Navigation – Plan du site

Towards an Analysis of Complex Motion Event Packaging in South Eastern Huastec (Maya, Mexico)

Ana Kondic

Résumés

Cet article présente une forme préalablement sous-étudiée de l’expression du chemin dans les événements de mouvement complexes en huastèque du sud-est (HSF), une langue maya parlée au Mexique. Les données collectées sur le terrain montrent que le HSF ne possède pas de morphèmes directionnels (contrairement à la plupart des langues mayas), ni de vraies adpositions; le HSF ne dispose que d’une préposition locative générale (typiquement maya) et de quelques noms relationnels. Il est aussi dépourvu de gérondifs. Ces propriétés amènent cette langue à segmenter les événements de mouvement complexes en plusieurs propositions. La segmentation des événements en HSF est intéressante d’un point de vue typologique car c’est l’une des rares langues connues à segmenter les évènements en sous-événements. Du point de vue de la linguistique comparative maya, les langues huastèques, comme celles du groupe yucatèque, se distinguent des autres langues mayas dans le fait que la grammaticalisation des morphèmes directionnels exprimant le chemin n’a pas eu lieu.

Haut de page

Texte intégral

1. Introduction

  • 1 Alternative names: Huastec of San Francisco, or Teenek de la Sierra de Otontepec. HSF has about 12, (...)
  • 2 I want to express my profound gratitude to two anonymous CogniTextes reviewers, and especially to A (...)
  • 3 The aim of this project was to examine the expression of Path of motion in typologically varied lan (...)
  • 4 A preliminary version of this paper was presented during a working session of the Project in Lyon i (...)

1This paper presents an underdescribed and typologically interesting way of encoding the Path component of complex motion events in South Eastern Huastec1 (in further text HSF, according to its Ethnologue code), a Mayan language spoken in Mexico.2 This is the first study on the expression of space and motion in Huastecan languages, based on the fieldwork in Mexico and carried out within the “Trajectoire” project funded by the Fédération de Typologie et Universaux Linguistiques (CNRS, France).3 The data presented in this paper was first analysed in my PhD thesis (Kondic 2012). The aim of this paper is to extend the analysis.4

2The following examples demonstrate the type of constructions to be discussed in this paper. The first construction demonstrates the HSF way of expressing a complex motion event (Source, Goal and Median) as three different clauses, where English, for example, has only one, like in Pablo went from Tantoyuca to Mexico City past Chontla:

(1)

Paablo

Pablo

kal‑ej

exit‑COM

ti

PREP

Tantoyuca,

Tantoyuca/

wat’‑ey

pass‑COM

ti

PREP

Bitxow,

Chontla/

ul‑ich

arrive‑COM

Meejico

Mexico City

(CiHSF4)

Lit: ‘Pablo left Tantoyuca, he passed Chontla, he arrived in Mexico City.’

Desired reading: ‘Pablo went from Tantoyuca to Mexico City past Chontla.’

3The construction shown in (1) is the only way of encoding this type of complex motion event in this language. The second construction to be discussed here is exemplified in (2) and (3). These examples show that the expression of Source as in A girl walked away from the tree or Goal as in She went into a corn field, which is done with one clause in English, consists of distinct clauses in HSF, with one clause describing motion, and another clause relating either to Source or Goal, describing the ‘background’ of the scene:

(2)

bel‑ej

walk‑COM

juun

one

i

NM

txithan

girl

//

(T32-4)

kub‑at

stand‑PPL

wik

PAST

t‑in

PREP‑E3SG

akan

feet

i

NM

te’

tree

Lit: ‘A girl walked, (she) was standing at the foot of a tree.’

Desired reading: ‘A girl walked away from a tree.’

(3)

taal

come

ti

R

bel‑al

walk‑INC

//

och‑ich

enter‑COM

ti

PREP

eemlam

corn.field

(T069‑26‑3)

Lit: ‘(S/he) comes walking, (she) entered the corn field.’

Desired reading: ‘She went into a corn field.’

  • 5 Alternative orthography: Tzeltal.

4Languages vary considerably in the lexical and syntactic means through which they encode such complex events, and in ways they package them syntactically. Bohnemeyer et al. (2007) investigated the encoding of complex motion events across languages in order to examine what kinds of constraints languages impose on event segmentation, and whether there are some universal principles of event encoding at the syntax-semantics interface. As will be discussed in more detail in section 3 of this article, the Bohnemeyer and al.’s research has shown that even closely related languages such as Yucatan and Tseltal5 (both Mayan) vary in their encoding of complex motion events. While Tseltal is a verb framed language that encodes Path in both the verb and the directional, Huastec and Yucatec do not possess directionals (see 6.4) and thus encode the path uniquely in verbs. This phenomenon is described in details in this article.

5Although this research provides an important contribution to our understanding of how languages segment complex motion events, the phenomenon of event segmentation remains underdescribed and we still lack the description of this phenomenon in individual languages. The aim of the present article is to contribute to the description of this phenomenon by investigating another Mayan language, South Eastern Huastec, and by examining how speakers of this language package complex motion events. This is the first study of this kind in Huastecan languages; although this analysis reflects work in progress I hope it still brings valuable insights into the system of space and motion encoding in the Huastecan languages.

6Section 2 outlines the context of this study of the expression of Path in HSF, including information on fieldwork, the HSF speakers and the data. The typological framework used in this analysis is described in section 3.  The previous studies on the expression of space across the Mayan family are presented in section 4 so as to put the HSF data in perspective. Section 5 deals with the basic notions of the HSF grammar. Section 6 describes the HSF means for encoding motion, while section 7 deals in detail with the expression of complex motion events in HSF. Finally, section 8 concludes the discussion on segmenting events in this language.

2. Methodology: fieldwork, speakers and the data

  • 6 Additional support to carry out the fieldwork was provided by the Mexican Government scholarship an (...)

7The data presented in this article comes from twelve months of fieldwork carried out by the author in the village of San Francisco Chontla (Veracruz, Mexico) between 2007 and 2011. A major focus of this fieldwork was the documentation of this endangered language, which was funded by the Hans Rausing Endangered Language Program of SOAS-London.6

8The particular data discussed in this paper is a combination of natural text material collected for the HRELP documentation project and data elicited with visual material specifically designed for the study of the expression of space. The data was collected using the following stimuli:

  • ‘Topological Relations Picture Series’ (Bowerman and Pedersen 1992), with 3 speakers

  • ‘Trajectoire’ video-clips (Ishibashi, Kopecka, and Vuillermet 2006), with 4 speakers

  • ‘Frog, where are you?’ (wordless book by Mercer Mayer, 1969), with 2 speakers

9The HSF speakers who contributed to this study were all fully bilingual: they are native speakers of both HSF and Spanish. They learned HSF natively from their parents and then subsequently Spanish, whilst at primary school. All of them were born and have lived all their life in the village of San Francisco Chontla. When these data were collected, Cirila was 42 years old, Goyo 50, Amelio 52 and Santiago was 47.

3. Typological framework

10The starting point for this analysis is the notion of Path as defined by Slobin (1997: 439), for whom the Path implies translational motion: “… a Path moves from a source to a goal, along or through some medium, passing one or more milestones”. For example, the Path expressed in the sentence ‘Maria went from her home to the supermarket crossing the bridge.’ can be divided into three parts (sequences or segments): Source (‘from her home’) – Median (‘crossing the bridge’) – Goal (to the supermarket).

11As mentioned earlier, languages vary in the kinds of morphosyntactic devices they use to express Path and in the way they distribute Path components in the sentence. First, as pointed out by Talmy (1985, 2000), while some languages encode Path in the verb (verb-framed type) others encode it in a verb satellite (satellite-framed type).  Furthermore, depending on lexical and grammatical resources, some languages (such as English) can express different portions of Path in one clause, others (such as Huastec, as we will see later) tend to distribute the different portions of Path in several verb clauses.

12The typological framework used for the present analysis is Bohnemeyer et al. (2007). This group of authors were interested in establishing a semantic typology of motion-event encoding, in other words, in studying ‘how conceptually comparable event representations are segmented across units of linguistic code’ (2007: 524), since events are encoded not just by lexical terms, but by verb phrases, clauses and larger discourse units that vary in the packaging of event information. Their research uncovered asurprising amount of variation in motion-event segmentation, caused by differences in lexicalisation and in the availability of syntactic constructions across languages.

13 For measuring event packaging across languages they proposed the notion of macro-event property (MEP). It refers to the property of languages to pack events in the way that temporal operators necessarily have scope over all sub-events (what they call the ‘tightness’ of packaging of sub-events in aconstruction). According to this typology the constructions that have the MEP present ‘an event in terms of a unique initial and/or final boundary, a unique duration, and a unique position on time line’ (2007: 524); in other words, a clause with a MEP would combine a departure, an arrival and a passing sub-event all in one clause with only one verb phrase, as in English ‘Floyd went from Nijmegen across the river to Elst’ (Bohnemeyer et al. 2007: 498). In many languages this is not possible.

