Available via license: CC BY-NC 4.0
Content may be subject to copyright.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 1
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024
doi: 10.20396/liames.v24i00.8676041
Counterfactual conditional strategies in some Amazonian languages
Jesús Olguín Martínez
University of Illinois, USA
https://orcid.org/0000-0002-4555-4213
Alonso Vásquez Aguilar
University of California, Santa Barbara, USA
https://orcid.org/0000-0001-5856-9863
ABSTRACT: Studies on individual Amazonian languages have shown that these languages can contribute to
informing and refining our theories of counterfactual conditional constructions. Still missing, however, is an
attempt at exploring this complex sentence construction across different genetic units of the Amazonia in a single
study. The paper explores counterfactual conditionals in a sample of 24 Amazonian languages. Special attention
is paid to the range of TAM markers and clause-linking devices used in counterfactual conditionals in the
Amazonian languages in the sample. As for TAM markers, it is shown that protases tend to be unmarked (they
do not occur with any TAM values), and apodoses tend to occur with irrealis or frustrative marking. As for clause-
linking devices, it is shown that most Amazonian languages in the sample contain counterfactual conditionals
occurring with non-specialized clause-linking devices. This means that the distinction between counterfactual
conditionals and other types of conditionals (e.g., real/generic) is not grammaticalized in clause-linking devices.
Instead, the counterfactual conditional meaning resides in the combination of specific TAM markers. The paper
also pays close attention to the distribution of TAM markers and clause-linking devices in counterfactual
conditional constructions in the Vaupés. In particular, special attention is paid to how Tariana counterfactual
conditional construction have been shaped by Tucanoan languages through language contact.
KEYWORDS: Counterfactual conditionals; Amazonian languages; Complex sentence; Tense-Aspect-Mood;
Clause-linking devices
RESUMEN: Estudios de diversas lenguas amazónicas han demostrado que estas lenguas pueden contribuir a nuestro
conocimiento teórico sobre las construcciones condicionales contrafactuales. Sin embargo, hasta el momento, no
hay investigaciones que tomen en cuenta lenguas amazónicas de diferentes bloques genéticos en una sola
investigación. El presente estudio explora las construcciones condicionales contrafactuales partiendo de una
muestra de 24 lenguas amazónicas. Se presta especial atención a la variedad de marcadores TAM y dispositivos
de vinculación de cláusulas utilizados en construcciones condicionales contrafactuales en las lenguas amazónicas
de la muestra. En cuanto a los marcadores TAM, se demuestra que la prótasis tiende a no estar marcada (no ocurre
con ningún valor de TAM) y la apódosis tiende a ocurrir con marcas irrealis o frustrativos. En cuanto a los
dispositivos de vinculación de cláusulas, se demuestra que la mayoría de las lenguas amazónicas en la muestra
contienen construcciones condicionales contrafactuales que ocurren con dispositivos de vinculación de cláusulas
no especializados. Esto significa que la distinción entre condicionales contrafactuales y otros tipos de
condicionales (real/genérico) no está gramaticalizada en los dispositivos de vinculación de cláusulas. Mas bien,
el significado condicional contrafactual reside en la combinación de marcadores TAM específicos. El artículo
también presta atención a la distribución de marcadores TAM y dispositivos de vinculación de cláusulas en
construcciones condicionales contrafactuales en el Vaupés. En particular, se presta atención a cómo lenguas de la
familia tucano han influido en la forma de la construcción condicional contrafactual de la lengua Tariana (Arawak)
a través del contacto lingüístico.
PALABRAS CLAVE: Condicionales contrafactuales; Lenguas amazónicas; Cláusula compleja; Tiempo-Aspecto-
Modo; Conectivos
1. Introduction
Various studies have explored conditionals crosslinguistically (e.g., Comrie 1986;
Haiman & Kuteva 2001; Olguín Martínez & Lester 2021; Xrakovskij 2005). However, only a
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 2
few studies have analyzed conditionals in specific regions (but see Nicolle 2017). The present
paper contributes to filling this gap by exploring conditionals in one specific region, that is,
Amazonia.
There are a number of publications that explicitly recognize conditional constructions
in particular Amazonian languages. They demonstrate that many languages of this region show
exceptions to wider typological generalizations. For instance, crosslinguistically, there seems
to be a strong correlation between counterfactual conditionals and irrealis marking because, as
explained by Mithun (1995: 384), when languages have a grammaticalized realis/irrealis
distinction, counterfactual conditionals tend to be encoded with irrealis markers. Interestingly,
a number of Southern Arawakan languages show that the picture may be more complex. In
many Southern Arawakan languages (e.g., Paunaka), irrealis is used in all types of conditionals
(i.e., conditional and main clauses occur in the irrealis; Danielsen & Terhart 2016: 14). There
are other Southern Arawakan languages (e.g., Terena) that also have a grammaticalized
realis/irrealis distinction. However, they differ from other Southern Arawakan languages in
that the conditional clause occurs with an irrealis marker and the main clause appears with a
realis marker, regardless of the type of conditional (simple, hypothetical, and counterfactual;
Danielsen & Terhart 2016: 14).
These studies indicate that Amazonian languages can contribute to inform and refine
our theories of conditional constructions. Still missing, however, is an attempt at exploring this
complex sentence construction across different genetic units of the Amazonia in a single study.
This type of investigation would be valuable as a general overview of conditionals, and would
be invaluable to those documenting and describing Amazonian languages, alerting them to
details to watch for and chronicle.
To keep the scope of the paper manageable, we only focus on one specific type of
conditional: counterfactual conditionals, as in (1). The reason behind this decision is that
crosslinguistically, counterfactual conditionals show formal and discourse properties that other
types of conditionals do not (Haiman & Kuteva 2001; Olguín Martínez & Lester 2021). This
seems to indicate that, within the realm of conditional constructions, counterfactual
conditionals have a special status.
Urarina (Isolate)
(1)
baana
it-a=ne
hananiane,
raj
kalaui-t
m-akat.
if
be.near-3SG=SUB
if
POSS
son-PL
catch-1PL.SBJ
‘If its creatures had been near, we would have caught it (about a peccary).’ (Olawsky
2006: 255).
Counterfactual conditionals originate from the human cognitive ability to compare
reality with what might have been (Olguin Martinez & Lester 2021). A counterfactual
conditional is a construction in which a condition is interpreted as contrary to fact, i.e., as
assumed to be true in a possible world that is incompatible with the real (actual) world
(Declerck & Reed 2001: 13). Protasis and apodosis are the most common ways to refer to the
counterfactual conditional clause and the main clause respectively. Counterfactual conditionals
may have present time reference (2a) or past time reference (2b). Given that most sources
contain information on past counterfactual conditionals, we focus on this pattern. This is a
construction that expresses a conditional relationship between two situations that failed to be
realized in the past (Dixon 2009: 16; Michael 2014).
(2) a. If she were here, she would help us.
b.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 3
The analysis is based on a sample of 24 languages. The goals of this article are
threefold. First, it has long been observed that, across a large number of unrelated languages
around the world, past tense markers tend to appear in counterfactual conditional constructions
(Comrie 1986). Different linguists have offered a possible explanation for the use of past tense
markers in counterfactual conditional constructions (e.g., von Prince 2019; Steele 1975). The
question is: what is the range of TAM markers that appear in the protasis and the apodosis of
a counterfactual conditional construction in the languages in the sample?
Second, counterfactual conditionals are encoded with different types of clause-linkage
patterns. These may be specialized in that they are only used for expressing a counterfactual
conditional meaning. In this scenario, the distinction between counterfactual conditionals and
other types of conditionals (e.g., real/generic) is grammaticalized in clause-linking devices.
Clause-linkage patterns may also be non-specialized in that they appear in counterfactual
conditional constructions, but also in other semantic types of conditionals (e.g., real, generic,
and hypothetical). It has also been noted that there are languages with no segmental lexeme or
morpheme that could translate as ‘if’ in counterfactual conditionals. These are paratactic
constructions in that they do not appear with any clause-linking devices. The question is: what
is the range of clause-linkage patterns by which counterfactual conditional constructions are
encoded in the languages of the sample?
The third and last goal of the article is areal in that it pays close attention to the
distribution of TAM markers and clause-linkage patterns in counterfactual conditional
constructions in the Vaupés. In particular, special attention is paid to how Tariana
counterfactual conditional construction have been shaped by Tucanoan languages through
language contact.
The organization of this paper is as follows. Section 2 introduces the sample used for
the present study. Section 3 documents the range of TAM values that tend to appear in both
the protasis and apodosis in the languages in the database. Section 4 investigates the range of
clause-linkage patterns by which counterfactual conditional constructions are encoded in the
languages in the sample. Section 5 explores the areality of counterfactual conditionals in the
Vaupés with special attention to Tariana and Tucanoan counterfactual conditional
constructions. Section 6 reviews the conclusions and implications of this paper and provides a
number of fruitful areas for future research.
2. Sample
Since this is primarily an explorative study that seeks to characterize a type of
construction traditionally neglected in the study of Amazonian languages, we tried to include
languages from each of the families found in the Amazonia. Because existing materials differ
tremendously in their delicacy and completeness with respect to the description of
counterfactual constructions, the present study takes into account a sample of 24 languages
belonging to 15 different language families listed in Table 1. Besides grammatical information
on counterfactual conditionals, the source also had to contain a detailed description of TAM
markers. In particular, how these markers are defined in the language of study. Moreover, the
source also had to contain a description of other conditional clauses to determine whether a
clause-linking device is specialized or non-specialized. Note that we take into account more
than one Arawakan language in the sample because many sources of languages belonging to
this family provide detailed information on counterfactual conditionals. In what follows, the
structure and motivations behind the selection of the languages for the current sample are
introduced.
A bottom-up method has been employed for building the sample of the present study.
Constructing a sample of this type means, in its simplest form, picking one language from
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 4
every family found in the Amazonia. Based on this, an attempt was made to find one language
from each family for which the available literature gives sufficient information on the grammar
of counterfactual conditional constructions. It was possible to find sufficient information on 24
languages, as is shown in Table 1. Languages from almost all Amazonian language families
and isolates are represented. Furthermore, it is important to mention that linguistic fieldworkers
on many languages of the sample have also been consulted to confirm certain analyses of the
data and/or discuss alternative analyses. By and large, this method of data collection has been
described as the ‘grammar-cum-dictionary method’ (Kortmann 1997: 53), i.e., the basic
information on counterfactual conditionals has been collected from available descriptive
grammars and dictionaries, and corrected and/or modified by linguistic fieldworkers.
