The acquisition of causatives in Q’anjob’al Maya

Abstract

Causatives have received considerable attention in first language acquisition. Of Mayan languages, acquisition of the causative has only been investigated for K’iche’ and Tzotzil, based on longitudinal and spontaneous data. K’iche’-speaking children do not acquire morphological causatives until the age of 3 years, while children acquiring Tzotzil start producing morphological causatives around the age of 2 years. The marked difference in the age of acquisition of causatives in K’iche’ and in Tzotzil has been explained through a morphological difference between causatives in the two languages. This paper, based on longitudinal and spontaneous data, examines acquisition of the causative in Q’anjob’al, a third Mayan language. The question is whether the findings in K’iche’ and Tzotzil are reproduced, or whether the acquisition of Q’anjob’al causatives follows a third, yet-unattested, trajectory.

The results show that three Q’anjob’al children, of the age range 1;9-3;0, 2;3-4;0, and 2;7-3;6, acquire the morphological and periphrastic causatives during the third year of life, although these children produce more periphrastic causatives than morphological causatives. Longitudinal and spontaneous studies in K’iche’ and Tzotzil have reported the acquisition of the morphological causative, but not the acquisition of periphrastic causatives. The Q’anjob’al children’s production of more periphrastic causatives might be due to their exposure to a special V1V2 construction, which is a general feature of Q’anjob’al. The Q’anjob’al child data show that even related languages in which causatives are expressed through similar morphemes can show strikingly different acquisition trajectories.

Keywords

alternation V1V2 construction non-derived verb derived verb input

Introduction

Causation is an event that involves external force that changes the status or position of another participant (Family & Allen, 2015). Crystal (2003) states that the causative is a linguistic form that refers to alternative versions of a clause. For example, the two clauses, ‘The cat killed the mouse’ and ‘The mouse died’ are related in that the transitive verb kill can be seen as the causative version of the intransitive verb die (The cat caused the mouse to die). Most languages have at least one way of expressing causation. There are four known possibilities for causative formation in most natural language: lexical, morphological, periphrastic, and a light verb construction (Family & Allen, 2015). In English, for example, there are two types of causatives: periphrastic and lexical. The data in (1) illustrate a periphrastic causative, which consists of two predicates, a causative predicate and an effect predicate (Kemmer & Verhagen, 1994). In a lexical causative, or direct causative (Álvarez, 2013), there is no morphological marking on the verb, as illustrated for English in (2) (Kemmer & Verhagen, 1994).

1. a. The man laughed. Intransitive verb

b. * The clown laughed the man. Transitive verb

c. The clown made the man laugh. Periphrastic causative

2. a. Juan broke the glass.

b. The sun dried the clothes.

In causative constructions, verbs can alternate between non-causative and causative in different ways (Family & Allen, 2015; Pye, 1990, 1993). Some intransitive verbs (3a) alternate only with the periphrastic causative (3b), but not with a lexical transitive (3c) (Marcotte, 2005, p. 8). In contrast, other transitive verbs do not form causative pairings at all, with either the intransitive or periphrastic, as in (4) (Marcotte, 2005, p. 8).

3. a. The audience laughed.

b. Andy made the audience laugh.

c. *Andy laughed the audience.

4. a. Henry cut the bread.

b. *The bread cut.

c. *Henry made the bread cut.

Research on the acquisition of causatives in Mayan languages has focused on K’iche’ (Pye, 1990, 1993) and Tzotzil (De León, 2013). Pye (1990) reports that K’iche’-speaking children do not acquire morphological causatives until the age of 3 years. In elicitation sessions, Pye (1991) reports that children produce periphrastic causatives with some kind of errors. In contrast, De León (2013) reports that children acquiring Tzotzil start producing the morphological causative around the age of 2 years.

Causatives in Mayan languages

In this section, I discuss elements of causatives in K’iche, Tzotzil, and Q’anjob’al¹ that are relevant for this study.² These three Mayan languages belong to different branches of the Mayan language family: K’iche’ belongs to the K’iche’an branch, Tzotzil to the Cholan-Tzeltalan branch, and Q’anjob’al to the Q’anjob’alan branch (Campbell & Kaufman, 1985).

In K’iche’, the morphological causative is marked with the suffix -isa. This suffix is productively used to mark causatives in the language and it attaches to intransitive verbs only, as illustrated in (5b). However, although the morphological causative is productive, there are two classes of K’iche’ intransitive verbs that do not take the causative suffix: the zero class and periphrastic class (Pye, 1991). The zero class has a causative that is not marked with the morphological causative morpheme; as shown in (6); it follows a different derivational process.³ It takes the suffix -ik in its intransitive form (6a), but takes the suffix -oh for its causative form (6b). The K’iche’ periphrastic causative does not require derivational morphology, as shown in (7).

5. a. X-ø-etz’an le ak’al. K’iche’ (Pye, 1991)⁴

com-3abs-play art boy

‘They boy played’.

b. Le achi x-ø-r-etz’an-isa-j le ak’al.

art man com-3abs-3erg-play-cau-tv art boy

‘The man made the boy play’.

6. a. X-in-tzaq-ik. K’iche’ (Pye, 1991)

com-1abs-fall-iv

‘I fell’.

b. X-in-a-tzaq-oh. Mario Marroquín (2019, p.c.)

inc-1abs-2erg-go_down-tv

‘You let me fall’.

7. a. K-in-muxan-ik. K’iche’ (Pye, 1991)

inc-1abs-swim-iv

‘I am swimming’.

b. K-ø-in-b’an k-at-muxan-ik

inc-3abs-1erg-do inc-2abs-swim-iv

‘I will make you swim’.

Tzotzil displays morphological and periphrastic causatives. The suffixes -es and -tas can be attached to intransitive and nominal stems to form the morphological causative (Cowan, 1969; Polian, 2017). The form -es is attached to a large number of intransitive verbs, while the form -tas is attached to a sub-set of intransitive verbs. Non-causal and causal forms with the suffix -es are given in (8) and non-causal and causal forms with the suffix -tas are given in (9) (Cowan, 1969, p. 101–102).

8. a. lok’ ‘to exit’ lok’-es ‘to take it out’

b. nopex ‘approach’ nopex-es ‘make it approach’

c. nox ‘full’ nox-es ‘fill it’

d. p’ol ‘increase’ p’ol-es ‘increase it’

e. kuxul ‘alive’ kus-es ‘cause to live’

9. a. lekub ‘become good’ lekub-tas ‘to take it out’

b. t’xulel ‘soul’ t’xul-tas ‘bless it’

c. anit’son ‘your son’ nit’xon-tas ‘cause to become one’s son’

d. biil ‘name’ biil-tas ‘name him’

The Tzotzil periphrastic causative is formed by the transitive verb ak’, which takes a sentential direct object as its complement (Aissen, 1983). According to Aissen, the verb ak’ means ‘give’, but when it takes a complement clause, it means ‘cause, make, let’. In non-finite complement clauses, a transitive verb does not take aspect marking (10a), while an intransitive verb takes the irrealis suffix -uk, which varies to -ik (10b) (Polian, 2017).

10. a. mu x-ø-y-ak’ ve’-ik-on. (Aissen, 1987, p. 214)

neg inc-3abs-3erg-give eat-irr-1abs

‘It doesn’t let me eat’.

b. ch-ø-k-ak’ av-il-ø li j-chob=e. (Haviland, 1981, p. 332)

inc-3abs-3erg-let2 erg-see-3abs art 1erg-cornfield=det

‘I am going to show you my cornfield’.