14After having examined 18 genetically and typologically diverse languages, Bohnemeyer et al. classified these languages into three types on the basis of how many and what kinds of location-change sub-events they can integrate into one macro-event construction. Apart from that, and as highlighted by the authors, the type under which a given language falls largely depends on “the interplay of two factors: (i) lexicalization – the expression of path or location-change functions in verbs, satellites (or ground phrases) or both; and (ii) the availability of certain morphosyntactic constructions” (Bohnemeyer et al. 2007: 516).

15Languages of the Type I of this typology “have clause- or phrase-level constructions that have the MEP and licence combinations of maximally one departure, arrival and passing sub-event each, as in English” (2007: 509); this is the case of languages such as Dutch (Indo-European, West Germanic), Ewe (Kwa, Ghana), Lao (Thai-Kadai, Laos), Marquesan (Polynesian, Marquesas Islands), Tiriyó (Carib, Brazil). That is, these languages can express complex motion events including the departure, the passing and the arrival sub-events in only one clause and with only one verb phrase.  Languages of this type are either satellite-framed (Talmy 1985, 2000) and permit multiple Path components in a single verb phrase (like English) or have multiverb constructions that form macro-event expressions. The examples below (4) from English and from Tiriyó (5) are both satellite-framed and can package several sub-events into one verb phrase:

(4)

Floyd went from Nijmegen across the river to Elst.

(5)

Kau

cow

wewe‑pisi

wood‑DIM

enee‑ja‑n

bring‑PRES‑EVID

wewe‑pǝe

wood‑from

ǝema‑tae

path‑along

kanawa‑pona.

vehicle‑toward

‘The cow is bringing the little stick from the tree along the path to the vehicle.’

  • 7 Serial verbs constructions are considered one clause.

16Other languages belonging to Type I, for example Lao in (6), may segment events like the type III languages (described below) using a separate verb phrase for each location-change sub-event. However departure-passing-arrival-denoting verb phrases are combined in multiverb constructions that have a MEP,7 which is why these languages are classified as Type I:

(6)

Man2

[3

lèèn1

run

(qòòk5)

exit

caak5

from

hùan2

house]VP

taam

[follow

thaang2

path]VP

hòòt

[reach

kòòn4‑hiin3.

CL-rock]VP

‘He ran (exited) out of the house, followed the path, reached the rock.’

17Type II languages have “macro-event expressions that may combine a departure and an arrival event, but may require a separate macro-event expression for the encoding of passing events, depending on the type of passing event” (2007: 511). Languages such as Basque (isolate), Japanese (isolate), Hindi (Indo-Aryan), Arrernte (Pama-Nyungan, Australia) and Trumai (isolate, Brazil) belong to this type. The languages of this group express the event ‘‘Floyd went from Nijmegen across the river to Elst’ using two verb clauses as in ‘Floyd went from Nijmegen to Elst, crossing the river’, putting departure and arrival into one MEP expression, and the passing into a separate MEP expression. These are languages that encode Path in both the verb root (verb-framed) and in the ground phrase (double-marking strategy). The following example is from Basque, where the departure and arrival are expressed in one clause but the passing event is singled out (for timing) into a separate clause:

(7)

Atzo

Yesterday

Arrasate‑tik

Arrasate‑ABL

Oinati‑ra

Oñati‑ALL

joan

go.PERF

zen

AUX.3SG

mendi‑ak

mountain.PL.ABS

eguerdian

noon.LOC

zeharkatu‑ta.

cross.PERF‑CON

‘Yesterday (s)he went from Arrasate to Oñati, crossing the mountains at noon.’

18In Hindi, single clauses comprising departure, passing and arrival events are possible (as in the Type I languages). However, Hindi requires a converbial construction if the route is not coextensive with the Path:

(8)

Voh

he.NOM

ghar=se

home=ABL

dukaan

store

ho‑kar

be‑CON

daftar

office(DAT)

gayaa.

go.SG.M.PERF

‘He went from home to the office via (being at) the store.’

19Finally, the languages belonging to Type III (and the ones that we are most concerned with here) “require a separate VP for encoding each location-change sub-event that involves a distinct ground” (Bohnemeyer et al. 2007: 515). These languages lack multi-verb constructions that combine multiple location-change-denoting verb phrases into a single MEP and, hence, express the change of location exclusively in verb roots. Languages that encode complex motion in this way include Kilivila (Austonesian, Papua New Guinea), Jalonke (Mande, Guinea), Saliba (Austronesian, Papua New Guinea), Yeli Dnye (East Papuan, Papua New Guinea), Tidore (West Papuan, Indonesia), Zapotec (Oto-Manguean, Mexico), Yucatec and Tseltal (Mayan, Mexico).

20In these languages, all the three sub-events (departure, passing, arrival) of the event ‘Floyd went from Nijmegen across the river to Elst’ are expressed by three different clauses, i.e. ‘Floyd left Nijmegen, crossed the river, and arrived in Elst’ or ‘The circle rolled from the square, then passed the house-shaped object, and finally reached the triangle’ (Bohnemeyer et al. 2007: 510). In these languages, a complex motion event and its sub-events are hence divided into several components (or sequences). The examples from Jalonke (9) and from Kilivila (10) illustrate this phenomenon:

(9)

a

3sg

keli

leave

wuri‑n’ii’,

tree‑DEF.LOC

a

3sg

siga

go

(haa)

until

gɛmɛ‑n’ii’.

rock‑DEF.LOC

‘He left the tree, (and) went as far as the rock.’

(10)

Kaukwau

[dog

e‑kaitau

3‑carry

bunukwa

pig]s

e‑la

[3‑go

va

ITI

vaya

creek]s

e‑lupeli

[3‑cross

e‑la

3‑go

va

ITI

kai.

tree]s

‘The dog carries pork, it goes to the creek, it crosses it, it goes to the tree.’

21The following example is from Yucatec, a Mayan language from Mexico, a close relative of HSF. This example shows how a complex motion event is segmented with insertion of time reference then between the segments (Bohnemeyer et al. 2007: 515):

(11)

le=chan

DEF=DIM

ba'l

thing

chak=o',

red(B.3SG)=

D2]

k‑u=bin

[IMPF‑A3=go

u=balk'=e',

A3=roll=TOP]

k‑u=ts'o'k-ol=e'.

[IMPF‑A.3=end‑INC=TOP]

‘The little thing that's red, it went rolling, and then

(12)

k‑u=màan

[IMPF‑A.3=pass

y=iknal

A.3=at

hun‑p’éel

one‑CL.IN

chan

DIM

ba’l

thing

chak

red(B.3SG)

xan=e’,

also=TOP]

it passes by a little thing that's also red,

k-u=ts’o’k‑ol‑e’,

[IMPF‑A.3=end‑INC‑TOP]

k‑u=k’uch‑ul

[IMPF‑A.3=

arrive‑INC

y=inknal

A.3=at

le=tràangulo

DEF=triangle

àasul=o’.

blue(B.3SG)=

D2]

and then it arrives at the blue triangle.’

22We might note that the authors also provide the parameters for testing the MEP: first, the possibility of insertion of temporal adverbs, conjunctions or intonational breaks between the segments, and second, the negation having scope over just one single verb phrase (2007: 516). The HSF examples have been tested (see §7.4 for details) according to these parameters. As it will be demonstrated below, this language together with Yucatec and Tseltal clearly belongs to Type III of this typology.

23Before examining the way HSF expresses complex motion events, I firstly provide a brief overview of the research on the expression of space in Mayan language and then some basic information about the structural properties of HSF.

4. Space studies in Mayan languages

  • 8 For the past sixteen years the Mayan space expression (in Yucatec, Mopan, Tsotsil and Tseltal) has (...)

24Some studies on the expression of space in Mayan languages focus on the presence of positionals and their use (Brown 1994, Martin 1994, Grinevald 2006). Others focus on the widespread phenomenon of directional (which can be realized as suffixes, post verbal particles or prefixes), their verbal origin, their various levels of grammaticalization, and their structural characteristics (Brown 2006, Craig 1994, Grinevald 2006, Grinevald 2010, 2011, Haviland 1991, 1993, De Leon and Levinson 1992, Mateo-Toledo 2004, Martin 1994, Larsen 1994, Quizar 1994, Zavala 1993, among many others).8 However, less has been written on the phenomenon of segmenting complex motion events, which thus far has been attested only in Yucatec (Bohmeneyer et al 2007) and Tseltal (Brown 2006). The present study contributes new information to a language in the Huastecan branch.

5. Basic notions about South Eastern Huastec grammar

25South Eastern Huastec is a polysynthetic language with a rich verbal morphology. The languages of the Huastecan branch are the only languages of the Maya family that have developed an inverse system of person marking embedded into the traditional Mayan ergative system. Aspectual markers are suffixed; the progressive and future are pre-posed. The HSF features of interest for this topic are outlined below. They will permit us see that the different segments of the complex motion event in HSF are actually separate clauses of a complex sentence, in which the subject is marked by a person marker and the verb is marked by an aspectual marker. The structure of the HSF verbal complex is: person marker – verbal root – suffixes (voices/applicatives/causative – transitive suffix - aspect)

5.1 Person marking in HSF

26HSF has three sets of personal markers: (a) Absolutive that are used for encoding the intransitive subjects, (b) Ergative that are used for encoding the transitive subjects and possessors of nouns, and (c) Inverse that encodes one part of the transitive agreement, namely specific subject-object combinations (for more see Kondic 2012):

a. Absolutive set

SG

PL

1

in

U

2

it

Ix

3

a/Ø

Ip

b. Ergative set

SG

PL

1

U

I

2

A

I

3

In

I

  

(12)

nanaa’

I

in

A1SG

kal‑ej.

leave‑COM

‘I left.’

