TRANSITIVITY INDICATORS - Yl./ AND -WI IN HUALAPAI

The distribution of Hualapai auxiliary verbs "-yu" and "-wi" can not be explained only by the presence or absence of an object, or by the active or stative feature of the matrix verb. It can be explained in terms of transitivity, in that "-wi" corresponds to high transitivity and "-yu" to low transitivity of the clause. Several grammatical processes (causative, applicative, and passive) interact with the transitivity of the clause, which correlates with auxiliary alternation. The speaker's attitude with regard to the nature of the event also affects the evaluation of transitivity in Hualapai. This suggests that the degree of transitivity is determined subjectively by the speaker. Among the transitivity parameters, the existence of second and affected participants and the activity indicated by the clause seem to be crucial components for deciding the degree of transitivity. More detailed observation about the precise nature of each grammatical process involved would be revealing. (Author/MSE) *********************************************************************** Reproductions supplied by EDRS are the best that can be made from the original document. *********************************************************************** Ir "PERMISSION TO REPRODUCE THIS MATERIAL HAS BEEN GRANTED BY TO THE EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC)." U.S. O(PARTMINT IDUCAllON Office of Educsoonal Research end Improvement EDUCATIONAL RESOURCES INFORMATION CENTER IERIC) Q'This document has been reproduced as received from the person or organization originating it 0 MinOr changes have been made to improve reproduction Quality Points of UMW Oropinions stated in this clomp mint do not necesseniy represent OERI position or policy TRANSITIVITY INDICATORS Yl./ AND -WI IN HUALAPAI

(WBY] Burt-Sub ijumpluddenly-3-Aux 'Burt is jumping.' (13) Jibay-ch (WBY] bird-Sub f1y-suddeuly-3-Ans 'The bird flew away.' In the examples above, verbs indicate locomcdon of the subject; that is, they do not denote the state of the subject but the action controlled by the subject.These phenomena demonstrate that the choice between -pa and cannot be explained only by the pretence/absence of an object of the verb, or by the activeistative feature of the verb. In this paper, I will su ,ggest that the distribution of -yu and in Hualapai may be explained in tams of 'Transitivity' in the sense introduced by Hopper and Thompson (1980), to the exteat that when the overall clause shows ielatively high Transitivity, it takes -wi marking, and when it shows low Transitivity, it takes -yu marking.The degree of Transitivity can be determined based upon several parameters Mating to the effectiveness or intensity with which the action is transferred from one participant to another (Hopper and Thomson 1980:252).These parameters include the number of participants, the activity indicated by the verb, the telicity and punctuality of the verb, the volitionality of the agent, the agency of participants, and the referentiality and degree of affectedness of the object.
In examples (3) through (9), even though there is a second participant in each clause, it is an object of a cognitive verb and not a 'true' patient; that is, the object is not really affected by the action of the subject.These verbs denote an interval process, not an out-going action, and the volitionality of the subject might be low.sTherefore, these clauses can be said to have a low degree of Transitivity, thus coded as 'intransitive' with -yu.As for (10) through (13), though the clauses have some high-Transitivity features (e.g., action, volitional), they lack the patient participant which is a aucial component of Transitivity.Alternatively, it can be said that the only participant in each of the sentences has the semantic role of patient or theme, in the sense that it itself undergoes the movement, as well as that of agent, in the sense that it is the instigator of the movement.6In any case, there is no transfer of action from one participant to another (different), participant and this reduces the degree of Transitivity of these clauses so that they are low-Transitivity marked by -yu in Hualapai.
If the auxiliaries -wi and -yu really correlate with high and low Transitivity of the clause, a change in the degree of Transitivity should be aozompanied by an alternation in the auxiliary.In the following sections, I examine several grammatical processes in Hualapai which manipulate Transitivity features, and show how they motivate the selection of auxiliary -yu or -wi.

CausatiysSonstruction
Hualapai has several causative vabal prefixes.According to the classification of Watahomigie, Benda and Yamamoto (1982), they am et-(general causative), j -(expressing some 9uality change), v -(expressing some movement change), s -(indicating some action by hand), and g -(expressing duality with partner(s)).The causative prefix changes the meaning of the stern verb into 'make someone/something do' or 'cause someone/something to do'.In this process, an agentive argument is added which takes over the subject marking -ch, while the original subject is demoted to object position and takes the object marking (null).The schematic representation of this process is: - As a result of this process, the number of participants is increased (mostly from one to two) and the clause comes to have an explicit agent (causer) and patient (causee).The original clause typically expresses the state of the subject, while the suffixed clause describes the action of the subject causing such a state.Thus, the causative process increases the degree of Transitivity of the clause.In fact, through this process, -yu verbs (the majority of the verbs which undergo this process) The girl let the little boy fall.'

