Lingua 122 (2012) 470480

Contents lists available at SciVerse ScienceDirect

Lingua
journal homepage: www.elsevier.com/locate/lingua

Revisiting the present perfect: Semantic analysis of Andean colonial

documents
Anna Mara Escobar *
School of Literatures, Cultures, and Linguistics, University of Illinois at Urbana-Champaign, 4080 Foreign Languages Building, 707 S. Mathews Avenue, Urbana, IL
61801, United States

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 4 July 2009
Received in revised form 4 October 2011
Accepted 15 October 2011
Available online 21 November 2011

The present perfect (PP) exhibits great variation in use, while concurrently following the
same general evolutionary path across languages. The Spanish PP is no exception, with
some varieties being more conservative (e.g. Northwestern Spain, Mexico City),
than others (e.g. Alicante, Andes). As little is known of the evolution of the PP in the
Andean region the focus of this paper a detailed semantic analysis of perfect
constructions in the Andean colonial period is presented.
Judicial complaints are chosen for the analysis, as they represent controlled eventstructured narratives, where the author is the complainant, and the audience is
the Spanish administration. As expected, PPs are common in the description of the
wrongdoing, since this section narrates events close to the experience of the complainant,
that is, events that have affective charge.
The analysis suggests that during this period, the PP exhibited semantic functions
found in the Spanish of the time, although constrained by discourse strategies attributable
to differences in the social status of individuals in colonial society. A unified development
of the PP is found to have taken place during this stage of the evolution of the PP in this
region, including resultative and current relevance functions.
2011 Elsevier B.V. All rights reserved.

Keywords:
Perfect
Semantic analysis
Structured narrative
Spanish
Andean

1. Introduction
Extensive work has been done on constructions expressing the perfect, both cross-linguistically as well as historically
within specific languages.1 The present perfect, in particular, is described in the literature as exhibiting great variation. The
grammaticalization path leading to present perfects progresses through the following semantic stages: resultative > anterior > perfective/past (Bybee et al., 1994: Chapter 3), although it can develop somewhat differently in distinct
languages (cf. Comrie, 1976; Harris, 1982; Dahl, 1985, 2000; Dahl and Hedin, 2000; Lindstedt, 2000; Detges, 2006). Variation
in the present perfect in modern Spanish is described as representing different evolutionary stages (Meier, 1968:603; Harris,
1982; Squartini and Bertinetto, 2000), and presenting the greatest verbal variation across dialects (Meier, 1968; cf. Howe,
2006). Some varieties are considered more conservative (e.g. Northwestern Spain [Meier, 1968; Harris, 1982], Mexico City
[Lope Blanch, 1983; Squartini, 1998; Schwenter and Cacoullos, 2008], Buenos Aires [Burgos, 2004; Rodrguez Louro, 2009],
ero Pin
ero, 2000]), while others are considered more innovative (e.g. Alicante, Madrid [Schwenter, 1994;
Las Palmas [Pin
Schwenter and Cacoullos, 2008], Andean Spanish [Escobar, 1997; Sanchez, 2003; Howe, 2006]). In some varieties, the

* Tel.: +1 217 333 3390; fax: +1 217 244 8430.

E-mail address: aescobar@illinois.edu
1
Perfect constructions are defined as those composed of an auxiliary and a past participle.
0024-3841/$ see front matter 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.lingua.2011.10.005

A.M. Escobar / Lingua 122 (2012) 470480

471

Table 1
Description of Andean colonial documents.
Colonial period

Origin of complainant

Documents (N = 111)

Source

Early 17th century

Early 17th century
Late 17th century

Spaniard
Indigenous
Indigenous

37
28
46

Fontanella de Weinberg (1993)

Escobar; Rivarola (2000) and Fontanella de Weinberg (1993)a
Escobar; Rivarola (2000)b

a
Three documents come from Fontanella de Weinberg (1993), seven from Rivarola (2000), and the remaining 18 were collected by the author from the
National Archive of Peru and the Archive of the Archbishop of Lima.
b
Twenty-two documents come from Rivarola (2000), and the remaining 24 are documents collected by the author from the National Archive of Peru and
the Archive of the Archbishop of Lima.

Spanish present perfect expresses hodiernal function and contrasts with the preterit (e.g. Alicante, Madrid, cf. Meier, 1968;
Schwenter, 1994; Serrano, 1994; Howe, 2006; Schwenter and Cacoullos, 2008), while in others it expresses evidential
function and contrasts with the pluperfect (e.g. Ecuadorean and Peruvian Andean Spanish; cf. Bustamente, 1991; Escobar,
1997; Howe, 2006; Jara, 2006). Studying the Spanish present perfect, in particular, requires, then, detailed semantic analysis,
paying special attention to the linguistic variety under study (e.g. Jara, 2006; Rodrguez Louro, 2009).
Several questions arise regarding the evolution of the Spanish present perfect (PP), especially in the 15th and 16th
centuries, with the inception of the Spanish-speaking settlements in the Americas and the diversification of Spanish in
diaspora. The present study focuses on the Andean region, where modern varieties of Andean Spanish have been found to use
innovative functions of the PP not found in other varieties of Spanish (Bustamente, 1991; Escobar, 1997; Sanchez, 2003;
Howe, 2006; Jara, 2006). This paper explores the early stages of the formation of the PP during the Andean colonial period
with the goal of uncovering the early functions of the PP in this region.
2. Spanish in the 15th and 16th centuries
During the period of the Spanish conquest and settlement in the Americas (the 15th and 16th centuries), Spanish in the
Peninsula was going through a period of great linguistic variation, with different norms coexisting side by side, as noted in
historical (e.g. Alonso, 1962:306; Cano Aguilar, 1992:225227) and historical sociolinguistic studies (Tuten, 2003; Medina,
2005). Cano explains that this situation was due to the fact that the Spain that emerged in 1479 was mainly a confederation
of smaller kingdoms under the same rulers (1992:222), where medieval variants could be found alongside more modern
ones in the same individuals (cf. Medina, 2005).2 The absence of a single norm in Peninsular Spanish enabled this stable
coexistence of variants, making this an ideal period for the study of Spanish variation (Medina, 2005:11).
Latin American Spanish during the colonial period (16th to early 19th centuries) evolved distant from and with little contact
with the imperial metropolis, although different Peninsular dialects were majority dialects in different urban centers of
Hispanic America (Boyd Bowman, 1968:XVff.). Viceroyalty capitals mainly Mexico City (founded in 1519), Lima (founded in
1535) and Caribbean ports (Cartagena, Veracruz) had more frequent contact with Peninsular norms because they were seats
of their respective Viceroys or entry ports. After Lima and Mexico City (and entry ports), capitals of later Viceroyalties (Bogota,
Ro de la Plata) had less contact, while the majority of the regions of the Spanish territory were considered areas of little or no
contact (e.g. Central America, Venezuela, Paraguay, Santiago de Chile) (de Granda, 1995). In the Americas, then, a more complex
linguistic variation situation ensued after the 16th century settlement period (cf. Cano Aguilar, 1992:225ff.).
In order to account for the variation found during this period, I focus on data from the early 17th century from the Peruvian
Viceroyalty.3 This time period is at least two generations after the creation of this administrative region, and during a
well-known stable and prosperous political and economic period. The documents written during this time and analyzed in this
study consist of judicial claims filed with the judicial or ecclesiastical authorities by individuals living in different regions of the
Peruvian Viceroyalty (Table 1). Somewhat more than half of the complainants are of Spanish origin, while the others are of
Indigenous origin; origin is indicated in this manner at the beginning of each document. A second group of late 17th century
claims (16591679) of Indigenous origin is included in the study to determine changes in patterns of use in the documents.
Claims are chosen because they represent a controlled event-structured narrative (cf. Labov and Waletsky, 1967; cf. Ritz
and Engel, 2008). They all contain an introduction of the complainant (e.g. name, ethnicity, origin), a description of the
wrongdoing done to the plaintiff, the actors involved, a petition for justice, a closing and a signature. Moreover, they all
have the complainant as the author, the Spanish administration as their audience, the same topic (a complaint), and the
same purpose: to obtain justice. Documents will be referred to as monolingual or bilingual documents, depending on
whether the origin of the complainant is Spanish or Indigenous, respectively.
This paper centers on the description of the wrongdoing, because it is in this section that the complainant describes the
events that he/she was involved in, in the temporal sequence in which they took place, and, more importantly, according to
his/her own viewpoint (Fleischman, 1990; Fleischman and Waugh, 1991). Research has shown that the present perfect is a
linguistic variable sensitive to event and discourse structure (cf. Hovav and Levin, 2001; Portner, 2003), both in
conversational (Hernandez, 2006; Ritz and Engel, 2008; Rodrguez Louro, 2009) and narrative styles (Givon, 1982; Rodrguez
2
3

