L. & Rafael Girard, Vocabularios lencas en Honduras (Anales del
Museo Nacional David J. Guzmán 6, San Salvador), 1951.
⇩ Gavarrete, E. Nataka Opalaca (Aprendamos lenca), 2003.
⇩ Hernández, Eusebio & A.-L. Pinart, Pequeño vocabulario de la lengua lenca
(Dialecto de Guajiquiro), Paris, 1897.
⇩ Herranz Herranz, Atanasio, El lenca de Honduras: una lengua moribunda
(Mesoamérica 14), 1987.
⇩ King, Alan R. “Preliminary report on Honduran Lenca” (December 2015).
⇩ King, Alan R. “A sentential marker for Guajiquiro Lenca?” (December 2015).
⇩ King, Alan R. “Lenca verbal morphology” (December 2015).
⇩ King, Alan R. “A preliminary proposal of unified Honduran Lenca” (August
2016).
⇩ King, Alan R. “Lenca dialect profile: Similaton” (September 2016).
⇩ King, Alan R. “Some puzzles in Lenca reconstruction” (November 2016).
⇩ King, Alan R. “Lenca dialect profile: Opatoro” (November 2016).
⇩ King, Alan R. “Lenca dialect profile: Santa Elena” (November 2016).
⇩ King, Alan R. "Subordination in Honduran Lenca" (January 2017).
⇩ ⇩ ⇩ Lardé, J. & Larín, B., “Diccionario lenca-español”. [1925]/1951.
⇩ Lehmann, W. Zentral-Amerika. Teil I. Die Sprachen Zentral-Amerikas in ihrer
Beziehung zueinander sowie zu Süd-Amerika und Mexiko. 1920.
[“Lenca-Dialekte”, pp. 668-722.]
⇩ Membreño, Alberto, Hondureñismos. Vocabulario de los provincialismos de
Honduras, Tegucigalpa, 1897.
⇩ Moreno Maldonado, Adela, “Apuntes del dialecto que usaron nuestros
aborígenes centroamericanos y que a la vez lo hablan varios
viejecitos del Municipio de Guajiquiro, Departamento de La Paz”,
1924, en Conrado Bonilla, Honduras en el pasado, 1949.
⇩ Peccorini, A. “Dialecte Chilanga.” Ed. by Lehmann, in Lehmann (1920).
⇩ Squier, Ephraim George, The states of Central America, 1858, pp. 253-5.

Versión 27/06/2017

44