Table 1. Languages of the sample
Language family
Language(s)
Sum
Arawakan
Asheninka Perené (Mihas 2015), Baure (Danielsen
2007), Paresi (Brandão 2014), Resígaro (Allin
1976), Tariana (Aikhenvald 2003), Yine-Piro
(Hanson 2010)
6
Arauan
Jarawara (Dixon 2004)
1
Boran
Bora (Thiesen & Weber 2012; Elvis Walter
Panduro Ruiz, pers. comm.)
1
Cariban
Hixkaryana (Derbyshire 1979)
1
Chapacuran-Wanham
Wari (Everett & Kern 1997)
1
Chicham
Aguaruna (Overall 2017; Simon Overall, pers.
comm.)
1
Huitotoan
Murui (Wojtylak 2020; Katarzyna Wojtylak, pers.
comm.)
1
Isolates
Kwaza (van der Voort 2004), Mosetén, (Sakel
2002; Jeanette Sakel, pers. comm.), Puinave
(Girón 2008), Urarina (Olawsky 2006)
4
Nadahup
Hup (Epps 2008; Patience Epps, pers. comm.)
1
Nambikuaran
Mamaindé (Eberhard 2009)
1
Nuclear-Macro-Je
Krahô (Maxwell Gomes 2014)
1
Panoan
Kakataibo (Zariquiey 2018; Roberto Zariquiey,
pers. comm.)
1
Takanan
Ese Ejja (Vuillermet 2012)
1
Tucanoan
Tucano (West 1980)
1
Tupi-Guarani
Paraguayan Guarani (Estigarribia 2020)
1
Zaparoan
Iquito (Michael 2009; Lev Michael, pers. comm.)
1
Total
24
Map 1 illustrates the geographical distribution of the languages of the present study.
All languages are located within the Amazonia geographical limits. The Amazonia is the
lowland region drained by the Amazon and Orinoco Rivers and extending to the northern and
eastern littorals of the continent (Dixon & Aikhenvald 1999: 4). It is bordered by the Andes
mountains to the west, the Caribbean and Atlantic oceans in the north and east, and the drier
regions of the Gran Chaco to the south (Epps & Michael 2017: 935). Note that defining the
area where Amazonian languages are spoken is not an easy task. This stems from the fact that
some families have members inside and outside the Amazonia.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 5
Map 1. Geographical distribution of the languages of the sample
Before we leave the present section, mention should be made of the following
methodological issue. As will be shown in the remainder of the paper, there are counterfactual
conditional examples that show positive or negative polarity. In many cases, the difference
between a negative clause and a corresponding affirmative clause gives rise to a number of
formal asymmetries. In particular, the TAM categories used in affirmative counterfactual
conditionals may not be the same as those used in negative counterfactual conditionals (see
Miestamo 2005 for a detailed discussion of various formal and functional asymmetries between
affirmation and negation). Ideally, the language sample should only contain counterfactual
conditionals showing positive polarity. However, a number of sources do not contain
information on counterfactual conditionals showing positive polarity. We decided to include
these languages and we are aware that this methodological decision is not without problems.
Nonetheless, these problematic cases are rather few and do not detract from the validity of the
overall conclusions.
3. TAM markers in counterfactual conditionals
It is a well-known fact that the semantics of TAM markers may harmonize with the
semantics of different types of adverbial clauses (Cristofaro 2003: 111). For instance,
Schmidtke-Bode (2009: 43) explains that since purpose clauses are future-oriented, they tend
to take irrealis marking. As for ‘after’ clauses, Hetterle (2015: 76-77) shows that,
crosslinguistically, temporally subsequent constructions tend to appear with past or perfective
marking in the adverbial clause. This stems from the fact that ‘after’ clauses tend to be past-
oriented and the proposition that they convey precedes the proposition of the main clause, and
it is completed at the onset of the main clause situation. ‘Before’ clauses also show systematic
patterns. ‘Before’ clauses express a situation that takes place posterior to the main clause
situation. Put another way, the situation expressed by the ‘before’ clause is not yet realized at
the time of the main clause situation. In many languages around the world, the semantics
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 6
translates directly into the coding properties of this adverbial relation in that ‘before’ clauses
tend to occur with future tense markers (Hetterle 2015: 77). What this seems to indicate is that
there are very systematic crosslinguistic correlations between TAM marking and the meaning
of adverbial clauses.
Counterfactual conditional constructions also seem to show well-motivated
correlations. Crosslinguistically, counterfactual conditionals tend to appear with TAM markers
whose semantics is appropriate to the counterfactual conditional context, such as irrealis
markers and counterfactual mood markers, among others (Mithun 1995: 384; Olguin Martinez
& Lester 2021). For instance, Mithun (1999: 173) mentions that “the irrealis portrays situations
as purely within the realm of thought, knowable only through imagination.” Given that
counterfactual conditionals express non-actualized situations, the semantics of irrealis markers
is appropriate to the counterfactual conditional context. However, it has long been observed
that, across a large number of unrelated languages, past tense markers, and other TAM markers
whose semantics does not harmonize with the counterfactual conditional meaning (e.g.,
perfective, completive), may appear in counterfactual conditional constructions (Comrie 1986;
Karawani 2014; Olguin Martinez & Lester 2021). This is a clear mismatch for the reason that
past tense and perfective marking tend to occur in situations that are actualized and
counterfactual conditionals express non-actualized situations. To explain this mismatch, Steele
(1975) and von Prince (2019) mention that past and counterfactuality share a semantic core of
distance from the actual present. Put another way, in this scenario, the connection between past
tense and counterfactual conditionals is that the past tense marker has as its basic meaning not
past tense but something distant from present reality. Karawani (2014: 15) mentions that the
connection between past tense and counterfactual conditionals stems from the fact that there is
an inherent nature of the past as being closed and therefore the condition is impossible or false.
In what follows, we analyze the range of TAM markers that occur in the protasis and
apodosis of counterfactual conditional constructions in the Amazonian languages in the
sample.
3.1 TAM values of the protasis
In most languages in the database, counterfactual conditional protases are unmarked
(13 languages). By unmarked is meant that they do not occur with any TAM values. This is
not common crosslinguistically in that protases tend to be marked with irrealis in the languages
of the world (Olguin Martinez & Lester 2021: 167). The Urarina counterfactual conditional
construction in (3) contains an unmarked protasis in that it does not appear with any TAM
values. In these languages, the counterfactual conditional meaning is achieved by the TAM
values of the apodosis (commonly marked with TAM values that harmonized with the
counterfactual context) and/or by a specialized clause-linking device. Crosslinguistically, in
many languages, the distinction between counterfactual conditionals and other types of
conditionals (e.g., real/generic) is grammaticalized in clause-linking devices (see Section 4).
Urarina (Isolate)
(3)
baana
it-a=ne
hananiane,
raj
kalaui-t
m-akat.
if
be.near-3SG=SUB
if
POSS
son-PL
catch-1PL.SBJ
‘If its creatures had been near, we would have caught it (about a peccary).’ (Olawsky
2006: 255).
In four languages in our sample, protases are marked with irrealis, as is shown in the
Mosetén example in (4). One remark on the irrealis category is in order here. A source of
potential confusion in any discussion on irrealis is that it has been applied to different concepts
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 7
and constructions in languages from many areas of the world. It is therefore important to clarify
what is meant when using this term. In the present paper, we consider irrealis as specific
markers (rather than notional descriptions of non-encoded meanings of constructions) in the
forms of verbal affixes and clausal enclitics (Brooks 2018: 4). There seems to be a strong
correlation between counterfactual conditionals and irrealis marking because, as explained by
Mithun (1995: 384), when languages have a grammaticalized realis/irrealis distinction,
counterfactual conditionals tend to be encoded by irrealis marking. This study supports this
theoretical claim in that most languages in the sample that have a grammaticalized
realis/irrealis distinction tend to be marked with irrealis.
Mosetén (Isolate)
(4)
mö-ya
obra-in
dejar-ye--nä-wï-rä’
obra-in,
3PL-ADESS
work-PL
leave-VS-3SG.OBJ-FOC-CERT-IRR
work-PL
mo--nä-wï-
progreso.
F-ADESS-FOC-CERT-IRR
progress
‘If they had left works, there would have been progress.’ (Sakel 2002: 441)
In three languages, protases are nominalized in that they appear with nominalizing
morphology.
1
Note that although the protasis verb is nominalized in these languages, it may
retain specific verbal categories. An example illustrating this pattern is found in Kwaza. In this
language, the protasis of a counterfactual conditional construction is marked with the
nominalizer - (5). The fact that counterfactual conditional protases may be nominalized
seems not to be surprising in that “probably the most common subordination strategy in South-
American languages is nominalization” (van Gijn et al. 2011: 10).
Kwaza (Isolate)
(5)
-daa-h-kywy-'ta,
dai-'he-da-rydy-ki.
know-1SG-NMLZ-if-COSUB
take-NEG-1SG-IRR-DECL
‘If I had known (the bulb was so weak), I would not have bought it.’ (van der Voort
2004: 631)
In one language, frustrative markers occur in the protasis of a counterfactual conditional
construction. The frustrative is a “grammatical marker that expresses the non-realization of
some expected outcome implied by the proposition expressed in the marked clause” (Overall
2017: 479).
In Iquito, the protasis occurs with the frustrative marker =ti (6). Typologically, frustrative as a
grammatically-marked category is not commonly found across languages; however, it does
seem to be a category employed by a fair number of languages of the greater Amazon region
(Overall 2017). This could be the result of a combination of areal diffusion and genetic
inheritance (see Section 3.3 for a more detailed discussion of frustratives and counterfactual
conditionals).
1
As correctly pointed by one reviewer, nominalized verb forms must be considered a subset of actualized and
non-actualized patterns because nominalized verb forms can be used for both actualized and non-actualized
situations.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 8
Iquito (Zaparoan)
(6)
ca=quias=ti,
quia-cúhuaaja
-quiáana.
NEG=2SG=CF=wake-MOM
2SG-heart
CF=2SG.IRR=tell-PFV.REP
‘Had you not awakened, your heart would have warned you.’ (Michael 2009: 157)
The use of past tense or perfective markers in counterfactual conditional protases is rare
in the languages of the database. In Bora, ca-clauses occur in the past (7). Another example is
found in Resígaro. In this language, the protasis of a counterfactual conditional construction
occurs with the past tense marker -mí (8). Note that the Resigaro example in (8) differs from
the Bora example in (7) in that the past tense marker -mí also appears with the irrealis marker
-ma. These are two semantically conflicting verbal inflections: the past tense marker -mí
(expected to occur in actualized situations) and the irrealis marker -ma (expected to occur in
non-actualized situations). This type of mixed pattern in counterfactual conditional protases
seems to be common in Eurasian languages (see Olguin Martinez & Lester 2021: 167).