(13)

nanaa’

I

u

E1SG

ch’a’–iy

buy‑TS(COM)

juun

one

i

NM

kwita’.

chicken

‘I bought a chicken.’

c. Inverse set

Marker

Subject

Object

tin

2 SG/PL, 3SG/PL

1SG

ti

3 SG/PL, 1PL

2SG

tu

2SG, 3SG/PL

1PL

1SG

2SG

tixi

3SG/PL, 1PL

2PL

tuxu

1SG

2PL

2PL

1PL

  

(14)

Tin

2SG>1SG

chuuj

see.COM

ti

PREP

we’eel

yesterday

nuu’

there

ti

PREP

plaasa

market

(AmArch204)

‘You saw me yesterday at the market.’

5.2 Aspect marking

27In HSF there are five different aspectual forms that indicate the type of action or time; three of them are marked on the verb by particular aspect markers and the other two are expressed periphrastically. These aspectual markers are always placed after the verb root (in intransitive verbs) or after the stem (in transitive verbs). According to Bohnemeyer (1997: 22), “there is no grammaticalized expression of absolute time reference, i.e. reference to a moment in time as defined with respect to coding time.” The meaning of time in an HSF verb is retrieved pragmatically from context.

28There are three different verbal aspects in HSF: completive, incompletive and perfect. Their forms vary depending on the voice used. Progressivity is conveyed by the auxiliary verb, exom which can be used with both transitive and intransitive verb stems. If the auxiliary verb exom is followed by an Absolutive person marker, it is preceded by a subordinator t-. Progressivity can also be encoded by other verbs, namely verbs of motion:

(15)

ip

A3PL

xe’ech

go.INC

t‑up

R‑A3PL

paajn‑aax

chase‑REC.INC

(CirRec65)

‘They are chasing each other.’

29Future is expressed in a periphrastic construction with the grammaticalized verb of motion ne’ech go,plus the subordinator t- (glossed Realis, withincompletive aspect)or k (glossed Irrealis, used with completive aspect). It could be noted here that in HSF t- has three different functions, namely: general locative preposition (pan Mayan), realis subordinator, and inverse alignment marker.

30Particles with temporal/aspectual meaning convey different nuances of the meaning, e.g., wik PAST, ich already, ej perdurative, chap again.

5.3 Basic word order: SVO with flexibility

31Unlike in other languages of the family, the word order in HSF appears to be rather flexible. Four word orders were found in the HSF corpus: SVO, VOS and OVS, which are frequent in the data, and VSO, which is also possible but less frequent. The following examples illustrate the first three possibilities. We may note however that (16a) and (16b) are identified by the native speakers as the most acceptable, and that (16c), although perfectly grammatical, is not very common.

(16)

a.

na

HUM

Beatriis

Beatris

in

E3SG

koo’‑y‑al

have‑TS‑INC

juun

one

i

NM

mixtun

cat

SVO

‘Beatriz has got a cat.’

b.

juun

one

i

NM

mixtun

cat

in

E3SG

koo’‑y‑al

have‑TS‑INC

na

HUM

Beatriis

Beatriz

OVS

‘Beatriz has got a cat.’

c.

in

E3SG

koo’‑y‑al

have‑TS‑INC

juun

one

i

NM

mixtun

cat

na

HUM

Beatriis

Beatriz

VOS

‘Beatriz has got a cat.’

(CirE6‑12)

32The free placement of the subject and the object in (16) is possible partly because of the ‘natural state of the things’, as, for example, a cat cannot possess Beatriz. Examples (17) illustrate the same phenomenon. All the combinations shown below are possible because of the ‘natural state of the things’; an interpretation like ‘I gave a man to some money’ would not make any sense.

(17)

a.

u

E1SG

pith‑a’

give‑TS(COM)

i

NM

meeluj

money

juun

one

i

NM

inik

man

SVOO

‘I gave some money to a man.’

(CirE6‑11)

b.

u

E1SG

pith‑a’

give‑TS(COM)

juun

one

i

NM

inik

man

i

NM

meeluj

money

SVOO

‘I gave some money to a man.’

c.

i

NM

meeluj

money

u

E1SG

pith‑a’

give‑TS(COM)

juun

one

i

NM

inik

man

OSVO

‘I gave some money to a man.’

d.

juun

one

i

NM

inik

man

u

E1SG

pith‑a’

give‑TS(COM)

i

NM

meeluj

money

OSVO

‘I gave some money to a man.’

5.4 Complementation

33Complementation constructions are multi-clausal in HSF. They can contain a complementiser baal that or kee (Sp) that, or a subordinator t- or k-:

(18)

tu

1SG>2SG

utx‑a‑al

say‑TS‑INC

baal

that

k‑a

IRR‑E2SG

cho’oobn‑a’

know‑TS(COM)

(Arch225)

‘I am telling you this so that you know.’

(19)

a

E2SG

wit’‑a‑al

can-TS-INC

t‑it

R-A2SG

laatx‑um

swim-AP(INC)

(SantArch218)

‘Can you swim?’

(20)

u

E1SG

uktx‑iy

forge‑TS(COM)

k‑u

IRR‑1ESG

map‑uy

close‑TS(COM)

an

DEF

wii’lep

door

(Arch262)

‘I forgot to close the door.’

34It is easy to identify the Subject in these cases because these are multi-clausal constructions where person marking is obligatory. Compare the above example (18) where each clause has its own (different) subject with (19) and (20) where both verb clauses share the same subject.

5.5 Subordination

35Subordinate clauses in HSF express temporal, purposive, conditional, and causal relations to the main clause. Manner adverbial clauses are expressed in HSF by means of complement-taking predicate constructions or depictive constructions. One of the HSF subordinate clause types is a temporal clause, as exemplified in the following complex sentence:

(21)

na

HON

jwaan

Juan

wa

A3SG

way‑al

sleep‑INC

wik

PAST

taam

when

t‑it

R‑A2SG

uli‑ch

come‑COM

(AmArch565)

‘Juan was sleeping when you arrived.’

36Typical Mayan motion-cum-manner and motion-cum-purpose constructions constitute a sub-class of HSF complex sentences. Motion-cum-manner construction, as illustrated in (22) and (23), is a combination of a verb of motion and a verb that describes the manner of the motion.

(22)

ip

A3PL

kalej

leave‑COM

tip

R‑A3PL

jumnal.

fly‑MID‑INC

(AmRana39‑11)

‘They left flying.’

(23)

naa’

there

ich

already

jayeej

he.too

wa

A3SG

kalel

leave‑INC

ti

R

laatxum

swim‑AP(INC)

k’ayuum.

slowly

(AmR54‑156)

‘Here he is coming swimming slowly.’

37The irregular verb taal come can participate in this construction. The example (24) illustrates a scene from the Frog Story describing setting:

(24)

an

DEF

kwitool

boy

jee’

there

taal

come

ti

R

ch’ub‑eb‑eel

swing‑RED‑POSIT

eep

outside

an

DEF

kotop

ravine

‘The boy is falling down swinging through the air, in the ravine’

38As its name says, the motion-cum-purpose constructionis a combination of a verb of motion with another full verb that denotes a purpose:

(25)

ut’‑ey

approach‑COM

[

baal

for

tin

3SG>1SG

kwath‑a’

hit‑TS(COM)

]

(IrE3‑34‑25)

‘He approached (me) to hit me.’

(26)

k’ath‑iy

climb‑COM

[

k‑in

IRR‑E3SG

aln‑a‑al

search‑TS‑INC

[

maax

if

koo’

SPEC

taa’

there

ak

IRR

k’waj‑at

be‑INC

an

DEF

k’wa’

frog

]

(CirR151‑61)

‘He went up to see if the frog was there.’

39The verbs in both matrix and subordinate clauses are marked with a personal marker. Regarding the placement of elements in these motion-cum-manner and motion-cum- purpose constructions, it should be mentioned that in most cases they have a fixed position: as seen in the examples above, the manner element immediately follows the main verb. Nevertheless, the manner element can sometimes precede the main verb, in which case a cleft construction is triggered, as in (27b) (focusing strategy):

(27)

a.

xaxaa’

you.pl

ix

A2pl

txi’‑ich

come‑COM

t‑ix

R‑A2pl

bel‑al

walk‑INC

(AmArch404)

‘You came walking.’

b.

xaxaa’

you.pl

ix

A2pl

bel‑al

walk‑COM

an

DEF

t‑ix

R‑A2pl

txi’ich

come‑COM

(AmArch404)

Lit: ‘It is by walking that you came?’

‘You came walking?’

5.6 Absence of gerunds

40HSF does not have gerunds. Instead complex sentences are used, as the examples (28a) and (28b) demonstrate. The events encoded in these two examples are simultaneous:

(28)

a.

in

E3SG

thutx‑a‑al

write‑TS‑INC

an

DEF

kaarta,

letter(Sp)

in

E3SG

ach’‑a‑al

drink‑TS‑INC

an

DEF

kapee.

coffee(Sp)

(AmE3)

Lit: He is/was writing the letter, he is/was drinking coffee.