The study centers on language use in picaresque novels, where characters are considered to be speaking a variety of the time (2005:36).
The documents were written between 1595 and 1646.

A.M. Escobar / Lingua 122 (2012) 470480

472

Louro, 2009). Because of the nature of these documents, there is an additional subjective and affective connection between
the perspective of the complainant and his/her description of the wrongdoing. Although the wrongdoing occurs in the past,
its consequences still affect the complainant at the time of speech. The description of the wrongdoing, then, resembles
personal narratives, which are found to favor the use of the present perfect (Labov and Waletsky, 1967; Givon, 1982; cf.
for Spanish, Hernandez, 2006; for Australian English, Ritz and Engel, 2008). As expected, a high incidence of the present
perfect is found in these documents, and only in the description of the wrongdoing. This is not surprising, since the
present perfect (different from the preterit) includes the viewpoint of the speaker (Fleischman, 1990; Fleischman and Waugh,
1991). It is important to also highlight that in complaint documents the roles of the author (the complainant) and that of the
intended addressee (the judicial authority) are well known to both. After all, the purpose of the document is for the author to
denounce or complain about a wrongdoing, and for the intended addressee to use his power to do something about it. In terms of
shared assumptions and expectations, we find that complaints represent ideal contexts for what Portner calls common
ground between speaker/writer and receiver/reader (2003:480), which are described as ideal contexts for (haber) perfects
since they can be used for extended-now presupposition (505), which can give rise to current relevance uses (503). The type of
document analyzed in this study allows us to study the PP in a controlled-event structured narrative which, in addition, has high
affective content.
The sociolinguistic distinction between complainants of Spanish and Indigenous origin during the colonial period is of
special relevance as well. While Spaniards were protected by the laws imposed by the Spanish crown, members of the
Indigenous population were treated as second-class citizens and were obliged to pay a special tax to the crown, not paid by
individuals of Spanish origin. This differential treatment affects the narrative viewpoint (Fleischman, 1990; Fleischman and
Waugh, 1991) that these individuals had towards the Spanish authority. This difference in perspectives has already been
found to be linguistically relevant in the use of performative verbs (communicative and directive) in these documents
(Escobar, 2007, 2008). Since the PP is a speaker-oriented verbal expression (cf. Fleischman, 1990; Fleischman and Waugh,
1991), it is hypothesized that the discourse patterns of use of the PP in these narratives will also be constrained by the
narrative viewpoint taken by the complainant, according to his/her ethnic origin and consequent social status. The objective
of this paper is to uncover differences in the grammatical functions of this verbal expression to gain insight into the early
stages of the development of the modern Andean present perfect.
3. Spanish perfect constructions
The modern Spanish present perfect developed from the stative possession verb haber (to have) + PAST PARTICIPLE. Romance
perfects which evolve from stative verbs follow the resultative > anterior > perfective/past path of development (Bybee et al.,
1994: Chapter 3).4 The resultative is described as signal[ing] that a state exists as a result of a past action, as in the
resultative reading of English to be, The house is (now) painted; while the anterior signals that the situation occurs prior to
reference time and is relevant to the situation at reference time, as in The train has arrived (Bybee et al., 1994:54).5 As the
PP evolves historically from a resultative into an anterior, it appears first, most frequently, with verbs of change or verbs of
state, later gradually appearing with other semantic types (69). Bybee et al. further explain that a typical use of the anterior
will be for indefinite temporal reference, which favors non-use of temporal adverbs (6162). The Spanish anterior is
considered to have developed in the 16th century (Harris, 1982:55). In an historical study of the PP in Peninsular Spanish,
Copple finds that in 15th century texts, it could be found mainly with resultative function and a developing anterior function
(2009:121, 125). However, in Peninsular texts of the 17th century, the anterior function was established, appearing mainly in
indeterminate temporal contexts. This suggests that at the time of the American Spanish settlement in the 16th century, the
Peninsular Spanish PP had both resultative and anterior readings.
In the documents analyzed here, the past participle is also found with ser to be, estar to be, and tener to have. The
present perfect with haber appears in (1) with decir to say in participial form (dicho). Also appearing with dicho said are ser
(2), estar (3) and tener (4).6
(1)

el susodicho las notifico a dona fran[cis]ca de Figueroa mi mujer y a dicho muncho mal de m en mi avsencia [FON, mono,
text 34]
the aforementioned [individual] notified it to Ms. Francisca de Figueroa my wife and has said many bad things
about me in my absence

(2)

vna yiegua castana que llaman yja de la losada la qual conpre de grabiel de la rreguera con la d[ic]ha su madre que
como d[ic]ho es se llama la losada [FON, mono, text 13]
a brown mare that is called Daughter of La Losada which I bought from Gabriel de la Reguera with the mother
of the mentioned [mare] that as has been said she is called La Losada

4
Romance languages that have a PP in the resultative stage are Sicilian and Calabrian (Harris, 1982). Languages that have PP in the perfective/past stage
are Standard French, Standard Rumanian, Sardinian, and Northern Italian (Harris, 1982; Squartini and Bertinetto, 2000).
5
Neither English nor Spanish have a modern reading of resultative in their have-perfects.
6
Each example is followed by the name of the compiler of the collection from which the text comes, the type of document according to its author, and the
identification number of the text. The translations are as close as possible to the original syntax.

