Grammar of the Xinkan languages

I have written about various aspects of the Xinkan languages elsewhere and will not repeat those analyses here (Rogers 2022; 2016a; 2020; 2018; 2014; 2010; 2016b; Rogers & Hamp 2020; Rogers 2019). In large part, my thinking about the Xinkan grammatical system(s) is the same as represented in that research with a few exceptions. For ease, here I present the orthographic system I use for the Xinka languages and a sketch of my current thinking about word categories in the Xinkan languages, all other information can be found in my other publications and in the work produced by Lyle Campbell or Frauke Sachse (for example, Sachse 2010; 2018; Campbell 1998; 1973; 1975; Campbell, Kaufman & Smith-Stark 1986; Campbell & Muntzel 1989).

Orthography

The orthographic system I use for the Xinkan languages follows closely that proposed by the Academia de Lenguas Mayas de Guatemala. Punctuation and capitalization (where used) follow the Spanish language orthographic tradition. Table 1 lists correspondences between the letters used and their counterparts in the International Phonetic Alphabet – but note that as discussed in (Rogers 2022) the glottalized sounds (e.g., pʼ, tʼ, and kʼ) are more likely to be laryngealized rather than ejective or implosive.

Table 1. Xinkan orthography

Word Categories

Different from the analysis I presented in (Rogers 2016a), this dictionary is based on the idea that word categories (such as nouns, verbs, adjectives, etc.) are only functionally necessary in the Xinkan languages and are, in and of themselves, not necessary for the underlying/psychological/inferred system used by the Xinkan speakers. For example, any lexical item can be used fluidly as a noun or a verb depending on the choice of the bound morphology used and not because of the inherent semantics of the stem. However, there does seem to be a necessary distinction between referents and predicates in that some roots appear to be used more frequently as one or the other without bound morphology (as their default function). Consequently, I see this distinction as essential for the description of the Xinkan grammatical system.

Referents are those morphs that refer to tangible or intangible real-world experiences (e.g., dog, corn, and love) while predicates are those morphs that are real-world experiences (referents) but which can be used to further delimit the reference of referents; that is they have some form of argument structure (e.g., eat, be tall, and green). In the Xinkan languages predicates link to semantic macro-arguments such as actors, undergoers, and instruments rather than grammatical arguments such as subject, direct object, or indirect object.

In the dictionary, I use traditional labels, such as transitive verb or intransitive verb,  for various types of predicates for purposes of crosslinguistic comparison but it is important to note that the labels do not represent the actual function of the predicate or the organization of the grammatical system. In fact, I now consider Xinkan roots to be organized largely without specific categories (contra Rogers 2016; see Hieber (2021) for information about precategoriality).

Furthermore, the form of any given referent or predicate is determined by three pragmatic parameters: (1) Completeness, (2) Predicate argument saliency, and (3) Referent activation.

Completeness

Completeness refers to the real-world completion of a specific relationship for either referents or predicates. For example, completely eating something or possessing something completely or absolutely are both examples of not being able to do more of the specified relationship (eating or possessing cannot be interrupted or halted). In both cases, it is not the action but the relationship between elements that is complete.

In Xinkan, referents and predicates are both marked for completeness of their relationships in the same way. Complete relationships are indicated through suffixation and incomplete relationships are indicated through prefixation. However different paradigms are used for either predicates (i.e., verbs) or referents (i.e., nouns). These different paradigms are largely identical but are distinguished in the third-person forms. In Rogers (2016) I referred to these as completive/incompletive aspects for the predicates and alienable/inalienable possession for the referents.

Predicate argument saliency

The saliency of the predicate arguments is an important part of the Xinkan grammatical system. By saliency, I mean that some semantic argument of the predicate is more prominent or important in a given utterance. Four types of saliency are marked in the Xinkan languages.