Desired reading: ‘He is/was writing the letter drinking coffee.’

b.

in

E3SG

thutx‑a’

write‑TS(COM)

an

DEF

kaarta,

letter(Sp)

in

E3SG

ach’‑a‑al

drink‑TS‑INC

an

DEF

kapee.

coffee(Sp)

(AmE3)

Lit: He wrote the letter, he was drinking coffee.

Desired reading: ‘He wrote the letter drinking coffee.’

41This phenomenon is directly related to the way HSF encodes complex motion events. The simultaneous reading is due to the use of the incompletive aspect marker in the second clause in these examples (28a, 28b.). Two consecutive events would be both encoded by a completive marker, as in (28c):

(28)

c.

in

E3SG

thutx‑a’

write‑TS(COM)

an

DEF

kaarta,

letter(Sp)

in

E3SG

ach’‑a’

drink‑TS(COM)

an

DEF

kapee.

coffee(Sp)

(AmE3)

Lit: He wrote the letter, he drank coffee.

Desired reading: ‘He wrote the letter (and then) he drank coffee.’

42The following example demonstrates a complex motion event in the past with the passing segment encoded by incompletive aspect:

(29)

n‑u

DEM‑E1SG

txithaan‑il

daughter‑POSS

k’al‑ej

go‑COM

an

DEF

ti

PREP

k’imaa’

house

nujneel

merchandise

/

in

E3SG

jaluk’n‑a‑al

cross‑TS‑INC

an

DEF

beel

road

(NarHSF4)

Lit: ‘My daughter went to the store, she crossed the road.’

Desired reading: ‘My daughter went to the store crossing the road.’

43This is the only way to express that complex motion event in HSF. As mentioned above, Mayan languages do not have tenses and lack syntactic relators, anaphoric connectives and temporal prepositions, so that the temporal reference is inferred from context.

6. HSF means for encoding space: position, location, motion

44The HSF means for encoding space, including position, location and motion are presented in this section.

6.1 Expression of movement: Intransitive verbs of motion

45Verbs of motion are often very similar, if not identical, across the languages of the family.Table 2 shows a list (not expected to be complete) of HSF verbs of motion:

belej

walk

pa’ay

descend

ch’akay

rise (get up)

pujk’in

fall down

ch’a’ey

fall down

taal

come

kalej

exit, go out, leave

txi’ich

come

k’alej

go

ulel

arrive

k’athiy

ascend

wat’ey

pass

kwajlan

fall down

witxiy

return

ne’ech

go

xe’ech

walk

ochich

enter

Table 2: HSF Intransitive Verbs of Motion

46The verb taal come (towards the speaker), shown in (30), is an irregular verb in HSF: it is never inflected for aspect and this form is its only form. This word is present in other Mayan languages as a directional: for example, in Tseltal tal come, arrivehere (Brown 2006) or Mocho taacoming or bringing along toward here’ (Martin 1994: 141).

(30)

baa’

no

u

E1SG

cho’oop

know

tamaa’

who

ni

DEM

taal

come

(CirArch251)

‘I don’t know who is coming.’

47The verbs pujk’i fall down, kwajlan fall down and jilk’on stay, remain (with the semantics of the lack of movement) are in Middle voice, which is their only form (they do not have an active voice):

(31)

na

HUM

jwaan

Juan

kwajl‑an

fall‑MID(COM)

an

DEF

ti

PREP

moom

well

(AmArch460)

‘Juan fell into the well.’

(32)

in

A1SG

ne’ech

FUT

txi’‑ich

go‑COM

wik

PAST

k‑in

IRR‑A1SG

kal‑ej

exit‑COM

ti

PREP

merkaadu

market

pero

but(Sp)

in

A1SG

jilk’‑on

remain‑MED(COM)

tee’

here

an

DEF

ti

PREP

ataaj

home

(AmArch312)

‘I was going to go (out) to the market, but I stayed here at home.’

48Some verbs of motion do not refer strictly to motion and have, like jilk´on stay, remain secondary or metaphoric meaning:

(33)

baal

so.that

k‑a

IRR‑A3SG

jilk’‑on

remain‑MID(COM)

ja’lith

wet

(GoyC2‑87‑29)

‘.... so that it gets wet.’

49Some other motion verbs, like ne’ech go and xe’ech go around, have been grammaticalized as auxiliaries. The verb ne’ech go encodes the meaning of future and habitual, as shown in (34) and (35), while xe’ech go around, as in (36), forms complex sentence constructions with a progressive meaning and a nuance of  movement:

(34)

ne’ech

FUT

xee’

now

k-u

IRR‑E1SG

lejk‑iy

prepare‑TS(COM)

an

DEF

k’ooyej

dough

(BerV9‑1)

‘I am going to prepare the dough now.’

(35)

och‑k‑an‑al

enter‑DM‑MID‑INC

a

HON

k’iixaaj

sun

ne’ech

HAB

k‑u

IRR‑A1PL

kal‑ej

go‑COM

(LeC2‑102‑19)

‘We used to go (home) when the sun sets.’

(36)

ip

A3PL

xe’-ech

go-INC

t-up

R-A3PL

paajn-aax

chase-REC(INC)

(CirRec65)

‘They are chasing (around) each other.’

50The most common auxiliary verb in HSF is the verb k’wajat be (located) (like the Spanish estar):

(37)

n‑a

DEM‑E2SG

paa

father

k’waj‑at

be‑INC

t‑in

PREP‑E3SG

k’imaa’

house

(AmArch38)

‘Your father is at home.’

51The HSF intransitive verbs of movement given in Table 2 can be divided into sub-groups on the basis of their nuances in semantics:

Directional motion verbs (which convey motion and Path)

a.

Source-oriented:

kalej

exit

b.

Goal-oriented:

ulel

arrive

txi’ich

come

ochich

enter

witxiy

return

k’athiy

ascend

pa’ay

descend

c.

Figure-passing:

jeka’

cross

krusaariy (Sp)

cross

Manner-oriented verbs in HSF that imply motion

laatxum

swim

aathil

run

tik’on

jump

juuman

fly

Verbs without a specified direction of movement

wat’ey

pass by

belej

walk

Verbs marking lack of movement

jilk’on

stay, remain

k’wajat

be located

52Verbs like fly, swim, flee, chase are considered in Tzetal ‘Manner and Path’ verbs as they convey not only motion but also ‘a kind of manner associated with the scene’ (Brown 2000: 70). It is interesting to mention that the verbs swim and run in HSF are historically transitive verbs (swim a swim, run a race, have a bath), their only form in contemporary HSF is in the antipassive voice. Verbs of this type also occur in other Mayan languages (England 1983). The verbs juuman fly, t’ikon jump, and  jilk’on stay, remain in HSF are Middle voice forms (their only form).

6.2 Expression of Ground: General prepositions, deixis and relational nouns

53HSF does not have a copula. The HSF data shows that there are two basic locative constructions in this language:

  • an existential predicate used for location too: wa’ach ‘there is’

  • a verb of location (like Spanish estar): k’wajat ‘to be (located)’

54These two location encoding verbs appear in HSF with a preposition.

6.2.1 Prepositions

55There is only one preposition in HSF: ti, a typical Mayan preposition with a very general meaning of location. The locative preposition does not encode Path meaning, which is wholly encoded within the verbs. That is, the HSF is a strictly verb-framed language (Talmy 1985). This general preposition ti is used to convey both location and motion, as the following examples describing location (38, 39, 40), Source (41) or Goal (42) show:

(38)

chanak’w

bean

k’waj‑at

be‑INC

an

DEF

ti

PREP

patx

pan

(IrArch55)

‘The beans are in the pan.’

(39)

an

DEF

mixtun

cat

bux‑uul

squat‑INC

an

DEF

ti

PREP

mesa

table

(GoyBowP)

‘The cat is sitting under the table.’

(40)

n‑u

REL‑E1

paa

father

k’waj‑at

be‑INC

ti

PREP

ale’

corn.field

(IrArch40)

‘My father is in a corn field.’

(41)

n‑u

REL‑E1

paa

father

kal‑ej

exit‑COM

ti

PREP

ataaj

house

(SanArch453)

‘My father went out of the house.’

(42)

na

HUM

Jwaan

Juan

kwaj‑l‑an

fall‑DM‑MID(COM)

an

DEF

ti

PREP

moom

well

(AmArch460)

‘Juan fell into the well.’

56This is a common situation in languages that lack an elaborate set of prepositions, such as Mayan languages. Because of its general meaning, the pan-Mayan preposition ti allows a number of different readings; its ‘function simply consists in relating any kind of peripheral participant to the event core expressed by the verbal complex’ and may be translated as ‘with the respect to’ (Bohnemeyer & Stoltz 2006: 286). As noticed by Bohnemeyer (1997), this preposition is ‘neither sensitive to the source-goal distinction nor to the dynamicity of the event core’.