In Q’anjob’al, morphological and periphrastic causatives are also found. The morphological causative is marked with the morpheme -tze(ne) (Mateo Toledo, 1999; Raymundo González et al., 2000), as illustrated in (11b). The periphrastic causative involves a single predicate formed by two consecutive verbs, one indicating causation and the other indicating effect (Zavala, 1992, p. 316), as illustrated in (12).

11. a. Max-ø b’ey-i. Non-causal form

com-3abs walk-iv

‘She walked’.

b. Max-ø s-b’ey-tze(ne)-j. Causative form

com-3abs 3erg-walk-cau-tv

‘She made him walk’. (Mateo Toledo, 1999, p. 127)

12. a. Y-uj k’al k’am ch-ø-y-a’ lo-w naq no mis

3erg-rn always neg inc-3abs-3erg-give eat-ap pro clf cat

ix ix tu’. . .⁵

clf woman dem

‘Only because he does not feed that woman’s cat. . .’

Tzotzil and Q’anjob’al do not have a zero class causative like that of K’iche’ (Pye, 1991), illustrated in (6). The zero form in K’iche’ usually has a morphological or periphrastic causative counterpart in Tzotzil and Q’anjob’al.

Although K’iche’, Tzotzil, and Q’anjob’al all feature both morphological and periphrastic causatives, not all verbs alternate in the same way in these three languages. For example, for the causative of the intransitive verb atin ‘to bathe’, K’iche’ uses the causative suffix -isa (13b), but Tzotzil uses the periphrastic causative (14b). For the same verb, Q’anjob’al uses both the morphological and periphrastic causative, as shown in (15b) and (15c).

13. a. X-ø-atin ri ak’al. K’iche’

com-3abs-bathe art boy

‘The boy bathed’.

b. Ri ixoq x-ø-r-atin-isa-j ri ak’al.

art woman com-3abs-3erg-play-cau-tv art boy

‘The woman bathed the boy’.

14. a. ø-ø-atin li k’ox kerem=e. Tzotzil

com - 3abs -bathe art adj youth.man=encl

‘The boy bathed’.

b. La y-ak’-ø atin-uk k’ox kerem li ants=e.

com 3erg -give- 3abs bathe- irr adj youth.man det woman=encl

‘The woman bathed the boy’.

15. a. Max-ø achin-w-i naq unin. Q’anjob’al

com-3abs bath-der-iv clf child

‘The boy bathed’.

b. Max-ø y-achin-ne-j ix ix naq unin.

com-3abs 3erg-bath-cau-tv clf woman clf child

‘The woman bathed the boy’.

c. Max-ø y-aq’ achin-w-oq ix ix naq unin.

com-3abs 3erg-give bath-der-inf clf woman clf child

‘The woman bathed the boy’.

The short, non-exhaustive list of some high-frequency causatives in Table 1 shows that in K’iche’ there are more morphological causatives than periphrastic causatives, while in Q’anjob’al, there are more periphrastic causatives than morphological ones. In Tzotzil, there are equal numbers of morphological and periphrastic causatives. While this is not a large number of lexical items, it suffices here to illustrate some differences between morphological and periphrastic causatives in the three Mayan languages.

Table 1.

Morphological and periphrastic causatives in K’iche’, Tzotzil, and Q’anjob’al.

	K’iche’		Tzotzil		Q’anjob’al
Verb	MC	PC	MC	PC	MC	PC
tickle	1	0	1	0	1	0
bathe	1	0	0	1	1	0
light	1	0	1	0	1	0
sleep	1	0	1	0	1	1
become small	1	0	1	0	1	1
die	1	0	1	0	0	1
boil	1	0	0	1	0	1
go	1	0	0	0	0	1
arrive there	1	0	0	1	0	1
shine	1	0	0	1	0	1
wet	1	0	1	0	0	1
pass	1	0	0	1	0	1
grow	1	0	0	1	0	1
cry	0	0	0	1	0	1
dream	0	0	0	0	1	0
Total	13 (100%)	0 (0%)	6 (46%)	7 (54%)	6 (35%)	11 (65%)

MC: morphological causative; PC: periphrastic causative.

Studies on causatives in K’iche’ (e.g. Gluckman, 2015; Pye, 1990) and Tzotzil (Aissen, 1983; Cowan, 1969; Polian, 2017) show that in these two languages, the morphological causative is more productively used than the periphrastic causative. In Q’anjob’al, in contrast, the periphrastic causative is more productive than the morphological causative. That is, the morphological causative in Q’anjob’al is used with fewer verb types than the periphrastic causative.

Causatives have received attention in studies on first language acquisition, especially in well-known languages, for example, English (Ambridge et al., 2006; Bowerman, 1988). Bowerman (1988) reports that in English, children make errors in the causative alternation, as in the example from Christy (2;9): ‘I come it closer so it won’t fall’. In the adult grammar of English, one would expect ‘I made it come closer’ or ‘I moved it closer’. Bowerman (1988) also reports that in English, children show preference for lexical causatives over periphrastic causatives. Studies on the acquisition of causatives have shown that children acquire the causative system of the language that they are exposed to, for example, Bowerman (1988). Family and Allen (2015), report that of the four types of causatives in Persian children initially acquire two types: lexical and light verb causatives. They start with the lexical causatives and progressively move to the light verb constructions. For Inuktitut, Allen (1998) reports that children start by acquiring lexical causatives, then morphological causatives.

Research on the acquisition of causatives in Mayan languages has focused on K’iche’ (Pye, 1990, 1993) and Tzotzil (De León, 2013). Based on longitudinal and spontaneous data of three children (Al Tiya:n, Al Cha:y, and A Carlos) acquiring K’iche’ of the age ranges: 2;1–2;10, 2;9–3;4, and 3;1–3;8, Pye (1990) found that these children produce morphological causatives around the age of 3 years. Although this type of causatives appear late, Pye reports not finding children’s errors in the use of the causative alternation in K’iche’. That is, these children do not use the causative suffix -isa with transitive verbs or intransitive verbs that do not allow this suffix, as in the case of pet ‘come’. However, in this type of data, children did not produce periphrastic causatives (Pye, 1990).

De León (2013), also using longitudinal and spontaneous data, studied the acquisition of the verb morphology of two children acquiring Tzotzil. The data come from a period of 6 months (19 to 25 months old), in which the two children were audio- and video-recorded in natural interaction with members of the family. De León found that around age 1;10, the two children productively produced the causative suffix -an (among other suffixes, for example, stative, imperative, and inchoative) with roots like va’-al ‘standing on two feet’, chot-ol ‘seated’, kot-ol ‘standing on four legs’. In Tzotzil, neither of the children produce periphrastic causatives.

In summary, based on longitudinal and spontaneous data, these studies in K’iche’ and Tzotzil report the acquisition of the morphological causatives. These studies did not find periphrastic causatives in the child data of K’iche’ and Tzotzil. Only Pye (1991), an elicitation-based investigation of causative overgeneralizations in K’iche’ with children aged 4–7 years, 8–11 years, and 12–13 years, includes any data on K’iche’ child periphrastic causative production. Children in these age groups were able to produce periphrastic causatives, as well as morphological, and zero derivation causatives. However, in some cases, these children applied the zero derivation to verbs in the periphrastic class, for example, muxanik ‘swim’ to muxa:j, which is not expected in the adult grammar of K’iche’.