2
Bora (Boran)
(7) ú-pée u -ca, muurá ávyétá -í-yó teéne.
2SG-PST 2SG die-if CONF almost sad-FRUST-FUT DEM
‘It would have been sad if you had die.’ (Elvis Walter Panduro Ruiz, pers. comm.)
Resígaro (Arawakan)
(8)
anepuu
aá
ee
kha-tshí-ma-mí,
ka
va.
much
father
fish
do-if-UNREAL-REC
well
we-eat
‘If my father had caught a lot of fish, we would have eaten well.’ (Allin 1976:
261)
3.2 TAM values in the apodosis
In most languages in the sample (10 languages), apodoses are marked with irrealis. In
Jarawara, the counterfactual conditional apodosis appears with the irrealis marker -ne (9).
Jarawara (Arauan)
(9)
faao
tee
ka-jawi
jaa,
faha
mee
water(F)
1NSG.A
APPL-be.jealous.over.NOM
PER
water(F)
2NSG.A
kii
re-ne.
look.at
NEG-IRR
‘If you had protected (your) waters (lit. been jealous over your waters), they
wouldn’t have fished the waters (lit. looked at the waters).’ (Dixon 2004: 215)
In six languages in the database, frustrative markers occur in the apodosis of a
counterfactual conditional construction (10). Note that the frustrative may also appear in
combination with other TAM markers, as in the Piro example in (11), where the frustrative
marker -maka must also be accompanied with the completive marker -na.
2
As correctly pointed out by one reviewer, this is also true of many Arawakan languages, especially those of the
Kampan group.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 9
Puinave (Isolate)
(10)
a-padatá
ka-dikú,
a-kuk-nók
brasit-á
.
1SG-money
3PL-DAT.EXIST
1SG-FRUST-go
Brasil-ALL
now
‘If I had been rich, I would have gone to Brazil.’ (Girón 2008: 416-417)
Piro (Arawakan)
(11)
t-ma
m--,
n-nika-na-lo-na-pa-maka-ni.
3SG.F-NEG.do
PRIV-run-ANTIC
1SG-eat-COMPL-3SG.F-COMPL-ELAT-FRUST-AFFCT
‘If she had not escaped, I would have eaten her.’ (Hanson 2010: 357)
In four languages in the sample, past tense or perfective markers occur in combination
with irrealis markers or conditional mood markers, as in shown in the Barasano example in (12).
Barasano (Tucanoan)
(12)
rioho
goro
b
goti-hab,
b-re
ha-beti-boo-ri-a-da
y.
straight
truly
2SBJ
tell-if
2S-OBJ
hit-NEG-IRR-PST-3-agree
1SBJ
‘If you had told the truth, I wouldn’t have hit you, right?’ (Jones & Jones 1991: 124)
Unlike unmarked counterfactual conditional protases (see Section 3.1), unmarked
apodoses are scarce in the database. This pattern is only attested in three languages in the
sample. In Murui, counterfactual conditional meanings are expressed with a construction in
which the apodosis is unmarked (13).
Murui (Huitotoan)
(13)
kue
mare-d-kue-na,
bi--kue.
1SG
good.ATT-LK-1SG-if
come-LK-1SG
‘If I had been well, I would have come.’ (Wojtylak 2020: 504)
In one language, a counterfactual conditional apodosis is nominalized. In Baure, the
apodosis is nominalized in that it appears with the nominalizing suffix -no, as in (14). Danielsen
(2007: 418) mentions that this nominalizing suffix sometimes is used to refer to a situation
completed in the past.
Baure (Arawakan)
(14)
vi=ka--po-no,
ver
vi=poto~poto-he-no.
INTERJ
1PL=go-IRR-PFV.REFL-NMLZ
CONJ
1PL=be.wet~INTENS-DISTR-NMLZ
‘Phew, if we had gone (home), we would have got completely wet.’ (Danielsen 2007: 418)
3.3 Discussion
The previous subsections have shown that protases tend to be unmarked and apodoses
tend to occur with irrealis marking in the Amazonian languages in the sample. Furthermore, it
has been shown that in a number of such languages, counterfactual conditional constructions
may be encoded with frustrative markers.
As for frustrative markers, it has been claimed that if an Amazonian language contains
a frustrative marker, it will tend to appear in counterfactual conditional contexts (Muller 2013:
159; Overall 2017). This is attested in Alto Perene, Bora, Hixkariana, Hup, Iquito, Paresi,
Puinave, and Yine Piro in the database. The fact that counterfactual conditionals occur with
frustrative markers is not surprising. Given that the frustrative is used for indicating the non-
realization of a situation, this harmonizes with the counterfactual meaning of counterfactual
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 10
conditional constructions. The question is: are there any languages in the sample that have a
frustrative marker that is not used in the expression of counterfactual conditional meanings? If
so, how are counterfactual conditional constructions encoded in these languages?
Table 2. Amazonian languages in the sample with a frustrative marker not used in the expression
of counterfactual conditional meanings
Languages
Frustrative marker
Protasis TAM
Apodosis TAM
Aguaruna
-takama
Perfective marking
Potential mood marker
-mai
Barasano
-boa
Unmarked
Irrealis marker -boo and past
tense marker -ri
Ese Ejja
-ʼaxa
Unmarked
Potential mood marker -me or
potential mood marker -kyae
Kakataibo
-
Unmarked
Conditional mood marker -tsin
and perfective marker -a
Kwaza
-le
Nominalizer -
Irrealis marker -rydy
Mosetén
-
Irrealis marker -ri
Irrealis marker -ri
Paresi
zaore
Irrealis marker =iya
Irrealismarker =iya (optional)
Tariana
-tha
Unmarked
Past conditional marker
-buhtaka
As is shown in Table 2, there are a number of Amazonian languages in the sample in
which frustrative markers are not used for expressing counterfactual conditional meanings. An
example is found in Paresi. In this language, the frustrative zaore indicates that the goal of an
action was not achieved (15). Counterfactual conditionals are not encoded with this marker.
Instead, protases and apodoses occur with the irrealis marker =iya (16).
Paresi (Arawakan)
(15)
ka-kikitsa-ke-heta
zaore
w=aika-hena
acordo
tyoma
wi=kakoa
ATT-separate-TH-PFV
FRUST
1PL=say-TRS
agreement
tyoma
1PL=COM
nikare-hare-ta
wa=sofre-hitiya
hoka.
like.this-M-IPFV
1PL=suffer-ITER
then
‘We wanted to kick him out, but he made an agreement with us, and we are
suffering.’ (Brandão 2014: 307)
(16)
no=tyoma-re=iya
hoka
no=waini=iya.
1SG=do-NMLZ=IRR
then
1SG=die=IRR
‘If I had done this, I would have died.’ (Brandão 2014: 398)
Another example is attested in Mosetén. In this language, the marker -indicates
that something did not turn out as expected or did not happen in relation to the context (Sakel
2002: 410). Usually, this marker appears in combinations of clauses, but can also mark a
contrast in a single clause, as in the example in (17). Counterfactual conditionals are not formed
with the frustrative marker -. Instead, they occur with the irrealis marker -, as is shown
in the example in (18).
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 11
Mosetén (Isolate)
(17)
-tsa’--
tsin
achae-i
kaechh-ae--ki
añe-
now-FRUST-NEC-IRR
1PL
dog-VI
go.on-VI-PROG-CON
rain-VI.F.S
‘We should have gone hunting with dogs now, but it goes on raining.’ (Sakel
2002: 410)
(18)
mö-ya
obra-in
dejar-ye--nä-wï-rä’
obra-in
3PL-ADESS
work-PL
leave-VS-3SG.OBJ-FOC-CERT-IRR
work-PL
mo--nä-wï-rä’
progreso.
F-ADESS-FOC-NEC-IRR
progress
‘If they had left works, there would have been progress.’ (Sakel 2002:
441)
As can be seen in Table 2, the Amazonian languages in the sample, in which frustrative
markers are not used in the expression of counterfactual conditional meanings, tend to use other
TAM markers. In particular, they tend to use TAM markers that harmonize with the
counterfactual meaning of this type of conditional construction, such as irrealis markers.
4. Clause-linkage patterns
In many languages around the world, clause-linking devices are among the most
important means to establish subordinative and coordinative relations (Hetterle 2015: 106).
These markers may sometimes shed light on the type of semantic relation holding between
clauses in that they serve as elements for labeling complex sentence relations like causal,
conditional or temporal relations (Verstraete 2014: 195). Harder (1996: 94) mentions that of
all grammatical elements in an adverbial clause construction, clause-linking devices are the
most necessary element to get the message across; “you can do fairly well without articles and
tense and auxiliaries, but if you mess up the clause-linkers you really leave your listener in the
dark.” Counterfactual conditionals show an interesting picture. While in a number of
languages, clause-linking devices played an important role in the expression of a counterfactual
conditional meaning, in other languages, TAM markers seem to play a more important role
than clause-linking markers. Besides clause-linking devices, there are other languages in which
counterfactual conditionals are realized with a paratactic pattern. In the present study, we use
the term ‘clause-linkage patterns’ as a cover term to refer to both clause-linking devices and
paratactic patterns. For this study, we classify clause-linkage patterns in the following way.
First, specialized clause-linking devices refer to items that are only used for expressing
counterfactual conditional meanings. This indicates that the distinction between counterfactual
conditionals and other types of conditionals (e.g., real/generic) is grammaticalized in clause-
linking devices. In Bora, the clause-linking device -ca is only used for encoding counterfactual
conditional constructions (19). Note that other types of conditionals appear with another
marker (20). Accordingly, -ca is a specialized clause-linking device. In four languages in the
sample, counterfactual conditionals are encoded with specialized clause-linking devices.
Bora (Boran)
(19) ú-pée u péé-ca muurá, teene ímí-iyá-hi.
2SG-PST 2SG go-if CONF DEM good-FUT-FRUST
‘It would have been good if you had gone.’ (Elvis Walter Panduro Ruiz, pers. comm.)
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 12
(20)
imít
-ʔàxtʃhí:
-mà
ó
ph
-.
you
want-if
you-with
I
go-FUT-<t>
‘If you wish, I will go with you.’ (Thiesen & Weber 2012: 371)
Second, non-specialized clause-linking devices refer to items that appear in
counterfactual conditionals and other semantic types of conditionals (e.g., real, generic, and
hypothetical). In Aguaruna, all semantic types of conditionals occur with -ka, as can be
observed in (21) and (22). On these grounds, -ka is non-specialized. Most languages in the
sample contain counterfactual conditionals occurring with non-specialized clause-linking
devices (18 languages). This means that in these languages, the counterfactual conditional
meaning resides in the combination of specific TAM markers.