1. Neutral salience (actor and undergoer are equally salient). In the dictionary, I label these as transitive verbs since they most often have two arguments but note that some of these only have a single argument. They are formally unmarked.
2. Actor salient. In the dictionary, I label these actor intransitives since they only have one argument and this must always be the actor. They are marked by the suffix -lhaʔ in the completive aspect.
3. Undergoer salient. In the dictionary, I label these undergoer intransitives since they only have one argument and this must always be the undergoer. They are always marked by the suffix -ʔ in the completive aspect. They are also further marked by long vowels in an initial syllable in the completive aspect (see Rogers 2016).
4. Predicate salient (actor and undergoer are equally non-salient). In the dictionary, I label them as predicate intransitives since they always require only one argument (whether it is an undergoer or an actor). They are always marked by the suffix -kʼi. This suffix can be used with predicate or referent stems, when used with predicate stems the meaning is something like “doing X” and when use with referents the meaning is something like “becoming or doing things like an X” where X is the stem the suffix is attached to.

Referent activation

Referent activation is only recognizable in Xinkan sentences and so is not represented in the lexical organization depicted in the dictionary. However, since many entries include example sentences, I briefly mention referent activation.

Clausal word order in the Xinkan languages is pragmatically determined. The clausal predicate is the most important element and the position of the referents depends on their semantic role and pragmatic function. Four pragmatic functions are important:

1. activated referent – meaning the referent that is the center of consciousness
2. focused referent – meaning the referent is being introduced into the participant’s consciousness
3. non-activated referent – meaning the referent is available in the shared common ground of the conversation but is not the center of consciousness
4. neutral referent – meaning that the referent is semantically necessary but pragmatically unimportant (in the Xinkan languages this is almost always an underoger when an actor is present).

The word order template according to these pragmatic functions is:

activated – (negation) – focus – verb – (neutral) – non-activated

Deriving predicates from referents

It is possible for those stems which default as referents to be used as predicates. The most common way for this to happen is to add the predicate saliency suffix -kʼi discussed above. Though, much less commonly, it is possible to simply use the predicate affixes on any referent.

Deriving referents from predicates

It is also possible to use a stem that defaults as a predicate as a referent. There are many ways of doing this.

• An abstract referent is formed from predicates by simply using the incompletive referential prefixes on the stem. These are labeled ‘verbal nouns’ in the dictionary.
• An agent referent is derived from predicates by adding the suffix -Vlha to the stem. These are labeled ‘agent nouns’ in the dictionary.
• There are two ways for a patient referent to be derived from predicates. The first is done by adding the suffix -wa to the stem. The second is only possible for some predicates where the undergoer salient incompletive form has become frozen as a referent, e.g., axuka’ ‘food’ [lit. it is eaten]. Both of these types of derivations are labeled ‘patient nouns’ in the dictionary.
• An instrument referent is derived from predicates by adding the suffix -k to the stem. These are labeled ‘instrument nouns’ in the dictionary.

Causation

Predicates and referents can also be used as causative predicates by the addition of one of four possible causative suffixes, as described in Rogers (2016:118). These are labeled causative transitive verbs in the dictionary.

All of this results in a fluid network of lexical items and a lot of homophony and indeterminacy. One morph can have many forms depending on the functions it is put to in a sentence. For example, in Guazacapan Xinka the following are possible forms of the lexeme //waya// ‘work in cornfield’ with italics in the gloss showing argument saliency:

• Functions of the lexeme //waya// ‘work in cornfield’

Organization of the dictionary

This dictionary has three parts a comparative dictionary, Spanish to Xinkan lookup index, and English to Xinkan lookup index. The format for the entries in each part are as follows:

• Entry format for the comparative Xinkan dictionary

Comparative Headword [prounciation] (CX : GX : JX : YX) {root} cf. cross reference

            POS || Definition [SOURCE] (Scientific Name)

• example sentence translation

notes and comments

etymology

derived forms

• Entry form at for the Spanish to Xinkan lookup index

Spanish gloss Xinkan form [POS]