This study, based on longitudinal and spontaneous data, seeks to expand on these early studies by examining the acquisition of causatives in Q’anjob’al, a third Mayan language, in which causatives are similar to that of K’iche’ and Tzotzil. Contrary to the findings in K’iche’ and Tzotzil, in the Q’anjob’al child data causatives are rare. However, children produce more periphrastic causatives than morphological causatives. The production of more periphrastic causatives might be due to the children being exposed to a V1V2 construction, which Mateo Toledo (2008) considers as a favorite construction in the adult grammar of Q’anjob’al. The findings in this study suggest that children acquiring Q’anjob’al, K’iche’, and Tzotzil may follow different patterns in the acquisition of causatives.

Background of Q’anjob’al

Q’anjob’al belongs to the Q’anjob’alan branch of the Mayan language family (Campbell & Kaufman, 1985); it is spoken in San Juan Ixcoy, San Pedro Soloma, Santa Cruz Barillas, and Santa Eulalia, Huehuetenango, Guatemala. There are about 99,000 speakers of Q’anjob’al (Richards, 2003). Thousands of Q’anjob’al speakers migrated to Mexico, the United States, and Canada during the civil war in Guatemala (Peñalosa, 1992). The Mayan language family consists of 30 different languages with more than 6 million speakers. Mayan languages are found in Mexico, Guatemala, Belize, and Honduras.

Q’anjob’al is an agglutinative, Verb Subject Object (VSO), and head marking language; it shows an ergative agreement system on the verb. Ergative morphemes cross-reference transitive subjects (16a), while absolutive morphemes cross-reference transitive objects (17a) and intransitive subjects (17b). In this article, I follow the analysis that tense/aspect and absolutive agreement appear in a separate word preceding the verb, which bears only ergative and status markers (e.g. Mateo Toledo, 2008).

16. a. Max-ach s-kol-o’.

com-2abs 3erg-help-tv

‘She helped you’.

17. a. Max-ach w-il-a’.

com-2abs 1erg-see-tv

‘I saw you’.

b. Max-ach way-i.

com-2abs sleep-iv

‘You slept’.

Status suffixes on the Q’anjob’al verb mark transitivity, phonological shape, and position of a verb in a clause. The suffixes -V’/-j appear with transitive verbs, while the suffixes -i/-oq appear with intransitive verbs. The suffix -V’ appears in final position only with non-derived transitive verbs (18a), while the suffix -j appears in non-final and final positions with derived transitive verbs (19).

18. a. Max-ach y-il-a’.

com-2abs 3erg-see-tv

‘She saw you’.

b. Max-ach y-il ix ix.

com-2abs 3erg-see clf woman

‘The woman saw you’.

19. a. Max-ach s-way-tze-j.

com-2abs 3erg-sleep-cau-tv

‘She made you sleep’.

b. Max-ach s-way-tze-j nani.

com-2abs 3erg-sleep-cau-tv now

‘She made you sleep now’.

Similarly, the suffixes -i/-oq appear in final position only with non-derived intransitive verbs (20). However, both suffixes appear in non-final and final positions with derived intransitive verbs (21).

20. a. Max-ach way-i.

com-2abs sleep-iv

‘You slept’.

b. Max-ach way ewi.

com-2abs sleep yesterday

‘You slept yesterday’.

21. a. Ch-ach b’it-n-i.

inc-2abs song-der-iv

‘You sing’.

b. Ch-ach b’it-n-i yekal.

inc-2abs sing-der-iv tomorrow

‘You will sing tomorrow’.

In Q’anjob’al, there are only two types of causatives: morphological and periphrastic.

In this article, I follow Zavala’s (1992) description of the causative suffix in Akateko as taking four different forms: -b’e, -ne, -te, and -tze. These forms of the causative suffix are also found in Q’anjob’al. They all occur in the same morphological slot, after the verb root and before the derived-transitive status suffix. The form -b’e is attached to positional roots (22a), while the form -ne is attached to intransitive verbs and to a reduced number of nouns and adjectives (22b).

22. a. Max-ø hin-xoy-b’e-j aj no’ manguera.

com-3abs 1erg-round-cau-tv dir clf hose

‘I wound up the hose’.

b. Tol chi-ø hin-chej-ne-j no’.

compl inc-3abs 1erg-horse-cau-tv animal

‘I ride it’.

The forms -te and -tze indicate control by the participants in the act of causation (Zavala, 1992). The form -te occurs when the causee shows some resistance in the realization of the event (23a), while the form -tze occurs when the causee does not show such resistance (23b). As Zavala (1992) reports for Akateko, in Q’anjob’al, the causative morpheme can also occur with verbs of indeterminate verb roots.

23. a. Chi-ø s-mon-te-j naq ach’ej ix q’opoj.

inc-3abs 3erg-entertain-cau-tv clf young clf girl

‘The young man makes the young lady love him.’

b. Chi-ø s-b’ey-tze-j ix q’opoj ix unin.

inc-3abs 3erg-walk-cau-vt clf girl clf child

‘The young lady made the girl walk.’

In the Q’anjob’al periphrastic causative, there is no morphological marking on the verb; instead, there is a combination of two predicates, one indicating causation and the other indicating effect (Zavala, 1992, p. 316). Mateo Toledo (2008) divides the Q’anjob’al periphrastic causative into three types: causative complex predicate, biclausal causative construction, and semi-integrated causative. Lexicalized periphrastic causatives are also very common in the adult grammar of Q’anjob’al (Francisco Pascual, p.c.).

A causative complex predicate consists of a transitive clause with the word order VSO and with a V1V2 construction: the V1 inflects, indexing both participants in the causation act, while the V2 is in infinitive form and describes the caused event, state or action, as illustrated in (24a). The biclausal causative construction is made up of a main verb and an aspect-less complement clause, as illustrated in (24b). The semi-integrated causative involves two verbs, like the complex causative, but each of the verbs inflects to index one participant in the causation action. In (24c), for example, the caused event ante- ‘to cure’ takes the ergative morpheme y-, referencing its own subject.

24. a. Y-uj k’al k’am ch-ø-y-a’ lo-w naq no mis

3erg-rn always neg inc-3abs-3erg-give eat-ap pro clf cat

ix ix tu’. . .⁶

clf woman dem

‘Only because he does not feed that woman’s cat. . .’

b. Max-on y-aq’ obligar ix Malin [ko-lo-w-i].

com-1abs 3erg-give force clf Malin 1erg-eat-ap-iv

‘Malin forced us to eat.’ (Mateo Toledo, 2008, p. 28)

c. Max-ach w-aq’ [y-ante-j ix Malin] ewi.

com-2abs 1erg-give 3erg-cure-tv clf Malin yesterday

‘I made Malin cure you yesterday.’ (Mateo Toledo, 2008, p. 29)

In (25), I provide a comprehensive list of V1 verbs that are involved in complex predicate and semi-integrated periphrastic causative formation in the adult grammar of Q’anjob’al (Mateo Toledo, 2008, p. 302). The following list will be a reference to evaluate the V1 verbs that children produce in the periphrastic causative in Q’anjob’al:

25. V1 verbs and semi-integrated causatives in Q’anjob’al

a. aq’ ‘to give/to cause, to do’

b. b’eq ‘free/leave, stop’

c. cha’ ‘receive/leave, allow ‘

d. jeq ‘push/do something to annoy’

e. etb’inej ‘accompany doing’

f. jutx ‘drag/force pulling’

g. iptzej ‘force to do . . ., animate . . .’ (animated object)

h. uqtej ‘pursue/do it fast’ (animated object)

i. tx’em ‘provoke/do something to annoy’ (animated object)

j. cheq ‘use as messenger/order to make’ (animated object)

k. montej ‘convince to do . . .’ (animated object)

l. ketz ‘take of the hand/do pull by’ (animated object)

m. iqb’aj ‘accompany and animate. . .’ (animated object)

Finally, the lexicalized periphrastic causative consists of a contraction of the first verb and the second verb in the causative complex predicate, as illustrated in (26). In this example, the first verb aq’ ‘to give’ and the second verb ok ‘to enter’ are contracted and become as ak.