Aguaruna (Chicham)
(21) wi kaini wi-a-ku-nu-ka,
1SG.SBJ tomorrow go-IPFV-SIM-1SG.SS-COND
taka-sa-ta-tata-ha-i.
work-ATT-NEG-FUT-1SG.SBJ-DECL
‘If I go tomorrow, I wonʼt work.’ (Overall 2017: 391)
(22) nu=na washi=na dushiki-a-cha-ku-un=ka,
ana=ACC spider.monkey=ACC laught.at-IPFV-NEG-SIM-1SG.SS=COND
‘If I hadn’t laughed at that monkey,
tuku-mai-inu awak-ka-ha-i.
shoot-POT-NMLZ overcome-PFV-1SG-DECL
I would have been able to shoot it.’ (Overall 2017: 495)
Given that in these languages, the clause-linking marker is non-specialized and does
not contribute to the counterfactual interpretation of the construction, there are Amazonian
languages in which the clause-linking device is optional and can be omitted. In Baure,
counterfactual conditionals appear with the non-specialized clause-linking device ver (23).
This marker can be omitted without affecting the counterfactual meaning between clauses (24).
In this regard, Danielsen (2007: 418) mentions that the clause-linking device ver “is used as a
neutral connector, which separates the two subsequent predicates from another and makes clear
that we are dealing with two clauses.” The optionality of constructional properties from
adverbial clause constructions has not gone unnoticed. In the context of adverbial clauses,
Hetterle (2015: 108) shows that in many languages, adverbial clause constructions can dispense
with any constructional property (e.g., TAM markers, clause-linking devices) as long as the
semantic relation holding between clauses is sufficiently cued by the remaining constructional
properties of the construction (Schmidtke-Bode 2009: 33). In the recent typological and
psycholinguistic literature, such patterns have attracted increasing attention under the label of
‘redundancy management in grammar.’
Baure (Arawakan)
(23)
vi=ka--po-no,
ver
vi=poto~poto-he-no.
INTERJ
1PL=go-IRR-PFV.REFL-NMLZ
CONJ
1PL=be.wet~INTENS-DISTR-NMLZ
‘Phew, if we had gone (home), we would have got completely wet.’ (Danielsen
2007: 418)
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 13
(24)
nka
pi=pa-a=ri-no
yiti
nka
ri=ka-wo-no.
NEG
2SG=give-IRR=3SGF-NMLZ
chili
NEG
3SGF=go-COP-NMLZ
‘If you had not given her chili, she would not have gone.’ (Danielsen 2007: 418)
Third, parataxis refers to those instances in which counterfactual conditionals do not
appear with any clause-linking device, as can be observed in the Iquito example in (25). Two
languages in the sample contain paratactic counterfactual conditionals. In these languages,
counterfactual conditionals are formed with irrealis markers, counterfactual mood markers, or
frustrative markers. This indicates that in this type of counterfactual conditional, TAM markers
serve as triggers of the counterfactual conditional interpretation. Put another way, if we add up
the meanings contributed by each of the TAM markers in the combinations, this leads to an
interpretation that includes the feature of reversal of polarity, i.e., it yields the feature of non-
occurrence (but it did not happen) (Van Linden & Verstraete 2008). This is in line with Mauri
& van der Auwera (2012: 396), who mention that in paratactic counterfactual conditionals not
all is left to inferential processes. Rather, if a language expresses counterfactual conditionals
with paratactic constructions, the clauses must be marked as irrealis (by means of irrealis,
dubitative, or hypothetical elements) in order for the counterfactual conditional relation to be
inferable.
Iquito (Zaparoan)
(25)
ca=quias=ti,
quia-cúhuaaja
ɨtɨ=quiao=átuu-quiáana.
NEG=2SG=CF=wake-MOM
2SG-heart
CF=2SG.IRR=tell-PFV.REP
‘Had you not awakened, your heart would have warned you.’ (Michael 2009: 157)
5. Areality
This section pays close attention to the distribution of TAM markers and clause-linking
devices in counterfactual conditional constructions in the Vaupés area. In particular, special
attention is paid to how Tariana counterfactual conditional construction may have been shaped
by Tucanoan languages through language contact. Exploring the internal diversity of Tucanoan
and North Arawak languages is important to conduct this task. This type of analysis is known
as the ‘intra-genetic typological approach’ (see Comrie 1993: 10; Kibrik 1998: 61). Bickel
(2008) mentions that for many typological research questions, it has become crucial to study
intra-genetic variance.
3
This is essential, for example, if one wants to estimate historical
stability, transition probabilities, and direction of spread of a pattern.
5.1 The Vaupés
The Vaupés region of the Brazilian and Colombian Amazon is relatively well
established as a linguistic area, characterized by considerable indirect diffusion of grammatical
categories and patterns but little direct borrowing of forms (see Aikhenvald 1996, 2002: ch.
10; Epps 2007: 267). Languages from four different families (Tucanoan, North Arawak
languages, Naduhup, and Kakua-Nɨkak) form the Vaupés linguistic area (Epps & Michael
2017: 938). The area of the Vaupés River basin is an intensive contact area within the Upper
Rio Negro basin, and it has received the most in-depth attention of any South-American contact
area. Some of the linguistic characteristics that the languages of this zone share are the
3
Kibrik (1998: 61) notes that the extragenetic typological approach must be enriched by the intragenetic approach
since this will enable us to make more fine-grained typological generalizations. Typologists should not blind
themselves to the fact that important insights into crosslinguistic variation can also be gleaned from the
examination of variation among languages genetically related (Comrie 1993: 10).
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 14
following: (i) a remarkably low number of lexical borrowings in basic vocabulary, (ii) a
pervasive calquing of the local lexicon (such as names and ethnonyms, flora and fauna, and
items of material and ritual culture), (iii) a significant degree of morpheme-to-morpheme and
word-to-word intertranslatability caused by areal diffusion, (iv) many similarities in serial verb
constructions, and (v) similarities on the expression of spatial relations, among others (Epps &
Michael 2017: 938).
The Vaupés area is distinct from many of the world’s other linguistic areas in that it has
apparently been shaped largely through unilateral, rather than multilateral language contact
(Epps 2007: 270). This socio-cultural context seems to be a case of asymmetric bilingualism,
defined as a “situation whereby a community speaking language A tends to become bilingual
in another language B, while the reverse is not true. Because speakers of B tend not to learn
language A, this increases the social pressure upon A speakers to eventually shift to language
B” (François 2012: 99).
As for clause-combining, it has been proposed that different types of complex sentence
constructions and discourse patterns have diffused through language contact in this area. For
instance, Tariana, Hup, and many Tucanoan languages, express precautioning situations in
similar ways (e.g., ). In a precautioning
construction, the main clause typically has directive illocutionary force, with the aim of
preventing the probable and undesirable situation from happening (e.g., Take your umbrella so
). While in many languages around the world, precautioning situations
are signaled with a purpose clause in combination with a negative marker (e.g., Put the food
there so that the ants do not eat it), the Amazonian languages mentioned before display a
special morphology for expressing negative purpose, as is the case of avertive ‘lest’ markers
(Olguín Martínez & Vásquez-Aguilar 2022: 12).
Another example is the following. Tariana contains recapitulative and summary tail-
head linkage construction that show different discourse functions.
4
Aikhenvald (2019: 488)
notes that these patterns are the result of relatively recent areal diffusion from East Tucanoan
languages into Tariana. She supports this hypothesis by explaining that Tariana and East
Tucanoan languages are spoken in the same region and are not genetically related. The transfer
of discourse patterns through contact is not uncommon (Mithun 2008: 208). Discourse
preferences are particularly prone to diffuse much more quickly and easily than grammatical
features (Beier et al. 2002: 123). Hup also contains similar tail-head linkage constructions and
is also involved in the Vaupé’s contact situation (Epps 2007: 285). Given that Hup has
undergone contact-induced restructuring of its discourse organization under the influence of
East Tucanoan languages (Epps 2007: 268), it seems reasonable to assume that tail-head
linkage constructions are the result of contact with Tucanoan languages. Furthermore, other
Naduhup languages seem not to have tail-head linkage constructions (e.g., Dâw; Martins 2004;
Yuhup; Ospina Bozzi 2002).
5
4
Tail-head linkage refers to a construction which contributes to discourse cohesion and structuring in that it “links
sentences or paragraphs together, usually by repetition of at least part of the previous clause” (Thurman 1975:
342). Two types have been traditionally recognized: recapitulating and summary constructions (see de Vries 2005:
364; Olguin Martinez 2023). First, recapitulative constructions involve the repetition of the predicate of one clause
(the tail clause) in the following clause (the head clause) (de Vries 2005: 364). Second, summary tail-head linkage
constructions involve the replacement of the lexical verb of the tail clause by a generic or light verb (de Vries
2005; Guérin 2015; Guérin & Aiton 2019).
5
Obert (2019: 5) mentions that Dâw has a specific type of tail-head linkage construction in which locative
adverbial clauses in sentence initial position can be exact replicas of a clause at the beginning of the following
sentence: ‘After this, we arrived (at the place) where there is the pupunha tree. Where there is the pupunha tree,
we lived close to our late uncle who moved.’ She mentions that tail-head linkage in Dâw only manifests itself in
the repetition of locative adverbial clauses.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 15
The goal of the present section is to analyze in more detail how counterfactual
conditional constructions are encoded in this contact area. Given that it was not possible to
obtain detailed descriptions of Naduhup and Kakua-Nɨkak counterfactual conditionals, the
discussion will only be based on Tucanoan and North Arawak languages.
5.2 Tucanoan
The Tucanoan family includes some 29 languages. They are found in a large area
extending from the northern part of the Brazilian-Colombian border region in the northern and
eastern extreme of their distribution to the Ecuadorean-Colombian upper Putumayo River basin
in their western extreme and to the Peruvian Napo River Basin in their southern extreme
(Chacon & Michael 2018: 63). Two major branches of the family are recognized: Eastern and
Western Tucanoan (Barnes 1994: 325). Of these, Eastern Tucanoan languages (e.g., Tucano,
Wanano, Desano, Tuyuca, Barasano, and Siriano) are spoken on the Colombian and the
Brazilian sides of the Vaupés area.
A closer look reveals that counterfactual conditional constructions in Eastern Tucanoan
languages show striking formal and functional similarities. They tend to have unmarked
protases and apodoses that appear with irrealis markers and past tense markers.
6
They also tend
to occur with non-specialized clause-linking devices.
In Tucano, counterfactual conditionals are expressed with the non-specialized clause-
linking device - (26). The protasis is unmarked, and the apodosis occurs with the irrealis
marker -bo and the past tense marker -a.