BEST COPY AVAILABLE
The auxiliary alienation here reflects the change in the degree of Ttunsitivity.By contrast, -wt verbs, which already show high Iivity, .More examples of non-causative/causative pairs:

Agyligatimauffia
The valency of the predicate can also be increased by attaching the Applicative verbal suffixes -o (after a consonant) / -yo (after y) / -wo (elsewhere).In this case, however, the added argument is non-agentive and is put into the object position while the subject remains in its original position.8(20) Malinds-Sub 1/3-sing-Appi-3-Aux 'Malinda is singing for me.' 'Yesterday John did not feed the bone (make the bone eat): Tbe interesting point hire is that if there akeady exists a [-human] object in the clause, the added (+human) object takes over control of the person prefix9 on the verb, i.e., the added [+human] argument, instead of the [-human] object, is indicated as the object by the person prefix)0 Further examination would be necessary to decide whether 1) it is always the case that the added object (indirect' object) is indicated in the person prefix despite its semantic featunts, or 2) the (+human] object has some kind of priority due to, e.g., i) an 'animacy hierarchy' between objects, or spealcer's 'empathy' (cf.Mithun 1990:13).The added argument can be eitha overt or coven (indicated only by a person prefix on the verb).'I am buying yQu brae.read for gap' (24) a) Owe le-gamy-ay-1. [WHY)   something 311-0.11-Irresl-Aux'I am going to tell a sicsy: b) Owe be nyl-ganav-o-y-i. (WEY]   some(hing P1 -Appl-Irreal-Aux am going to tell a story for you all: Through this process, the number of participants is increased.The clause which did not have an object before WITH% to have an affected object.As a result, this process may cause the degree of Ttansitivity to be increased.Accordingly, the clause which formerly took -yu alternates its auxiliary with -wi as follows: (25) a) 'Had nye nyi-hele-ch we-sevley-k-yu. (WM/ dog I Re1-pet-Sub bect-bosoran-3-Aux rlify dog is mean.' b) 'Han nya ny1-haria-ch wa-nyl-sayley-yo-k-w1.

EaLlinalittlitinadhablacsigazaLSMA
The verbal suffix -v has two main functions.One of them is to detopicalize the agent argument (by deleting it from subject position) and promote the patient argument (object) to subject position with subject marking -ch.Functionally, , this is the (agent-less) 'passive' (cf.Yamada 1979).'When the suffixed clause describes the state resulting from the action or event, it may also be called the Istative.It is difficult to distiquish between the 'passive' and 'stative' based or only the English translation.( 28) is the schematic representzlion of this process, which shows the mirror image of the causative process.11When the subject refers to a single entity, the reading must be reflexive sin= the mimesl interpretation needs at least two entitio.However, when the subject refers to more than one entity, if semantically plausible, it is presumed that the interpretation is ambiguous Ws/ m. the reflexive and reciprocal.
(33) Bosa-ch  The cats are scratching themselves / ? each other.'Also, the verb + -v construction can be interpreted either as passive or as teflexive even when the particle yevm 'self is added.
[Yanomoto 1989:12] John-Sub self kick-Pau/Rell-3-Aux 'John got lacked (and he was all by himself but it was his own fault): or 'Jolm kicled himself.' However, Yamamoto (1989) has pointed out that when the subject is [-human], this ambiguity does not occur nd the only possible interpretation is the reflexive.
[Yamamoto 1989:12] bone-Dem-Sub self kick-Re11-3-Anz horse kicked itself.' but *Ibe horse got kicked.' As we shall see in examples (36) through (38), the v-construction with [-human] subjects is not necessarily restricted to the reflexive interpretation like (35).We must ask whether the (b) samaces in (36) through (38) indicate not just the state itself but also the process which has brought about the state, e.g., not just 'John's car is in the state of being broken' but also 'John's car g21 broken' for (36b).If not, it would be hypothesized that ee more active (less stative) interpretation (the 'got' interpretation) is associated only with the l+human] subject, while the [-human] subject gets either the reflexive/mciptucal interpretation (Nmimate]) or the stative interpretation ([-animate]).The exact natme of correlation between the semantic features of the subject, the semantic nanny of the verb, and the possible interpretation of the suffix -v should be further examined.
The above-mentioned fimctions of the suffix -v influence the degtee of Transitivity of the clause; Transitivity becomes lower than that of the corresponding clause without -v.Through the passtve/stative process, the agent becomes covert and the only expressed participant is the patient.In addition to decreasing the number of participants, the focus of the clause changes from the action/event itself to the state resulting from the action or event.In the reflaiverecipmcal construction, the action transfer takes place internally; that is, the agent and the patient refer to one imd the same participant ore group of cipants, and the action transfer is confined to these partscipsnts.As noted Hopper and (1980), the non-distinctneas ('non-indi nature of the patient from tl agent is also one component of low nansitivity.As expected, this diange in the degree of Transitivity coincides with the auxiliary aberration, that is, the auxiliary -wi chases into -yu.It should be noted that the demonstrative suffix, which semantically corresponds to and syntactically co-occurs with the possessive ponoun (e.g., -v Ibis very close one' : 'my', -ny 'this/that very one' : 'you', and -it 'that one over there' : 'hisiba), is often attached to these tams in order to be felt natural (Watahomigie, Bender and Yamamoto 1982:185).If this demonstrative suffix can be presumed to be the reflex of tbe obligatory personal affix fortune lams, they may be embodied as inalienables' and have somewhat distinctive status in Hualapai.The peculiar nature of these terms might explain the exception of the auxiliary alternation hae. 5.