26. q-ø-ø-ak ha-chan mija xhi.

pot-3abs-3erg-give.enter 2erg-skirt my daughter said

‘Daughter, you will put on your skirt, she said’.

In this section, we have seen that there are two types of causatives in Q’anjob’al: morphological and periphrastic. The morphological causative is marked with the morphemes -b’e, -ne, -te, and -tze. The periphrastic causative is a combination of two predicates. Within the periphrastic causatives, lexicalized periphrastic causatives are also found, which are very common in the adult grammar of Q’anjob’al.

Acquisition of causatives in Q’anjob’al

This study evaluates the acquisition of morphological and periphrastic causatives in Q’anjob’al, in which the morphological and periphrastic causatives are similar to that of K’iche’ and Tzotzil. Although in the Q’anjob’al child data causatives are rare, I will show that these children produce more periphrastic than morphological causatives. This preference for the periphrastic causatives may be due to the use of a V1V2 construction found in the adult grammar and also in child-directed speech in Q’anjob’al.

The data

The Q’anjob’al child data discussed in this article come from the project Documentation of the Acquisition of Mayan Languages (www.almaya.org) (Mateo Pedro, 2015; Pye, 2017; Pye et. al, 2017). The data were collected longitudinally: children were audio/video recorded every 2 weeks for approximately an hour, throughout a year (Mateo Pedro, 2010, 2015). For the purpose of this article, I analyzed data from three children acquiring Q’anjob’al of the following age ranges: Xhuw (1;9–3;0), Xhim (2;3–4;0), and Tum (2;7–3;6).

Xhuw is a girl of the age range 1;9–3;0. She is the first child of the family and spends most of her time with her cousin as the caregiver. Xhim is a boy of the age range 2;3–4;0; he is the only child of the family and lives with his grandparents. He spends more time with his two aunts, and less time with his uncles and grandparents. Tum is a girl of the age range 2;7–3;6 and the first child of the family. She spends time with her younger sister and her mother. These children are from the town of Santa Eulalia, Huehuetenango, Guatemala. They are exposed to Q’anjob’al the most, although they are also exposed to Spanish because their parents are bilingual in Q’anjob’al and Spanish, as in the case of Xhuw, or because they are exposed to television programs in Spanish. In Guatemala, Spanish is the official language and it is used in different contexts, for example, education, religion, justice, health, and so on. Q’anjob’al is used in reduced contexts, for example, family and community activities (Mateo Toledo, 2008). The recordings and transcriptions were done by native speakers of Q’anjob’al, who were trained in data collection, digitization, and transcription. For the extraction of the child data for this study, I used Pye’s (2019) qancord program, which is accessible at http://pyersqr.org/minimal/.

Criteria for selecting causatives

I counted morphological causatives when children use verbs with the causative suffixes -b’e, -ne, -te, -tze. I included as morphological causatives instances in which the causative suffix is attached to indeterminate verb roots from the causative morpheme, as shown in the section ‘Background of Q’anjob’al’. In the section ‘Morphological causative’, I separate indeterminate and determinate verb roots to evaluate if indeterminate verb roots are acquired differently by children in Q’anjob’al.

I counted periphrastic causatives when children use a caused verb (V2) together with a causing verb (V1). In Table 2, I group lexicalized periphrastic causatives together with other periphrastic causatives. However, in the section ‘Periphrastic causative’, I separate lexicalized periphrastic causatives from periphrastic causatives to evaluate what type children produce the most. Verbs without any of these features were excluded from the study. The causatives counted in the study are either the children’s own productions or responses to adults’ questions. Repetitions were excluded.

Table 2.

Q’anjob’al children’s morphological and periphrastic causatives.

	Morphological		Periphrastic
Child	Types	Tokens	Types	Tokens
Xhuw (1;9–3;0)	3	95	35	85
Xhim (2;3–4;0)	19	43	97	242
Tum (2;7–3;6)	16	18	44	87

Children’s causatives

The three children’s morphological and periphrastic causatives, given in types and tokens, are summarized in Table 2. As mentioned earlier, in Table 2, I group lexicalized periphrastic causatives together with other periphrastic causatives. With the exception of Xhuw, these children produce more periphrastic causatives than morphological causatives.

These children’s types and tokens of morphological and periphrastic causatives, age by age, are given in Table 3 for Xhuw, Table 4 for Xhim, and Table 5 for Tum. Table 3 shows that Xhuw starts producing the morphological causative (causative suffix -ne) around the age of 2;9 and the periphrastic causative around the age of 2;0. Table 4 shows that Xhim starts producing the morphological causative (causative suffix -ne) around 2;4 and the periphrastic causative around the age of 2;6. Table 5 shows that Tum starts producing morphological causatives (causative suffix -ne and -te) around the age of 2;9 and the periphrastic causative around the age of 2;11.

Table 3.

Xhuw’s morphological and periphrastic causatives: types and tokens.

	Morphological		Periphrastic
Ages	Types	Tokens	Types	Tokens
1;9	0	0	0	0
1;11	0	0	1	1
2;0	0	0	1	5
2;1	0	0	1	5
2;2	0	0	0	0
2;3	0	0	1	1
2;4	0	0	0	0
2;5	0	0	2	4
2;6	0	0	0	0
2;7	0	0	2	4
2;8	0	0	4	5
2;9	1	2	3	4
2;10	1	46	6	9
2;11	0	0	0	0
3;0	1	47	0	0
Total	3	95	21	38

Table 4.

Xhim’s morphological and periphrastic causatives: types and tokens.

	Morphological		Periphrastic
Ages	Types	Tokens	Types	Tokens
2;3	0	0	0	0
2;4	3	3	1	2
2;5	0	0	0	0
2;6	0	0	0	0
2;7	0	0	1	5
2;8	0	0	3	5
2;9	1	1	1	1
2;10	0	0	2	3
2;11	1	1	4	6
3;0	2	4	5	5
3;1	0	0	11	29
3;2	0	0	8	19
3;3	3	20	10	12
3;5	3	6	8	13
3;8	1	1	17	68
3;9	2	3	14	55
3;10	0	0	0	0
3;11	1	1	0	0
4;0	2	3	0	0
Total	19	43	85	223

Table 5.

Tum’s morphological and periphrastic causatives: types and tokens.