Tucano (Tucanoan)
(26)
acó
pejáti-c,
m
yá
boa.
yesterday
rain
fall.NEG-if
POSS.3S
house
?
would.have.burned
‘If it had not rained yesterday, your house would have burned down.’ (West 1980: 45)
A similar counterfactual conditional construction is also found in Barasano (27). In this
language, counterfactual conditional protases are realized with the non-specialized cluse-
linking device -, protases are unmarked and apodoses occur with the irrealis marker -boo
and the past tense marker -ri.
Barasano (Tucanoan)
(27)
rioho
goro
b
goti-hab,
b-re
ha-beti-boo-ri-a-da
y.
straight
truly
2SBJ
tell-COND
2SBJ-OBJ
hit-NEG-IRR-PST-3-agree
1SBJ
‘If you had told the truth, I wouldn’t have hit you, right?’ (Jones & Jones 1991: 124)
In Tatuyo, the distinction between counterfactual conditionals and other types of
conditionals (e.g., real/generic) is not grammaticalized in clause-linking markers. This
language indicates counterfactual conditional meanings with the non-specialized clause-
linking marker -ata (28) (Whisler 1977: 235). Note that the -ata ‘if’ clause does not appear
with any TAM values, and the apodosis occurs with the irrealis marker -bo and the past tense
marker -ricu.
6
Other terms that have been used to refer to irrealis markers in Tucanoan languages are conditional mood markers,
potential mood markers, and subjunctive mood markers (Ramirez 1997: 191).
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 16
Tatuyo (Tucanoan)
(28)
c
y
catata,
aáqueja!
c
y
boricʉ.
him
I
see.if
go.NEG
him
I
said.would.have
‘If I’d have seen him, I’d have said to him, “don’t go!” (Whisler 1977: 236)
The non-specialized clause-linking marker -ta not only occurs in Carapana
counterfactual conditionals, but also in other types of conditionals (Metzger 1981: 203). As
can be seen in (29), the protasis is unmarked, and the apodosis appears with the irrealis marker
-bujio and the past tense marker -ro.
Carapana (Tucanoan)
(29)
to
ca-ta,
peti-coa-bujio-rica-ro.
it
burn-if
waste-completely-IRR-probably-PST
‘If it had burned, it probably would have been completely lost.’ (Metzger 1981: 48)
Tuyuca counterfactual conditionals and other types of conditionals (e.g., real/generic)
are formed with the unspecialized clause-linking marker - (30) (Barnes 1999: 216). The
protasis is unmarked and the apodosis occurs with the irrealis marker -bo and the past tense
marker -.
Tuyuca (Tucanoan)
(30)
j
okó
peá-ri-at
-ja-wii.
yesterday
water
fall-NEG-DEP
2PL-POSS-CLF.building
h-bo-ahju.
burn-IRR-REC-PST
‘If it had not rained yesterday, your house probably would have burned.’ (Barnes 1999: 216)
The examples discussed above indicate that apodoses in Eastern Tucanoan
counterfactual conditional constructions play an important role in the expression of
counterfactuality. This stems from fact that protases are unmarked, and the distinction between
counterfactual conditionals and other types of conditionals (e.g., real/generic) is not
grammaticalized in clause-linking devices.
Western Tucanoan languages seem to show a similar picture to Eastern Tucanoan
languages. An example is attested in Koreguaje. In this language, counterfactual conditionals
apodoses are formed with the irrealis marker - and the perfective marker - (31). Note that
protases do not occur with any TAM values but only with the non-specialized clause-linking
marker - (Cook & Criswell 1993: 98).
Koreguaje (Tucanoan)
(31)
ape-r
-
ikha-tho
theana
r
before-time
1SG-PAT
to.talk-if.DS
same.time
a.lot
-raʔ-siʔ-kh-a-.
to.ask-almost-PFV-M.SG-VBLZ-M.SG
‘If he had asked me first, right then and there I would have asked him for a lot
(money).’ (Cook & Criswell 1993: 98)
A similar exposition can be given for Siona counterfactual conditionals. In this
language, apodoses appear with the irrealis marker -and the past tense marker -w (32).
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 17
Protases are deprived of TAM values and occur with the non-specialized clause-linking device
-to (Bruil 2014: 218)
Siona (Tucanoan)
(32)
-to,
trabaha-da’-wɨ.
be.good-if
work-IRR-PST.ASS
‘If they had been healthy, they would have worked.’ (Bruil 2014: 218)
There are a number of Western Tucanoan languages that slightly differ from the
Tucanoan counterfactual conditional constructions discussed before. In Secoya, counterfactual
conditionals are formed with the non-specialized clause-linking marker -ni (Johnson &
Levinson 1990: 62) and apodoses are realized with the irrealis marker - and the past tense
marker - (33). Note that Secoya differs from other Tucanoan languages in that the perfective
marker -ci appears in the protasis of a counterfactual conditional construction.
Secoya (Tucanoan)
(33)
sai-ci
pa-ni,
yëquë
-ra’-huë.
to.go-PFV
PL
to.be-CONN
1PL.EXCL
to.arrive-IRR-PST
‘If we were walking, we would have arrived.’ (Johnson & Levinson 1990: 62)
Máíhĩkì counterfactual conditionals bear formal and functional resemblance to
counterfactual conditionals in other Tucanoan languages in that this complex sentence
construction is realized with a non-specialized clause-linking device (i.e., the marker -tu;
Farmer 2015: 77) and protases are unmarked, as in (34). Note that apodoses in Máíhĩkì appear
with the frustrative marker -ra and the past tense marker -b. Most Tucanoan languages contain
frustrative markers (Aikhenvald 2012: 185), and can appear in different communicative
scenarios (see Ramírez 1997: 151). However, they tend not to be used to express counterfactual
conditional meanings, as has been shown in the present section. Accordingly, the fact that
Máíhĩkì counterfactual conditional apodoses are realized with a frustrative marker differs from
other Tucanoan apodoses.
Máíhĩkì (Tucanoan)
(34)
yì
ká-huna
k-ma-tu,
1SG
DIST.ANAPH.DEM-CLF
tell.story-NEG-if
yété-ma-ra-bɨ
yì.
learn-NEG-FRUST-1SG.PST.DECL
1SG
‘If they hadn’t taught me, I wouldn’t have learned.’ (Farmer 2015: 77)
5.3 North Arawak languages
The Arawakan family constitutes one of the largest linguistic families of the Americas,
with more than 40 languages still spoken by around 500,000 speakers (Aikhenvald 1999: 72).
The Arawakan language family is usually divided into Northern and Southern Arawakan, and
then further into subgroups that cluster according to their grammatical similarities.
North Arawak is a group of languages consisting of Tariana, Baniwa of
Içana/Kurripako, Piapoco, Warekena, Achagua, Yucuna, and Resigaro (Aikhenvald 2019:
460). Of these, Tariana is the only language within the multilingual Vaupés Basin linguistic
area. This language contains a counterfactual conditional construction similar to the Tucanoan
construction discussed in Section 5.2. The traditional Vaupés River Basin linguistic area is
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 18
characterized by language-based exogamy between speakers of Tariana and those belonging
to the East Tucanoan subgroup (Aikhenvald 1999). In the example in (35), the protasis does
not appear with any TAM values. It only appears with the clause-linking marker -ka. This
marker is non-specialized in that it can be used in counterfactual conditionals and other types
of conditional constructions. The apodosis occurs with -buhtaka. This marker conflates
conditional mood and recent past tense. The apodosis can also appear with -buhtana (36).
Unlike -buhtaka, -buhtana conflates conditional mood and remote past tense.
Tariana (Arawakan)
(35)
heku
iya
di-wha-ka,
amaku
pu
-buhtaka.
yesterday
rain
3SG.N.F-fall-SUB
hammock
be.wet-COND.REC.PST
‘If rain had fallen yesterday, the hammock would have been wet.’ (Aikhenvald 2003: 391)
(36)
palipa-nipe
sede-ka,
iya
wa-na
pu
-buhtana.
IMP.cover-NMLZ
NEG.EXIST-SUB
rain
1PL-OBJ
be.wet-COND.REM.PST
‘If we had not had the cover, rain would have made us wet (a week or so ago).’
(Aikhenvald 2003: 391)
It is instructive to compare the Tariana counterfactual conditional construction with its
equivalent in other North Arawak languages. Baniwa of Içana/Kurripako is closely related to
Tariana, but it is spoken outside the Vaupés area. The Baniwa of Içana/Kurripako dialect
continuum is spoken in the Içana river basin and along its tributaries, bordering on the Vaupés
and extending further to the north into Venezuela and north-east into Colombia. In Baniwa of
Içana, counterfactual conditionals occur with the non-specialized clause-linking marker -ka
(37). This pattern is similar to the Tariana clause-linking marker. However, unlike Tariana,
TAM values appear in both clauses. In this regard, the irrealis marker -mitha occurs in the
protasis and the apodosis of a counterfactual conditional construction.
Baniwa of Içana (Arawakan)
(37)
pi-kapa-mitha
pi-kaite-ka,
no-lhio
wheekodza
no-aa-mitha
pi-lho.
2SG-see-IRR
2SG-tell-SUB
1SG-to
yesterday
2SG-give-IRR
2SG-to
‘If you had told me yesterday, I would have given you.’ (Ramirez 2001: 255)
As for Kurripako, it has not been possible to find a source providing a detailed
description of counterfactual conditionals. Granadillo (2006: 82) briefly describes
counterfactual conditionals in this language. However, she does not give any examples
appearing with apodoses, but only with protases (38). Interestingly, the Kurripako
counterfactual conditional protasis seems to differ from the Tariana protasis in that it appears
with the conditional mood marker -apa and the frustrative marker -ya.
Kurripako (Arawakan)
(38)
un-sru-ka-dan-tha
apa-ya
carro.
1SG-have-TA-COND-FRUST
one-CLF
car
‘If only I had a car.’ (Granadillo 2006: 82)
Another North Arawak language closely related to Tariana is Piapoco. This language
is spoken to the west of the Upper Rio Negro area in Colombia. The counterfactual conditional
pattern is somewhat similar to the Tariana pattern in that it also occurs with a non-specialized
clause-linking device (i.e., the verbal form -). However, unlike Tariana, Piapoco contains
a counterfactual conditional construction in which both clauses occur with TAM values.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 19
Protases are realized with the remote past tense marker - in combination with the
counterfactual mood marker -, and apodoses are formed with the remote past tense marker -
(39).