6.2.2 Deixis and locational (adverbial) words

57There are some deictic words in HSF that participate in encoding space:

Proximal

Distal

Adverb

tenchee’

taa’

Demonstrative

axee’

naa’, naja’

58Locational words (adverbial words) encode location (static) and are often used together with another word indicating location, as in (45) and (46).

alk’ith

down there

junaxch’ejel

in the middle

altaa’

inside

k’unat

close

eblith

on, on the surface/top of

t’ek’at

up there

6.2.3 Relational nouns

59The relational nouns are a pan Mayan and Mesoamerican phenomenon. They represent possessive constructions with a locative meaning, like for example, ‘at his back’ meaning ‘behind him’ or“at its top the stick” meaning ‘on the top of the stick’:

t‑in

PREP‑E3SG

kuux

back

t‑in

PREP‑E3SG

txuum

top

an

DEF

te’

tree

‘behind him’

‘on the top of the stick’

  • 9 In data elicited with the Bowerman and Pedersen questionnaire (1992) the relational nouns appeared (...)

60The relational nouns are numerous in Mayan languages, but in HSF they are rather scarse.9 The following examples illustrate their usage in locational expressions:

(43)

bal‑ith

insert‑PPL

t‑in

PREP‑E3

wii’lep

door

k’imaa’

house

an

DEF

pik’o’

dog

(AmBP71C)

‘The dog is at the door of his house.’

61Often, a locational word is used with a relational noun to emphasize the expression of space:

(44)

k’waj‑at

be‑INC

jee’

here

t‑u

PREP‑E1SG

kuux

back

(AmArch45)

‘(S/he) is sitting here behind me.’

62In examples (45) and (46) the spatial relation is expressed by a locational word t’ek’at, which encodes the static location (up), plus a relational noun to signal the direction t-in juntxal (in the direction of). The notion of Path is therefore expressed by the combination of an adverbial word that encodes a point of localization, and a relational noun that encodes direction. This is an instance of a Path without movement (Grinevald 2011).

(45)

k’waj‑at

be‑INC

t’ek’at

up.there

t‑in

PREP‑E3SG

juntxal

direction

an

DEF

meesa

table(Sp)

(AmBPed13)

‘(The lamp) is upstairs in the direction of the table.’

(46)

pal‑at

hang‑PPL

we’

small

i

NM

tokow

cloud

t’ek’at

up.there

t‑in

PREP‑E3SG

lujtal

direction

an

DEF

cheen

hill

(BowPed36C)

‘There is a small cloud up there in the direction of the hill.’

6.3 Expression of Figure: Positionals and verbs of posture

63In HSF, like in many other Mayan languages, position/posture is an important element of the description of motion, and position is often integrated together with motion into one expression.

  • 10 Brown (2006:246) calls them dispositionals.
  • 11 Positionals have even been recognized in Mayan hieroglyphics.

64Positionals are a pan-Mayan phenomenon.10 They are also present in many Mesoamerican languages. Positionals convey, as their name says, position or posture, as well as manner or some other specific meaning like size, shape or orientation toward the ground. While positionals form an important word class in majority of Mayan languages where their number can number in the hundreds (Brown 2006: 246 for Tseltal; Martin 1994: 39 for Mocho; Ameka & Levinson 2007),11 they are not numerous in HSF: only 7 positionals have been attested innaturalistic data, and about a dozen through formal elicitation. Some of HSF positionals (derivational suffix – VVl ) are as follows:

buxuul

squat

emeel

lean

kutuul

sit

  

(47)

bux‑uul

squat‑POSIT

an

DEF

ti

PREP

maplap

wall

an

DEF

pik’o’

dog

(BowPed6C)

‘The dog is squatting next to the wall.’

  • 12 The combination of positionals with motion verbs, directional and auxiliaries in spatial language i (...)

65According to Brown (2000: 69), positionals bring manner-like information into the verbal clause:12

(48)

an

DEF

pik’o’

dog

metx‑eel

lay.mouth.down‑POSIT

altaa’

inside

an

DEF

ti

PREP

ataaj

house

‘The dog is laying mouth down inside his house.’

(BowPed71B)

66As shown in the example below, positionals indicate the Figure’s position in the description of his or her movement, where motion and spatial configuration are combined:

(49)

ani

and

an

DEF

kwitool

boy

jee’

there

taal

come

ti

R

ch’ub‑eb‑eel

fly‑RED‑POSIT

eep

outside

an

DEF

kotop

ravine

‘The boy is coming ‘waving’ down the ravine.’

67Positionals are not numerous in HSF. Their function is often fulfilled by adverbs:

(50)

an

DEF

kwitool

boy

je’at

mouth.up

an

DEF

ti

R

ne’ech

FUT

k‑a

IRR‑A3SG

ch’a’‑ey

fall‑COM

(AmR118)

‘The boy is going to fall down on his back.’

(51)

an

DEF

pik’o’

dog

chepat

head.first

jayej

also

an

DEF

ti

R

ne’ech

FUT

k‑a

IRR‑A3SG

pa’‑ay

descend‑COM

(AmR117)

‘The dog is also descending (falling down) head first.’

68In the descriptions of static relations elicited using the Bowerman-Pederson (1992) stimuli, participles are abundant (out of 211 clauses, participles are used 137 times; compare this to only 25 positionals):

(52)

an

DEF

fleecha

arrow

xap‑at

put‑PPL

an

DEF

ti

PREP

mansaana.

apple(Sp)

(BowPed31A)

‘The arrow is (put) at the apple.’

(53)

an

DEF

kwaadru

painting(Sp)

pal‑at

hang‑PPL

an

DEF

ti

PREP

pareeth

wall(Sp)

(BowPed33B)

‘The painting is hanged at the wall.’

(54)

an

PREP

tool

fish

k’waj‑at

is(located)‑INC

an

DEF

ti

PREP

fraasko

jar(Sp)

(BowPed34A)

‘The fish is in the jar.’

(55)

pun‑at

put.up‑PPL

t‑in

PREP‑E3SG

ook’

head

an

DEF

ataaj

house

juun

one

i

NM

inik

man

(BowPed30B)

‘A man is ascended/gone up the roof of the house.’

69The following examples from the Frog Story in HSF further illustrate the usage of the participles in an expression that focuses on manner and setting description:

(56)

an

DEF

pik’o’

dog

chep‑at

head.down‑PPL

jayej

also

an

DEF

ti

R

ne’ech

FUT

k‑a

IRR‑A3SG

pa’‑ay

descend‑COM

‘The dog is also falling down head first.’

(57)

an

DEF

kwitool

boy

je’at

legs.up‑PPL

an

DEF

ti

R

ne’ech

FUT

k‑a

IRR‑A3SG

ch’a’‑ey

fall‑COM

‘The boy is going to fall on his back.’

70As seen in the above examples, HSF uses participles formed from intransitive verbs that fulfil the role of the positionals:kubat ‘standing’ from kubey ‘stand’, kochat ‘lying’ from kochiy ‘lie’.

6.4 Absence of directionals in HSF

71As already mentioned, Huastecan and Yucatecan languages do not have directionals. However, even the languages of the family with elaborated system of directionals (like Tseltal, for example) package complex motion events in sequences in the very same way. Typical Mayan directionals are illustrated here with inventories from Mam and Tseltal of two distinct branches of the family:

Mam (England 1983)

Tseltal (Pollian 2006)

aj

return from here

beel

towards there

ajtz

return from there

jajch’el

getting up

b’aj

complete

Jilel

without movement

el

out

julel

arriving here

etz

out toward

k’axel

passing

ex

out away

k’oel

reaching the goal

iky’

passing

koel

downwards

iky’tz

passing to this side

lok’el

towards outside, leaving/going out

iky’x

passing to other side

moel

upwards

jaw

up

ochel

towards inside, entering

jatz

up toward

sujt’el

returning

jax

up away

tel

towards here

kub’

down

ku’tz

down toward

ku’w

down away

kyaj

remaining

ok

in

okx

in away

oktz

in toward

pon

here to there

tzaj

toward

ul

there to here

xi

away

Table 3: Directionals in Mam and Tseltal

72According to England (1983: 303), directionals express ‘some ideas which in other languages might require complex sentence structures’. The usage of the directionals in encoding Path in Mam and Tseltal is illustrated in the following examples:

(58)

MAM

itzaajal

will.bring

a’

water

wu’na

by.me

maax

up.there

jaw‑nax

DIR‑ADV

(England 1983: 229)

‘I will bring the water up there.’

(59)

TZEL

wil‑Ø

fly‑B3

lok’el

DIR:exit

moel

DIR:go.up

beel

DIR:go

(Polian 2006: 158)

‘He flew out far up.’

73An example from Jakaltek-Popti (Grinevald 2011) further illustrates how directionals function in the majority of Mayan languages:

(60)

JAK

a.

xil‑ah‑toj

saw‑DIR2‑DIR3

naj(REF)

CL/man

tet

to

ix

CL/woman

‘He saw her [up.away].’

b.

xil‑ay‑tij

saw‑DIR2‑DIR3

ix

CL/woman

naj(REF)

CL/man

‘She saw him [down.toward].’

74In the above example the verb xil see is used with two directionals to precise the direction of the act (60a. upwards, 60b. downwards). According to Grinevald (2011:90), the usage of directionals in Jacaltec Popti’ is similar to the usage of verbal particles in English, ‘the difference being a more systematic use of the directional in individuated referential scenes.’ Comparing the HSF data in (1), (2), (3) to those from other Mayan languages with directionals, as in (58), (59), (60) it seems plausible to argue that the Yucatec and Huastec way of encoding space and motion events could represent an early stage of the expression of trajectory in the Mayan family, before the grammaticalization of motion verbs into directionals took place in majority of Mayan languages.