	Morphological		Periphrastic
Ages	Types	Tokens	Types	Tokens
2;7	0	0	1	1
2;8	0	0	0	0
2;9	2	2	0	0
2;10	2	2	2	3
2;11	1	1	5	7
3;0	1	1	5	7
3;1	2	2	8	13
3;2	2	2	2	4
3;3	4	6	5	10
3;4	1	1	0	0
3;5	1	1	0	0
3;6	0	0	5	7
Total	16	18	33	52

These tables show individual differences in the age of acquisition of the morphological and periphrastic causatives in Q’anjob’al. Because of the age differences and because causatives are rare in the data here, it is difficult to establish when children start producing causatives. However, based on the data, one can assume that these children have acquired both morphological and periphrastic causatives before the ages shown in these tables. The differences among the three children seen in Tables 3 –5 may be the result of the combination of different ages of the recordings of each child and the rarity of causatives. It is worth noting that Xhuw is younger than the other two children and shows few productions of causatives. Xhim is more talkative than Tum; this could explain why he has more types and tokens of causatives in some of his sessions. For example, in the session at age 3;8, he produced 17 types of periphrastic causatives. This session was about playing with a truck, motorcycle, and bicycle. So, the conversation was about making the truck drive, the truck not working anymore so it was thrown away, and so on. Later in the session, he started to talk about braking the motorcycle and someone getting killed by the motorcycle. Some of the questions that he received from the input were the following: Who gave you the truck? What happened to the truck?

Although causatives are rare in the child data of Q’anjob’al and there are individual differences, Figure 1 shows that Xhuw’s overall number of periphrastic causative tokens is clearly divergent from those of the other two children. While Xhuw’s total morphological causative counts (in Table 3 and Figure 1) are higher than her total periphrastic causative counts, her elevated morphological causative token counts are due to very high token numbers of a single type each in the sessions at 2;10 and 3;0. In fact, Table 3 shows that Xhuw begins producing the periphrastic causative well before the morphological causative, and with greater diversity of types. Tables 4 and 5, as well as Figure 1, show that the other two children produce more periphrastic causative tokens overall than morphological ones.

Figure 1.

Children’s Morphological and Periphrastic Causatives.

Morphological causative

In this section, I discuss the acquisition of the morphological causative, comparing the use of causative morphemes with determinate and indeterminate verb roots. As in the adult grammar, children produce verb roots that are distinguishable from the causative morpheme and produce other verb roots that are not.

Children begin to produce the morphological causative in Q’anjob’al around the age of 3;0. These children produce the causative suffixes -b’e, -ne, -te, but not the suffix -tze. They produce the suffix -ne more than the other suffixes. The children’s production of the causative suffix -ne is illustrated in (27).

27. a. Jah kowahneh. (Xhuw 2;10)

= ja’ ko-watx’-ne-j

yes 1erg-good-cau-tv

‘Yes, let’s fix it.’

b. Hinwanej karitera ch’en het. (Xhim 3,8)

= hoq-ø hin-watx’-ne-j carretera ch’en het

pot-3abs 1erg-good-cau-tv road pro yours

‘I will fix the road of yours.’

c. Tol hinchejnej no’. (Xhim 3;8)

= tol chi-ø hin-chej-ne-j no’

compl inc-3abs 1erg-horse-cau-tv animal

‘I ride it.’

Compared to other roots, Xhuw and Xhim produced the adjective root watx’ ‘good’ the most, which received the causative suffix -ne, as illustrated in (28) and (29).

28. a. Jah kowahneh. (Xhuw 2;10)

= ja’ ko-watx’-ne-j

yes 1erg-good-cau-tv

‘Yes, let’s fix it.’

b. Wanej hipelota. (Xhuw 2;10)

= watx’-ne-j hin-pelota

good-cau-tv 1erg-ball

‘Fix my ball.’

c. Kun ta hawaneh yet. (Xhuw 2;10)

= jun la hoq-ø ha-watx’-ne-j y-et

one dem pot-3abs 2erg-good-cau-tv 3erg-rn

‘You will fix the one that belongs to this.’

29. a. Tom wach’ wutej? (Xhim 3;3)

= tom watx’ ch-ø-w-ute-j

perhaps good inc-3abs-1erg-do-tv

‘Perhaps I do it well?’

b. Nwanej ok yetoX naX. (Xhim 3;3)

= hoq-ø hin-watx’-ne-j ok y-etoq naq

pot-3abs 1erg-good-cau-tv dir 3erg-rn pro

‘I will fix it with him.’

c. Maj kowaneh ok chen. (Xhim 3;3)

= maj ko-watx’-ne-j ok ch’en

neg 1erg-good-cau-tv dir metal

‘We will not fix it.’

Tum produces other roots that take the causative suffix -ne (30a), -te (30b), and -b’e (30c). She produces different roots other than watx’ and other forms of the causative suffix, and not only the suffix -ne.

30. a. Waaneh ayti. (Tum 2;10)

= chi-ø hin-watx’-ne-j ayin ti

inc-3abs 1erg-good-cau-tv me dem

‘Me, I fix it.’

b. Wantej hinb’a. (Tum 3;1)

= chi-ø w-an-te-j hin-b’a

inc-3abs 1erg-plant-cau-tv 1erg-refl

‘I cure myself.’

c. Lekb’aj no’. (Tum 3;1)

= lek-b’e-j aj no’

stand-cau-tv dir animal

‘Make it stand up.’

As already mentioned, in Q’anjob’al, children also produce causative suffixes that are attached to indeterminate verb roots, as illustrated in (31).

31. a. Kojatneh xhi. (Xhuw 2;10)

= ko-jat-ne-j xhi

1erg-root-cau-tv says

‘Let’s fix it, she says.’

b. YuXtej heb ka la. (Xhim 3;0)

= chi-ø y-uq-te-j heb’ kayla

inc-3abs 3erg-root-cau-tv pl here

‘They root it here.’

c. Tane kon te kona. (Tum 2;9)

= tay-ne-j kan te ko-na

root-cau-tv dir clf 1erg-house

‘Take care of our house.’

A comparison of the children’s production of determinate and indeterminate verb roots with causative morphemes is given in Table 6. This table shows that Xhim and Tum produce more determinate verb roots than indeterminate verb roots, which means they productively use causative morphemes. In contrast, Xhuw produces one type of determinate verb root with high tokens and two types of indeterminate verb roots, but with low tokens. One may argue that Xhuw uses the causative morphemes with the indeterminate verb roots as frozen forms. However, it does not seem the case. She produces indeterminate verb roots later than determinate verb roots, as illustrated in (31a) . This is also true for Xhim and Tum.

Table 6.

Children’s causative morphemes with verb roots.

	Determinate verb roots		Indeterminate verb roots
Child	Types	Tokens	Types	Tokens
Xhuw (1;9–3;0)	1	93	2	2
Xhim (2;3–4;0)	15	29	4	14
Tum (2;7–3;6)	13	13	3	5

Periphrastic causative

The children’s periphrastic causative token and type numbers were given in Tables 3, 4 and 5 earlier. These tables show that each child starts producing the periphrastic causative at different ages; Xhuw around the age of 2;0, Xhim around the age of 2;6, and Tum around the age of 2;11. Children begin to produce periphrastic causatives around the age of 2 years, as Table 3 shows for Xhuw. In (32), Xhuw produces periphrastic causatives although with missing morphology. In the second verb (V2) way ‘to sleep’ in (32a), the infinitival morpheme -oq is missing, while in (32b), in the second verb (V2) kam ‘to die’, the child uses the suffix -i instead of the infinitival morpheme -oq. Xhim’s example of periphrastic causative is given in (33).