A.M. Escobar / Lingua 122 (2012) 470480

(3)

473

Esto q[ue] Esta dicho Es muy importante y necesario a su officio. [FON, mono, text 21]
This that has been said is very important and necessary to your work

(4)

y que esto que dicho tiene es La verdad publico y notorio [ESC, bil-2, text 16]
and this that he has said is the public and well-known truth

Historically, these three latter perfect expressions (with ser, estar, and tener) are semantically related to the haber perfect.
In the early stages of the evolution of Spanish, ser + PART appeared mainly with reflexive and intransitive verbs, while haber
+ PART appeared mainly with transitive verbs (Alonso, 1962:134, 304; Vincent, 1982; Aleza Izquierdo, 1987:21), as in other
Romance languages.7 By the 16th century, ser + PART and haber + PART overlapped with respect to the verbs they could each
appear with. As haber + PART expanded its use to new verbs, Spanish evolved from a two-way system using ser and haber
perfects, into a single perfect auxiliation system with only haber by the 18th century.8 Haber originally appeared with verbs
that expressed change (e.g. of state, location) and later expanded to durative and stative verbs (Aleza Izquierdo, 1987:63), as
to reflexive and intransitive verbs (Alonso, 1962:304). In the colonial documents, the haber perfect seems to be more
grammaticalized than the other perfects, judging by the fixed word order in the periphrasis and non-intervening elements
(cf. Rini, 1990; Harre, 1991:111119), which the other perfects allow, but only in the documents of the early 17th century.
Compare ser and tener in (2) and (4), and estar and tener in (5).
(5)

a.

el dicho padre bino con nosotros a esta Ciudad a donde esta agora escondido (ESC, bil-1, text 2)
the aforementioned father came with us to this city where he is now hidden.

b.

ya tengo. A v[uestr]a m[erced] avisado (FON, mono, text 8)

I already have your highness given notice.

Aleza Izquierdo explains that the two perfects had different functions. Ser + PART centered on the result brought about by an
event and focused on the subject who is affected by the result (1987:60). Haber + PART centered on the event AND the
consequent state (1987:73; cf. Harris, 1982), focusing on the action and its object (1987:64).
Spanish perfect expressions with the copula verbs ser and estar have a related semantic history (cf. Pountain, 1982). In the
Medieval period (1215th century), ser + PART made reference to the result brought about by the event, while estar + PART
pointed to the cause of that result (Aleza Izquierdo, 1987:63). By the early 17th century, ser + PART and estar + PART both
signaled resultative function (63), overlapping with one of the present perfects functions. During the 15th and 16th
centuries, both perfects could appear with the same verbs, first with verbs of movement, then with verbs of mental state,
reflexive, and pronominal verbs, and much later with statives (Aleza Izquierdo, 1987:63, 8990). However, in the 17th
century, ser + PART became less frequent, as estar + PART expanded (63).
Haber + PART and tener + PART were also semantically connected since their emergence in Vulgar Latin (Harris, 1982:59). In
Medieval Spanish, haber and tener competed as verbs of possession.9 The desemanticization which took place in both verbs
as they grammaticalized involved haber + PART marking the beginning of the change (inchoative function) including the
result, and tener + PART expressing durative function in addition to the result (Aleza Izquierdo, 1987:100101).10 While the
first focused on the event and its object, the latter focused on the subject of the verb (1987:99). Tener + PART expressed
resultative function (1987:61) since Cantar de mio Cid in the 12th century. Hence, tener + PART overlapped in function with
estar + PART, and haber + PART during the period of the documents.11
The overlapping functions of the Spanish perfects during the first century of the Spanish settlement in the Americas
suggest that an analysis of the functions of the PP during the early 17th century requires consideration of the semantic and
syntactic characteristics of the main verb. These factors include its lexical aspect (Aktionsart, cf. Vendler, 1967), its semantic
class (following mainly Van Valin, 2005; but also Martin, 1998; Sorace, 2000; McKoon and Macfarland, 2000), and its valency
(transitivity), which are described further below. Due to the social differences between the individuals filing the complaints
and their distinct places in colonial society, factors referring to the viewpoint (or subjectivity) of the complainant
(cf. Fleischman, 1990; Fleischman and Waugh, 1991) are also considered. For example, which actors appear in subject
position and which actors are given agency will give insight into the complainants point of view in the narration of the
wrongdoing. Finally, factors referring to the marking of the time of speech, i.e. when the writing of the complaint took place,
7

This distinction is also described as between unaccusatives with ser and unergatives with haber (Romani, 2006).
Similar to Spanish are Catalan, Portuguese, Sicilian, and Romanian. Other Romance languages have two-way (French, Italian) and three-way contrastive
systems (some dialects of Veneto, Trentino, Lombardo, Sardinian, and Castrovillarese) (Loporcaro, 2007).
9
In the 13th century, tener appeared more frequently than haber as a lexical verb (Harre, 1991:97). Tener also meant to hold or to sustain (98). Haber also
meant to exist.
10
In standard modern Spanish, the tener perfect appears frequently in iterative situations with adverbials such as muchas/varias veces many times or
mucho a lot (Harre, 1991:7072; Squartini, 1998:161), as in tengo ledo este libro muchas veces I read this book many times.
11
El Cid or Cantar de mio Cid, an epic poem of the 12th century, is considered to be the first written text in Spanish, since Latin was the only (or main)
language used in writing until the 13th century (Cano Aguilar, 1992:62). The Kingdom of Castile was created in 1035, but the creation of a standard Spanish
is attributed to Alfonso X the Learned, king of Castile and Leon between 1252 and 1284 (Penny, 2002:1415).
8