Piapoco (Arawakan)
(39)
ca-úláica-caalí-té
pi-atúa-cá,
né-íse
càmi-ta-té
uu-wènìa
ATT-disease-if- REM.PST
2-mother-CF
there-from
not-EMPH-REM.PST
3F-buy
wáluma-tá.
cloth-CLF
‘If you mother had been sick, then she would not have bought cloth.’ (Klumpp 2019:
326)
There are other North Arawak languages that also differ from the Tariana
counterfactual conditional pattern. In Achagua and Resígaro, counterfactual conditional
protases appear with specific TAM values. While the Achagua counterfactual conditional
protasis is realized with the irrealis marker -kta (40), the Resígaro counterfactual conditional
protasis occurs with the irrealis marker -ma in combination with the recent past tense marker -
mi (41).
Achagua (Arawakan)
(40)
páablu
íinu-kta-ta-i,
hó-ka-ta
wa-táwahaa.
Pablo
come-IRR-CONTR-SIG
NEG-AFF-CONTR
1PL-work
‘If Pablo had come, we would not have worked.’ (Wilson 1992: 164)
Resígaro (Arawakan)
(41)
anepuu
aá
ee
kha-tshí-ma-mí,
ka
va.
much
father
fish
do-if-UNREAL-REC
well
we-eat
‘If my father had caught a lot of fish, we would have eaten well.’ (Allin 1976: 261)
The grammatical differences between counterfactual conditionals in Tariana and other
North Arawak languages spoken outside the Vaupés area seem to reflect the Tucanoan impact
on Tariana.
5.4 Discussion
It has been shown in the previous subsections that Tucanoan languages and Tariana
contain similar counterfactual conditional constructions. This seems to be the result of
language contact. Long-term interaction among the languages spoken in the Vaupés has
resulted in extensive diffusion of many grammatical and other linguistic features (Epps 2006,
2016). The traditional Vaupés River Basin linguistic area is characterized by language-based
exogamy between speakers of Tariana and those belonging to the East Tucanoan subgroup
(Aikhenvald 2012: 75). East Tucanoan languages within this traditional marriage network
include Tucano, Wanano, Desano, Piratapuya, and Tuyuca, among others. This ensures
obligatory multilingualism. Note that Tucano is the major language of the Brazilian Vaupés,
and most Tariana use this language on a day-to-day basis (Aikhenvald 2019: 470). This has
increased Tucano’s impact, especially in syntax and discourse Tariana patterns. Accordingly,
it is likely that the Tariana counterfactual conditional pattern developed under the influence of
this Tucanoan language.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 20
Further evidence that the counterfactual conditional pattern diffused through language
contact in this area is the following. Most Tucanoan languages contain frustrative markers that
are not used in the expression of counterfactual conditional meanings (see Section 3.3 for other
languages that show the same pattern). Interestingly, Tariana (Aikhenvald 2003: 380) also have
frustrative markers that are not used in the expression of counterfactual conditional meanings.
It is also very likely that Tariana copied other counterfactual constructions from
Tucanoan languages. Crosslinguistically, it has been shown that a counterfactual simple clause
construction (e.g., I would have gone) may be structurally similar to the apodosis of a
conditional counterfactual construction (Van Linden & Verstraete 2008: 1888). In Ik,
counterfactual simple clause constructions are encoded with the realis marker -a, the
hypothetical marker a, and the past tense marker =naa, as in (42). In a similar fashion, the
apodosis of a counterfactual conditional construction is marked with the same TAM values, as
in (43). However, there are also many languages in which simple counterfactuals cannot be
equated with conditional counterfactuals with an elided protasis (Van Linden & Verstraete
2008: 1889).
Ik (Kuliak)
(42) -a ƙa=naa barats-o=nák.
come-1SG-REAL HYP=PST morning-INS=DEM.SG.PST
‘I would have come this morning.’ (Schrock 2014: 516)
(43) ár
m-a bra--k,
CONJ=HYP=PST insecurity-NOM not.be-3SG-SEQ
‘If insecurity had not been there,
a-í-ísin-a ƙa=nakᵃ.
go-PL-1PL.INCL-REAL HYP=PST
we would have gone regularly.’ (Schrock 2014: 517)
Most Tucanoan languages encode counterfactual simple clause constructions and
counterfactual conditional apodoses in the same way. As was shown in Section 5.1,
counterfactual conditional apodoses in Tucano are realized with the irrealis marker -bo and the
past tense marker -a. Counterfactual simple clause constructions are also marked in the same
way in that they appear with the irrealis marker -bo and the past tense marker -a, as is shown
in the example in (44).
Tucano (Tucanoan)
(44)
ní-bo-a-p.
be-IRR-PST-1SG.SBJ
‘I would have been there.’ (West 1980: 45)
Similarly, Tariana contains counterfactual simple clause constructions and
counterfactual conditional apodoses occurring with the same TAM values. Counterfactual
conditional apodoses in Tariana are formed with -buhtaka or -buhtana (see Section 5.3).
Counterfactual simple clause constructions also occur with the same markers in this language,
as seen in the example in (45).
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 21
Tariana (Arawakan)
(45)
ikasu-bothaka
pi-ñare
phia,
di-a-pidana
di-na
now-COND.REC.PST
2SG-dissapear
you
3SG.NF-say-REM.PST.REP
3SG.NF-OBJ
ne:ri-ne.
deer-FOC
‘Now you would have disappeared, said the deer.’ (Aikhenvald 2003: 142)
6. Final remarks
The present study has shown that counterfactual conditional protases in Amazonian
languages tend to be unmarked (they do not occur with any TAM values), and apodoses tend
to occur with irrealis or frustrative marking. It has also been shown that most Amazonian
languages in the sample contain counterfactual conditionals occurring with non-specialized
clause-linking devices. This means that the distinction between counterfactual conditionals and
other types of conditionals (e.g., real/generic) is not grammaticalized in clause-linking devices
in the Amazonian languages in the sample. Instead, the counterfactual conditional meaning
resides in the combination of specific TAM markers.
The present research has shown that Tariana counterfactual conditional construction
have been shaped by Tucanoan languages though language contact. It remains to be explored
whether other languages spoken in the Vaupés area (i.e., Naduhup and Kakua-Nɨkak
languages) express counterfactual conditional meanings in the same way and the role of
language contact. This will enable us to determine whether counterfactual conditionals can be
used as a diagnostic feature of this contact area.
As a sobering note, this study barely scratches the surface. There are a number of
aspects relevant to the study of Amazonian counterfactual conditional clauses that this study
could not address. Accordingly, they remain to be investigated by future studies, and in what
follows some potentially fruitful areas are mentioned.
First, as was shown in this paper, in a number of Amazonian languages, a counterfactual
simple clause construction (e.g., I would have gone) is structurally similar to the apodosis of a
conditional counterfactual construction. Whether this is pervasive in Amazonian languages can
only be answered by future work with natural discourse data from diverse languages.
Moreover, although counterfactuals are typically associated with the kind of grammatical
construction discussed in this paper, they may also show up in other guises. For instance,
hypothetical manner constructions (e.g., she treats me as if I were a stranger) portray a
counterfactual situation (see Olguín Martínez 2021) and seem to be found in a number of
Amazonian languages (e.g., Asheninka Perené; Mihas 2015: 285; Piapoco; Klumpp 2019:
332). Crosslinguistically there are languages that have a construction that could be regarded as
a counterfactual conditional construction with an elided apodosis (e.g., if only Hans had come).
These instances are known in the literature as ‘counterfactual wishes’ and seem to be the result
of insubordination, defined as the “recruitment of main clause structures from subordinate
structures, or synchronically as the independent use of constructions exhibiting prima facie
characteristics of subordinate clauses (like English If you could fill this out, please or That he
could say such a thing!)” (Evans & Watanabe 2016: 2). Counterfactual concessive conditionals
may also express a counterfactual meaning: Even
have been paid the same wages as the others. These counterfactual patterns form a ‘family (of
constructions).’ In recent years, the notion of family has established itself in Construction
Grammar as a label for sets of constructions with a similar meaning or function, often despite
striking differences in form (Diessel 2019: 199-200; Leuschner 2020; Ruiz de Mendoza Ibáñez
et al. 2017). In the usage-based approach, grammar consists of constructions interconnected by
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 22
various links that reflect the language users’ experience with particular grammatical patterns
(Croft 2001; Diessel 2015, 2017, 2019). It remains to be explored how counterfactuals are
encoded in Amazonian languages in a single study. This will enable us to explore connections
among these constructions and provide hypotheses regarding their directionality of historical
development. Shared morphosyntactic properties can be explained by analogical connections
between constructions in the grammar network. Family resemblances should be considered a
synchronic reflection of the ongoing diachronic emergence of the constructions in question
(Croft & Cruse 2004: 318).
Second, as was discussed in the present research, a number of Amazonian languages
contain counterfactual conditionals with optional clause-linking devices. This goes against
Harder (1996: 93), who mentions that of all grammatical elements in an adverbial clause
construction, clause-linking devices are the most necessary element to get the message across.
Intriguingly, there are other Amazonian languages in which TAM values (e.g., irrealis markers)
are optional and can be omitted without affecting the adverbial meaning holding between
clauses. The question is: what are the factors that may lead speakers to omit TAM values from
a counterfactual conditional construction? The optionality of clause-linking markers and TAM
values in counterfactual conditionals is unexplored territory and open to future research.
Third, the diachronic sources of counterfactual conditional clause-linking devices in
Amazonian languages seems like another interesting area for future research. From a
crosslinguistic perspective, it has been proposed that conditional clause-linking devices tend
to be derived from: adverb(ial)s meaning ‘then’, verbs meaning ‘to say’, words for modality
(especially epistemic and optative), copular constructions, interrogatives, words that mark
something as known or given (including topic markers and demonstratives), and words
temporal in origin (Martowicz 2011: 188; Olguin Martínez & Lester 2021; Traugott 1985:
292). The question is: do Amazonian languages show a similar picture?
We hope that this study will be valuable as a general overview of counterfactual
conditional constructions in Amazonian languages and that it will help linguistic researchers
come up with more accurate descriptions in the future.