• Entry format for the English to Xinkan lookup index

English gloss Xinkan form [POS]

Note that not all elements of the comparative dictionary entry template are used for all entries in the dictionary (dependent on available information). The pronunciation is always given using the International Phonetic Alphabet. The Yupiltepeque Xinka forms all come from Calderón (1908) and Lehmann (1920). Part-of-speech (POS), definitions, example sentence translations, and notes and comments are given first in Spanish (preceded by es) and second in English (preceded by en). Definition sources provide the box and page number corresponding to the Xinkan information recorded by Terrence Kaufman and Lyle Campbell as organized in AILLA, preceded by fn (fieldnotes). Etymologies include sources to locations where they have been mentioned previously, but I have relied heavily on Sachse (2018) as the main source for these (Hull 2005; Campbell & Kaufman 1976; Barrera Vásquez, Bastarrachea Manzano & Brito Sansores 1980; Kaufman 2003; Campbell 1978; 1972; 1998; 1971 inter alia). Most of the suggested etymologies require more discussion and many of them are not reliable semantically or grammatically, but I have included them in an effort to be as thorough as possible and to make the information available to others who might find them useful.

Additionally, the definition is always as specific as the information available to me. I have included specific names for animals, insects, fish, and plants where I could determine what was being indicated, but occasionally I simply define them as plant species or type of insect. It is unfortunate that we may never know more about these specific meanings.

There is more information in Terry’s and Lyle’s Xinkan fieldnotes that are not included in this dictionary. These omissions are choices based on the way I understand the available Xinkan information and grammar and fall into three groups.

1. I have not included all the example sentences recorded for any of the Xinkan languages but have opted only for representativeness where possible. The rest can be found in the original fieldnotes in AILLA and glossed at (https://languageconservation.org/index.php/projects/xinkan). I have not included any example sentences from Yupiltepeque Xinka in the dictionary, though these are in the database the dictionary is based on, but these can all be found in Calderón (1908) and Lehmann (1920).
2. I have not included information directly from Maldonado de Matos (Maldonado de Matos 1770) since this is available in Sachse’s work (Sachse 2010) and have not included any information from the so-called zeeje manuscript since this is available in Rogers (2020).
3. I have not included all the lexical items recorded in Terry and Lyle’s fieldnotes. Specifically, this dictionary does not list all the transparent loans from Spanish used by the Xinkan speakers. For example, the fieldnotes list ebra from Spanish hebra [ebra] as a Xinka lexeme, but because of the social pressure to switch to Spanish early on during the colonial period (see Roges 2020 for more discussion of this) and fluency of the Xinkan speakers represented in the fieldnotes, it is unclear if these are loanwords, replacements for forgotten words, or the speakers simply speaking Spanish. The dictionary does contain loans from Spanish or other Mesoamerican languages where an interesting change in pronunciation occurred (all of the potential loanwords, even those not included in this dictionary, are in the Xinkan fieldnotes and in Sachse 2010 and 2018).
4. Lastly, I have not included phrases that are transparent syntactically (i.e., their meanings are determined by the syntactic relationships of their parts) and which are not lexical in nature. This includes phrases such as humu kawayu ‘stud, stallion’ (lit. male horse) where the meaning is simply a modified noun. Though I have included phrases that are more idiomatic, such as ut’ah pu ‘thumb’ (lit. mother of the hand), even though these also adhere to regular morphosyntactic patterns.

Speaker variation

The Xinkan materials this dictionary is based on contain significant amounts of inter- and intra-speaker variation. I have opted to not hide that variation or to produce an overly analyzed standard (such as fixing obvious errors in consonant voicing or glottalization) since these represent an important part of the legacy of the Xinkan speakers. This variation shows that the language is adaptable, and fluid and that to be a speaker it is less important that you follow a prescribed standard and more important that you use the language. I sincerely hope many contemporary Xinkas will benefit from this legacy.