A.M. Escobar / Lingua 122 (2012) 470480

474

are especially relevant in the study of the PP, such as taking note of temporal adverbials which include the time of
speech (Portner, 2003). The linguistic factors used in the discourse analysis of the complaints are described in the next
section.
4. Linguistic factors
In a study of the verbal expressions used in Cantar de mio Cid (12th century), Moreno de Alba (2000) finds that the use of
ser/haber + PART had a function similar to that of the historical present, and that such expressions were used for events
referring to the discussed world (Weinrich, 1974:96; cf. Alarcos Llorach, 1984:125126; see also Fleischman, 1990;
Fleischman and Waugh, 1991). As such, it contrasted with the past expressions, the imperfect and preterit, that were used for
referring to events of the narrated world.12 More specifically, Moreno found that the historical present was used for the voice
of the narrator, while the perfect constructions were used for the voices of the characters, and as a consequence, expressed
more subjectivity. Moreno explains that the function of both perfects and of the historical present in this text is to bring the
event to the narrative present, and, hence, closer to the experience of the reader, bringing the reader into the story. The PP at
this time, then, was already serving a discourse function.
The description of the wrongdoing constitutes what Labov and Waletsky (1967) call the complication action of the narrative.
In this section the speaker narrates or retells the events he was involved in and does so in the sequence in which they occurred.
Present perfects are found in this section, although the preterit is considered to be the verbal expression favored in (strict)
sequenced past events (Weinrich, 1974; Alarcos Llorach, 1984). Detail on the order in which the events took place in the
description of the wrongdoing is not necessarily always clarified in the texts, as very few temporal adverbials are used.
Consequently, the order in which the events are presented is taken as the order in which the events took place, following earlier
work.
The presence of deictic adverbials, which include the speech time of the narration, has been found helpful in
distinguishing uses of the PP (e.g. Portner, 2003; de Swart, 2007; for Spanish, Meier, 1968). The PP with anterior function can
be accompanied by imperfective adverbials, which include the speech time (cf. Portner, 2003; de Swart, 2007; Depraetere,
2007).13 The fact that few adverbials were found in the documents might be a consequence of the documents being highly
controlled-event narratives.
Quantified noun phrases are also known to promote a reading which can include the speech time (cf. Depraetere, 2007;
Zagona, 2008). While specified noun phrases promote a perfective reading (expressed in count-NPs, e.g. cuatro anos four
years, las penas the sad events, mi casa my house), unspecified noun phrases promote an imperfective reading (expressed
in bare plurals and mass nouns, e.g. dineros money, cosas things, memoria memory, derecho right), which can include the
speech time (cf. for Spanish, Schwenter and Cacoullos, 2008; Zagona, 2008). Quantified noun phrases are taken into
consideration in the analysis. However, as with temporal adverbials, they are not always present in the data.
In addition to temporally-related elements of the discourse, the subjectivity or point of view of the speaker/writer (cf.
Fleischman, 1990; Fleischman and Waugh, 1991), or in this case, the claimant, is considered in the analysis. Point of view is
defined here as related to the degree of agency that human subjects (the complainant, the accused, and the judicial authority)
express in the narration. Who appears as the subject of the utterance, and who is given the highest degree of agency says
something about the point of view of the author. It is hypothesized that since we are analyzing complaints, the accused will
be given the highest agency, while the complainant will be given the lowest agency, appearing more often as the patient of
the actions performed by the accused.
The previous overview makes clear that in the analysis of perfect constructions, verbal, discourse, and subjectivityrelated elements all need to be taken into consideration. For the main verb, the lexical aspect, the valency, and semantic class
of the verb are relevant. The lexical aspect of the verb follows the broad distinction of telic (closed event) and atelic (open
event) verbs (cf. Vendler, 1967). The valency factor distinguishes between transitive and intransitive verbs. The semantic
class is defined following Van Valin (2005), but also Martin (1998), Sorace (2000), and McKoon and Macfarland (2000). The
verbs are distinguished by degree of agency expressed by its subject, from highest to lowest degree. Verbs with agent
subjects are verbs of activity (write, dance), of movement (walk, climb), and of self-controlled localization (get up, kneel) (see
6). Verbs with experiential subjects, expressing less agency in their subjects, are verbs of cognition (think, know), of
perception (see, hear), and psych-verbs (fear, admire) (see 7). Verbs that take patient subjects are statives (have, exist), verbs
of change of state (grow, die), and change of location (fall, arrive) (see 7).
(6)

Agent:

Activity verbs, e.g. escribir to write, hacer to do or make

Communication verbs, e.g. decir to say, expresar to express
Controlled verbs of movement, e.g. ir to go, seguir to follow

12

The historical present (84%) and the ser/haber + PARTICIPLE (16%) accounted for 38% of the verbs in El Cid.
Bull (1971:87) describes this as a systemic function of the Spanish PP, especially with imperfective and iterative adverbials. Some Spanish examples are
temporal adverbials such as hoy today, esta semana this week (Alarcos Llorach, 1984:24; cf. Howe, 2006:8889), as well as habitual, durative or repetitive
adverbials: siempre always, muchas veces many times (cf. Bull, 1971:8689; Alarcos Llorach, 1984:27; Schwenter and Cacoullos, 2008).
13

A.M. Escobar / Lingua 122 (2012) 470480

475

Table 2
Linguistic factors considered in the analysis.
Linguistic factors
Verb
Aktionsart
Valency
Semantic class
(by degree of agency)

Discourse
Deictic adverbials
Quantification
Speech time
Subjectivity
Syntactic role
Semantic role of subject
Narrative referent
Human
Inanimate

(7)

Experiencer:

Categories

Based on

Atelic, telic
Intransitive, transitive
Activity, verb of movement, self-controlled localization
> cognitive, perception, psych-verbs > statives,
change of state, change of location

Vendler (1967)
Van Valin (2005)
Van Valin (2005); but also Martin (1998),
Sorace (2000) and McKoon and Macfarland (2000)

Speech time included or not

Specified (count NPs) or not (bare plurals, mass N)
Included or not

Gueron (2008) and de Swart (2007)

Depraetere (2007) and Zagona (2008)
Meier (1968) and de Swart (2007)

Subject > object > other

Agent > experiencer > recipient > patient

Van Valin (2005)

Weinrich (1974) and Moreno de Alba (2000)

Complainant, accused, judicial authority

Wrongdoing, complaint, laws

Cognition verbs, e.g. sonar to dream, pensar to think of

Perception verbs, e.g. ver to see, oir to hear

Patient:

Stative/existence verbs, e.g. estar to be, ser to be

Change of state, e.g. convencer to convince, morirse to die
Change of place or location, e.g. venir to come, caerse to fall

Regarding discourse factors, the presence of deictic adverbials and quantified noun phrases provide information on
whether the speech time of the narration is being included or not in the reading (cf. Gueron, 2008; de Swart, 2007). Finally,
information on subjectivity (or viewpoint) expressed in the use of the PP is uncovered in the analysis of the semantic roles of
human subject participants in the narration, i.e. whether their role is of agent, experiencer, or patient. To summarize, the
semantic and syntactic behavior of the main verb, together with the discourse properties of the event structure it is
associated with, can serve as the foundation for an event structure-based analysis (cf. Hovav and Levin, 2001; Hernandez,
2006; Ritz and Engel, 2008). Table 2 displays the linguistic factors considered in the analysis.
Because of the relatively low number of present haber/ser/estar/tener perfect tokens in the 111 documents (317 of a total
of 3981 verbs), statistical measures are inappropriate, so only general frequency counts and a detailed qualitative linguistic
analysis are performed on the data.
5. Andean colonial documents
A frequency count of the verbal forms in the documents (see Table 3) shows that in each type of document perfects account
for 8% of verb forms, that the haber perfect represents the majority in each case, but that the size of that majority and, more
importantly, the proportions of the other three perfect constructions, are more variable. Variations in the use of the preterit and
within the three perfects (with tener, estar and ser) suggest differences in the use of these verbal constructions in the documents.
The haber perfect is the preferred type in the three sets of documents, making up from 55% to 69% of the total number of
perfect constructions. In the monolingual documents, the three non-haber perfects occur with very similar frequencies. The
early 17th century bilingual documents, however, seem to favor the estar-perfect construction. The late 17th century
bilingual documents, on the other hand, suggest a convergence towards the monolingual use: tener and estar perfects now
have similar low frequencies, and the frequency of the haber perfect is now slightly greater than in the monolingual set. The
ser perfect must be considered a special case in the analysis of the late 17th century documents, since by this time ser + PART
had started to disappear (cf. Aleza Izquierdo, 1987).14
In all the Andean colonial documents, the haber perfect appears preferably with transitive verbs (72%, 147/253) (8), and with
some intransitive verbs (28%, 58/253), especially telic verbs of movement (venir to come, llegar to arrive) (9) and statives
(estar/ser to be, haber to have, vivir to live) (10). The expansion of haber over ser perfects is already present in this data.
(8)

y de como injustamente pidiendo terminos i mas terminos me a Entretenido El Pleito dos anos [FON, mono, text 35]
and how unjustly by asking [new] deadlines and more deadlines he has kept the dispute [going] for two years

14

Unfortunately, at this time, no monolingual documents of the same period (late 17th century) are available for comparison.