7. Expression of complex motion events in HSF

75When expressing motion, Mayan languages generally use the verbs of motion and their derived directionals; therefore, the majority of Mayan languages typologically are both verb-framed and satellite-framed languages (see Brown 2006 for Tseltal), where verbs provide motion plus Path information, and satellites (directionals) provide  ‘adverbial modifications which are satellite-like in contributing separate path information which can be added to the meaning of a predicate of any form’ (Brown 2006: 251). In contrast to majority of Mayan languages, and as mentioned above, Huastecan and Yucatecan languages do not possess directionals. These groups of languages are strictly verb-framed and this is reflected in the ways they codify motion.

76In this section we are going to concentrate on the way HSF segments complex motion events: each segment (or sequence) of a complex motion event is encoded in this language by a separate clause (i.e. verb phrase).

7.1 Motion-cum-manner

77One kind of segmentation of events in HSF is between motion and manner of motion. It is represented in the particular constructions motion-cum-manner and motion-cum-purpose. As mentioned in section 5.5 on HSF subordination, these complex constructions involve two different clauses, where the first one encodes motion by a verb of motion, and the second one (the dependent one) has an embedded manner verb introduced by the subordinator ti:

(61)

nanaa’

I

wat’‑ey

pass‑COM

t‑in

R‑A1SG

aath‑il

run‑AP(INC)

(Tr23‑4b)

Lit: ‘I passed running.’

Desired reading: ‘I ran past.’

(62)

kal‑ej

exit‑COM

ti

R

bel‑al

walk‑INC

an

DEF

ti

PREP

jool

cave

(Tr23‑40)

Lit: ‘She came out walking from the cave.’

Desired reading: ‘She walked out of the cave.’

(63)

naa’

DEM

i

NM

chikam

boy

och‑ich

enter‑COM

ti

R

laatx‑um

swim‑INC

Lit:’ That boy entered (the water) swimming.’

Desired reading: ‘That boy swam into (the water).’

78It is not possible to understand these two clauses as two temporally separated sequences. They are one event, ‘splitting’ the component of motion in one segment and the component of manner into another. This is the only way of expressing these events in HSF; expressions like English ‘I ranpast’ or ‘That boy swaminto the water’ are simply not possible in this language. Because HSF does neither have dynamic prepositions of motion (like past and into, as well as towards,away) nor gerunds, the only way to express the above meaning (swim into, run past) in this language is to associate the verbs swim  and  pass with a verb of motion in a motion-cum-manner construction.

7.2 Segmentation of events expressing Source and Goal

79The basic way to express Source (the initial point) or Goal (the final point) in South Eastern Huastec is with an adpositional phrase introduced by the same basic location preposition ti. Thisindicates that it is the verb, and not the preposition, that bears the meaning of Path. The following examples illustrate the expression of these two portions of Path.

(64)

a. goal

och‑ich

enter‑COM

ti

PREP

juun

one

i

NM

jool

cave

(Tr70)

‘(S/he) went into a cave.’

b. source

kal‑ej

exit‑COM

ti

PREP

juun

one

i

NM

jool

cave

(Tr73)

‘(S/he) came out of a cave.’

(65)

a. goal

och‑ich

enter‑COM

an

DEF

ti

PREP

alte’

bush

(Tr56)

‘(S/he) went into the bush.’

b. source

kal‑ej

exit‑COM

an

DEF

ti

PREP

alte’

bush

(Tr73)

‘(S/he) came out of the bush.’

80HSF has several verbs that are goal-oriented, like utey approach, ulel arrive, txi’ich come, pujk’in fell down, or like in the above examples ochich enter, and several verbs that are source-oriented, e.g. kalej exit, go out. The verb k’alej go is sometimes used as a source oriented (‘he left’, from here/there), and sometimes as a goal-oriented verb. Note however that in the expression of Goal k’alej is used without any preposition (like in English ‘go home’), as in (66):

(66)

n‑u

DEM‑E1SG

maa

mother

k’al‑ej

go‑COM

Tantoyuca

Tantoyuca

(AmE6)

‘My mother went to Tantoyuca.’

81As is the case in Yucatec (Bohnemeyer & Stoltz 2006: 299), an expression like He went from X to Y would be encoded in HSF as two clauses (He was in X. He arrived in Y), and sometimes three: He was in X. He left. He arrived in Y.The expression of complex motion involving a passing component, as in He went from X past Y  to  Z, is typically encoded in three clauses: He was in X, he passed Y, he arrived in Z. Therefore, Source and Goal cannot be expressed in a same clause in HSF. This phenomenon is illustrated in the following example:

(67)

Paablo

Pablo

kal‑ej

exit‑COM

ti

PREP

Tantoyuca,

Tantoyuca/

ul‑ich

arrive‑COM

Meejico

Mexico City

(CiHSF4)

Lit: ‘Pablo left Tantoyuca, he arrived to Mexico City.’

Desired reading: ‘Pablo went from Tantoyuca to Mexico City.’

82Source and Goal with a passing element is already shown in example (1) in the introduction of this article and repeated here for convenience:

(1)

Paablo

Pablo

kal‑ej

exit‑COM

ti

PREP

Tantoyuca,

Tantoyuca/

wat’‑ey

pass‑COM

ti

PREP

Bitxow,

Chontla/

ul‑ich

arrive‑COM

Meejico.

Mexico City

(CiHSF4)

Lit: ‘Pablo left Tantoyuca, he passed Chontla, he arrived in Mexico City.’

Desired reading: ‘Pablo went from Tantoyuca to Mexico City past Chontla.’

83After some consideration, consultants provided us with another possibility, shown in (68), using Spanish preposition, which is an obvious calque:

(68)

in

A1SG

k’al‑ej

go‑COM

k’aal

with

u

E1SG

akan

foot

San Francisco

San Francisco

axta

till(Sp)

ti

PREP

Bitxow

Chontla

(AmE6‑70)

Spanish: Fui a pie desde San Francisco hasta Chontla.

‘I went on foot from San Francisco to Chontla.’

(69)

chaap

two

i

NM

oora

hour(Sp)

u

E1SG

t’aj‑a-al

do‑TS‑INC

San Francisco

San Francisco

a

to(Sp)

Tantoyuca

Tantoyuca

(AmE6‑70)

Spanish: Dos horas toma desde San Francisco a Tantoyuca.

‘It takes two hours from San Francisco to Tantoyuca.’

84The above examples display the usage of a zero preposition in HSF to encode the Source, while the Spanish prepositions hasta (HSF form axta) till, until, and a to are used to encode the Goal.

85Complex Paths in HSF cannot be encoded in one clause, while other complex semantic combinations like Path with position (expressed by a verb of motion plus a positional) are often found in one clause.

7.2.1 Motion towards Goal

86The expression of motion with a goal is another area where HSF behaves interestingly, partly due to the lack of the preposition towards. The example (70) shows that a simple juxtaposition of two clauses is used to convey the meaning of motion oriented toward the Goal:

(70)

kub‑ey

stand‑COM

//

ut’‑ey

approach‑COM

t‑in

PREP‑E3SG

akan

feet

an

DEF

te’

tree

(Tr61‑75)

Lit: ‘She stood up, she approached the tree trunk.’

Spanish: ‘Se paró, se acercó  a la mata del árbol’ and not ‘Fue hacia el árbol’

Desired reading: ‘She went towards the tree.’

87The following example demonstrates a complex sentence with three segments: a motion-cum-manner clause and a juxtaposed clause encoding the motion with goal event:

(71)

taal

come

ti

R

bel‑al

walk‑INC

//

och‑ich

enter‑COM

an

DEF

ti

PREP

alte’

bush

(Tr52‑3)

Lit: (He) is coming walking, (he) entered the bush.

Spanish: ‘Viene caminando y entra en el monte.’ and not ‘Entró en el monte’

Desired reading: ‘He walked into the bush.’

88Besides juxtaposition, a relative clause can be used, as demonstrated in (72) and (73):

(72)

ip

A3PL

k’al‑ej

go‑COM

//

xoo’

REL

ti

PREP

wa’‑ach

exist‑INC

i

NM

ja’

water

(Tr76‑1)

Lit: ‘They went there where there was water.’

Spanish: ‘Se fueron adonde había agua’ and not ‘Se fueron hacia la agua‘.

Desired reading: ‘They went towards the water.’

89It should be mentioned that the translation into Spanish (kept in the above three examples for illustration as it clearly reflexes the HSF structure) given by the consultants clearly segmented the events, although they were all native speakers of Spanish too.

90In (73) there are four clauses: the first describes the posture the boy had before starting walking; the second describes the fact that he started walking, the third that he entered the place, while the fourth is a relative clause describing the final destination. Note that the same event can be expressed by just one clause in English (as suggested by the desired reading):

(73)

axee’

DEM

i

NM

chikam

little

kwitool

boy

pun‑at

climb‑PPL

wik

PAST

an

DEF

ti

PREP

t’ujup

stone

//

tayiil

later

bel‑ej

walk‑COM

//

ani

and

och‑ich

enter‑COM

//

xoo’n

REL

ti

R

k’waj‑txik

be‑PL

an

DEF

paktha’

big

t’ujup

stone

(Trj31‑4)

Lit: ‘This little boy was standing on the stone(s), then he walked and he entered where there were big stones.’