32. a. Toh ya way. (Xhuw 2;0)

=x-ø-toj y-aq’ way-oq

com-3abs-go 3erg-give sleep-inf

‘She went to make her sleep’.

b. Ihaj pitz’la kami. (Xhuw 2;10)

=ijan x-ø-pitz’-lay kam-oq

almost com-3abs-tight-pas die-inf

‘She was almost killed’.

33. Hima’ poj xhi’ xhi jab’ papa. (Xhim 2;8)

=chi-ø hin-maq’ poj si’ xhi jab’ papa

inc-3abs 1erg-hit break firewood says adj father

‘I split wood, father says’.

Of the list of verbs in the complex causative construction provided in (25) for the adult grammar of Q’anjob’al (Mateo Toledo, 2008), only two occur in the Q’anjob’al child data: aq’ ‘give/cause, do’ (34) and b’eq ‘free/let, stop’ (35).

34. a. Toh ya way. (Xhuw 2;0)

=x-ø-toj y-aq’ way-oq

com-3abs-go 3erg-give sleep-inf

‘She went to make her sleep’.

b. Ya’ poj naq si’. (Xhim 2;8)

=chi-ø y-aq’ poj naq si’

inc-3abs 3erg-give break clf firewood

‘He splits firewood’.

35. a. IpeX iloX wakax no’. (Xhuw 2;7)

=max-ø hin-b’eq el no’ wakax

com-3abs 1erg - let exit clf cow

‘I let the cow fall’.

b. Wen, heb’ unin lan tit beqon aytoq heb’ nene’. (Xhim 3;8)

=bueno heb’ unin lanan tit heb’ b’eq-on ay-toq heb’ s-nene’

good pl child prog mov pl let-dep fall-dir pl a3s-toy

‘Well, the children are coming to drop their toys’.

Table 7 shows the children’s periphrastic causative types. Xhuw and Xhim use the same verbs for V1 (aq’ ‘give’, b’eq ‘let’), but they use different verbs for V2. Xhim and Tum use three similar verbs for V2. Tum produces different verbs for V1 and V2.

Table 7.

The children’s complex periphrastic causatives.

Periphrastic causatives		Xhuw (1;9–2;8)	Xhim (2;3–4;0)	Tum (2;7–3;6)
aq’ way	‘make sleep’	+	−	−
aq’ saqch	‘make play’	+	−	−
aq’ taj	‘make cook’	+	−	−
aq el	‘let fall’	+	−	−
aq’ ok	‘put on’	−	+	+
aq’ kan	‘leave’	−	+	+
aq’ ay	‘put down’	−	+	+
aq’ poj	‘make break’	−	+	−
aq’ pul	‘make spill’	−	+	−
aq’ low	‘make eat’	−	+	−
aq’ man	‘make buy’	−	+	−
aq’ noj	‘fill’	−	+	+
b’eq ay	‘let fall’	−	+	−
aq’ aj	‘put on’	−	−	+
i el	‘take off’	−	−	+
maq’ ok	‘hit on’	−	−	+
ten ok	‘push’	−	−	+
ten aj	‘lift’	−	−	+
pen ek’	‘push fall’	−	−	+

Lexicalized periphrastic causatives

These children also produce lexicalized periphrastic causatives, such as ak (36a, b), which is a combination of aq’ and ok ‘put on, turn on’. The example in (36c) illustrates the use of aj, which is a combination of aq’ and aj ‘put on’.

36. a. Ak kotele. (Xhuw 2;1)

=aq’-ok ko-tele

give-enter 1erg-television

‘Turn on our television’.

b. Cha waktoq. (Xhim 2;8)

=cha w-aq’ ok-toq

you 1erg-give enter-dir

‘You, I put it in’.

c. Wa nwonit. (Xhim 2;11)

=waj hinwonit

w-aq’ aj hin-wonit

1erg-give up 1erg-hat

‘I put my hat on’.

Table 8 shows the children’s periphrastic and lexicalized periphrastic causatives. Xhuw and Tum produce more tokens of lexicalized periphrastic causatives than other periphrastic causatives, in contrast to Xhim. However, when combining the children’s periphrastic causatives with their lexicalized periphrastic causatives, one finds that overall these children produce more periphrastic causatives than morphological causatives, as illustrated in Table 2.

Table 8.

Children’s periphrastic and lexicalized periphrastic causatives.

Child	Periphrastic		Lexicalized		Total
	Types	Tokens	Types	Tokens	Types	Tokens
Xhuw (1;9–3;0)	21	38	14	47	35	85
Xhim (2;3–4;0)	85	223	12	19	97	242
Tum (2;7–3;6)	33	52	11	35	44	87

Another component of the acquisition of causatives in Q’anjob’al involves the causative alternation of a verb, although there are not many examples of this type in the dataset. In (37a), for example, the intransitive verb way ‘to sleep’ is the caused event (V2) of the causative verb aq’ ‘to give’ (V1). In contrast, in (37b) the intransitive verb way, which is V2 in (37a), appears in its base form. These two utterances are from the same session. The child used both forms, when she was trying to convince her father to let her take a nap, while her father was trying to convince her to keep talking. A little later, their conversation turned to the topic of putting someone to sleep.

37. a. Yu li wa’ wayu la. (Xhuw 2;0)

=y-ul ti hoq-ø w-aq’ way-oq la

3erg-rn dem pot-3abs 1erg-give sleep-inf dem

‘Look, I will make her sleep inside’.

b. Toh ya way. (Xhuw 2;0)

=x-ø-toj y-aq’ way-oq

com-3abs-go 3erg-give sleep-inf

‘She went to make her sleep’.

Non-causative two-verb predicates in the child production data

Complex predicate constructions involving two verbs are also found in other functions in Q’anjob’al, for example, resultative, directional, and motion constructions (Francisco Pascual, 2013; Mateo Toledo, 2008). Having exposure to these constructions could enable children to extract specific environments for the periphrastic causative. In other words, these other uses of the two-verb complex predicate construction could compliment the causative use. Studies on the acquisition of Mayan languages have shown that when children are exposed to different contexts of use of a specific feature of the adult grammar, it is an advantage for them to acquire such a feature, for example, the acquisition of split ergativity and status suffixes in Mayan languages (Pye et al., 2017), the acquisition of ergative prefixes, which are used with nouns and verbs in Mayan languages (e.g. Brown, 1998 on Tzeltal).

The data below illustrate directional (38) and motion (39) two-verb complex predicates. These two types involve a single actor completing multiple actions, or a single action with multiple dimensions.

38. a. A txahiloX. (Xhuw 2;1)

=as tx’aj el-oq

go wash dir-inf

‘Go wash it’.

b. Tek’ ayoX. (Xhim 2;4)

=tek’ ay-oq

kick dir-inf

‘Step on it’.

39. a. Xhi lu xha mamin. (Xhuw 2;7)

=si’ ch-ø-’ul y-a’ mamin

firewood inc-3abs-come 3erg-give grandfather

‘Firewood, grandfather came to deliver’.

b. Ay b’ay xhb’et wil heb’. (Xhim 3;8)

=ay b’ay chi-ø b’et w-il heb’

exst prep inc-3abs mov 1erg-see pl

‘Sometimes I go see them’.

Overall, the children’s use of two-verb complex predicates in other contexts is summarized in Table 9.

Table 9.

Children’s two-verb complex predicate constructions in different contexts.