A.M. Escobar / Lingua 122 (2012) 470480

476

Table 3
Perfect constructions in the Andean colonial documents.

Verbal expression

Tokens
N = 3981

Preterite
Perfect

957
317

24
8

Haber + PART
Tener + PART
Estar + PART
Ser + PART

205
40
46
26

65
13
14
8

(9)

Bilingual 17a
N = 848
(/28 doc. = 30)

Bilingual 17b
N = 1628
(/47 doc. = 35)

24%
8%

31%
8%

N = 125

N = 67

N = 125

66%
12%
10%
12%

55%
15%
24%
6%

Monolingual 17a
N = 1505
(/37 doc. = 41)
17%
8%

69%
12%
13%
5%

estos malos chrestianos endios que han binido de Lima [RIV, bil-2, text 12]
these bad Christian Indians who have come from Lima

(10)

todo lo qual a sido En g[r]an desaCato de la j[ustici]a Eclesiastica que V[uestra] m[erce]d administra
[ESC, bil-1, text 1]
all of which has been in great contempt of the Ecclesiastical justice that your Highness administers

Verbs that appear in tener (11) and estar (12) perfects are all transitive in all documents. With ser (13), verbs are
exclusively transitive only in the monolingual documents, suggesting a passive reading, which is a more grammaticalized
function of the ser + PERFECT (cf. Alonso, 1962:197, 223). In the bilingual documents, however, intransitive verbs can also be
found, suggesting that in these latter documents earlier functions of ser + PERFECT with intransitives are still present.
Nonetheless, since the numbers are small, no conclusions can be drawn regarding ser.
(11)

ya por muchas tengo escrito como con vn juan lopez harriero inbie dozientos p[es]os En rreales [FON, mono, text 6]
already through several [writings] I have written how with a Juan Lopez Harriero I sent 200 pesos in coins

(12)

se les paga en cada vn ano de la caxa de los censos de los indios como esta declarado por el Senor Juez de los
d[ic]hos Censos [ESC, bil-1, text 15]
every year they are paid from the treasury of the census of the Indians as it is established [declared] by the
Lord Judge of the said census

(13)

vna yiegua castana que llaman yja de la losada la qual conpre de grabiel de la rreguera con la d[ic]ha su madre
que como d[ic]ho es se llama la losada [FON, mono, text 13]
a brown mare that is called Daughter of La Losada which I bought from Gabriel de la Reguera with the
mother of the aforementioned [mare] that as has been said she is called La Losada

Both ser and estar appear with the same verbs, sometimes in the same document (14), without a clear difference in
meaning. This seems to signal again that the use of ser with perfects was in decline.15 Ser perfects are therefore not further
discussed here.
(14)

a.

digo que mi querellu contra Santiago Guaranga, indio natural deste pueblo de Ambar sobre que susudicho
esta amansebado con poco temor de Dios Noestro Senor con vna endia bioda llamada Luisa Caxa
[RIV, bil-2, text 20]
I say that my complaint against Santiago Guaranga, natural Indian of this town of Ambar about
[the fact] that the mentioned [individual] is cohabiting with little fear of God our Lord with a
widowed Indian named Luisa Caxa

b.

que la susudicha endia se llama Lionor Mara, con quen es mansebado el dicho Alonso de Todo Santos
mucho timpo desta parte [RIV, bil-2, text 27]
that the mentioned Indian is named Leonor Mara, with whom has been cohabiting the mentioned
Alonso de Todo Santos [for] a long time to this date

Focusing on the lexical aspect of the main verb, the monolingual data suggest that haber perfects constitute a different
kind of perfect from the others. While non-haber perfects appear more frequently with transitive telics in the monolingual
15
The loss of semantic contrast between the two verbs seems to also be reflected in its use with adjectives, e.g. estoy cassique I am a cacique [an indigenous
leader], where estar appears in a context where ser is expected.

A.M. Escobar / Lingua 122 (2012) 470480

477

data (around 67% each), haber perfects appear with transitive verbs regardless of telicity. The bilingual data, on the other
hand, show a different pattern. In the early 17th century documents, tener perfects (but not haber perfects) appear with
transitive verbs regardless of telicity, while haber and estar perfects show a preference for transitive telic verbs (75%). The
data from the late 17th century show yet a third pattern: both haber and tener perfects favor transitive telic verbs (6771%),
while estar perfects appear with transitive telic verbs only in 39% of the data. What these patterns seem to suggest is that
telicity and valence alone cannot explain the uses of these perfects. Discourse-related functions seem to be the underlying
factors that influence the use of the perfects in the three varieties. We therefore turn to a consideration of semantic class and
subjectivity-related linguistic factors.
A look at the semantic class of the verb by degree of agency of its subject suggests that haber and tener are used differently
in the monolingual and the bilingual documents. The difference is in the degree of agency expressed in the subject of each
periphrasis. Tener perfects are found in all documents with main verbs of activity and communication (e.g. decir to say,
hablar to speak) (15), and verbs of change (of state or location; e.g. embargar to take away, errar to make a mistake) (16).
(15)

con tanta crueldad Con dos Ramales crudios [sic] de cuero que me dexaron las Espaldas llagadas y llenas de sangre
lo qual chico [sic: hizo] como tengo dicho sin Causa ni Racon mas de tan solam[en]te porque Jure verdad
[ESC, bil-1, text 1]
with great cruelty with two raw strands for whips that left my back [with] wounds and full of blood
which he did as I have said without cause or reason but only because I swore the truth

(16)

tenimos embargados todo lo poco que ellas tinen, aunque dos endias dellas case nada tinen [RIV, bil-2, text 10]
we have had all the little that they [we] have seized, although two of the Indian women have almost nothing

In all documents, the subjects of tener perfects take human subjects overwhelmingly the complainant (6789%), but also
the accused (in one case, it refers to the judicial authority). By choosing the complainant as the subject, the tener perfect
stresses the resulting state of an action performed by the complainant. In these cases, as the main verb is either a
communicative verb or a verb of change, the tener perfect stresses what the complainant said (e.g. tengo dicho I have said) or
the change that the complainant experienced, as in example (16) (their things have been taken from them). When the
subject is the accused (1133%), the use of tener perfects stresses the resulting state of an action performed by the accused
(e.g. tiene cometido he has committed) against the complainant. The remnant possessive meaning of tener places the focus of
the result on the subject. As a consequence, the subjects in these cases have a non-agentive reading.
Similarly, haber perfects take human subjects in 8089% of the cases. The accused is usually the subject in all documents
(5183%), although the complainant is also favored in the bilingual documents (49%). Haber perfects are used to refer to
actions performed by the accused, which are at the center of the complaint narrative. Bilingual documents favor telic verbs
(7071%), which further emphasize actions performed by the accused (17).
(17)

y Siete patacones en plata q[u]e quito a la d[ic]ha india de la Bolsa maltratandola en lo qual ha cometido delito
digno de castigo [ESC, bil-1, text 3]
and seven coins in silver that he took from the mentioned Indian (female) from the (her) bag, mistreating
her, by which he has committed a crime worthy of punishment

When the complainant is the subject in the bilingual documents, it appears mainly with verbs of change in 7086% of the
cases. This preference is not found in the monolingual documents, which do not favor any type of verb. The verbs of change in
the bilingual documents refer to actions caused by the accused, which affected (changed) the everyday life (or state) of the
complainants, as in (18, 19).
(18)

Muchos indios se an muerto sin confision por no querer venir el dicho padre Meja a confesarlos [RIV, bil-1, text 4]
many Indians have died without confession because the mentioned priest Meja did not want to come
to confess them.