Desired reading: ‘The little boy walked into the stony place.’

7.2.2 Motion in relation to Medium Point (Milestone)

91We find similar segmentation of events for expressing motion past a mid-point (Milestone), or along a trajectory. In (74), the girl passed running (motion with manner) and crossed a tree (Medium):

(74)

wat’‑ej

pass‑COM

ti

R

aath‑il

run‑AP(INC)

ani

and

in

E3

jek‑a’

cross‑TS(COM)

juun

one

i

NM

te’

tree

(Tr67‑3)

Lit: ‘(She) passed running and crossed a tree.’

Desired reading: ‘She ran past a tree.’

92The same phenomenon is observed in (75) where there are four different segments:  motion-cum-manner (which is always perceived as one simultaneous event), Medium passing motion, and boundary crossing with a goal:

(75)

wat’‑ey

pass‑COM

/

ti

R

bel‑al

walk‑INC

/

ani

and

in

E3

crusaar‑iy

cross‑TS(COM)

oox

three

i

NM

persoona

people

/

ani

and

och‑ich

enter‑COM

an

DEF

ti

PREP

jool

cave

(Tr20)

Spanish: ‘Vino caminando, cruzó a tres personas, y entró en la cueva.’

Lit: ‘He passed he walked, he crossed three people and went into the cave.’

Desired reading: ‘He walked into the cave past three people.’

  • 13 Just to remind the reader that in this example, the first, the second and the fourth clause do not (...)

93It is shown here again that on the level of the lexical presentation of events the lack of the prepositions like into, past, from, towards and the lack of gerunds lead this language to segment events into several components, as is the case in example (75).13

7.2.3 Motion from Source

94As with the expression of motion with respect to a Goal and a Medium, the expression of motion with respect to a Source too requires the use of several clauses. As the example (76) shows, while the first clause typically describes the fact of motion itself, the second clause depicts the posture and the initial location of the Figure before the motion:

(76)

bel‑ej

walk‑COM

juun

one

i

NM

txithan

girl

//

kub‑at

stand‑PPL

wik

PAST

t‑in

PREP‑E3SG

akan

feet

i

NM

te’

tree

(Tr32‑4)

Lit: ‘A girl walked, she was leaning at the tree.’

Desired reading: ‘A girl walked away from the tree.’

  • 14 Verbs of motion with directionals are underlined and positionals are italicised.

95Another Mayan language, Tseltal, also segments complex motion events. The following example from the Tseltal Frog Story (Brown 2004:43) reveals a language possessing several motion verbs and an elaborate system of trajectory-indicating directionals to still to still segment Path into each clausal component:14

96Comparing the examples in HSF with those from Yucatec and Tseltal reveals that in describing motion, attention is often drawn to ‘dispositional configurations if the position is non-canonical (e.g. ‘fallen down’, or ‘tipped over’,  Brown (2006)). According to Brown, motion semantics enter into static descriptions ‘when a static configuration is described from the perspective of how things got into that particular position’ (Brown 2006: 263). In South Eastern Huastec, just like in Tseltal, many responses to where-questions produced while recounting the Frog Story or while describing ‘Trajectoire’ stimuli included a description of the state as a result of a directional action, as (77) exemplifies:

(77)

ne’‑ech

go‑INC

ti

R

bel‑al

walk‑INC

an

DEF

ti

PREP

toomlam

field

//

kal‑ej

exit‑COM

an

DEF

ti

PREP

alte’

bush

(Tr55‑3‑19)

Lit: ‘(He) goes walking across the field, (he) came out of the bush.’

Desired reading: ‘He walked out of the bush.’

97In (77) ‘going walking across the field’ is a result of ‘coming out of the bush’. The main motivation for using directional verbs in these contexts, according to Brown (2006), is to import a deictic reference point.

7.3 Order of the event segments

98How does pragmatic inferencing of temporal relations between sub-events work? The order of the segments in the package of a complex motion event can be different.  In many instances, as is the case in the example (78), the first clause describes the initial posture of the Figure at the Source location and the last clause expresses motion. In other instances, as is the case in the example (79), the motion of the Figure is expressed in the first clause and the posture at the initial location in the second clause.

(78)

a.

juun

one

i

NM

inik

man

kub‑at

stand‑PPL

xoo’n

REL

ti

R

koch‑at

lie‑PPL

juun

one

i

NM

txithan //

girl

tayiil later

an

DEM

inik

man

bel‑ej

walk‑COM

(AmTr35‑57)

Lit: ‘A man is standing where a girl is lying, later the man walked.’

Desired reading: ‘A man walked away from the girl lying there.’

b.

an

one

inik

man

bel‑ej //

walk‑COM

kub‑at

stand‑PPL

wik

PAST

xoo’n

REL

ti

R

koch‑at

lie‑PPL

juun

one

i

NM

txithan

girl

(CirTr35‑57)

Lit: ‘A man walked, he was standing where a girl was lying.’

Desired reading: ‘A man walked away from the lying girl.’

(79)

a.

kal‑ej

exit‑COM

an

DEF

ti

PREP

eem //

corn.field

ani

and

wat’‑ey

pass‑COM

an

DEF

ti

PREP

beel

way

Lit: ‘(She) exited maize, and she crossed the way.’

Desired reading: ‘She came out of the maize field crossing the way.’

b.

wat’‑ey

pass‑COM

an

DEF

ti

PREP

beel //

way

kal‑ej

exit‑COM

an

DEF

ti

PREP

eem

corn.field

(GoyTr38‑51)

Lit: (She) crossed the way, she exited maize.’

Desired reading: ‘She crossed the street coming out of the maize field.’

99The above examples (78) and (79) illustrate Bohnemeyer et al.’s macro-event linking principle: ‘temporal interpretation of the sub-event is independent of the order of path phrases’ (2007: 525). The variation shown in these two examples (description of the same scene given by two different consultants) demonstrates that the event sequences are in fact independent.

100With regard to the order of the elements (segments), it should be noted that (as in Yucatec, Bohnemeyer 1997: 72), event order and Path are not lexicalized in HSF. As mentioned in 5.6, syntactic relators like subordinative connectives (after, before, while), anaphoric connectives (afterwards, beforehand, meanwhile) and temporal prepositions (after, before, during) are all absent in HSF. This language, like the Mayan languages in general, do not possess such markers. Although Spanish loan word are sometimes used (despwees after, aantes before, mieentras while), temporal coherence is normally not expressed in HSF and relative ordering of events is interpreted contextually.

101Bohnemeyer (1997) discusses the absence of Path lexicalization in Yucatec Maya as aligning with the absence of event order lexicalization: “Yucatec displays a striking absence of event order relators in both syntax (connectives, adpositions) and inflection (tense). Time reference and temporal coherence in discourse instead rely heavily on inferences from aspectual and modal information, discourse structure and world knowledge. In line with this, relations of motion of a spatial figure with respect to a spatial ground (‘path’) are not lexicalized in Yucatec” (Bohnemeyer 1997: 73).

7.4 Testing the MEP on HSF

102Linguistic tests that Bohnemeyer et al. used to demonstrate whether an expression has the MEP or not, mentioned in section 3.3 (the possibility of insertion of temporal adverbs, conjunctions or intonational breaks, and negation that has scope over just one single verb phrase) have been applied to HSF.

7.4.1 Insertion of temporal adverbs

103The possibility of insertion of a temporal adverb like then or later or a conjunction and between the segments of a complex motion expression demonstrates that the complex event expression does not have a MEP, as (73-75), (78a), and (79a) illustrate. These examples prove that these sub-events are separated segments that have their own MEP. In the examples found in this article several illustrations of all the different kinds of interruptions (adverbial, conjunctions, intonation pauses) are attested. The HSF data demonstrate that an adverbial can be placed in between these segments, for example tayiil then, later, a conjunction ani and, or there is an intonation break.This is a proof that in an expression of complex motion in HSF these segments are independent. In other words, this proofs that there is no MEP in the HSF expressions of complex motion and that all of these segments are separate expressions (different clauses with independent predicates that represent different segments), as illustrated in the mentioned examples above. The presence of temporal adverbs in these examples shows also that the temporal interpretations of the sub-events are independent of the order of these segments.

7.4.2 The scope of negation

104In English, for example, the whole event (the MEP expression) is negated: the following clause Philippe went from Paris to Nice past Lyon, will have the whole event negated in: Philippe didn’t go from Paris to Nice past Lyon. In HSF the equivalent sentence Philippe left Paris, he passed Lyon, and he arrived to Nice can have each segment (clause) negated: Philippe didn’t leave Paris, he didn’t pass Lyon, and he didn’t arrive to Nice. All these segments of one complex motion event are independent clauses. In our initial example, (1), all the three units have clausal status on the negation criteria, as each one can be negated independently (1a); therefore, each of these small units has the MEP:

(1a)

Paablo

Pablo

baa’ no

kal‑ej exit‑COM

ti PREP

Tantoyuca, Tantoyuca/

baa’ no

wat’‑ey

pass‑COM

ti

PREP

Bitxow,

Chontla/

baa’

no

ul‑ich

arrive‑COM

Meejico.