SatakettAllillideletaradiguadibekiegt
So far we have examined the canektion between the Tonsitivity alternation and various grammatical processes.In addition, some clauses can show the auxiliary alternation without undergoing any ovat grammatical process.In this case, when a dause takes a -yu auxiliary, it is said to indicate what is the case, while when a clause takes -wf, it is said to imply an accompanying action along with another covert participant's involvemait in the action.
(50) a) Misi:-ch walba:-k-yu-ny.The girl believed it to be true (and that's why she acted that way).' (51) a) Nya-ch sme.'-yu.[WHY] 1-Sub miss-Aux 'I miss him.'b) Nya-ch sme:'-w L. [WHY] 'I miss him (and tiiat's why I am doing it -eg.getting mad at everyone): (52) a) Cindy-ch wi'ham jikba.-k-yu.All data used in this paper are individual sentences.Since the degree of Transitivity of a clause seems to be affected also by the context in which the clause is contained, it is necessary in further research io look at texts, and examine how the auxiliary selection is made in wider contexts.NOTES I I would like to thank members of the Seminar on the Structure of Hualapai (University of Kansas, Fall 1990).Discussions in the class evoked some of the ideas in this paper.I would especially like to thank Dr. Akira Yamamoto for his comments, suggestions and constant encouragement.I would also like to thank Dr. Sandra A. Thompson and Dr. Marianne Mithun for their valuable comments.
2 Hualapai is an SOV language.The subject and object may be overtly expressed or indicated only by the person prefix on the verb.
3 Hualapai examples are written using their orthographic system: ch40] but it" as a subject marker; ny=[ri]; e*[t]; ge).Since I have not conducted fieldwork on this language myself, most of examles cited in this paper are taken from Watahomigie, Bender and Yamamoto (1982) (indicated as EWBY)) or Hualapai Dictionary ([D]).Other sources will be specified.4 Auxiliary verbs in Hualapai still retain some properties of lexical verbs.The matrix verb takes the same subject marker which indiaates that the subject- reference is identical with that of the following (auxiliary) verb; and the auxiliary verb takes a person prefix.The sequence of the same subject marker and person prefix has been omitted totally (in the case of 1st person), changed phonologically (2nd person) or retained (3rd person), and is now reinterpreted as a person marking for auxiliary verbs: 1) 1st person : V-Aux, 2) 2nd persore V-ng(4.-k+m-)-Aux, and 3) 3rd person: V-k(4)-Atix.
5 As for example (6), it is not clear whether 'take care of' expresses an emotional state or actual nursing.
6 But note that the 0-criterion in Generative Grammar does not allow two thematic roles to be assigned to the same single argument 7 Not all verbs may undergo the causative process.The examination of the range of verbs which have causative counterparts and their semantic or syntactic properties are subjects for further research.
8 Here I assume that the Applicative can be distinguished from the causative based on the thematic role (non-agent vs. agent) and grammatical function of the added argument (object vs. subject).Further research is needed to determine the semantic difference between them since sometimes the causative-like translation ('make someone/something do) is obtained as may be hinted at by_the English translation of the Applicatrve construction (see examples (22b), (25b), (26b) and (27b)) .According to the data available, it seems that verbs which take the Applicative suffix and those which take the causative prefix are in complementary distribution.If this is the case, these two affixes may be unified as having the function of adding another argument to the predicate, and it may be determined by, e.g., semantic properties of verbs, whether thLy take either the Applicative or causative.Whether or not this is indeed the case, what kind of verbs take the causative/Applicative affixes and what Itind of verbs do not, and what determines the thematic role of the added argument and the semantic interpretation of the resultant clause ate left to further investigation.

['
Cindy is climbing the hill and you can see it.'Itis difficult to translate this alternation ptecisely into English.Comparal to the corresponding (a) sentences, (50b) and (51b) indicate not only the state of mind of the subject, but also some accompanying action by die subject (which possibly affects some other covert participants).In (52a), Cindy's climbing is reported just as the fact, while in (52b), it is irdcated u the progressive action which the speaker and other participants (havers) are observing.Because of the indication of the additional activity involving another participant, the (b) sentences can be said to have a higher degree of Transitivity than the corresponding (a) sentences, as marked by a -wi auxiliary.
Transitive) categories allow auxiliary alternation more easily than those at the extreme ends of the Transitivity contimnim.A certain &pee of thine in Transitivfty might be enough for 'marginal' clauses 2o go across the boundary between 'high' and low' categories.'The same degree of change, however, might not be enough for 'extreme' clauses.Also 'true marginal' clauses may fluctuate between 'high' and low' categories, that is, alternate the auxilim, possibly without changing any meaning.If the degree of Transitivity of crausa and changed degree of Transitivity are somehow quantified, these conelations can be represented numerically.