(19)

el dicho cap[ita]n don gaspar . . . entro a gorvernar [sic] que abra quarenta anos y los derechos de sepolturas
de indios forasteros que han enterrado en la dicha iglesia an entrado en Su poder
the mentioned captain mister Gaspar . . . started governing, must be 40 years ago And the rights of burial
of the foreign Indians that they have buried in the mentioned Church, have entered under his poder

The data from both types of individuals suggest, then, a strong correlation between the semantic roles of the subjects and
the semantic class of the main verb of the perfects that accompany them. While haber and tener perfects have human
subjects in 88100% of the examples, estar (and ser) perfects prefer inanimate subjects (the wrongdoing, the judicial
document, the law). However, while tener perfects appear preferably with the complainant as a patient or experiencer, haber
perfects appear with the accused as the agent subject or the complainant as the non-agent subject (see Table 4). Both
documents seem to use haber perfect, the focus of this study, in a similar fashion. All documents use the haber perfect mainly

A.M. Escobar / Lingua 122 (2012) 470480

478
Table 4
Patterns of use of haber perfects in the data.

Valency
Lexical aspect
Semantic class
Degree of agency
Subjects
Complainant
Accused
Speech time

Monolingual

Bilingual 17a

Bilingual 17b

Mainly transitive
All
All
All
Human
Non-agent
Agent
Yes

Mainly transitive
Mainly telic
Mainly V of change
Mainly non-agent
Human
Non-agent
Agent
Yes

Mainly transitive
Mainly telic
Mainly V of change
Mainly non-agent
Human
Non-agent
Agent
Yes

with transitive verbs with human subjects, and include the speech time. However, there are also differences between the
monolingual and bilingual documents, which have to do with the types of main verbs that appear with the haber perfect.
While the monolingual documents show no preference for the lexical aspect, semantic class and type of verb by degree of
agency expressed in the subject, the bilingual documents do show a preference, as they tend to favor telic verbs, especially
verbs of change when the subject is the complainant. When the subject is the accused, the haber perfect is used to highlight
the actions performed by the accused that constitute the core of the wrongdoing (see 1, 6, 21). The complainant appears more
often as subject when accompanied by verbs of change, which highlights the consequences suffered by the complainant by
the actions of the accused (see 22, 23).
As second-class citizens during this period, it was imperative for individuals of indigenous origin to be more assertive in
denouncing the wrongdoings done to them by the accused. The higher incidence of haber perfects with telic verbs, mainly
verbs of change, in their documents signals their desire to highlight the wrongdoing actions themselves, bringing these
actions closer to the judicial authority, and therefore making them more real. The data suggest that the use of the haber and
tener perfects is to highlight either the action of the agent subject (especially in the case of the accused with haber perfects),
or the resulting state affecting the subject (especially in the case of the complainant with tener perfects and with verbs of
change with haber perfects), or the resultant state itself, which came about due to the events that took place (especially with
ser/estar perfects). Nonetheless, in all documents the effects of the wrongdoing are presented as (still) in force and
(still) highly relevant at the time the official complaint is made. Since the actions of the wrongdoing itself are relevant to the
speech time when the complaint is being made, one could argue that the PP is being used to make more salient, more vivid
(cf. Ritz and Engel, 2008), and more real, the actions being referred to.
In the data, there are also uses of the PP with clear past-time function, with no state resulting from the action that is still
ongoing at speech time. In 3949% of the examples (from all the documents), the event took place in the past and does not
include the speech time in its reading (20, 21).16
(20)

a mi noticia es venido que baltazar Ticllay auri yn[di]o del pueblo de chacos a presentado vna peticion
ante v[uestra] m[erce] declarando en ella lo que avia depuesto contra el lic[encia]do Alonso peres de llanos
[ESC, bil-1, text 12]
I received news that Baltazar Ticllay Auri, Indian of the town of Chacos, has presented a petition to your
Highness declaring the statement he had made against Alonso Peres de Llanos

(21)

y se an dado a las calles con vn yndio pregonero [RIV, bil-1, text 3]

and they have taken [themselves] to the streets with an Indian town crier

The fact that the preterite is used in these documents for past-time events raises questions regarding the function of haber
perfects with perfective events.17 While 39% (32/82) of the haber perfects in the monolingual documents fall into this
category, almost half of the haber perfects in the bilingual documents appear with perfective events.18 The subjects are
mainly the accused, but the complainant also appears as the subject. However, the events have in common that when
the accused is the subject, the present perfect is used with verbs that refer to actions done by the accused that constitute the
heart of the wrongdoing, e.g. asotar to whip, maltratar to mistreat, quitar to take away, amenazar to threaten, forzar to
force someone. Likewise, when the complainant is the subject, the PP appears with verbs that refer to past events highly
relevant to the narration, such as tener noticias to receive news, or actions done in their defense, ir a Lima to go to Lima,
escribir a vuestra merced to write to the authority, huir to escape, darse a las calles to go out into the streets. In both
instances, the PP is used to highlight events that are highly relevant to the narration of the wrongdoing.
This seems to suggest that the use of haber perfects in perfective past events in these documents is a discourse strategy to
bring to the experience of the reader the event itself or to bring the reader closer to the event itself. That is, the purpose is to
16
Only one example of this type occurs with estar in the bilingual documents, questan ensserrada de vna despenssa that [she] are [was] locked up in a
pantry.
17
No deictic adverbials or unspecified quantified noun phrases are present that could be said to bring the event to the speech time.
18
The percentages are 49% (18/37) in the early bilingual and 47% (40/86) in the late bilingual documents.