Child	Directional	Causative	Motion	Total
Xhuw (1;9–3;0)	18 (35%)	33 (63%)	1 (2%)	52
Xhim (2;3–4;0)	211 (70%)	74 (25%)	15 (5%)	300
Tum (2;7–3;6)	40 (41%)	56 (57%)	2 (2%)	98

The input

In addition to exploring the child data, we also want to know what the input can tell us about the acquisition of causatives in Q’anjob’al. For this purpose, I counted morphological and periphrastic causatives from the adult input following the criteria applied to the child data, described in the section ‘Criteria for selecting causatives’. Table 10 shows the distribution of morphological and periphrastic causatives of the input. The input is child-directed speech of the three children’s parents and relatives. The data for each parent come from five sessions over 5 months. Xhuw’s father’s data were extracted when Xhuw was in the age range 2;0–2;4; Xhim’s grandmother’s data were extracted when Xhim was in the age range 2;3–2;7; and Tums’ mother’s data were extracted when Tum was in the age range 2;7–2;11.

Table 10.

Morphological and periphrastic causatives in the input.

Input	Morphological		Periphrastic
	Types	Tokens	Types	Tokens
Xhuw’s father	6	23	32	57
Xhim’s grandmother	7	13	28	47
Tum’s mother	23	41	14	17

According to Table 10, Xhuw’s father and Xhim’s grandmother produce more periphrastic causatives than morphological causatives. In contrast, Tum’s mother produces more morphological causatives than periphrastic causatives. Examples of morphological and periphrastic causatives from the input are given in (40) and (41), respectively.

40. a. ¿Tze ch-ø-a-wa’-ne-j b’ay tu? Xhuw’s father

what inc-3abs-2erg-good-cau-tv prep dem

‘What do make/fix there?’

b. Xib’-te-j heb’ lim. Xhim’s grandmother

scare-cau-tv pl quick

‘Scare them, quick’.

c. Aw-te-j naq. Tum’s mother

scream-cau-tv pro

‘Call him’.

41. a. J-ak ko-tele tu b’ela. Xhuw’s father

1erg-give.enter 1erg-television dem later

‘We will turn on our television later’.

b. Ha-b’eq ta el-oq. Xhim’s grandmother

2erg-let neg exit-inf

‘Do not let it fall’.

c. Q-ø-ø-ak ha-chan mija xhi.⁷

pot-3abs-2erg-give.enter2 erg-skirt my.daugher says

‘You will put your skirt on my daughter, she says’.

The data in Table 10 suggest that children acquiring Q’anjob’al are exposed to both morphological and periphrastic causatives, although with variation. Xhuw and Xhim are exposed more to periphrastic causatives than morphological causatives. In contrast, Tum is exposed to more morphological causatives than periphrastic causatives. A comparison of the child data with the input is provided in the tables below. Based on these data, it is difficult to establish whether the input has a strong effect on the type of causatives children produce in Q’anjob’al. For instance, Table 11 shows that Xhuw produces tokens of periphrastic causatives only, although in the input there were tokens of morphological causatives. In contrast, Xhim (Table 12) and Tum (Table 13) produce more periphrastic causative tokens than morphological causative tokens. In addition, Table 13 shows despite her exposure to more morphological causatives than periphrastic causatives, Tum produces more periphrastic causatives than morphological causatives, following the other two children’s pattern.

Table 11.

Morphological and periphrastic causatives: input and Xhuw.

Ages	Input				Xhuw
	Morphological		Periphrastic		Morphological		Periphrastic
	Types	Tokens	Types	Tokens	Types	Tokens	Types	Tokens
2;0	0	0	6	8	0	0	1	5
2;1	0	0	8	11	0	0	1	5
2;2	2	10	3	9	0	0	0	0
2;3	1	6	11	23	0	0	1	1
2;4	3	7	4	6	0	0	0	0
Total	6	23	32	57	0	0	3	11

Table 12.

Morphological and periphrastic causatives: input and Xhim.

Ages	Input				Xhim
	Morphological		Periphrastic		Morphological		Periphrastic
	Types	Tokens	Types	Tokens	Types	Tokens	Types	Tokens
2;3	1	1	12	22	0	0	0	0
2;4	0	0	2	3	3	3	1	2
2;5	4	5	1	1	0	0	0	0
2;6	0	0	8	9	0	0	0	0
2;7	2	7	5	12	0	0	1	5
Total	7	13	28	47	3	3	2	7

Table 13.

Morphological and periphrastic causatives: input and Tum.

Ages	Input				Tum
	Morphological		Periphrastic		Morphological		Periphrastic
	Types	Tokens	Types	Tokens	Types	Tokens	Types	Tokens
2;7	4	9	2	2	0	0	1	1
2;8	2	2	0	0	0	0	0	0
2;9	11	24	2	3	2	2	0	0
2;10	2	2	4	4	2	2	2	3
2;11	4	4	6	8	1	1	5	7
Total	23	41	14	17	5	5	8	11

Figures 2 and 3 provide a summary of the tokens of input and child production of morphological and periphrastic causatives, respectively. Both figures show that these children are exposed to morphological and periphrastic causatives and they all produce more periphrastic than morphological causatives in the period when input was examined (first 5 months for each child). Notice though that in this period of comparison between the input and the child data, Xhuw does not produce morphological causatives, as Figure 2 shows. After this period of comparison, she produces more tokens of morphological than periphrastic causatives, as seen in Table 3.

Figure 2.

Morphological Causatives: Input and Child Production.

Figure 3.

Periphrastic Causatives: Input and Child production.

Conclusion

Causatives have received attention in first language acquisition research, but few studies have been done on the acquisition of this topic in Mayan languages, for example, Tzotzil (De León, 2013) and K’iche’ (Pye, 1990, 1993). In this article, I have shown that, while causatives are relatively rare in child productions in Q’anjob’al, two of these children produce more tokens of periphrastic causatives than morphological causatives. While the third child produces overall more tokens of morphological than periphrastic causatives in the study period, her elevated morphological causative token numbers are solely due to large numbers of tokens of a single morphological causative type in two of the 15 recording sessions. In addition, this child produces only periphrastic causatives for a sustained period before ever producing morphological causatives, reinforcing the notion that periphrastic causatives are more accessible than morphological causatives to children acquiring Q’anjob’al. The children’s production of more periphrastic causatives might be related to the presence of the same, two-verb complex predicate, construction in other functions in the grammar of Q’anjob’al. The Q’anjob’al adult input for two of the children includes fewer morphological causatives than periphrastic causatives. This could also help explain why children acquiring Q’anjob’al produce more periphrastic causatives than morphological causatives. Furthermore, children do not show errors in the production of causatives in Q’anjob’al, as Pye (1990, 1993) has reported for K’iche’. This is in contrast to children acquiring English, for example, who do make errors in the production of causatives (Marcotte, 2005).

This is the first study on the acquisition of causatives in Q’anjob’al. Previous studies on the acquisition of causatives in the Mayan languages K’iche’ (Pye, 1990, 1993) and Tzotzil (De León, 2013) did not examine children’s naturalistic production of periphrastic causatives. The previous studies established that in K’iche’ and Tzotzil, children productively produce morphological causatives. This study has shown that in Q’anjob’al, children produce both types and in fact produce more periphrastic causative tokens and types than morphological causative tokens and types, especially early in the acquisition process.