Mexico City

(CiHSF4a)

‘Pablo didn’t leave Tantoyuca, he didn’t pass Chontla, he didn’t arrive to Mexico City.’

8. Discussion and conclusions

105Some languages package three sub-events of a motion event (departure, passing, arrival) into a single clause representing a single MEP expression (Type I). Others package two sub-events (departure, arrival) into a single expression and a third sub-event (passing) separately (Type II). Finally, there are languages that package all the three sub-events in three separate clauses which all have an MEP (Type III). The study of the temporal and spatial organization of events in HSF reveals that HSF belongs clearly to Type III in the Bohnemeyer et al. (2007) classification. In this language any location change with respect to a Ground must be encoded in a separate macro-event expression by a separate verb phrase. That is, rather than saying ‘Floyd arrived from Nijmegen to Elst crossing the river’ HSF packages the same information in three verbal phrases as in ‘Floyd left Nijmegen, crossed the river, and arrived in Elst’. As with Yucatec (Bohnemeyer at al 2007) and Tseltal (Brown 2006), HSF does not permit the integration of the three sub-events into a single macro-event expression. This way of encoding complex motion events seems to be a result of the lack of prepositions (towards, from, away, past) as well as gerunds (she was reading drinking tea) and directionals. Besides these lexical and grammatical constraints, there is also this typical Mayan way of emphasising description of posture and setting. Moreover, in Yucatecan and Huastecan language groups there are no multi-verb constructions (serial verbs) that would combine multiple location-change-denoting verbal phrases into a single MEP serial construction.

106Mayan languages seem to be very similar in the way they encode the Path. According to Bohnemeyer & Stoltz (2006: 295), Path, i.e. the translational motion along an extended trajectory from Source to Goal, is expressed neither by a morpheme nor by a construction in Yucatec Maya, but‘left to pragmatic inference’. They claim it is not clear whether Path, in Talmy’s sense, is encoded at all in Yucatec. The same applies to HSF. In the Frog Story examples, as well as in many others in this article, Path is not explicitly encoded. The ‘desired reading’ in our examples represents in fact the way the English language expresses the Path, capturing the meaning of the corresponding segments of HSF, where Path is left to pragmatic inference.

107It can also be mentioned that what has been said about event segmentation leads to Bohnemeyer’s et al. claim of a universal principle of event segmentation (“unique vector constraint”, Bohnemeyer et al. 2007), that is, languages don’t allow two distinct vectors of a motion trajectory to be expressed in the same clause: ‘left and to the north” cannot be expressed as one word. Unique vector constraint concerns specifically the encoding of direction information in macro-event expressions.

108This ‘over-emphasis’ on posture and settings would look unusual or simply unnecessary to a speaker of, for example, Indo-European languages. Nevertheless, Slobin (1996: 204), comparing spatial expressions from the English and Spanish Frog Stories, notices that languages seem to differ in their ‘relative allocation of attention to movement and setting’, and that the speakers of Spanish often elaborate descriptions of settings, leaving Paths to be inferred. For this author it is a rhetorical style that favors separate clauses for each segment of a complex motion event. Besides, he notes that the speakers of Spanish use more descriptive locative constructions, like relative clause constructions, to encode the setting (background description) of events. This applies in totality to the way Path is encoded in Mayan languages. While being merely a rhetoric style in Spanish, in Mayan languages it is the only way of encoding complex motion events. This area requires much more analytic attention if we want to fully understand this interesting phenomenon. There are 30 Mayan languages, the majority of them is totally undescribed in this matter. Some Mayan languages have serial verb constructions (Zavala 2006: 274) and it would be very interesting to see how they encode complex motion events.

109This topic also addresses some interesting structuralists’ claims including the following from Boas (1911: 22) that ‘each language, from the point of view of another language, may be arbitrary in its classification’, or that by Sapir (1931: 578) that languages in fact ‘differ very widely in their systematization of fundamental concepts’ and ‘tend to be only loosely equivalent to each other as symbolic devices and are, as a matter of fact, on the whole, incommensurable’. It will be very interesting to see what further research on expression of complex motion events in other Mayan languages reveals.

A brief summary

110The data collected demonstrated that HSF does not have directionals (present in most Mayan languages), that there are no real adpositions (from, away, towards, by, etc.). Instead HSF has a typical Mayan general locational preposition, and some relational nouns. HSF also lacks gerunds (instead complex sentences are employed). The absence of directionals, adpositions and gerunds leads this language to segment complex motion events into multiple clauses. On the other hand, the absence of Path lexicalization seems to engender some need to import the deictic reference point to the description of the scene settings.

111South Eastern Huastec event segementation is interesting typologically firstly because this language is one of thefew that are knownto segment events into sub-events, and secondly, for the comparative Mayan Linguistics. This phenomenon may represent the Proto-Maya way of expressing the Path: other Mayan languages, like above mentioned Mam, Tseltal and Jacaltec-Poptí, may have lexicalised one of these segments into a directional that expresses Path. Yucatecan and Huastecan languages had the lexical, resources but in these two language groups, the grammaticalization process did not take place.

112This is the first insight into the expression of space and motion of the South Eastern Huastec and in Huastecan languages more generally. While this study is far from being exhaustive and the analysis is still to be refined, it contributes new data to the study of space in Mayan languages.

Haut de page

Annexe

Abbreviations:

A

Absolutive

INT

intensifier

AP

antipassive voice

MID

middle voice

CAUS

Causative

NM

nominal modifier

COM

completive verbal aspect

PASS

passive voice

DAT

dative (applicative)

PAST

past

DEF

determinator (‘definite article’)

PERF

perfect verbal aspect

DEM

demonstrative

PL

plural

DM

derivational morpheme

POSIT

positional

E

Ergative

POSS

possessive

EP

linking consonant

PREP

preposition

FUT

Future

PROGR

progressive

HAB

Habitual

R

subordinador realis

HON

honorific (article)

REL

relative particle

INC

incompletive verbal aspect

RED

reduplication

INSTR

instrumental (applicative)

QUOT

quotative

IRR

subordinator irrealis

SG

singular

irr

Irregular

SP

stative participle

TS

transitive suffix

1,2,3

person

Haut de page

Notes

1 Alternative names: Huastec of San Francisco, or Teenek de la Sierra de Otontepec. HSF has about 12,000 speakers in the region of la Sierra de Otontepec, Veracruz, Mexico. It is one of the least known languages of the Mayan family. The language is not passed to new generations and, hence, is endangered.

2 I want to express my profound gratitude to two anonymous CogniTextes reviewers, and especially to Anetta Kopecka, whose comments helped me to improve this paper.

3 The aim of this project was to examine the expression of Path of motion in typologically varied languages and to investigate the hypothesis in the expression of Source and Goal of motion from a cross-linguistic perspective. For more details about the project and the methods used for collecting the data see Fortis et al. (2011), and Kopecka & Ishibashi (2011).

4 A preliminary version of this paper was presented during a working session of the Project in Lyon in May 2011, and at the AFLiCo IV conference in May 2011.

5 Alternative orthography: Tzeltal.

6 Additional support to carry out the fieldwork was provided by the Mexican Government scholarship and the DDL (CNRS & Lyon 2).

7 Serial verbs constructions are considered one clause.

8 For the past sixteen years the Mayan space expression (in Yucatec, Mopan, Tsotsil and Tseltal) has been studied at the Max Planck Insitute for Psycholinguistics. Brown (2006) summarizes the results of this research for Tseltal.

9 In data elicited with the Bowerman and Pedersen questionnaire (1992) the relational nouns appeared only 25 times in 221 clauses describing position.

10 Brown (2006:246) calls them dispositionals.

11 Positionals have even been recognized in Mayan hieroglyphics.

12 The combination of positionals with motion verbs, directional and auxiliaries in spatial language is described in detail for Tseltal (Brown 2004, 2006), and Mocho (Martin 1994).

13 Just to remind the reader that in this example, the first, the second and the fourth clause do not contain the 3rd person Absolutive marker as it is zero, see more in 4.1

14 Verbs of motion with directionals are underlined and positionals are italicised.

15 No glosses or original text in Tseltal are given in the source article.

Haut de page

Pour citer cet article

Référence électronique

Ana Kondic, « Towards an Analysis of Complex Motion Event Packaging in South Eastern Huastec (Maya, Mexico) », CogniTextes [En ligne], Volume 9 | 2013, mis en ligne le 10 juin 2015, consulté le 17 octobre 2017. URL : http://cognitextes.revues.org/736 ; DOI : 10.4000/cognitextes.736

Haut de page

Auteur

Ana Kondic

Centre for Mexican and Mesoamerican Studies (CEMCA), Sierra Leona 330, Lomas de Chapultepec, Miguel Hidalgo, MX-11000 Ciudad de México DF, Mexico
ana.kondic@cemca.org.mx

Haut de page
  • Logo DOAJ – Directory of Open Access Journals
  • Logo ERIH PLUS | NSD
  • Logo AFLiCo – Association française de linguistique cognitive
  • Les cahiers de Revues.org