A.M. Escobar / Lingua 122 (2012) 470480

479

present these events as most relevant to the present narration, and to highlight them in the description of the wrongdoing.
The fact that the mentioned events are the core of the description of the wrongdoing suggests that the haber perfect has
present relevance function in these examples, regardless of how long ago they took place. These examples represent uses of
the PP in colonial Spanish for anterior function.
6. Final discussion
An analysis of the PP was presented with the goal of uncovering early functions in its evolution in the Andean region, by
contrasting two classes of colonial documents, one authored by individuals who presented themselves as being of Spanish
origin, the other authored by individuals who described themselves as of Indigenous origin. Complaint narratives were the
focus of the study because they represent a controlled event-structured narrative with clear assumptions and clear
elements: a complainant (the author), an accused, a wrong suffered by the complainant at the hands of the accused, and the
same audience a judicial or ecclesiastical authority of the Spanish administration. The analysis shows that the controlled
event-structured narrative of a complaint document is an ideal corpus to study the PP, since the description of the
wrongdoing has an affective charge, which favors its use.
The PP is used in all the documents to highlight events in the description of the wrongdoing. However, the two types of
documents differ in the verb types that are employed when highlighting these main events. Although the subject was more
frequently the accused, the bilingual documents had the complainant as the subject almost as frequently as the accused, and in
higher frequencies than the monolingual documents. When the accused was the subject, highlighted events constituted the
core of the wrongdoing, that is, actions performed by the accused as a subject agent. When the subject was the complainant, he/
she appeared as a non-agent subject to whom the wrongdoing had been done. In the bilingual documents, the complainant
appears as a subjectpatient mainly with verbs of change and stative verbs, while the accused appears as the subjectagent
mainly with activity and communication verbs. That the accused and the complainant appear as the subject at similar rates in
the bilingual documents is attributable to the viewpoint from which the indigenous complainant chooses to describe the
wrongdoing done to him/her. The PP is used here to bring closer to the reader the actions of the accused and the sufferings of the
complainant. Moreover, the bilingual documents seem to take advantage of the possibility to express affect in the PP, by using it
more with telic verbs with all subjects, and with verbs of change and stative verbs with complainant-subjects.
Although most of the events with PP included the speech time, a substantial proportion of events (39% in the monolingual
documents and almost 50% in the bilingual documents) are described with a PP that makes reference to perfective past
events without any result existing in the speech time. By bringing these past events to the experience of the reader, the
authors are using the PP for present relevance, again to highlight the affective charge. This suggests that the PP has anterior
function in this variety of Spanish, and that it is being used for present relevance in these documents.
To summarize, the analysis suggests that during the Andean colonial period, the semantic functions of the PP were similar
in the monolingual and bilingual documents. The uses of the PP of the time suggest a unified development, as an anterior
which marks current relevance. The discourse differences found in the data correspond to the different viewpoints of
monolinguals and bilinguals in colonial society, attributable to their social standing. However, in the bilingual documents,
the preferred narrative of the events favors contexts that are more sensitive to subjectivity and speech time-related factors,
resulting in higher frequencies of use of the PP for verbs of certain types. This would explain why we find similar frequencies
of the accused and the complainant as subjects in the bilingual documents, but not in the monolingual documents, where
there is a preference for having the accused as the subject. The favoring of certain contexts in the bilingual documents is no
doubt connected to their authors lower standing in colonial society. This social dynamic continued until the socioeconomic
changes of the 20th century, when the innovative evidential function of the PP appears (cf. Escobar, 1997). This suggests a
hypothesis connecting both the use of the PP in certain types of structures of narratives and the viewpoint of the narrator to
the later development of the PP in this region. Studies of event-structured narratives of the intervening centuries can help
shed further light on the emergence of the modern present perfect in the Andean region.
Acknowledgments
I would like to thank Bill Pagliuca, colleagues present at the 2009 Linguistics Association of the Southwest, where an early
version of this paper was presented, and two anonymous reviewers for their valuable and helpful comments and
suggestions.
References
ol, 3rd ed. Gredos, Madrid.
Alarcos Llorach, E., 1984. Estudios de gramatica funcional del espan
ol medieval). Universidad de Valencia, Valencia.
Aleza Izquierdo, M., 1987. Ser con participio de perfecto en construcciones activas no oblicuas (espan
ol desde el iberorromano hasta nuestros das. Aguilar, Madrid.
Alonso, M., 1962. Evolucion sintactica del espanol. Sintaxis historica del espan
oles de America en el siglo XVI. Academia Mexicana de Genealoga y
Boyd Bowman, P., 1968. Indice geobiografico de cuarenta mil pobladores espan
Heraldica, Ciudad de Mexico.
Bull, W., 1971. Time, Tense and the Verb. University of California Press, Berkeley.
Burgos, D., 2004. Anteriority Marking in British English, Standard German and Argentinean Spanish: An Empirical Examination with Special Emphasis on
Temporal Adverbials. Lincom, Muenchen.
ol quiten
o. Lexis 15, 195231.
Bustamente, I., 1991. El presente perfecto o preterito compuesto en el espan