The study enriches our understanding of the acquisition of causatives in Mayan languages. A review of the structure of causatives and the similar, naturalistic child data used for K’iche’, Tzotzil, and Q’anjob’al indicates that the difference in the acquisition of causatives may be either due to structural and distributional differences between causatives in these languages, or due to the sampling or the type of analysis done in each language. Beyond Mayan languages, this study provides evidence for the potential for early acquisition of periphrastic causatives, unlike in Persian (Family & Allen, 2015), for instance. This suggests that children’s acquisition of causatives is not uniform, but conditioned by the language to which they are exposed, confirming the expectations of Bowerman (1988).

Footnotes

Acknowledgements

The author would like to thank the parents, speakers of Q’anjob’al, who allowed him to work with their children in the project ‘Documentation of the Acquisition of Maya Languages’ (). He is grateful to the National Science Foundation (BCS 0515120 and BCS 0613120), UNAM-PAPIIT (IN-401207), CIESAS-Sureste, the KU Center for Latin American Studies, the Schiefelbusch Child Language Scholarship, and the Tinker Fellowship for the support of child data collection. A special thanks to the assistant researchers who helped him in the recordings and transcriptions of the child data: Francisco Pedro Mateo, Basilio Luin Bernabé, Flora García, Diego Adalberto Martínez Esteban, Pedro Martínez Esteban, Efraín Ramón de León, and Basilio Sebastian Basilio. He would like to thank Adán Francisco Pascual, B’alam Mateo Toledo, Maria Polinsky, Clifton Pye, Rodrigo Ranero, and Justin Royer for their comments and suggestions. He would also like to thank Jorge López-Bachiller Fernández, Marisa Fried, Carlos Alberto Haas, Sophie Kurschner, and Brian McNamara of the Winter Field School 2019 at the Field Station Guatemala, the audience of the Department of Spanish and Portuguese Studies of the University of Florida, and the audience of the Facultad de Lenguas y Letras of the Universidad Autónoma de Querétaro for their helpful comments and observations. He would like to thank Professor Chloë Marshall, Editor in Chief of First Language, Dr Hannah Sarvasy, the Guest Editor for their help in the revised version of the article. He also wants to thank two anonymous reviewers for their helpful questions and comments to early versions of this article. The remaining errors are the author’s responsibility.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author(s) received financial support for the authorship, and/or publication of this article from the Facultad de Lenguas y Letras, Universidad Autónoma de Querétaro.

ORCID iD

Pedro Mateo Pedro

Notes

References

Aissen

(1983). Indirect object advancement in Tzotzil. In: Perlmutter

D. M.

(Ed.), Studies in relational grammar (pp. 272–302). University of Chicago Press.

Aissen

(1987). Tzotzil clause structure. Reidel.

Allen

(1998). Categories within the verb category: Learning the causative in Inuktitut. Linguistics, 36(4), 633–677.

Álvarez

A. A.

(2013). Adquisición de construcciones causativas en el español [Unpublished master’s thesis]. National University of Comahue, Argentina.

Ambridge

Pine

Rowland

Young

(2006). The effect of verb semantic class and verb frequency (entrenchment) on children’s and adults’ graded judgements of argument-structure. Cognition, 106, 87–129.

Bowerman

(1988). The ‘no negative evidence’ problem: How do children avoid constructing an overly general grammar? In: Hawkins

J. A.

(Ed.), Explaining language universals (pp. 73–101). Blackwell.

Brown

(1998). Children’s first verbs in Tzeltal: Evidence for an early verb category. Linguistics, 36(4), 713–753.

Campbell

Kaufman

(1985). Mayan linguistics: Where are we now? Annual Review of Anthropology, 14, 187–198.

Can Pixabaj

T. A

. (2016). Una mirada a la alternancia de pares incoativo-causativo en k’iche’. Revista Pueblos Y Fronteras Digital. http://www.scielo.org.mx/pdf/rpfd/v11n22/1870-4115-rpfd-11-22-00138.pdf

10.

Cowan

M. M.

(1969). Tzotzil grammar. Summer Institute of Linguistics.

11.

Crystal

(2003). A Dictionary of Linguistics & Phonetics (5th ed.). Blackwell.

12.

De León

. (2013). From roots to edges: First steps into the acquisition of verb morphology in Tzotzil, Amerindia, 37(1).

13.

Family

Allen

(2015). The development of the causative construction in Persian child language. Journal of Child Language, 42(6), 1337–1378. https://doi.org/10.1017/S0305000915000057

14.

Francisco Pascual

. (2013). La construcción resultativa de verbo serial en q’anjob’al. In: Palancar

E. L.

Zavala Maldonado

(Eds.), Clases léxicas, posesión y cláusulas complejas en lenguas de Mesoamérica (pp. 199–216). CIESAS.

15.

Gluckman

(2015). Some Causative Alternations in K’iche’, and a Unified Syntactic Derivation. 10.20354/B4414110026.

16.

Haviland

(1981). Sk’op sotz’leb: el tzotzil de San Lorenzo Zinacantán. UNAM.

17.

Kemmer

Verhagen

(1994). The grammar of causatives and conceptual structure of events. Cognitive Linguistics, 5–2, 115–156.

18.

Marcotte

J.-P.

(2005). Causative alternation errors in child language acquisition [Unpublished doctoral dissertation]. Stanford University.

19.

Mateo Pedro

. (2010). Acquisition of inflection in Q’anjob’al: A longitudinal study [Unpublished doctoral dissertation]. University of Kansas.

20.

Mateo Pedro

. (2015). The Acquisition of inflection in Q’anjob’al Maya. John Benjamins.

21.

Mateo Toledo

. (1999). La cuestión Akateko-Q’anjob’al, una comparación gramatical (Unpublished undergraduate thesis). Mariano Gálvez University of Guatemala.

22.

Mateo Toledo

. (2008). The Family of Complex Predicates in Q’anjob’al Maya: Their Syntax and Meaning (Unpublished doctoral dissertation). University of Texas in Austin.

23.

Peñalosa

(1992). ¿Cuántos Idiomas Q’anjob’al Existen? Boletín De Lingüística, 5, 1–15.

24.

Polian

(2017). Tseltal and Tsotsil. In: Aissen

England

N. C.

Zavala Maldonado

(Eds.), The Mayan Languages (pp. 610–647). Routledge Taylor & Francis Group.

25.

Pye

(1990). The acquisition of Vo movement. In: Papers and reports on child language development (pp. 2996–2103).

26.

Pye

(1991). Acquiring lexical constraints on causatives in K’iche’ Maya. In: Papers and reports on child language development (pp. 3096–3103).

27.

Pye

(1993). A cross-linguistic approach to the causative alternation. In: Levy

(Ed.), Other Children, Other Languages (pp. 243–263). Erlbaum.

28.

Pye

(2017). The comparative method of language acquisition research. University of Chicago Press.

29.

Pye

(2019). Minimal coding page. http://pyersqr.org/minimal/

30.

Pye

Pfeiler

Mateo Pedro

(2017). Mayan language acquisition. In: Aissen

England

N. C.

Zavala Maldonado

(Eds.), The Mayan languages (pp. 19–42). Routledge Taylor & Francis Group.

31.

Raymundo González

Francisco Pascual

Mateo Toledo

Mateo Pedro

. (2000). Sk’exkixhtaqil yallay koq’anej: variación dialectal en Q’anjob’al. Editorial Cholsamaj.

32.

Richards

(2003). Atlas Lingüístico de Guatemala. Instituto de Lingüística y Educación-Universidad Rafael Landívar.

33.

Zavala

(1992). El Kanjobal de San Miguel Acatán. Universidad Nacional Autónoma de Mexico.