Abbreviations
1=first person, 2=second person, 3=third person, A=agent, ADESS=adessive, aff=affirmative,
AFFCT=affected argument, ALL=allative, ANAPH=anaphoric, ANTIC=anticipatory,
APPL=applicative, ART=article, ass=assertive, ATT=attributive, AUX=auxiliar,
CERT=modal marker of certainty, CLF=classifier, CMP=complement, CF=counterfactual
mood, COM=comitative, COMPL=completive, CON=contrastive, COND=conditional,
CONJ=conjunction, conn=connective, CONT=continuous, contr=contrast,
COSUB=cosubordination, DAT=dative, DECL=declarative, DEM=demonstrative,
DEP=dependent marker, DES=desiderative, DIST=distal, DISTR=distributive, DR=bivalent
direct, DS=different subject, DYNM=dynamic, ELAT=elative, EMPH.TAG=emphatic tag,
EV=evidential, EXCL=exclusive, EXIST=existential, F=feminine, FRUST=frustrative,
FUT=future, HYP=hypothetical, IMP=impersonal, INC=inclusive, INF=inferential evidential,
INTENS=intensifier, INTERJ=interjection, INS=instrumental, IPFV=imperfective,
IRR=irrealis, ITER=iterative aspect, LK=linker, M=masculine, N=non, NEC=modal marker
of necessity, NEG=negative, NMLZ=nominalizer, NOM=nominative, NR=subject
nominalizer, OBJ=object, OBL=oblique, PASS=passive, pat=patient PER=peripheral element,
PERF=perfect, PFV=perfective, PL=plural, POSS=possessive, POT=potential, PR=general
participial, PRIV=privative, PROG=progressive aspect, pron=pronominal,
PRX.CNTR=proximate contrast, PRS=present, PST=past, REAL=realis, REC=recent,
REF=referential, REL=relativizer, REM=remote, REP=reportative, SBJ=subject,
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 23
SEQ=sequential, SG=singular, sig=significant, SIM=simultaneous, S=single argument,
SS=same subject, SUB=subordinator, ta=tense, aspect TH=thematic suffix, TOP=topic,
TRS=transitivizer, unreal=unrealized, vblz=verbalizer, VI=sixth class, VS=verbal stem
marker, VT=verb terminating classifier.
References
Aikhenvald, Alexandra Y. (1999). The Arawak language family. In R.M.W. Dixon & Alexandra Y. Aikhenvald
(eds.), The Amazonian languages, pp. 65-105. Cambridge University Press.
Aikhenvald, Alexandra Y. (2002). Language contact in Amazonia. Oxford University Press.
Aikhenvald, Alexandra Y. (2003). A grammar of Tariana. Cambridge University Press.
Aikhenvald, Alexandra Y. (2019). Bridging linkage in Tariana, an Arawak language from Northwest Amazonia.
International Journal of American Linguistics 85(4): 455-496. https://doi.org/10.1086/704563
Allin, Trevor Reginald (1976). A grammar of Resígaro. Summer Institute of Linguistics International. Available
Online: http://hdl.handle.net/10023/1012
Barnes, Janet (1994). Tuyuca. In Peter Kahrel & René van den Berg (eds.), Typological Studies in Negation, pp.
325-342. John Benjamins.
Barnes, Janet (1999). Tucanoan. In R.M.W. Dixon & Alexandra Y. Aikhenvald (eds.), The Amazonian
Languages, pp. 207-226. Cambridge University Press.
Beier, Christine, Lev Michael, & Joel Sherzer (2002). Discourse forms and processes in indigenous lowland South
America: An area-typological perspective. Annual Review of Anthropology 31: 121-145.
https://doi.org/10.1146/annurev.anthro.31.032902.105935
Bickel, Balthasar (2008). A refined sampling procedure for genealogical control. Language Typology and
Universals 61: 221-233. https://doi.org/10.1524/stuf.2008.0022
Brandão, Ana Paula Barros (2014). A reference grammar of Paresi-Haliti (Arawak). University of Texas at Austin
doctoral dissertation. Available at: http://hdl.handle.net/2152/24847
Brooks, Joseph (2018). Realis and irrealis: Chini verb morphology, clause chaining, and discourse. University
of California, Santa Barbara doctoral dissertation. Available at:
https://escholarship.org/uc/item/36s8s9t4
Bruil, Martine (2014). Clause-typing and evidentiality in Ecuadorian Siona. LOT Dissertation Series. Link:
https://www.lotpublications.nl/Documents/350_fulltext.pdf
Chacon, Thiago Costa & Lev Michael (2018). The evolution of subject-verb agreement in Eastern Tukanoan.
Journal of Historical Linguistics 8(1): 59-94. https://doi.org/10.1075/jhl.16024.cha
Comrie, Bernard (1986). Conditionals: A typology. In Elizabeth Traugott; Alice ter Meulen; Judy Reilly, &
Charles Ferguson (eds.), On conditionals, pp. 77-99. Cambridge University Press.
Comrie, Bernard (1993). Language universals and linguistic typology: Data-bases and explanations.
Sprachtypologie und Universalienforschung 46: 3-14. https://doi.org/10.1524/stuf.1993.46.14.3
Cook, Dorothy M. & Linda L. Criswell (1993). El idioma koreguaje (Tucano Occidental). Asociación Instituto
Lingüístico de Verano. Available at: https://www.sil.org/resources/archives/18776
Cristofaro, Sonia (2003). Subordination. Oxford University Press.
Croft, William (2001). Radical construction grammar. Oxford University Press.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 24
Croft, William & Alan Cruse (2004). Cognitive linguistics. Cambridge University Press.
Danielsen, Swintha (2007). Baure: An Arawak language of Bolivia. Research School of Asian, African and
Amerindian Studies, University of Leiden.
Danielsen, Swintha & Lena Terhart (2016). Realis/irrealis as a basic grammatical distinction in Southern
Arawakan languages. Revue de Sémantique et Pragmatique 38: 97-120. https://doi.org/10.4000/rsp.897
Declerck, Rennat & Susan Reed (2001). Conditionals. A comprehensive empirical analysis. Mouton De Gruyter.
Derbyshire, Desmond C. (1979). Hixkaryana syntax. North Holland.
Diessel, Holger (2015). Usage-based construction grammar. In Ewa Dabrowska & Dagmar Divjak (eds.),
Handbook of cognitive linguistics, pp. 295-321. Mouton de Gruyter.
Diessel, Holger (2017). Usage-based linguistics. In Mark Aronoff (ed.), Oxford research encyclopedia of
linguistics, pp. 1-25. Oxford University Press.
Diessel, Holger (2019). The grammar network. How linguistic structure is shaped by language use. Cambridge
University Press.
Dixon, R.M.W. (2004). The Jarawara language of Southern Amazonia. Oxford University Press.
Dixon, R.M.W. (2009). The semantics of clause linking in typological perspective. In R.M.W. Dixon & Alexandra
Aikhenvald (eds.), The semantics of clause linking: A cross-linguistic typology, pp. 1-55. Oxford University
Press.
Dixon, R.M.W. & Alexandra Y. Aikhenvald (1999). Introduction. In R.M.W. Dixon & Alexandra Y. Aikhenvald
(eds.), The Amazonian Languages, pp. 1-22. Cambridge University Press.
Eberhard, David M. (2009). Mamaindê grammar: A Northern Nambikwara language and its cultural context.
LOT Dissertation Series.
Epps, Patience. (2007). The Vaupés melting pot: Tucanoan influence on Hup. In Alexandra Y. Aikhenvald &
R.M.W. Dixon (eds.), Grammars in contact: A cross-linguistic typology, pp. 267-289. Oxford University
Press.
Epps, Patience (2008). A grammar of Hup. Mouton de Gruyter.
Epps, Patience (2017). Subsistence pattern and contact-driven language change: A view from the Amazon basin.
Language Dynamics and Change 7: 47-101. https://doi.org/10.1163/22105832-00602004
Epps, Patience & Lev Michael (2017). The areal linguistics of Amazonia. In Raymond Hickey (ed.), The
Cambridge handbook of areal linguistics, pp. 934-963. Cambridge University Press.
Estigarribia, Bruno (2020). A grammar of Paraguayan Guarani. University College London Press. Available at:
https://www.uclpress.co.uk/products/138755
Evans, Nicholas & Honoré Watanabe (2016). The dynamics of insubordination: An overview. In Nicholas Evans
& Honoré Watanabe (eds.), Insubordination, pp. 1-38. John Benjamins.
Everett, Daniel L. & Barbara Kern (1997). Wari': The Pacaas Novos language of Western Brazil. Routledge.
Farmer, Stephanie J. (2015). . University of California, Berkeley doctoral
dissertation. Available at: https://escholarship.org/uc/item/0rn709mg
François, Alexandre (2012). The dynamics of linguistic diversity: Egalitarian multilingualism and power
imbalance among northern Vanuatu languages. International Journal of the Sociology of Language 214:
85-110. https://doi.org/10.1515/ijsl-2012-0022
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 25
Girón Higuita, Jesús Mario (2008). Una gramática del Wansöhöt (Puinave). LOT Dissertation Series
Granadillo, Tania (2006). An ethnographic account of language documentation among the Kurripako of
Venezuela. University of Arizona doctoral dissertation.
Guérin, Valérie (2015). Demonstrative verbs: A typology of verbal manner deixis. Linguistic Typology 19: 141-199.
https://doi.org/10.1515/lingty-2015-0006
Guérin, Valérie & Aiton, Grant (2019). Bridging constructions in typological perspective. In Valérie Guérin (ed.),
Bridging constructions, pp. 1-44. Language Science Press.
Haiman, John & Tania Kuteva (2001). The symmetry of counterfactuals. In Joan Bybee & Michael Noonan (eds.),
Complex sentences in grammar and discourse, pp. 101-124. John Benjamins.
Hanson, Rebecca (2010). A grammar of Yine (Piro). LaTrobe University doctoral dissertation. Link:
https://opal.latrobe.edu.au/articles/thesis/A_grammar_of_Yine_Piro_/21847407
Harder, Peter (1996). Subordinators in a semantic clause structure. In Betty Devriendt; Louis Goossens, & Johan
Van der Auwera (eds.), Complex structures. A functionalist perspective, pp. 93-118. Mouton de Gruyter.
Hetterle Katja (2015). Adverbial clauses in cross-linguistic perspective. De Gruyter Mouton.
Johnson, Orville E. & Stephen E. Levinson (1990). Gramática Secoya. Instituto Lingüístico de Verano. Available
at: https://www.sil.org/resources/archives/17548
Jones, Wendell & Paula Jones (1991). Barasano Syntax. Summer Institute of Linguistics and The University of
Texas at Arlington. Available at: https://www.sil.org/resources/archives/8775
Karawani, Hadil (2014). The real, the fake, and the fake fake in counterfactual conditionals, crosslinguistically.
Utrecht: Landelijke Onderzoekschool Taalwetenschap, Netherlands National Graduate School of
Linguistics.
Kibrik, Alexander (1998). Does intragenetic typology make sense? In Winfried Boeder; Christoph Schroeder, &
Karl Heinz Wagner (eds.), Sprache im raum und zeit: In memoriam Johannes Bechert, volume 2:
Beiträge zur empirischen sprachwissenschaft, pp. 61-68. Tübingen: Narr.
Klumpp, Deloris A. (2019). A grammar of Piapoco. Summer Institute of Linguistics International.
https://www.sil.org/system/files/reapdata/35/41/89/35418936594625772486468628979716143769/eBo
ok_71_Piapoco.pdf
Kortmann, Bernd (1997). Adverbial subordination: A typology and history of adverbial subordinators based on
European languages. Mouton de Gruyter.