480

A.M. Escobar / Lingua 122 (2012) 470480

Bybee, J., Perkins, R., Pagliuca, W., 1994. The Evolution of Grammar: The Grammaticalization of Tense, Aspect and Modality in the Languages of the World.
University of Chicago Press, Chicago.
ol a traves de los tiempos. Arco Libros, Madrid.
Cano Aguilar, R., 1992. El espan
Comrie, B., 1976. Aspect: An Introduction to Verbal Aspect and Related Problems. Cambridge University Press, Cambridge.
Copple, M.T., 2009. A diachronic study of the Spanish perfect(ive): tracking the constraints on a grammaticalizing construction. Doctoral Thesis. University
of New Mexico.
., 1985. Tense and Aspect Systems. Blackwell, Oxford.
Dahl, O
. (Ed.), 2000. Tense and Aspect in the Languages of Europe. Mouton, de Gruyter, Berlin.
Dahl, O
., Hedin, E., 2000. Current relevance and event reference. In: Dahl, O
. (Ed.), Tense and Aspect in the Languages of Europe. Mouton, de Gruyter, Berlin,
Dahl, O
pp. 386401.
ol. Un modelo interpretativo sociohistorico de variantes areales de contacto
de Granda, G., 1995. El influjo de las lenguas indoamericanas sobre el espan
lingustico. Revista Andina 13, 173198.
de Swart, H., 2007. A cross-linguistic discourse analysis of the perfect. Journal of Pragmatics 39, 22732307.
Depraetere, I., 2007. (A)telicity and intentionality. Linguistics 45, 243269.
Detges, U., 2006. Aspect and pragmatics. The passe compose in Old French and the Old Spanish perfecto compuesto. In: Eksell, K., Vinther, T. (Eds.), Change
in Verbal Systems. Issues on Explanation. Peter Lang, Frankfurt am Main, pp. 4772.
Escobar, A.M., 1997. Contrastive and innovative uses of the present perfect and the preterite in Spanish in contact with Quechua. Hispania 80, 859870.
Escobar, A.M., 2007. Performative verbs in Spanish monolingual and bilingual colonial court documents. In: Camacho, J., Flores-Ferran, N.,Sanchez, L., Deprez,
V., Cabrera, J. (Eds.), Romance Linguistics 2006. Selected papers from the 36th Linguistic Symposium on Romance Languages (LSRL), New Brunswick,
MarchApril 2006. John Benjamins, Amsterdam, pp. 121134.
Escobar, A.M., 2008. Deixis social en documentos jurdicos del virreinato del Peru: Analisis verbal de textos monolingues y bilingues. In: Company, C., Moreno
ola, Merida, Mexico, 48 Setiembre 2006. Arco Libros, Asociacion de
de Alba, J. (Eds.),Actas del VII Congreso Internacional de Historia de la Lengua Espan
ola, Madrid.
Historia de la Lengua Espan
Fleischman, S., 1990. Tense and Narrativity. University of Texas Press, Austin.
Fleischman, S., Waugh, L. (Eds.), 1991. Discourse-Pragmatics and the Verb: The Evidence from Romance. Routledge, London.
ola, Madrid.
Fontanella de Weinberg, M.B., 1993. (Comp.) Documentos para la historia lingustica de Hispanoamerica siglos XVI a XVIII. Real Academia Espan
Givon, T., 1982. Evidentiality and epistemic space. Studies in Language 6, 2349.
Gueron, J., 2008. On the difference between telicity and perfectivity. Lingua 118, 18161840.
Harre, C., 1991. Tener + Past PART: A Case Study in Linguistic Description. Routledge, London.
Harris, M., 1982. The past simple and the present perfect in Romance. In: Harris, M., Nigel, V. (Eds.), Studies in the Romance Verb. Croom Helm, London, pp.
4270.
Hernandez, J.E., 2006. Present perfect for preterit in Salvadoran narratives: the perfective expansion into narrative discourse. In: Sagarra, N., Almeida, J.T.
(Eds.), Selected Proceedings of the 9th Hispanic Linguistics Symposium. Cascadilla Proceedings, Project, Sommerville, MA, pp. 297307.
Hovav, M.R., Levin, B., 2001. An event structure account of English resultatives. Language 77, 766797.
Howe, C., 2006. Cross-dialectal features of the Spanish present perfect: a typological analysis of form and function. Doctoral Thesis. The Ohio State
University.
Jara, I.M., 2006. The use of the present perfect and the preterit in the Spanish of Lima. Doctoral Thesis. University of Pittsburgh.
Labov, W., Waletsky, J., 1967. Narrative analysis: oral versions of personal experience. In: Helm, J. (Ed.), Essays on the Verbal and Visual Arts. University of
Washington Press, Seattle, pp. 1244.
. (Ed.), Tense and Aspect in the Languages of Europe. Mouton, de Gruyter,
Lindstedt, J., 2000. The perfect-aspectual, temporal and evidential. In: Dahl, O
Berlin, pp. 365383.
ol de Mexico. Universidad Nacional
Lope Blanch, J., 1983. Sobre el uso del preterito en el espanol de Mexico. In: Lope Blanch, J. (Ed.), Estudios sobre el espan
Autonoma de Mexico, Ciudad de Mexico, pp. 131143.
Loporcaro, M., 2007. On triple auxiliation in Romance. Linguistics 45, 173222.
Martin, J., 1998. The lexico-syntactic interface: psych verbs and psych nouns. Hispania 81 (3), 619631.
McKoon, G., Macfarland, T., 2000. Externally and internally caused change of state verbs. Language 76, 833858.
Medina Morales, F., 2005. La lengua del siglo de oro. Un estudio de variacion lingustica. Universidad de Granada, Granada.
ola, peninsular e hispanoamericana. Actas III. Asociacion Internacional de Hispanistas, Soria, pp. 601610. http://cvc.
Meier, H., 1968. Sintaxis verbal espan
cervantes.es/obref/aih/pdf/03/aih_03_1_068.pdf

Moreno de Alba, J., 2000. Los preteritos de indicativo en el poema del Cid. Nueva Revista de Filologa Hispanica XLVIII, 275297.
Penny, R., 2002. A History of the Spanish Language. Cambridge University Press, Cambridge.
ero Pin
ero, G., 2000. Perfecto simple y perfecto compuesto en la norma culta de Las Palmas de Gran Canaria. Iberoamericana, Madrid/Vervuert, Frankfurt.
Pin
Portner, P., 2003. The (temporal) semantics and (modal) pragmatics of the perfect. Linguistics and Philosophy 26, 459510.
Pountain, C., 1982. Essere/stare as a Romance phenomenon. In: Harris, M., Vincent, N. (Eds.), Studies in the Romance Verb. Croom Helm, London, pp. 139
160.
Rini, J., 1990. Dating the grammaticalization of the Spanish clitic pronoun. Zeitschrift fur Romanische Philologie 106, 354370.
Ritz, M.A., Engel, D.M., 2008. Vivid narrative use and the meaning of the present perfect in spoken Australian English. Linguistics 46, 131160.
ol andino. Textos de bilingues de los siglos XVI y XVII. Iberoamericana, Madrid/Vervuert, Frankfurt.
Rivarola, J.L., 2000. Espan
Rodrguez Louro, C., 2009. Perfect evolution and change: a sociolinguistic study of preterit and present perfect usage in contemporary and earlier Argentina.
Doctoral Thesis. The University of Melbourne.
ola.
Romani, P., 2006. Tiempos de formacion romance I. Los tiempos compuestos. In: Company Company, C. (Cor.), Sintaxis historica de la lengua espan
Primera parte, La frase verbal, vol. 1. Fondo de Cultura Economica, Mexico, pp. 243348.
Sanchez, L., 2003. Quechua-Spanish Bilingualism: Interference and Convergence in Functional Categories. John Benjamins, Amsterdam.
Schwenter, S., 1994. The grammaticalization of an anterior in progress: evidence from a Peninsular Spanish dialect. Studies in Language 18, 71111.
Schwenter, S., Cacoullos, R., 2008. Defaults and indeterminacy in temporal grammaticalization: the perfect road to perfective. Language Variation and
Change 20, 139.
ol de Canarias y Madrid. Lingustica
Serrano, M.J., 1994. Del preterito indefinido al preterito perfecto: un caso de cambio y gramaticalizacion en el espan
ola Actual 16, 3757.
Espan
Sorace, A., 2000. Gradients in auxiliary selection with intransitive verbs. Language 76, 859890.
Squartini, M., 1998. Verbal Periphrases in Romance: Aspect, Actionality and Grammaticalization. Mouton, de Gruyter, Berlin.
. (Ed.), Tense and Aspect in the Languages of Europe.
Squartini, M., Bertinetto, P.M., 2000. The simple and compound past in Romance languages. In: Dahl, O
Mouton, de Gruyter, Berlin, pp. 403439.
Tuten, D., 2003. Koineization in Medieval Spanish. Oxford University Press, Oxford.
Van Valin, R., 2005. Exploring the SyntaxSemantics Interface. Cambridge University Press, Cambridge.
Vendler, Z., 1967. Verbs and times. In: Vendler, Z. (Ed.), Linguistics and Philosophy. Cornell University Press, Ithaca, NY, pp. 97121.
Vincent, N., 1982. The development of the auxiliaries habere and esse in Romance. In: Harris, M., Vincent, N. (Eds.), Studies in the Romance Verb. Croom
Helm, London, pp. 7196.
Weinrich, H., 1974. Estructura y funcion de los tiempos en el lenguaje. Gredos, Madrid.
Zagona, K., 2008. Perfective aspect and contained perfectivity. Lingua 118, 17661789.