Leuschner, Torsten (2020). Concessive conditionals as a family of constructions. Belgian Journal of Linguistics
34: 235-247. https://doi.org/10.1075/bjl.00049.leu
Martins, Silvana Andrade (2004). Fonologia e gramática Dâw. Vrije Universiteit Amsterdam doctoral
dissertation.
Martins, Valteir (2005). Reconstrução fonológica do Protomaku Oriental. LOT Dissertation Series.
Martowicz, Anna (2011). The origin and functioning of circumstantial clause linkers: A cross-linguistic study.
University of Edinburgh doctoral dissertation. Available at: http://hdl.handle.net/1842/6411
Mauri, Caterina & Johan Van der Auwera (2012). Connectives. In Allan Keith & Kasia M. Jaszczolt (eds.), The
Cambridge handbook of pragmatics, 347-402. Cambridge University Press.
Metzger, Ronald G. (1981). Gramática popular del Carapana. Instituto Lingüístico del Verano. Available at:
https://www.sil.org/resources/archives/19013
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 26
Miestamo, Matti (2005). Standard negation: The negation of declarative verbal main clauses in a typological
perspective. Mouton de Gruyter.
Michael, Lev (2009). Clause-linking in Iquito (Zaparoan). In R.M.W. Dixon & Alexandra Aikhenvald (eds.), The
semantics of clause-linking: A cross-linguistic typology, pp. 145-166. Oxford University Press.
Michael, Lev (2014). The Nanti reality status system: Implications for the typological validity of the realis/irrealis
contrast. Linguistic Typology 18: 251-288. https://doi.org/10.1515/lingty-2014-0011
Mihas, Elena (2015). A grammar of Alto Perené (Arawak). De Gruyter Mouton.
Miranda, Maxwell Gomes (2014). Morfologia e morfossintaxe da língua Krahô (família Jê, tronco Macro-Jê).
Universidad de Brasília doctoral dissertation. Available at:
http://repositorio2.unb.br/jspui/handle/10482/17796
Mithun, Marianne (1995). On the relativity of irreality. In Joan Bybee & Suzanne Fleischman (eds.), Modality in
grammar and discourse, pp. 367-388. John Benjamins.
Mithun, Marianne (1999). The languages of Native North America. Cambridge University Press.
Mithun, Marianne (2008). Borrowed rhetorical constructions as starting points for grammaticalization. In
Alexander Bergs & Gabriele Diewald (eds.), Constructions and language change. Trends in linguistics,
pp. 195-229. Mouton de Gruyter.
Muller, Neele (2013). Tense, aspect, modality, and evidentiality marking in South American indigenous
languages. LOT Dissertation Series. Available at:
https://www.lotpublications.nl/Documents/324_fulltext.pdf
Nicolle, Steve (2017). Introduction to special issue on conditional constructions in African languages. Studies in
African Linguistics 46(1/2): 1-15. https://doi.org/10.32473/sal.v46i1.107239
Obert, Karolin (2019). Locative adverbial clauses and tail-head linkage in Dâw narratives (Naduhup-AM).
Handout of paper presented at the workshop complex structures in Brazilian Languages, Universidad de
São Paulo.
Olawsky, Knut J. (2006). A grammar of Urarina. Mouton de Gruyter.
Olguín Martínez, Jesús (2021). Hypothetical manner constructions in world-wide perspective. Journal Linguistic
typology at the crossroads 1: 6-46.
Olguín Martínez, Jesús (2023). A typological study of tail-head linkage constructions. In Alessandra Barotto &
Simone Mattiola (eds.), Discourse phenomena in typological perspective, pp. 403-432. John Benjamin
[Studies in Language Companian Series 227].https://doi.org/10.1075/slcs.227.14olg
Olguín Martínez, Jesús & Nicholas Lester (2021). A quantitative analysis of counterfactual conditionals in the
world’s languages. Italian Journal of Linguistics 33(2): 147-182.
https://doi/org/10.26346/1120-2726-178
Olguín Martínez, Jesús & Alonso Vásquez (2022). The contribution of Amazonian languages to the typology of
purpose clauses. LIAMES: Línguas Indígenas Americanas 22. 1-21.
https://doi.org/10.20396/liames.v22i00.8670091
Ospina Bozzi, Ana Maria (2002). Les structures elémentaires du Yuhup Makú, langue de l'Amazonie
Colombienne: Morphologie et syntaxe. Université Paris doctoral dissertation.
Overall, Simon (2017). A grammar of Aguaruna (Iiniá Chicham). De Gruyter Mouton.
Ramírez, Henri (1997). A fala Tukano dos Ye'pâ-Masa. Inspetoria Salesiana Missionária da Amazônia.
Ramirez, Henri (2001). Uma gramática do Baniwa do Içana. Universidade Federal do Amazonas.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 27
Rijkhoff, Jan & Dik Bakker (1998). Language sampling. Linguistic Typology 2(3): 263- 314.
https://doi.org/10.1515/lity.1998.2.3.263
Rijkhoff, Jan; Dik Bakker; Kees Hengeveld, & Peter Kahrel (1993). A method of language sampling. Studies in
Language 17(1): 169-203.https://doi.org/10.1075/sl.17.1.07rij
Ruiz de Mendoza Ibáñez; Francisco José; Alba Luzondo Oyón, & Paula Pérez Sobrino (2017). Investigating the
construction. In Francisco José Ruiz de Mendoza Ibáñez; Alba Luzondo Oyón, & Paula Pérez Sobrino,
Constructing families of constructions: Analytical perspectives and theoretical challenges [Human
Cognitive Processing 58], pp. 1-13. John Benjamins. https://doi.org/10.1075/hcp.58.01rui
Sakel, Jeanette (2002). A grammar of Mosetén. Mouton de Gruyter.
Schmidtke-Bode, Karsten (2009). A typology of purpose clauses. John Benjamins.
Schmidtke-Bode, Karsten & Holger Diessel (To appear). The typology of non-argument clauses. In Manfred
Krifka & Mathias Schenner (eds.), The Oxford handbook of embedding. Oxford University Press.
Schrock, Terrill B. (2014). A grammar of Ik (Icé-. LOT
Dissertation Series.
Steele, Susan (1975). Past and irrealis: Just what does it all mean? International Journal of American Linguistics
41(3): 200-217. https://doi.org/10.1086/465362
Thiesen, Wesley & David Weber (2012). A grammar of Bora with special attention to tone. Summer Institute of
Linguistics International. Available at: https://www.sil.org/resources/archives/55005
Thurman, Robert C. (1975). Chuave medial verbs. Anthropological Linguistics 17(7). 342-352.
Traugott, Elizabeth C. (1985). On conditionals. In John Haiman (ed.), Iconicity in syntax, pp. 289-307. John
Benjamins.
van Gijn, Rik; Katharina Haude, & Pieter Musyken (2011). Subordination in South America: An overview. In
Rik van Gijn; Katharina Haude, & Pieter Musyken (eds.), Subordination in Native South American
languages, pp. 1-24. John Benjamins.
Van Linden, An & Jean-Christophe Verstraete (2008). The nature and origin of counterfactuality in simple
clauses: Cross-linguistic evidence. Journal of Pragmatics 40(11): 1865-1895.
https://doi.org/10.1016/j.pragma.2008.03.008
Verstraete, Jean-Christophe (2014). The role of mood marking in complex sentences: A case study of Australian
languages. Word 57: 195-236. https://doi.org/10.1080/00437956.2006.11432563
von Prince, Kilu (2019). Counterfactuality and past. Linguistics and Philosophy 42. 577-615
https://doi.org/10.1007/s10988-019-09259-6
Voort, Hein van der (2004). A grammar of Kwaza. Mouton de Gruyter.
Vries, Lourens de (2005). Towards a typology of tail-head linkage in Papuan languages, Studies in Language
29(2): 363-384. https://doi.org/10.1075/sl.29.2.04vrj
Vuillermet, Marine (2012). A grammar of Ese Ejja, a Takanan language of the Bolivian Amazon. Université
Lumière Lyon 2 doctoral dissertation.
West, Birdie (1980). Gramática popular del tucano. Ministerio de Gobierno, Dirección General de Integración y
Desarrollo de la Comunidad, División Operativa de Asuntos Indígenas.
Whisler, David (1977). Some aspects of Tatuyo discourse. In Robert E. Longacre & Frances Woods (eds.),
Discourse grammar: Studies in indigenous languages of Colombia, Panama, and Ecuador Part 3, pp.
207-52. Summer Institute of Linguistics Publications in Linguistics and Related Fields.
LIAMES, Campinas, SP, v. 24, 1-28, e024010, 2024 28
Wilson, Peter J. (1992). Una descripción preliminar de la gramática del Achagua (Arawak). Instituto Lingüística
del Verano.
Wojtylak, Katarzyna I. (2020). A grammar of Murui (Bue): A Witotoan language of Northwest Amazonia. Brill.
Xrakovskij, Viktor S. (2005). Conditional constructions: A theoretical description. In Viktor S. Xrakovskij (ed.),
Typology of conditional constructions, pp. 3-95. LINCOM. Studies in Theoretical Linguistics.
Zariquiey, Roberto (2018). A grammar of Kakataibo. De Gruyter Mouton.
CRediT - Taxonomy of Academic Collaboration Roles
Acknowledgements
We would like to thank all colleagues who have kindly discussed with us specific points or
have shared with us their data, as well as two anonymous reviewers whose comments have
contributed to greatly improve this paper in both content and form. Needless to say, all
remaining shortcomings are our own.
Declaration of conflict of interests
All authors declare that they have no conflicts of interest.
Authors’ contribution
Conceptualization: Jesús Olguín Martínez
Data curation: Jesús Olguín Martínez & Alonso Vásquez-Aguilar
Formal Analysis: Jesús Olguín Martínez & Alonso Vásquez-Aguilar
Investigation: Jesús Olguín Martínez & Alonso Vásquez-Aguilar
Methodology: Jesús Olguín Martínez
Project administration: Jesús Olguín Martínez & Alonso Vásquez-Aguilar
Resources: Jesús Olguín Martínez & Alonso Vásquez-Aguilar
Software: NA
Supervision: NA
Validation: NA
Visualization: Jesús Olguín Martínez & Alonso Vásquez-Aguilar
Writing – original draft: Jesús Olguín Martínez & Alonso Vásquez-Aguilar
Writing – review & editing: Jesús Olguín Martínez & Alonso Vásquez-Aguilar
Ethics in research with human beings
NA
Research funding
No research funding
Submissão recebida: 12/3/2024
Versão revista e corrigida 21/6/2024
Aceita: 2/7/2024
Publicado: 4/7/2024